Faculty of Science (ETDs)

Permanent URI for this communityhttps://hdl.handle.net/10539/37933

Browse

Subject: search.filters.subject.Remote sensing

Search Results

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    The application of machine learning methods to satellite data for the management of invasive water hyacinth
    (University of the Witwatersrand, Johannesburg, 2023-06) Singh, Geethe; Reynolds, Chevonne; Byrne, Marcus; Rosman, Benjamin
    Biological invasions are responsible for some of the most devastating impacts on the world’s ecosystems, with freshwater ecosystems among the worst affected. Invasions threaten not only freshwater biodiversity, but also the provision of ecosystem services. Tackling the impact of invasive aquatic alien plant (IAAP) species in freshwater systems is an ongoing challenge. In the case of water hyacinth (Pontederia crassipes, previously Eichhorniae crassipes), the worst IAAP presents a long-standing management challenge that requires detailed and frequently updated information on its distribution, the context that influences its occurrence, and a systematic way to identify effective biocontrol release events. This is particularly urgent in South Africa, where freshwater resources are scarce and under increasing pressure. This research employs recent advances in machine learning (ML), remote sensing, and cloud computing to improve the chances of successful water hyacinth management. This is achieved by (i) mapping the occurrence of water hyacinth across a large extent, (ii) identifying the factors that are likely driving the occurrence of the weed at multiple scales, from a waterbody level to a national extent, and (iii) finally identifying periods for effective biocontrol release. Consequently, the capacity of these tools demonstrates their potential to facilitate wide-scale, consistent, automated, pre-emptive, data-driven, and evidence-based decision making for managing water hyacinth. The first chapter is a general introduction to the research problem and research questions. In the second chapter, the research combines a novel image thresholding method for water detection with an unsupervised method for aquatic vegetation detection and a supervised random forest model in a hierarchical way to localise and discriminate water hyacinth from other IAAP’s at a national extent. The value of this work is marked by the comparison of the user (87%) and producer accuracy (93%) of the introduced method with previous small-scale studies. As part of this chapter, the results also show the sensor-agnostic and temporally consistent capability of the introduced hierarchical approach to monitor water and aquatic vegetation using Sentinel-2 and Landsat-8 for long periods (from 2013 - present). Lastly, this work demonstrates encouraging results when using a Deep Neural Network (DNN) to directly detect aquatic vegetation and circumvents the need for accurate water extent data. The two chapters that follow (Chapter 3 and 4 described below) introduce an application each that build off the South African water hyacinth distribution and aquatic vegetation time series (derived in Chapter 2). The third chapter uses a species distribution model (SDM) that links climatic, socio-economic, ecological, and hydrological conditions to the presence/absence of water hyacinth throughout South Africa at a waterbody level. Thereafter, explainable AI (xAI) methods (specifically SHapley Additive exPlanations or SHAP) are applied to better understand the factors that are likely driving the occurrence of water hyacinth. The analyses of 82 variables (of 140 considered) show that the most common group of drivers primarily associated with the occurrence of water hyacinth in South Africa are climatically related (41.4%). This is followed by natural land cover categories (32.9%) and socio-economic variables (10.7%), which include artificial land-cover. The two least influential groups are hydrological variables (10.4%) including water seasonality, runoff, and flood risk, and ecological variables (4.7%) including riparian soil conditions and interspecies competition. These results suggest the importance of considering landscape context when prioritising the type (mechanical, biological, chemical, or integrated) of weed management to use. To enable the prioritisation of suitable biocontrol release dates, the fourth chapter forecasts 70-day open water proportion post-release as a reward for effective biocontrol. This enabled the simulation of the effect of synthetic biocontrol release events under a multiarmed bandit framework for the identification of two effective biocontrol release periods (late spring/early summer (mid-November) and late summer (late February to mid-March)). The latter release period was estimated to result in an 8-27% higher average open-water cover post-release compared to actual biocontrol release events during the study period (May 2018 - July 2020). Hartbeespoort