- Title
- Relating vegetation distribution to cycles of erosion and deposition in the Kromme River wetlands
- Creator
- Jarvis, Samuel Cameron
- ThesisAdvisor
- Ellery, William
- ThesisAdvisor
- Meiklejohn, Ian
- Subject
- Biogeomorphology South Africa Kromme Estuary (Eastern Cape)
- Subject
- Earth observation
- Subject
- Remote sensing
- Subject
- Niche construction
- Subject
- Wetland ecology
- Subject
- Geomorphology
- Subject
- Ecological succession
- Subject
- Optical radar
- Subject
- Prionium serratum
- Date
- 2023-10-13
- Type
- Academic theses
- Type
- Master's theses
- Type
- text
- Identifier
- http://hdl.handle.net/10962/424582
- Identifier
- vital:72166
- Description
- The role of geomorphic disturbance has been increasingly recognized as fundamental in the creation and functioning of wetlands. This is true of the Kromme River wetland which has been formed through repeated cycles of erosion and deposition. However, the response – and influence – of wetland plants to these sorts of disturbance has not been investigated. This study sought to fill this knowledge gap by classifying vegetation communities over a range of hydrological and geomorphic disturbance regimes that have happened over the last few decades, and relating those vegetation communities to environmental factors. The study identified seven vegetation communities based on their species composition and abundance, which were related to geomorphic disturbance events. A conceptual model that accounts for vegetation distribution in the Kromme wetland was developed. Soil saturation was the most important factor explaining vegetation community distribution, which, in turn, is influenced by cycles of erosion and deposition. Following an erosional event on the valley floor, Prionium serratum dominated wetland is converted to a number of other vegetation communities. On the floodplain surface adjacent to the eroded gully, the Prionium serratum dominated wetland is transformed over time to Cynodon dactylon and Sporobolus fimbriatus communities. Prionium serratum clumps immediately adjacent to the recently incised gullies are able to persist, having sufficient access to water. Within the newly formed gullies, Juncus lomatophyllus colonizes the gully beds flooded to a shallow depth, Miscanthus capensis colonizes the gully bars and Setaria incrassata colonizes the exposed gully banks. Localised depositional features close to the thalweg in the gully are colonized by Prionium serratum seedlings and vegetative propagules. These plants represent the regenerating phase of Prionium serratum wetland, which also colonizes depositional floodouts downstream of the newly-formed gully. The Stenotaphrum secundatum community dominates drier, more elevated areas of the floodout. Over time, as the gully fills, Prionium serratum expands beyond the gully onto the valley floor, to replace the floodplain communities Cynodon dactylon and Sporobolus fimbriatus. Over time, Prionium serratum is thought to colonize the valley floor as the gully fills, stabilising it and promoting diffuse flow. Many restoration efforts in damaged palmiet wetlands have been focused on the preservation of intact palmiet communities upstream of erosional headcuts, with limited understanding of vegetation dynamics associated with the cut-and-fill cycles that naturally occur in these wetlands. Understanding the regeneration of Prionium serratum following erosional events is thus important for wetland restoration, as it should focus more attention on promoting palmiet restoration on depositional floodouts downstream of eroded gullies. A secondary aim of this study was to explore the possibility of mapping palmiet communities in Kromme River wetland using remote sensing techniques. Using a combination of ground-truthed data from this and previous studies in the Kromme River wetland, together with raster layers derived from a LiDAR survey, an overlay analysis was developed to effectively map the distribution of the Prionium serratum dominated community. The overlay was created using a machine learning library in RStudios known as Rpart. The results found that the model were 91% effective in classifying the distribution of the Prionium serratum community. A secondary finding was that the inclusion of a Relative Elevation Model in the overlay analysis allowed for the identification of Prionium serratum communities vulnerable to degradation following previous geomorphic disturbance events and those Prionium serratum communities that are likely to persist following a geomorphic disturbance event.
- Description
- Thesis (MSc) -- Faculty of Science, Geography, 2023
- Format
- computer, online resource, application/pdf, 1 online resource (125 pages), pdf
- Publisher
- Rhodes University, Faculty of Science, Geography
- Language
- English
- Rights
- Jarvis, Samuel Cameron
- Rights
- Use of this resource is governed by the terms and conditions of the Creative Commons "Attribution-NonCommercial-ShareAlike" License (http://creativecommons.org/licenses/by-nc-sa/2.0/)
- Hits: 623
- Visitors: 631
- Downloads: 17
Collections
RU Department of Geography
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details | SOURCE1 | JARVIS-MSC-TR23-150.pdf | 3 MB | Adobe Acrobat PDF | View Details |