Faculty of Science (ETDs)

Permanent URI for this community

https://hdl.handle.net/10539/37933

Browse

Now showing 1 - 20 of 247

3D seismic constraints on the strato-structural evolution of the deep-water Orange Basin, South Africa
(University of the Witwatersrand, Johannesburg, 2023) Maduna, Nombuso Gladys; Jinna, Zubair; Manzi, Musa
This research utilizes seismic attributes and advanced machine learning methodologies to analyse high-resolution 3D reflection seismic data from the deep-water Orange Basin, located offshore western South Africa. The primary goal is to gain valuable insights into the basin's tectonic setting, depositional environment, and hydrocarbon potential. Significant features are delineated within the basin including (1) a gravitational collapse system in the Mesozoic Late Cretaceous, (2) mass flow features in the Cenozoic, (3) natural gas and fluid escape structures, (4) a large slope-perpendicular submarine canyon cutting Oligocene strata, and (5) multiple slope-parallel, sinusoidal channel features in the Miocene. The Late Cretaceous succession exhibits a gravitational collapse system with a translational and compressional domain detaching on seaward-dipping Turonian shales. Gravitational collapse during margin uplift formed fold-and-thrust belts along the slope characterizing the compressional domain. As they are commonly linked to hydrocarbons, the compressional domain of these systems has been extensively studied, while the translational domain has been poorly constrained due to its structural complexity. In this research, the translational domain is shown to contain a mixture of extensional tectonics (normal faults) up-dip and compressional tectonics (thrusts) down-dip, with extensive oblique-slip faults cutting thrusts perpendicularly during the translation of sediment. Variance and chaos, conventional seismic attributes, were used to manually pick and interpret the >500 regional-scale faults arising from the gravitational collapse system. Fault-net, a convolutional neural network (CNN), was compared with these edge-enhancing seismic attributes for extracting faults from the seismic volume. The CNN offers several notable advantages over conventional seismic attributes, such as automation, accelerated analysis, and improved time-efficiency on large datasets. Analysing the distribution, type, and geometry of faults within the basin gave valuable insights into the potential hydrocarbon system at work. Numerous natural gas and fluid escape features are identified in the seismic volume including an elongated mud volcano, pockmarked surfaces, and polygonal faults. The stability of the evolving margin is influenced by the underlying structure of a Late Cretaceous gravitational collapse system, also referred to as a deep-water fold and thrust belt (DWFTB) system. The fault framework within provides primary migration pathways for hydrocarbons. Major seafloor slumping occurs directly above a syncline of the Late Cretaceous DWFTB system. This slumping surrounds a structurally controlled, 4.2 km long elongated mud volcano situated between the translational and compressional domains of the underlying DWFTB system. The late Campanian has the largest accumulation of hydrocarbons evidenced by (1) an anticline with a positive high amplitude anomaly situated at the intersection of the two domains, and (2) >950 pockmarks preserved on the palaeo-surface compared to the 85 pockmarks observed on the seafloor. In addition to tectonics, the onset of stratified oceanographic circulation patterns and climate played a large role in changing depositional trends since the mid-Cenozoic. The Oligocene is characterized by a ~2.3 km wide, >13 km long, slope-perpendicular canyon formed at ~30 Ma during a major sea-level fall by a turbidity current. The Miocene is characterized by a ~14 km wide zone of slope-parallel, sinusoidal channels between water depths of 1 200–1 500 m. The formation and preservation of these features during the Miocene are attributed to the erosive interaction between two distinct water currents: (1) the Antarctic Intermediate Water flowing northwards, and (2) the deep North Atlantic Deep Water bottom currents flowing southwards; and the effects of the Benguela Upwelling System and a dry climate prevailing in southwest Africa all intensifying around 11 Ma. While pre-Miocene hydrocarbons originate from Turonian and Aptian source rocks, the origin of hydrocarbons on the seafloor is likely biogenic, arising from organic-rich sediment in the Miocene
A Clot to Uncover: FOXP3 and SARS-CoV-2 Nucleocapsid Interactions and Their Effect on DNA Binding
(University of the Witwatersrand, Johannesburg, 2024) Mcinnes, Keiran; Fanucchi, Sylvia
During COVID-19, systemic coagulopathy, which can lead to strokes and embolisms, is often observed in COVID-19 patients and may also contribute to long COVID. This coagulopathy is the result of overactivated platelets in circulation that leads to inappropriate clot formation. FOXP3 is a transcription factor involved in platelet development and loss of FOXP3 function leads to platelets that resemble those seen during COVID-19. Thus, FOXP3 may be dysregulated in COVID-19. The SARS-CoV- 2 nucleocapsid (NC) is a multifunctional protein typically associated with viral genome packaging and virion assembly. However, it is also capable of binding DNA and is potentially able to alter regulation of host protein expression. Here, potential interactions between the DNA-binding forkhead domain (FHD) of FOXP3 and the SARS-CoV-2 NC were investigated. Identification of a novel interaction between FOXP3 and SARS CoV-2 NC may provide new clues as to the pathophysiology of COVID-19. To address this aim, both proteins were overexpressed in T7 E. coli, purified via immobilised metal affinity chromatography, and monitored for potential interactions in the absence and presence of DNA using pull-down assays and fluorescence anisotropy. A direct interaction was identified between the two proteins in the absence of DNA. Additionally, it was found that both proteins are capable of binding to DNA at the same time, but excess NC was found to cause FHD dissociation from the FHD- NC-DNA complex. This result implicates NC in FOXP3 dysfunction which may be associated with the coagulopathy and other symptoms seen during COVID-19. Additionally, NC DNA binding does not appear to be driven by the FOXP3 consensus sequence, indicating that FOXP3 may not be the only transcription factor potentially dysregulated by NC
A comparative analysis of mining environmental management programme reports following a change to the one environmental system
(University of the Witwatersrand, Johannesburg, 2022) Mathe, Lebogang; Watson, I.
