3. Electronic Theses and Dissertations (ETDs) - All submissions
Permanent URI for this communityhttps://wiredspace.wits.ac.za/handle/10539/45
Browse
8 results
Search Results
Item The use of geographic information systems (GIS) and remote sensing on the matchless amphibolite belt (MAB), gorob-hope area, Namibia(2018) Kaasheka, Elina KThe Matchless Amphibolite Belt (MAB) is a linear geological feature up to 3 kilometers wide and at least 350 kilometers long within the Damara Orogen of Namibia (Miller, 1983). The exact length of the belt is not well known due to the superficial cover of Kalahari sand, especially east of Windhoek, towards the border with Botswana. The Matchless amphibolite belt is associated with Neoproterozoic massive sulphide Cu-Au-(Zn-Ag) deposits (Breitkopf & Maiden, 1988; and Killick, 2000). Conventional geological and mineral exploration practices need enormous investments, a long time period plus intensive human labour especially in areas that are not easily accessible or with complex geological features such as the Gorob-Hope area. This study focused on the occurrence of mineralization in the Gorob-Hope area of the Matchless Amphibolite Belt and involved gathering, processing and interpretation of historic geological and remote sensing data using ENVI and ArcGIS software. These techniques were applied with a view to establish how the ore deposits are associated with the presence of the MAB and to establish the relationship between lithological units, the structures and the ore body localization. Application of ENVI and GIS techniques proved that it stands for an efficient and cost effective option of outlining mafic units, metasedimentary units, structures and ore body localization based on the available historical geological, and remote sensing data. The study also established that ore deposits found in the study area are controlled by structural features and are associated with MAB.Item Assessing the effects of land use/land cover changes on carbon storage of natural forest ecosystems in Southern Africa: a remote sensing approach(2018) Lembani, Reuben LThe extensiveness of Africa’s natural forests with approximately 20% of the continent’s land cover, makes them an important source of livelihood to over 100 million people. The exclusion of many natural forest types from the United Nations Reducing Emissions from Deforestation and forest Degradation Plus schemes suggest that very little is known about their contribution to global carbon sequestration storage. Furthermore, there is often significant underestimation of forest total carbon storage as most studies have focussed mainly on monitoring, reporting and evaluating carbon stocks in aboveground live biomass. This is partly because the belowground carbon pool is influenced by several spatially heterogeneous abiotic and biotic factors which are themselves not well studied. This thesis uses multi-temporal Landsat remotely sensed images to map land use/land cover changes and aboveground forest carbon stocks for the periods 1986-1998, 1998-2001, 2001-2004, 2004-2011 and 2011-2016 in Luanshya, Zambia, and the periods 1992-1998, 1998-2001, 2001-2006, 2006-2010 and 2010-2016 in Bushbuckridge, South Africa. Aboveground forest carbon stocks were estimated from the regression model relating in situ estimates in 30 x 30 m plots to spectral reflectance measurements for different time periods. The study also used PCA to assess 11 edaphic factors associated with the spatial and downward vertical distribution of SOM content in three forest subsoil layers. The calculated rate of forest regeneration was almost twice lower than the rate of deforestation, and thus over 68% of natural forests are secondary growth forests. From the ecological survey on 126 analogous forest sample plots of different succession stages, it was apparent that forest disturbances result in the replacement of bigger individual trees by smaller trees, and subsequent increase in tree density and reduction in species diversity. These anthropogenic and natural phenomena explained the mean changes in aboveground forest carbon stocks which were 44.20 t ha-1 in 1986, 43.85 t ha-1 in 1998, 54.81 t ha-1 in 2001, 53.02 t ha-1 in 2004, 46.83 t ha-1 in 2011 and 36.67 t ha-1 in 2016 for Luanshya, and 4.67 t ha-1 in 1992, 9.52 t ha-1 in 1998, 40.84 t ha-1 in 2001, 34.15 t ha-1 in 2006, 41.92 t ha-1 in 2010 and 24.95 t ha-1 in 2016 for Bushbuckridge. Conversely, the effects of these changes were also observed in the complexity in the spatial and downward vertical distribution of soil organic matter content. Mean soil organic matter stock in the three soil depth of 0-10 cm, 10-20 cm and 20-30 cm were 7056 g m-3, 5427 g m-3 and 6135 g m-3 in Luanshya, and 5065 g m-3, 4784 g m-3 and 4733 g m-3 in Bushbuckridge, respectively. Principal components analysis revealed that the percentage of clay and silt correlated consistently with increased soil organic matter content, while bulk density and sand% impeded its accumulation. The baseline data and the empirical models can be used as counterfactual scenarios for monitoring the emerging forest carbon projects in southern Africa. The patterns and strength of relationships can be used as input or validation datasets for improving soil carbon models and for devising more efficient strategies for forest management.Item Spatio-temporal trends in vegetation dynamics across a savanna-dominated national park(2018) Ernst, Yolandi-MariThe Kruger National Park (KNP) is a savanna-dominated landscape upon which complex processes interact to result in dynamic patterns. Evidently, climate change is already impacting the KNP landscape. A major objective in the KNP’s Strategic Adaptive Management approach (SAM) is therefore to maintain ecological integrity through an informed, objective-driven decision-making process that involves Thresholds of Potential Concern. Continuous and informative monitoring at the landscape-scale is therefore essential, but challenging to do for such dynamic savanna environments, over large spatial extents. To address this concerning issue, I aimed to develop and evaluate a remote