Dam, South Africa, is considered as a case study for improving the pre-existing management strategy used during the biocontrol of water hyacinth. The novel frameworks introduced in this work go a long way in advancing IAAP species management in the age of both ongoing drives towards the adoption of artificial intelligence and sustainability for a better future. It goes beyond (i) traditional small-scale and infrequent mapping, (ii) standard SDMs, to now include the benefits of spatially explicit model explainability, and (iii) introduces a semi-automated and widely applicable method to explore potential biocontrol release events. The direct benefit of this work, or indirect benefits from derivative work outweighs both the low production costs or equivalent field and lab work. To improve the adoption of modern ML and Earth Observation (EO) tools for invasive species management, some of the developed tools are publicly accessible. In addition, a human-AI symbiosis that combines strengths and compensates for weaknesses is strongly recommended. For each application, directions are provided for future research based on the drawbacks and limitations of the introduced systems. These future efforts will likely increase the adoption of EO-derived products by water managers and improve the reliability of these products.
  • Thumbnail Image
    Item
    Evaluating the spatiotemporal changes of urban wetlands in Klip River wetland, South Africa
    (University of the Witwatersrand, Johannesburg, 2023-09) Nxumalo, Nolwazi; Knight, Jasper; Adam, Elhadi
    This study assesses the impacts of land use / land cover (LULC) change in an urban wetland over the past 30 years utilizing machine learning and satellite-based techniques. This study looked at LULC distributions in the Klip River wetland in Gauteng, South Africa. The aims and methods used in this study were: (1) to conduct a comprehensive analysis to map and evaluate the effects of LULC changes in the Klip River wetland spanning from 1990 to 2020, employing Landsat datasets at intervals of 10 years, and to quantify both spatial and temporal alterations in urban wetland area. (2) To predict the change in urban wetland area due to specific LULC changes for 2030 and 2040 using the MOLUSCE plugin in QGIS. This model is based on observed LULC including bare soil, built-up area, water, wetland, and other vegetation in the quaternary catchment C22A of the Klip River wetland, using multispectral satellite images obtained from Landsat 5 (1990), Landsat 7 (2000 and 2010) and Landsat 8 OLI (2020). (3) For the results of this study, thematic maps were classified using the Random Forest algorithm in Google Earth Engine. Change maps were produced using QGIS to determine the spatiotemporal changes within the study area. To simulate future LULC for 2030 and 2040, the MOLUSCE plugin in QGIS v2.8.18 was used. The overall accuracies achieved for the classified maps for 1990, 2000, 2010, and 2020 were 85.19%, 89.80%, 84.09%, and 88.12%, respectively. The results indicated a significant decrease in wetland area from 14.82% (6949.39 ha) in 1990 to 5.54% (2759.2 ha) in 2020. The major causes of these changes were the build-up area, which increased from 0.17% (80.36 ha) in 1990 to 45.96% (22 901 ha) in 2020—the projected years 2030 and 2040 achieved a kappa value of 0.71 and 0.61, respectively. The results indicate that built-up areas continue to increase annually, while wetlands will decrease. These LULC transformations posed a severe threat to the wetlands. Hence, proper management of wetland ecosystems is required, and if not implemented soon, the wetland ecosystem will be lost.
  • Thumbnail Image
    Item
    Monitoring and evaluating urban land use land cover change using machine learning classification techniques: a case study of Polokwane municipality
    (University of the Witwatersrand, Johannesburg, 2023) Funani, Tshivhase; Mhangara, Paida
    Remote sensing is one of the tools which is very important to produce Land use and land cover maps through the process of image classification. Image classification requires quality multispectral imagery and secondary data, a precise classification technique, and user experience skill. Remote sensing and GIS were used to identify and map land-use/land-cover in the study region. Big Data issues arise when classifying a huge number of satellite images and features, which is a very intensive process. This study primarily uses GEE to evaluate the two classifiers, Support Vector Machine, and gradient boosting, using multi-temporal Landsat-8 images, and to assess their performance while accounting for the impact of data dimension, sample size, and quality. Land use/Landcover (LULC) classification, accuracy assessment, and landscape metrics comprise this study. Gradient Tree Boost and SVM algorithms were used in 2008, 2013, 2017, and 2022. Google Earth Engine was used for supervised classification. The results