The mining fraternity has recently seen a shift in terms of environmental requirements under the ambit of the Mineral and Petroleum Resources Development Act 28 of 2002 (MPRDA) to the National Environmental Management Act 107 of 1998 (NEMA). The objective was to ensure a streamlined environmental compliance and to put emphasis on integrated environmental management. The One Environmental System (OES) was introduced which denotes that all environmental related projects or activities, including mining, are regulated through an ambit of one system. NEMA is recognised as an overarching law which provides for co-operative environmental governance; therefore, all environmental related activities are regulated under its ambit and EIA regulations. In this research report, the Environmental Management Programme (EMP) reports and Environmental Authorisations that were issued under the MPRDA and the 2014 NEMA regulations respectively, were analysed using the respective regulations to compare the environmental management measures outlined in the reports to achieve better environmental results. The aim of the research was to assess whether the Environmental Authorisations granted in terms of NEMA reflect better environmental results as compared to those issued under the MPRDA. Better environmental results mean improved management plans with clearer assessment of impacts, more detailed mitigation measures with specific, achievable, relevant and time bound actions that do not pose harmful effects on the environment and communities. A mixed method of analysis was implemented which included a review and comparison of legislation, evaluation and scoring of 20 EMPs (10 submitted under MPRDA and 10 submitted under NEMA) and interviews with key informants, namely competent authorities and the Environmental Assessment Practitioners (EAPs) were conducted to support the findings of the reports. In conclusion, the reports analysed presented a significant improvement and better environmental results under NEMA. The interviews conducted with authorities and the EAPs have also substantiated the observations made in terms of Environmental Management Programme reports and Environmental Authorisations submitted. The findings indicated that there were similarities in the MPRDA and NEMA regulations as 4 | P a g e regards some environmental requirements. However, the MPRDA lacked to outline a detailed description of environmental requirements needed to manage environmental impacts. Thus, applicants provided limited information in terms of the management of impacts. NEMA has provided additional requirements, including baseline environmental information, specialist reports and public participation amongst others; these contributed to improved environmental results. The study revealed that the reports submitted under NEMA provided better environmental management measures as compared to the reports submitted under MPRDA.
A comprehensive analysis of urban river pollution – the case of the Hennops river in Gauteng Province, South Africa
(University of the Witwatersrand, Johannesburg, 2023) Letseka, Thabiso Esaiah; Chimuka, L.; Richards, L.H.
The water quality of rivers is declining at an alarming rate due to pollution from anthropogenic activities associated with urbanization. To ensure ecological restoration and management of rivers, engaging in pollutant source apportionment, evaluation, and monitoring of water quality is of great significance. The study delivers a comprehensive assessment of the state of pollution in the Hennops river catchment facing pollution threats from rapid urbanization. The water quality assessment of the Hennops river was performed through chemical, microbiological, microplastics analysis and ecotoxicological approaches, spanning from upstream region in Tembisa to the downstream Hartbeespoort Dam. Standard methods were employed to assess physiochemical properties of the river’s water. Electrical conductivity and pH fell within the accepted criteria based on the standard water quality guidelines. However dissolved oxygen (DO) levels were below acceptable limits, ranging from 1.53 mg L-1 to 6.47 mg L-1. This signifies a substantial demand for oxygen in the river, likely due to the discharge of sewage from leaking pipes and wastewater treatment plants. This sewage introduces a high volume of organic matter, leading to an increased oxygen demand in the water. Microbiological pollution indicators were employed to assess the microbial water quality of the river. The study's findings revealed elevated bacterial counts, with Escherichia Coli (E. coli) reaching up to 2 250 cfu mL-1 upstream and decreasing to 30 cfu mL-1 downstream. These high counts suggest faecal contamination in the river water. Similar trends were observed with total coliform counts, high coliform counts 170 000 cfu mL-1 in the upstream which remained detectable even downstream and beyond the Hartbeespoort Dam, despite the dilution effects within the dam. The dam was identified as the primary repository for pollution originating upstream. Grab sampling followed by solid phase extraction (SPE) and the passive sampling using a Polar Organic Integrative Sampler (POCIS), were employed as sample preparation methods for preconcentration of methocarbamol, etilefrine, nevirapine, carbamazepine and venlafaxine from river water with subsequent analysis on Liquid Chromatography coupled to quadrupole time of flight mass spectrometry. Both methods yielded good figures of merit with limits of quantification in the range of 0.57 to 2.12 ng mL-1 for POCIS and 0.19 to 1.82 ng mL-1 for SPE. The compounds were detected in the water but at low levels (µgL-1 ), with detected concentrations of carbamazepine in the range 0.62 ng mL-1 – 0.32 ng mL-1 , methocarbamol detected in the range 0.11 ng mL-1 - 0.14 ng mL-1 and venlafaxine 0.50 ng mL-1 – 0.44 ng mL-1 using POCIS. The detected concentrations using SPE were in the range 0.13 ng mL-1 – 0.19 ng mL-1 for carbamazepine, while nevirapine and venlafaxine were detected although below limit of quantification. This underscores the advantage of using passive samplers, which enable the detection of fluctuating contaminant concentrations over time, in contrast to the one-time measurements obtained through grab sampling. In the case of microplastics in the water and sediment samples, five polymer types were identified: polyethylene (PE), polypropylene (PP), high density polyethylene, (HDPE), polyester and polystyrene. The predominant polymer type in surface water was PE (48.6 %), and that in sediment was PP (52.7 %). PE and PP were the most abundant polymer types in both phases, and as these also the leading polymers in plastics production. 80% of the identified microplastics were found to be fibre with most dominant sizes of 1-2 mm for sediments and 0.5-1 mm in water samples. The conducted tests deemed the river water not suitable for irrigation, drinking or recreational purposes and not capable to support aquatic life.