sensing-based methodological approach for the assessment and monitoring of landscape-scale spatio-temporal trends in vegetation greenness dynamics in relation to the influence of precipitation, fire and elephants across the Kruger National Park. My approach is novel in its combination of methods to monitor dynamic landscapes over large extents. Chapter 2 comprises of a comparison of different NDVI data products within the framework of mean-variance and spatial persistence analyses, thereby building the foundation upon which I based data analyses in the following chapters. Comparison of NDVI data derived from different satellite sensors revealed that the overall mean-variance trajectories and spatial persistence were relatively similar, but some large variations were evident due to differences in sensor characteristics. In Chapter 3, I exploited the fact that the relationship between phenology and vegetation greenness are confounded by the effect of other drivers. Through subjecting NDVI, phenology metrics and precipitation to spatial persistence analyses and then quantifying the converging and diverging change trends, I was able to show that precipitation is the primary driver of the general spatial greenness trends across the park but that persistent changes in vegetation greenness at finer spatial scales were more likely caused by other factors, such as fire and herbivory during the period of observation. Following this, I evaluated the role of fire and elephants as drivers of vegetation change trends with the use of pheno-season (defined by phenology metrics) vegetation greenness in correlation analyses. Here, the results seemed to suggest a combined association of elephants and fire frequency with vegetation greenness change trends, although the results showed much variability. Elephant bull and herd densities iii also seem to play different roles in their effects on vegetation greenness Through further development and adjustment of data input the proposed monitoring approach could be optimised for the broad-scale monitoring of vegetation pattern and processes within the SAM framework.Item An assessment of the physical drivers for farm dam distribution in Midlands KwaZulu Natal, using GIS and remote sensing(2018) Tsoka, JonathanThe interest in farm dams emanates mainly from their use for livestock watering, irrigation and fisheries enhancement on a sustainable basis. While management information on large dams in South Africa is largely available, it is lacking for farm dams which cumulatively store large volumes of water. As a result they are barely considered as part of the water resources of a river basin. Data acquisition methods for obtaining information about farm dams are costly, time consuming and labour intensive. This study was an attempt to map farm dams and establish the factors driving their spatial distribution pattern in the Midlands, KwaZulu-Natal, South Africa, using cost effective, time saving and less laborious GIS and remote sensing techniques. A classified April 2017 Landsat 8 satellite image was used to identify all water bodies in the Umgeni River basin U2 quaternary catchment (U2) while Google Earth was subsequently employed for differentiating farm dams from other water bodies. There were approximately 2000 water bodies that were identified by the classification. These included large national dams, pools in golf courses, ponds and disused mine dumps. A total of 864 farm dams in the U2 region quaternary catchment was observed. Six physical factors, namely slope, aspect, elevation, land use, soil type and geology were assessed to establish to what extent they influenced the siting of farm dams. The results indicated the importance of soils and land use as farm dams were mainly found in clusters in areas where agricultural farm land is also found since water is required for crop and livestock production. The influence of other factors such as slope, geology and elevation were observed in the spatial distribution maps. They all gave significant p-values in their univariate analysis. Of the six variables only aspect gave non-significant results while the rest were significant. A binary multivariate logistic regression was created for forecasting future farm dam sites and to establish which sites are poorly sited. The other four factors were fitted to the model except aspect and geology type which had no significant p-values in the model. The model had an Akaike information criterion (AIC) score of 293.42 and had the best combination of variables relative to other models. It was validated using 500 farm dam sites and it predicted 86% correctly.Item Reconstructing the history of urban development in the mining town of Virginia, Free State between 1940 and 2015(2017) Ajayi, Paul OluwanifemiThe nature of urban development experienced by mining towns across the world has been a subject of concern among urban planners because of its transitory nature. Most times mining towns develop gloriously into booming urban centres that create employment, generate wealth and satisfaction. All these fades into oblivion as soon as the mines get depleted. Mining towns often go through a number of urban processes which have been considered an expression of ‘infrastructural violence’ especially in the earlier stage of urban growth, and continually persists throughout the town’s life span. This research sought to reconstruct the history of urban development in the mining town of Virginia, Free State, and to quantify the manifestations of infrastructural violence throughout its timeline using GIS and remote sensing. Hence, land use and land cover maps were produced from aerial photographs, topographical maps and Landsat images through manual on-screen digitizing and classification using supervised support vector machine algorithms. Land use change detection analysis was conducted on the produced images using the cross classification and tabulation tool of QGIS 2.18.4 and the post classification tool of ENVI 5.3. Landscape metrics were employed to calculate the dimensions of growth and change experienced by all the land use classes during the timeline