Doctoral Degrees (Geography)

Permanent URI for this collection

http://hdl.handle.net/11660/426

Browse

Browsing Doctoral Degrees (Geography) by Author "Adelabu, Samuel"

Now showing 1 - 1 of 1
  • Thumbnail Image
    ItemOpen Access
    Grass nutrients estimation as an indicator of rangeland quality using satellite remote sensing
    (University of the Free State, 2022) Mashiane, Katlego Kgabo; Adelabu, Samuel; Ramoelo, Abel
    South Africa's grasslands are known for their rich biodiversity, which makes for rangelands with expansive landscapes for grazing. Numerous studies have been conducted on how species variety and composition affect the sustainability and productivity of grasslands. The provision of ecosystem functions and services, mainly grazing for both domestic livestock and wild animals, is threatened by rapid biodiversity loss. Hence crucial to develop monitoring systems for biodiversity, especially considering the anticipated impacts of global environmental changes. Nutritionally sufficient grass swards strongly influence the distribution and abundance of grazing animals. Although estimates of crucial ecological and biodiversity indicators can be obtained from in situ and remote sensing data, remote sensing offers timely and cost-effective data that can be used for species monitoring. This study sought to measure essential biodiversity variables in GGHNP related to plant productivity and nutrient availability using in situ and remote sensing approaches. In addition, by estimating grass species nitrogen using remote sensing, the research intended to improve carrying capacity models and stocking rate in the park, thus ensuring efficient rangeland management. Assessments of species diversity and richness are pivotal for devising effective conservation strategies. These biodiversity metrics are good indicators of rangeland quality, health, and ecological response to disturbances. Data collected using in situ methods showed that species richness are virtually similar across the park under different richness of disturbance regime. However, species richness were higher at landscape richness than at the site richness. Albeit criticised for their use in conservation, species diversity metrics could be useful for measurements of rangeland quality. Due to its intolerance to harsh environmental circumstances, mountainous grassland vegetation is probably the most susceptible to environmental changes on a worldwide scale. Understanding the factors that influence species distribution in alpine grasslands will be crucial for identifying biodiversity and preserving it. This evident in that an increase in topo-edaphic variables negatively affects species richness, while slope and elevation showed an improvement of species richness. The influence of topography and other accompanying factors on species diversity is highlighted, emphasizing how topography affects species dispersion in mountainous grassland communities. Understanding the value of environmental conditions are to the geographical distribution of biodiversity has been the focus of the most active ecological research. The selection of appropriate modelling algorithms could be beneficial for gaining insights into biodiversity-environment relations. Non-parametric and parametric modelling frameworks were used to assess these relationships. Topographically controlled edaphic variables continued to be the most significant drivers of species richness and diversity in grassland plant communities in the park, despite higher prediction accuracies being attained using parametric models. Remote sensing permits rapid and inexpensive recording and assessment of vegetation over short to long-term periods at a local and global scales. This study sought to predict and model species richness and diversity in GGHNP. Near-infrared (NIR) was the most selected spectral interval for predicting species diversity, further ascertaining the efficiency of NIR in vegetation mapping. Grass species N (grass N) estimation is valueable for rangeland management because it determines their forage quality which has nutritional implications for grazing animals. I used remote sensing data to predict grass N in the park using Sentinel 2 Multi-Spectral Instrument (S2 MSI). The results showed red edge bands as the optimal bands for estimating grass N, which makes S2 MSI superior for modelling grass N throughout grass phenology and among seasons because of its multiple red edge bands. Grassland monitoring is imperative for both assessing global change impacts and the security of sustainable development goals. Grasslands serve as rangelands that supply forage for domestic and wild animals. Monitoring of forage quality and quantity is crucial for evaluating carrying capacity models and thus ensuring effective rangeland management. Plant species richness and species diversity are key indicators for plant primary productivity in rangelands and many other ecosystem health parameters. Due to the immensity of rangeland landscapes, remote sensing could be the effective technology for determining and keeping track of ecological parameters in grasslands.