A Continuous Reinforcement Learning Approach to Self-Adaptive Particle Swarm Optimisation
(University of the Witwatersrand, Johannesburg, 2023-08) Tilley, Duncan; Cleghorn, Christopher
Particle Swarm Optimisation (PSO) is a popular black-box optimisation technique due to its simple implementation and surprising ability to perform well on various problems. Unfortunately, PSO is fairly sensitive to the choice of hyper-parameters. For this reason, many self-adaptive techniques have been proposed that attempt to both simplify hyper-parameter selection and improve the performance of PSO. Surveys however show that many self-adaptive techniques are still outperformed by time-varying techniques where the value of coefficients are simply increased or decreased over time. More recent works have shown the successful application of Reinforcement Learning (RL) to learn self-adaptive control policies for optimisers such as differential evolution, genetic algorithms, and PSO. However, many of these applications were limited to only discrete state and action spaces, which severely limits the choices available to a control policy, given that the PSO coefficients are continuous variables. This dissertation therefore investigates the application of continuous RL techniques to learn a self-adaptive control policy that can make full use of the continuous nature of the PSO coefficients. The dissertation first introduces the RL framework used to learn a continuous control policy by defining the environment, action-space, state-space, and a number of possible reward functions. An effective learning environment that is able to overcome the difficulties of continuous RL is then derived through a series of experiments, culminating in a successfully learned continuous control policy. The policy is then shown to perform well on the benchmark problems used during training when compared to other self-adaptive PSO algorithms. Further testing on benchmark problems not seen during training suggest that the learned policy may however not generalise well to other functions, but this is shown to also be a problem in other PSO algorithms. Finally, the dissertation performs a number of experiments to provide insights into the behaviours learned by the continuous control policy.
A fully-decentralised general-sum approach for multi-agent reinforcement learning using minimal modelling
(University of the Witwatersrand, Johannesburg, 2023-08) Kruger, Marcel Matthew Anthony; Rosman, Benjamin; James, Steven; Shipton, Jarrod
Multi-agent reinforcement learning is a prominent area of research in machine learning, extending reinforcement learning to scenarios where multiple agents concurrently learn and interact within the same environment. Most existing methods rely on centralisation during training, while others employ agent modelling. In contrast, we propose a novel method that adapts the role of entropy to assist in fully-decentralised training without explicitly modelling other agents using additional information to which most centralised methods assume access. We augment entropy to encourage more deterministic agents, and instead, we let the non-stationarity inherent in MARL serve as a mode for exploration. We empirically evaluate the performance of our method across five distinct environments, each representing unique challenges. Our assessment encompasses both cooperative and competitive cases. Our findings indicate that the approach of penalising entropy, rather than rewarding it, enables agents to perform at least as well as the prevailing standard of entropy maximisation. Moreover, our alternative approach achieves several of the original objectives of entropy regularisation in reinforcement learning, such as increased sample efficiency and potentially better final rewards. Whilst entropy has a significant role, our results in the competitive case indicate that position bias is still a considerable challenge.
A Geospatial Approach to Mapping Jacaranda Tree Distribution in Johannesburg, South Africa
(University of the Witwatersrand, Johannesburg, 2023-11) Reddy, Rohini Chelsea; Fitchett, Jennifer
Accurate mapping of the spatial distribution of invasive species is vital for the implementation of effective monitoring and management strategies. In countries where resources are scarce and costly, citizen science provides a cost-effective and accurate alternative for large-scale data collection. Citizen’s familiarity with their environment contributes to aspects such as accurate identification of features on the landscape. Advances in a geographic information system (GIS) together with open-sourced photography from Google Street View, provide accurate methods for in-field and remote validation of citizen science data for invasive mapping and assists with the creation and compilation of maps to visualize the spatial distribution of invasive plants upon the landscape. In this study, the first spatial distribution maps for invasive tree species, Jacaranda mimosofolia (common name: Jacaranda), are created for the City of Johannesburg (CoJ). Jacaranda trees are well-known by citizens in the CoJ for their district purple flowers which blanket the landscape during springtime. A combination approach using citizen science, GIS, and Google Street View for data collection, analysis, and creation of the first spatial distribution map of exact location and prevalence of Jacaranda trees within certain suburbs of the CoJ, is produced. A total of 8,931 ground-truthing geopoints together with extensive Google Street View validation for Jacaranda tree presence, formed the basis of accurate spatial distribution maps. The first research question of this study focused on the spatial distribution of Jacaranda trees in the CoJ and was answered as a total of 54 suburbs were confirmed as having a large presence of Jacaranda trees in the CoJ. Citizen science data collected a total of 488 geotags for possible Jacaranda tree presence in the CoJ, over a 75-day online survey collection period. Although citizen science data provided a lower spatial resolution compared to successful fieldwork and Google Street View approaches, citizen science data provided very high accuracy for the identification and geolocation of Jacaranda tree presence in the CoJ which answers the second research question based on the effectiveness of the geospatial approach towards citizen science, ground-truthing and Google Street View as data collection methods. Since the accuracy of citizen science resulted in 66% of collected geotags within the categories of ‘very high’, ‘high’ and ‘moderate’ accuracy ranges of between <7-24m from a confirmed Jacaranda tree, together with the accuracy of 8,931 in-field collected geolocation of Jacaranda trees and Google Street View’s accuracy and capability of collecting street view imagery – it is concluded that the combined approach of ground-truthing, citizen science and Google Street View contribute not only to effective data collection, but also towards the successful mapping of Jacaranda tree presence in the CoJ.