of change detection showed that urbanization has occurred and most of the encroachments were on agricultural land. In this study, XG boost, and support vector machine (SVM)) were used and compared for image classification to oversight spatio-temporal land use changes in Polokwane Municipality. The Google Earth Engine has been utilized to pre-process the Landsat imagery, and then upload it for classification. Each classification method was evaluated using field observations and high-resolution Google Earth imagery. LULC changes were assessed, utilizing Geographic Information System (GIS) techniques, as well as the dynamics of change in LULCC were analysed using landscape matrix analysis over the last 15 years in four different periods: 2008–2013, 2018 and 2022. The results showed that XGBoost performed better than SVM both in overall accuracies and Kappa statistics as well as F-scores and the ratio of Z-score. The overall accuracy of gradient boosting in 2008 was 0.82, while SVM showed results of 0.82 overall accuracy and kappa statistics of 0.69. The average F-score for SVM in 2008 was from 0.58- 1.00, in 2013 an average of 0.86-0.97, and in 2022 it was 0.76. Z values were not statistically significant as all values were below the z score of 1.96. The ratios for the two classifiers were also taken to know which classifier performs the best. The results showed 212:212 which indicates that during 2008 SVM and XG boost performed the same way as they classified the same number of cases. During 2013 the ratio was 345:312 which shows that XGBoost performed better than SVM. The results of 2017 show 374:316 which shows that XGBoost performed better than SVM. Lastly, in 2022 the ratio was 298:277 which shows that XGBoost performed better than SVM. Overall zscores result show that XGBoost performs better than SVM. Overall, this study offers useful insight into LULC changes that might aid shareholders and decision makers in making informed decisions about controlling land use changes and urban growth
  • Thumbnail Image
    Item
    Potential Source for the Alluvial Gold Deposit in the Kapoeta Area, Eastern Equatoria State, South Sudan
    (University of the Witwatersrand, Johannesburg, 2023-10) Bali, Francis Khamis Alex; Woldai, Tsehaie; Kinnaird, Judith A.
    This study concentrates on the integrated approach of applying geochemical, and remote sensing data combined with field investigation to understand the source of the alluvial gold deposit in the Kapoeta area. The use of an integrated technique approach proves effective in mapping and outlining areas of potential gold mineralisation in Kapoeta. Kapoeta has been known for its endowment of gold deposits for quite some time. Landsat-8 (OLI) and SRTM-DEM satellite images were employed during this study to map lithological units and highlight areas of hydrothermal alteration and structural trends in Kapoeta to understand the potential source for the widely distributed alluvial gold in the area. The several image-processing techniques employed were effective in mapping the lithological units and outlining the major structural trends in the area. Mapping the hydrothermal alteration zones was challenging because of the overburden that masks most parts of the Neoproterozoic juvenile metavolcano-sedimentary sequences. Secondly, the vegetation cover is another factor affecting the mapping of the lithological units around the Didinga Hills and Dongotona mountains. Structural mapping identified four main trends N-S, NNW-SSE, NW-SE, and NE-SW. The image interpretation, together with geochemical occurrence and anomaly map showing Cu, Co, Ni, and Cr, shows that N-S and NNW-SSE are the potential controls on mineralisation in the area. The whole-rock geochemical data plotted on the TAS discrimination diagrams shows that the rocks analysed are dominantly tholeiitic with minor calc-alkaline units. Trace elements normalised to the primitive mantle show low concentrations of Zr, Hf, Nb and Ta. The REE normalised chondrite values indicates general enrichment of LREE elements relative to HREE with a negative Eu anomaly. On the tectonic discrimination diagrams, the results indicate that the rocks of Greater Kapoeta belong in a marine environment related to a spreading centre, oceanic island, oceanic ridges, and floor. The distribution of Cu, Co, Cr and Ni and their high concentration suggest that these elements are the potential pathfinders for gold in the area. The study concludes the following: Based on the remote sensing, field observations and geochemical data indicates that the gold mineralisation of Kapoeta is orogenic in nature with a strong element of epigenetic characteristics, however, a syngenetic origin cannot be ruled out. The host lithologies may vary from quartz veins to schists, amphibolites, quartzo-feldspathic gneisses, and ultramafics and the marbles related to the N-S, NNS-W-SSE and NW-SE trending shear zones that limit the greenstone belt.