under study. Results obtained from this study confirmed the thoughts and findings of several theories vis a vis the nature of mining towns. Results reveal a rapid growth in the urban formal land use class up until 1995 with urban expansion and sprawl happening in the years between 1986 and 1995 with metrics of CA, NP and ED multiplying to twice their initial values ten years earlier. The urban informal land use class also experienced its subtle growth throughout the timeline of the study with its own urban expansion also happening between 1986 and 1995 with double increase in CA, NP and ED metric values. However, unlike the formal class that experienced decline after this period of urban expansion, the informal class continued to experience growth up until the end of the study period. Infrastructural violence was measured using the fractal dimension index (AWMPFD) of the landscape metrics for the formal and informal LU class. The results reveal continuous fragmentation throughout the period of study but with higher values in the years in which urban development started.Item Assessment of changing urban dynamics in Johannesburg city regions as consequence of re-mining of the tailings dumps using Geographical Information System and remote sensing(2017) Mahao, Tseliso JohnThere is a growing interest in the reclamation of the old gold mines’ tailings dumps in Johannesburg city region driven by the economic value of the remaining gold resource. This reclamation activity is accompanied by various rehabilitation methods to reduce issues such as acid mine drainage and wind pollution. The impact of land use and land cover change (LULCC) can have an enormous impact on land development and planning. Monitoring of LULCC is very important in the planning and decision making processes. Remote sensing (RS) as the source of basic data for monitoring change is very highly recommended as tool to monitor changes occurring in the Johannesburg City Region as a result of rehabilitation and reclamation of the gold mines’ tailings deposits. The main aim of the study is to understand how the reclamation and rehabilitation of the historical tailings dumps in Johannesburg are changing land use patterns of the city and its precincts. The objectives are to quantify the changes in land use and land cover as the results of mining rehabilitation using Landsat earth observation data over a period of 30 years at five years intervals and; to recommend on how the land cleared of waste dumps could be used for looking at the surrounding environment spatially. Various Geographical Information Systems techniques are applied here for change detection analysis and monitoring of potential changes in urban dynamics patterns. The results show rate of rehabilitation and reclamation to be slow, taking several years to see a meaningful change. There is some form of bias towards transforming the reclaimed land into industrial zones as opposed to other activities. The success of tailings dumps rehabilitation through revegetation process is highlighted.Item Raster to vector conversion in a local, exact and near optimal manner(1991) Carter, John AndrewRemote sensing can be used to produce maps of land-cover, but to be of use to the GIS community these maps must first be vectorized in an intelligent manner. Existing algorithms suffer from the defects of being slow, memory intensive and producing vast quantities of very short vectors. Furthermore if these vectors are thinned via standard algorithms, errors are introduced. The process of vectorizing raster maps is subject to major ambiguities. Thus an infinite family of vector maps ccrresponds to each raster map. This dissertation presents an algorithm for converting raster maps in a rapid manner to accurate vector maps with a minimum of vectors. The algorithm converts raster maps to vector maps using local information only, (a two by two neighbourhood). the method is "exact" in the sense that rasterizing the resulting polygons would produce exactly the same raster map, pixel for pixel. The method is "near optimal" in that it produces, in a local sense, that "exacb" vector map having the least number of vectors. The program is built around a home-grown object oriented Programming System (OOPS) for the C programming language. The main features of the OOPS system, (called OopCdaisy), are virtual and static methods, polymorphism, generalized containers, container indices and thorough error checking, The following general purpose objects are implemented with a large number of sophistiated methods :- Stacks, LIFO lists, scannable containers with indices, trees and 2D objects like points, lines etc.Item The design and application of wayfinding in physically and virtual immersive environnments(2016) Haarhoff, ChanéThe focus of this research paper is on the relationship between wayfinding and interactive, immersive environments. Wayfinding systems inherently exist to assist people on a journey to a desired destination. Interactive immersive environments have introduced a new paradigm, in which users are becoming active participants within both physical and virtual environments. The merging of wayfinding with immersive environments could thus present potential to propel wayfinding to a new level. The effectiveness and applications of wayfinding systems within three immersive environments is therefore determined by looking at the design of informational wayfinding, the ease of navigability within immersive environments and the adaptation and use of wayfinding techniques within four case studies. The three immersive environments chosen to be examined within this paper are: physically immersive environments, digital physically immersive environments and virtual immersive environments. These environments are represented by four case studies which include Menlyn Park Shopping Centre's newly updated wayfinding system, Find Your Way Essex, a virtual wayfinding mobile application, StreetMuseum: Museum of London a purely digital mobile application and Honeywell - Virtual Museum Tour, a virtual wayfinding experience that relies fully on the GoogleCardboard device.