A GIS framework for the integrated conceptualisation, analysis and visualisation of Gauteng's complex historic and contemporary post-mining urban landscape
(University of the Witwatersrand, Johannesburg, 2023) Khanyile, Samkelisiwe; Esterhuysen, Amanda; Kelso, Clare
This research study applies assemblage theory as a philosophical lens. It proposes a framework for integrating contemporary and historical landscape characteristics of post-mining and urban landscapes for an integrated conceptualisation, mapping, and analysis of Gauteng, South Africa. The study utilises a mixed methods approach, incorporating spatial and non-spatial (literature and survey) data of varying formats to identify landscape characteristics. Additionally, it applies three multicriteria decision analysis (MCDA) and GIS mapping techniques, employing a simplified rationale to keep its complexity low. Descriptive and inferential statistics were used to analyse the quantitative data, while the qualitative data was analysed using a thematic analysis. The literature and survey analysis findings were used to inform the development of a framework demonstrating the integration of Gauteng's post-mining and urban landscape characteristics using a fuzzy overlay, weighted overlay and random forest classification, along with an accuracy assessment of the mapped results. Based on the proposed framework, the mapped results' performance was evaluated through four methods: confusion error matrix, cross-evaluation, areal coverage comparison, and an image differencing assessment. The literature and survey analysis findings, used to inform the framework, reveal that the two landscapes consist of an assemblage of characteristics and highlight differences in the characterisation of post- mining and urban landscapes. Distinctions were also apparent between literature-derived characteristics and those identified from local experts. The local expert-derived characteristics demonstrate context- specific characteristics of Gauteng's post-mining and urban landscape. At the same time, those based on the literature emphasise a more distinct and separate portrayal of post-mining and urban landscape characteristics (pages 115-116). The characteristics identified from local experts were less conservative (pages 117-118). They included urban-related characteristics in the description of post-mining landscapes and mining-related characteristics in the description of urban landscapes, presenting some similarities in the characterisation of these two landscapes in Gauteng. Moreover, the findings from local experts also revealed that literature and other written or mapped work informed most definitions of post-mining and urban landscapes. The framework for integrating landscape characteristics (pages 121-123) was spatially represented through the three mapping methods, visually demonstrating several findings providing insight into the Gauteng landscape's uniqueness. First, it demonstrates that the differences in the characterisation of these landscapes also impact how they are spatially represented. The maps of post-mining and urban landscape characteristics based on the literature presented a similar pattern to the traditional mapping of mining and urban landscapes in Gauteng. These mapping techniques show the highest values across the mining belt and at the province's core. These findings highlight the influence of literature on the representation of these two landscapes, which is consistent with local experts' reports. In all three mapping methods, the maps generated from local expert characterisations of post-mining and urban landscapes presented a larger spatial footprint than those based on literature-derived characteristics. This distinction was attributed to incorporating additional post-mining and urban landscape characteristics in the maps based on expert input and applying the three mapping techniques - using representation methods not commonly used in mapping these landscapes. Second, the integrated maps of post-mining and urban landscape characteristics suggested a variance in the presence of post-mining and urban landscape characteristics across the province in the maps generated using fuzzy and weighted overlay techniques. This indicates that some parts of the province have a higher or lower presence of post-mining or urban characteristics (pages 125-132). These findings were visible in the maps generated from literature and local experts, indicating the diversity of both landscapes and the co-existence of post-mining and urban landscape characteristics in the local expert maps. This implies an intricate relationship between these landscapes, challenging the idea of them being strictly separate, as indicated in maps presenting characteristics identified from the literature. Furthermore, a closer inspection of the areas showing the intersection between post-mining and urban landscape characteristics also points towards the porosity of boundaries of these two landscapes and alevel of spatial overlap, organisation and arrangement, which are prevalent at varying levels (pages 164- 168). Third, the maps generated using literature-derived characteristics achieved higher accuracy scores, attributed to using reference data traditionally used to map the two landscapes under investigation. This reference data only comprised classes that characterised the physical mining and urban classes, consistent with those identified in the literature. Consequently, it lacked additional factors characterising the post-mining and urban landscape identified from local experts. The fuzzy overlay maps informed by literature demonstrated an accuracy exceeding 70% for post-mining and urban landscape characteristics. In comparison, those reported by local experts scored 64. The weighted overlay and random forest classification resulted in accuracy rates exceeding 50% for post-mining landscape characteristics maps, regardless of whether literature or expert-derived characteristics were used. Additionally, urban landscape characteristics maps achieved an accuracy of over 76%, regardless of the characteristics used to inform the mapping. These findings were attributed to the different mapping techniques employed, with fuzzy and weighted overlay using a gradual range scale, while random forest classification employed a binary scale. This highlights how different mapping methods affect the representation of space. Additionally, it demonstrates the versatility of these mapping techniques in mapping complex spaces such as post-mining and urban landscapes. In this study, the fuzzy overlay accuracies exceeded 60% for all maps and emerged as the most suitable choice for integrating landscape characteristics due to its ability to represent blurred and porous boundaries between Gauteng's post- mining and urban landscapes. In conclusion, the study challenges the notion of post-mining and urban landscapes as distinct landscapes, emphasising the importance of considering the varying levels of spatial intersection between these two landscapes. With the proposed framework and the alternative representation of these landscapes, including contextual information, this research provides insights into new conceptualisations of urban, post-mining landscapes and mineralised urbanisations as assemblages of different landscapes and characteristics with porous boundaries. This enables a better understanding of Gauteng's post-mining and urban landscapes, which could benefit the representation, communication and management of these landscapes. Recognising the potential applications and limitations of frameworks such as the one developed for this study, the high-level recommendation arising from this study suggests a need for ongoing research into the contextual representation of landscapes and their characteristics. This can be achieved by incorporating input from communities, conducting research on quantifying intangible landscape characteristics and developing tools that facilitate the automation and alignment of such data with development plans.
A Phenotype Prediction Framework for Classifying Colorectal Cancer Patients’ Response to FOLFOX Treatment: An Integrated Approach
(University of the Witwatersrand, Johannesburg, 2024) Mashatola, Lebohang; Kaur, Mandeep
Colorectal cancer (CRC), characterised by its prevalence and heterogeneity, poses a significant challenge in understanding drug resistance, especially in the context of FOLFOX treatment. This study presents an innovative methodology that integrates diverse data analysis approaches to address the challenge of predicting the phenotype of CRC patients resistant or sensitive to FOLFOX. The initial analysis involved dierential and co-expression analyses, identifying pivotal hub genes crucial to drug resistance in CRC, regulating intricate molecular networks. Subsequent enrichment analysis revealed their significant roles in biological functions, particularly influencing DNA repair and nuclear division. To capture inherent topological characteristics within genetic expression data, a novel technique utilising topological data analysis (TDA) was employed. By applying persistence homology to generate persistence images, the Vietoris-Rips complex was constructed using the signed-topological overlap matrix, comprehensively capturing numerous topological features, including high-dimensional Betti-1 and Betti-2. This provided valuable insights into the structural patterns of gene expression between the hub genes. Furthermore, the integration of whole-slide images enhanced understanding of tissue anatomy, which is crucial for predicting cancer stages. Using a MobileNet architecture, a deep learning model classified cancer stages, contributing to a holistic understanding of colorectal tumor microenvironments. For predictive modelling of drug resistance, a multilayer perceptron applied topological summaries generated by TDA. The developed framework, GeTopology, exhibited remarkable performance metrics, achieving an overall 83% accuracy in predicting the FOLFOX response, demonstrating a 3% improvement over a previously published phenotype prediction framework (NSCLC ) that utilised similar data modes. Robust accuracies were consistently observed in independent datasets, classifying both cancer patients and healthy individuals. The results indicated an approximate 10% increase in model prediction accuracy compared to NSCLC, emphasising the potential clinical impact of this integrative approach. In conclusion, this study advances the understanding of drug resistance in CRC by proposing a novel approach that integrates topology with histopathological images, oering transformative insights into predictive modelling and precision medicine
A study of the support effect of carbon dots-derived graphene-like sheets on the autoreduction of cobalt nanoparticles for Fischer–Tropsch synthesis
(2022) Mokoloko, Lerato Lydia
The aim of this study was to synthesize and characterize carbon dots (CDs) and to use them as a support material for cobalt (Co) based Fischer-Tropsch synthesis (FTS) reactions. The CDs were chosen for this study due to their small size (< 10 nm), easy surface functionalization and synthesis. The small size of the CDs was required for the study of inverse catalyst support effects. An inverse supported catalyst (in this case, the Co/CDs catalyst) refers to the dispersion of a support material that has a small size (d < 5 nm) onto the surface of a metal catalyst with a similar small size (d > 8 nm). The synthesis of this proposed catalyst was successful. FTS studies on the Co ‘supported’ CDs were attempted. Extremely poor FT activity was observed. Post analysis of the catalyst revealed that the CDs did not retain their quasi-spherical and small particle size morphology under the FTS reaction conditions (temperature 220 °C, 10 bar P; H2:CO ratio = 2:1). Instead, upon exposure to a heat treatment, the CDs were transformed into layered structures with a unique resemblance to graphene-based nanosheets (GNSs). This transformation impacted on the use of these catalysts in the FTS reaction. However, this result indicated an unusual transformation of the CDs into another carbon shape. In light of the fascinating transformation phenomenon, annealing studies were then conducted to investigate the effect of annealing temperatures on the CDs structural changes. The CDs (average d= ~ 2.5 nm) used in this study were obtained from the microwave-assisted carbonization of L-ascorbic acid and subjected to a heat treatment (i.e. annealing) at temperatures between 200 and 700 ℃ in a horizontal CVD apparatus under an inert nitrogen gas. It was observed that annealing transformed the CDs from 0-D qausi-spherical nanoparticles to 3- D multi-layered carbons (at 300-600 ℃) and finally 2-D layered materials (at 700 ℃). Furthermore, annealing at 700 ℃ yielded a 2-D single-layered material with comparable properties to traditionally reduced graphene oxide (rGO). A wide range of characterization techniques were used to gain an insight into the physicochemical properties of these novel CDs-derived allotropes as well as to rationalize their mechanism of formation. After evaluating the properties of these materials, it was clear that the surface oxygen functional groups, observed from XPS, 13C NMR and other studies, were responsible for the CDs to rGO transformation. It was proposed that the CDs are assembled to form rGO (and other CDs-rGO derivatives) by either the Ostwald ripening (in which the carbons agglomerated via a gas phase) or a solid phase reaction (involving reaction of CD edges). To further investigate the effect of annealing on the evolution of CDs to layered carbon structures, N-doped CDs (or NCDs) were also studied. The method used to make the pristine CDs was modified by incorporating urea as a nitrogen source to make the NCDs. Annealing the NCDs at temperatures between 200 and 700 ℃ also transformed the quasi-spherical NCDs (average d = ~ 4.1 nm) to multi-layered carbon sheets at temperature as low as 200 ℃. The CD transformation was also associated with the loss of surface functional groups, with % O and N contents of ca. 17 and 16 % (pristine NCDs) being reduced to ca. 8 and 7 % for NCDs annealed at 700 ℃. A similar mechanism for the formation of these N-doped layered carbon structures by annealing was also proposed here. For these samples, it was also observed that the N-bonds, especially the sp3 type nitrogen bonds found on the edges of the NCDs, also took part in the coalescence of the NCDs to give the layered materials. XPS data suggested that in the process, these sp3 type nitrogen bonds were transformed into sp2 pyrrolic-N, pyridinic-N and GraphiticN groups. The annealed CDs products were used to support Co (called Co3O4/T250, Co3O4/T400 and Co3O4/T700 where T is the temperature at which the CDs were annealed) for use in FT studies. Studies were conducted to evaluate the effect Co hydrogen reduction temperatures verses autoreduction temperature, catalyst thermal stability and performance in the FTS reaction at 220 °C (10 bar P; H2:CO ratio = 2:1). Upon investigation of the reduction behaviour of the Co/CDs derivative catalysts using in situ PXRD, it was found that these materials can successfully facilitate autoreduction of Co3O4 to Co face-centered-cubic (fcc) at temperatures > 400 ℃ by a reduction pathway similar to that observed using conventional H2 reduction conditions. As expected, the reduction under H2 took place at a lower activation temperature (> 250 ℃) than the autoreduction process. It was also noted that these novel carbon support derived from CDs gave reduced FTS performance compared to the unsupported Co, especially towards C5+ yields (< 30 % for all Co supported catalysts). These novel CDs-derived allotropes were found to have limited use as supports in Co-based FTS, due to Co agglomeration. These NCDs-derived allotropes (annealed at 200 ℃, 400 ℃ and 700 ℃) were incorporated as active layers in the fabrication of chemoresistive sensing device detection of volatile organic compounds (VOCs). These layered showed enhanced chemical vapour sensing properties, especially for methanol and ethanol detection at room temperature. Therefore, although there are great limitations for applications of these CDs-derived layered allotropes in FTS reaction, these materials show a much better potential for applications in facile and cost effective VOC sensors. Further studies on this will be conducted.
A symmetry perspective of third-order polynomial evolution equations
(University of the Witwatersrand, Johannesburg, 2024) Gwaxa, Bongumusa; Jamal, Sameerah
In this thesis, we analyse the full class of ten Fujimoto-Watanabe equations. In particular, these are highly nonlinear third-order and two fifth-order equations. With the aid of computer algebra software such as Mathematica, we calculate symmetries for these equations and we construct their commutator tables. The one dimensional system of optimal subalgebras is obtained via adjoint operators. Finally, we reduce these higher-order partial differential equations into ordinary differential equations, derive their solutions via a power series solution method and show how convergence may be tested. Lastly, we determine some conservation laws
An Assessment of Beauty Waste Management Practices: A Case Study of Rustenburg Beauty Salons
(University of the Witwatersrand, Johannesburg, 2024) Knight, Jasper; Knight, Jasper
The beauty salon industry is one of the fastest growing industries and is a significant waste generator in South Africa. Waste that results from beauty salons is a thorny environmental issue because it spans from different waste types and sources. Futhermore it requires waste treatment and different disposal methods. In view of this, this study seeks to assess waste management practices of beauty salons in Rustenburg, South Africa, in order to identify the types of waste salons generate and to identify recommendations that can assist beauty salons to be environmentally sustainable by improving their waste management practices. The aim of the research is to understand how beauty salon waste is discarded and to what degree beauty salon personnel understand the impacts of waste on the environment. The objectives of this research are to (1) determine the total amount of waste produced by selected beauty salons in Rustenburg, (2) identify the waste management practices undertaken by the beauty salons, (3) explore the challenges the beauty salons face in relation to waste management, and (4) identify recommendations of how beauty salons in South Africa can further improve their waste management practices. This study employed a mixed methods design through quantifying the amount of waste the selected beauty salons generate over a two month period, and by interviewing salon personell on their views on salon waste and waste management practices. Fifteen salons were surveyed. Results were analyzed using thematic analysis. The results show that the all the beauty salons combined produce a total annual estimated waste of 4732.2 kg. Through interviews, the study identified waste management practices of the beauty salons to be primarily premised on discarding waste in dustbins for municipal collection, burning waste, or dumping waste in unregulated dumping sites when waste service delivery poses challenges. These three practices are the most common modes of waste disposal in the Rustenburg beauty salons. Issues of waste management facing beauty salons includes lack of waste facilities and lack of knowledge about waste management. The major recommendation from beauty salons and civil (professional) organisations was for government to provide beauty salons with better infrastructure for waste sorting, recycling, pick-up and disposal. Promotion of good practice and awareness campaigns were cited as recommendations to improve waste management practices in beauty salons
An Essay on Branching Time Logics
(University of the Witwatersrand, Johannesburg, 2024) Marais, Chantel
In this thesis we investigate the Priorian logics of a variety of classes of trees. These classes of trees are divided in to irreflexive and reflexive trees, and each of these has a number of subclasses, for example, dense irreflexive trees, discrete reflexive trees, irreflexive trees with branches isomorphic to the natural numbers, etc. We find finite axiomatisations for the logics of these different classes of trees and show that each logic is sound and strongly / weakly complete with respect to the respective class of trees. The methods use to show completeness vary from adapting some known constructions for specific purposes, including unravelling and bulldozing, building a network step-by-step, filtering through a finite set of formulas, as well as using some new processes, namely refining the filtration and unfolding. Once the logics have been shown to be sound and complete with respect to the different classes of trees, we also show that most of these logics are decidable, using methods that include the finite model property, mosaics and conservative extensions. Lastly, we give a glimpse into the available research on other languages used to study branching time structures, including the Peircean and Ockhamist languages, and languages that include additional modal operators like “since” and “until”
An essential variable approach for integrated social-ecological systems monitoring to determine sustainability in a South African catchment
(University of the Witwatersrand, Johannesburg, 2024-05) Itzkin, Adela; Clifford-Holmes, Jai Kumar; Coetzer, Kaera; Scholes, Mary
This thesis presents an essential variable (EV) approach tailored for integrated social ecological systems (SES) monitoring within the Tsitsa River Catchment (TRC) in South Africa. The study addresses the conceptual and methodological challenges inherent in SES integration and monitoring to provide insights into sustainable landscape management. The research methodology employed a transdisciplinary social learning process, integrating systems thinking and participatory research methods across three related publications. In the first publication, systems diagramming, qualitative interviews, and participatory data collection provided a systemic snapshot of the interconnected social and biophysical drivers of land degradation in the TRC. The findings underscored the dual benefits of changes in land use and grazing practices for landscape improvement and sustainable livelihoods, informing the identification of monitoring variables crucial for sustainable land management. In the second publication, a social learning process, termed participatory self-observation, explored approaches to enhance integration and monitoring of biophysical and social data for adaptive management. The process identified data integration, overload, scale, learning oriented monitoring, and relationship-building as key challenges in SES monitoring Recommendations included participatory approaches focusing on applied work, identifying essential data for SES monitoring, and improving transdisciplinary collaboration. In the third publication, a transdisciplinary process guided by EV development identified Essential Social-ecological System Variables (ESEVs) for the TRC, emphasising the relational connection between social and ecological aspects of SESs. ESEVs were prioritized based on essentiality scores and participant consensus, facilitating integrated planning and management at the catchment scale. The ESEVs identified for the TRC include 'soil erosion related to human actions on the land,' 'participation in natural resource governance,' 'grazing and rangeland sustainability,' and 'land cover and condition. Participants proposed three additional ESEVs, 'access to water,' 'local natural resource governance system,' and 'human well-being in the landscape'. The study concludes that collaborative learning informed by diverse transdisciplinary perspectives can guide adaptive monitoring approaches, with lessons from the TRC applicable to diverse contexts. The ESEV approach offers wider application value, addressing the question of the minimum variables needed for integrative landscape management in complex SESs. Overall, this thesis pioneers a transdisciplinary mixed-methods approach for efficiently monitoring social-ecological sustainability outcomes. Through advancements in conceptualizations of integrated, contextually grounded monitoring, it offers a novel perspective for understanding the implications of development interventions, bridging critical gaps in SES monitoring.
An ethnobotanical study of indigenous knowledge of the medicinal plants used by traditional healers in the rural communities of Nkomazi Local Municipality, Mpumalanga province
(University of the Witwatersrand, Johannesburg, 2024) Khoza, Nompendulo; Dukhan, Shalini; Ramalepe, Phillemon; Risenga, Ida
Traditional medicine continues to significantly impact many people’s lives amid all the advancements in modern medicine. Many rural communities in Mpumalanga province depend on indigenous traditional medicines to manage various ailments. The available research on the traditional usage of medicinal plants among rural communities in Mpumalanga is highly fragmented and under-researched. The decline of medicinal plant populations has led stakeholders to take various initiatives to counteract over-exploitation, including cultivation as a viable conservation approach. However, the scientists' inadequate understanding of the acceptance of cultivated medicinal plants by traditional healers is one of the issues contributing to the failure of medicinal plant cultivation programs. Consequently, this study aimed to document medicinal plants utilised by the Nkomazi Local Municipality's traditional healers and assess opportunities and constraints for medicinal plant conservation in the Nkomazi Local Municipality. The ethnobotanical data was obtained through semi-structured questionnaires and guided field walks with traditional healers. Individual interviews were conducted with ten traditional healers from eight villages across Nkomazi during field visits between July 2021 and February 2022. The study employed qualitative and quantitative approaches to understand traditional healers' perspectives concerning the ethnobotanical significance and medicinal plant conservation. The study found that the indigenous knowledge of medicinal plants in the Nkomazi Local Municipality is diverse, encompassing 111 species from 59 different families employed to treat 70 ailments. Most of the reported medicinal plants for this study are of Least Concern. Additionally, the top fourteen most reported species in the Nkomazi Local Municipality included commercially valuable plants such as Psidium guajava, Ricinus communis, Sclerocarya birrea, Aloe ferox, Aloe maculata, Leonotis leonurus, and Moringa oleifera. Most of the Nkomazi Local Municipality's traditional healers did not know about protected plant species and the National Environmental Management Act (NEMA). Traditional healers were aware of the decline in wild populations of medicinal plants, which they attributed to various factors such as overharvesting. Diviner’s and herbalist perception of using cultivated plants did not differ significantly (𝜒2=0.4762, df=1, P= 0. 490). The study provided a comprehensive inventory of medicinal plants utilised by Nkomazi traditional healers and essential data for future assessments of the use local use of indigenous medicinal plants.
An integrated approach for detecting and monitoring the Campuloclinium macrocephalum (Less) DC using the MaxEnt and machine learning models in the Cradle Nature Reserve, South Africa
(University of the Witwatersrand, Johannesburg, 2024) Makobe, Benjamin; Mhangara, Paidamoyo
The invasion of ecosystems by invasive plants is considered as one of the major human- induced global environmental change. The uncontrolled expansion of invasive alien plants is gaining international attention, and remote sensing technology is adopted to accurately detect and monitor the spread of invasive plants locally and globally. The Greater Cradle nature reserve is a world heritage site and intense research site for archaeology and paleontology.It was accorded the world status by the United Nations Educational, Scientific and Cultural Organizations (UNESCO) in 1991 due to its variety of biodiversity present and carries information of significance about the evolution of mankind. The invasion of Campuloclinium macrocephalum (pompom weed) at the Cradle nature reserve is downgrading the world status accorded to the site, lowers the grazing capacity for game animals and replaces the native vegetation. This research study explored the capability of Sentinel-2A multispectral imagery in mapping the spatial distribution of pompom weed at the nature reserve between 2019 and 2024. The non-parametric classification models, support vector machine (SVM) and random forests (RF) were evaluated to accurately detect, and discriminate pompom weed against the co-existing land cover types. Additionally, the species distribution modelling MaxEnt Entropy was incorporated to model spatial distribution and pompom weed habitat suitability. The findings indicates that SVM yielded 44% and 50.7% spatial coverage of pompom weed at the nature reserve in 2019 and 2024, respectively. Whereas, the RF model indicates that the spatial coverage of pompom weed was 31.1% and 39.3% in 2019 and 2024, respectively. The MaxEnt model identified both soil and rainfall as the most important environmental factors in fostering the aggressive proliferation of pompom weed at nature reserves. The MaxEnt predictive model obtained an area under curve score of 0.94, indicating outstanding prediction model performance. SVM and RF models had classification accuracy above 75%, indicating that they could distinguish pompom weeds from existing land cover types. The preliminary results of this study call for attention in using predictive models in predicting current and future spatial distribution of invasive weeds, for effective eradication control and environmental management.
An investigation into high gear and low gear propulsion in human gait and its relation to metatarsal diaphyseal geometric cross-sectional properties
(University of the Witwatersrand, Johannesburg, 2023-06) Reyneker, Mark Brenden; Carlson, Kristian J.; Zipfel, Bernhard
This study investigates the relationship between metatarsal bone form, as quantified by cross-sectional geometric properties, and its relationship to high (medial forefoot loading) versus low gear (lateral forefoot loading) push-off during the propulsion phase of the gait cycle. The objective being to assess whether forefoot loading may be variable or whether high gear loading occurs in higher frequencies, as depicted in theoretical foot function models. The study sample (n=53), made up of three broad groups, include Later Stone Age southern Africans, post-industrial individuals from South Africa, and the Jomon of Japan. Metatarsals 1-5 cross-sectional geometric properties (CSA, Ix, Iy, Imin, Imax, Zx, Zy, Zp, Zmin, Zmax) taken from CT scans at 25%, 35%, 50% and 65% metatarsal diaphyseal biomechanical lengths are grouped into high gear (metatarsal 1-2) and low gear (metatarsal 2-5) for comparison. The combined population analysis reveals that the high gear metatarsal diaphysis exhibit significantly higher strength and rigidity driven mainly by the post-industrial individuals from South Africa and the Later Stone Age southern Africans. In contrast, the Jomon of Japan, exhibit no significant differences between high and low gear metatarsals except for CSA, Imax, and Zmax. Furthermore, metatarsal 1 and 5 differ far less in cross-sectional geometric properties in the Jomon of Japan compared to the other populations except for medial-lateral strength (Zy) and torsional and average bending strength (Zp) where metatarsal 5 is significantly higher. The study findings indicate that forefoot loading demonstrates variability during the propulsion phase of gait, while also suggesting a higher frequency of occurrence for high gear push-off. This challenges current theoretical models of foot function that emphasise high gear push-off as typical and normal for striding bipedalism.
Analysing RNA-sequence data for pancreatic ductal adenocarcinoma tissue samples to identify potential biomarkers
(University of the Witwatersrand, Johannesburg, 2023-09) Jamal, Khadija Sanober; Kaur, Mandeep
Pancreatic ductal adenocarcinoma (PDAC) accounts for approximately 90% of pancreatic cancer and is the fourth leading cause of death with a five-year survival rate of less than 10%. Patients are asymptomatic until detection is observed at a metastatic stage, hence contributing massively towards the high mortality rate. This study was conducted to explore PDAC and its two main subtypes, the classical and basal-like subtype, in an in-depth level via bioinformatic analysis. Bioinformatics is a computational approach to evaluate biological data by analysing omics data including genomic expression and proteomic sequences. A workflow consisting of programmes and web-tools was used to analyse PDAC RNA-sequence data. The sample sets were grouped according to tumour, stage, and subtype. The workflow began with quality control using FastQC and Trimmomatic. Alignment of sequencing files and counts were done through HISAT2 and HTSeq. The main component of this workflow was differential gene expression analysis to identify differentially expressed genes (DEGs), statistically significant genes, per compared conditions. WGCNA was used for co-expression analysis to identify the hub genes involved in regulating the biological network. Lastly, in-silico validation was done by using available web tools to support the findings of this workflow. The identified tumour genes included S100A11, PKM, GPRC5A, LAMC2 and ITGA2, which may represent as universal biomarkers as sample extraction was performed from data generated from individuals belonging to 8 different countries. KRT13 and IL6 were identified in the advanced stage and their role in cancer progression have been explored in this current study. The basal-like subtype had CAV1, DCVLD2 and TGFB2 genes that contribute to treatment resistance. The common dysregulated genes in the basal-like subtype and advanced stage were analysed to evaluate the link between subtype and stage which included WNT3A, TP63, KRT13 and IGF2BP. Coexpression analysis revealed hub genes for tumour (KIF4A, SPAG5, RRM2 and AURKA), basal-like subtype (BUB1, DEPDC1 and KIF14) and classical-subtype (PTPRN and CAMK2B). Through a machine learning model, recall, precision and accuracy scores per sample conditions for the DEGs were all above 94%. These potential biomarkers all have significant roles in promoting cancer progression, aggression and resistance. Hence, these may serve as a less invasive screening method for PDAC as DEGs were classified based on tissue or blood (extracellular vesicle) biomarkers. However, further wet laboratory validation is required for these biomarkers.
Analysis of some convergence results for inertial variational inequalities problem and its application
(University of the Witwatersrand, Johannesburg, 2023) Kunene, Thembinkosi Eezy
Some core aspects of nonlinear analysis, which is a major branch of mathematics, are the optimization problems, fixed point theory and its applications. These concepts, that is, optimization theory, fixed point theory and its applications are widely applied in several fields of science such as networking, inventory control, engineering, economics, policy modelling, transportation and mathematical sciences to mention but a few. Due to its relevance to different fields, the theory of optimization and fixed point has been a popular field of research for a long time. Given its expansive nature, researchers continue to make new discoveries and advancements, contributing to its enduring significance across various disciplines. The goal of this dissertation is to explore some convergence iterative methods for approximating optimization problems. We propose a new modified projection and contraction algorithm for approximating solutions of a variational inequality problem involving a quasi-monotone and Lipschitz continuous mapping in real Hilbert spaces. We incorporate the technique of two-step inertial into a single projection and contraction method and prove a weak convergence theorem for the proposed algorithm. The weak convergence theorem proved requires neither the prior knowledge of the Lipschitz constant nor the weak sequential continuity of the associated mapping. Under additional strong pseudomonotonicity, the R-linear convergence rate of the two-step inertial algorithm is presented. Finally, some numerical examples are given to illustrate the effectiveness and competitiveness of the proposed algorithm in comparison with some existing algorithms in the literature
Analyzing the performance and generalisability of incorporating SimCLR into Proximal Policy Optimization in procedurally generated environments
(University of the Witwatersrand, Johannesburg, 2024) Gilbert, Nikhil; Rosman, Benjamin
Multiple approaches to state representation learning have been shown to improve the performance of reinforcement learning agents substantially. When used in reinforcement learning, a known challenge in state representation learning is enabling an agent to represent environment states with similar characteristics in a manner that would allow said agent to comprehend it as such. We propose a novel algorithm that combines contrastive learning with reinforcement learning so that agents learn to group states by common physical characteristics and action preferences during training. We subsequently generalise these learnings to previously encountered environment obstacles. To enable a reinforcement learning agent to use contrastive learning within its environment interaction loop, we propose a state representation learning model that employs contrastive learning to group states using observations coupled with the action the agent chose within its current state. Our approach uses a combination of two algorithms that we augment to demonstrate the effectiveness of combining contrastive learning with reinforcement learning. The state representation model for contrastive learning is a Simple Framework for Contrastive Learning of Visual Representations (SimCLR) by Chen et al. [2020], which we amend to include action values from the chosen reinforcement learning environment. The policy gradient algorithm (PPO) is our chosen reinforcement learning approach for policy learning, which we combine with SimCLR to form our novel algorithm, Action Contrastive Policy Optimization (ACPO). When combining these augmented algorithms for contrastive reinforcement learning, our results show significant improvement in training performance and generalisation to unseen environment obstacles of similar structure (physical layout of interactive objects) and mechanics (the rules of physics and transition probabilities).

Browse

Browsing Faculty of Science (ETDs) by Title

Results Per Page

Sort Options