- Title
- Bush clumps as indicators of thresholds of change in Arid Thicket mosaic piospheres
- Creator
- Schmidt, Anton George
- Subject
- Biodiversity conservation Ecological processes
- Date Issued
- 2017
- Date
- 2017
- Type
- Thesis
- Type
- Doctoral
- Type
- PhD
- Identifier
- http://hdl.handle.net/10948/20811
- Identifier
- vital:29403
- Description
- In the literature, Arid Thicket transformation by domestic meso-herbivores is purported to follow a state-and-transition model, whereby the ecosystem is initially resilient to herbivory until some threshold is crossed, where after there is a rapid shift to a new alternative stable state. My study represents a first attempt to verify this hypothesis by searching for evidence of stable ecosystem states separated by structural, functional and degradation thresholds in Pruim-Spekboomveld, a variation of Arid Thicket found in the southern Cape of South Africa. In this variation, thicket bush clumps ranging in size from approximately 1 m2 to 100 m2 are interspersed in a matrix of Succulent Karoo vegetation. I correlate structural and functional attributes of the thicket bush clumps and the soil environment beneath them (indicators of ecosystem function) to distance from artificial watering points in four piosphere treatments that have had a different rangeland management history. I mostly fit linear, exponential and sigmoid regression models to my data and use Akaike’s Information Criterion to select the best fitting model. When analysing data on changes in the variation of vegetation production with distance from water using the Normalized Difference Vegetation Index, I additionally fit Generalized Additive Models to the data to take into account more complex predicted patterns of change. Furthermore, I compare the end regions of my data in the piosphere treatments to each other and to similar data collected in a transformed and untransformed reference site. I use both the pattern and extent of transformation within and between the treatments and the reference sites, to verify the existence of stable states and structural, functional and degradation thresholds. I also subject my entire dataset to multivariate analysis using ordination and permutation procedures, in an attempt to corroborate my results from the univariate analysis and to assess the response of bush clump community composition to herbivory. My results indicate that at the scale of a bush clump, a threshold reduction in species richness, functional type diversity, succulent shrub redundancy and cover, palatable shrub cover and bush clump leaf mass below 1.5 m occurs in regions intensively used by domestic meso-herbivores. Furthermore, at the landscape scale there is a threshold reduction in canopy tree cover and bush clump density. In addition to these structural threshold changes, results from a Landscape Function Analysis indicate that soil nutrient cycling, soil water infiltration and soil stability thresholds are crossed at the landscape scale. The soil nutrient cycling threshold is affirmed by soil chemical analyses which indicate that the most rested treatment in the study area has low carbon and nitrogen levels. Furthermore, the keystone species, Portulacaria afra, is unable to re-establish itself in any of the treatments. High pH and potassium levels in all the treatments, relative to the untransformed reference site, indicate that Portulacaria afra (which prefers moderate soil conditions) is unlikely to re-establish in the study area. This finding suggests that all my treatments have crossed a degradation threshold and therefore represent an alternative stable state to the untransformed reference site. My multivariate analysis supports this finding and further indicates that the sampling zones within 25 m of the watering points in the most utilized treatments have crossed a second degradation threshold. This finding is supported by my results that indicate that structural and function thresholds (related to the size, density and ability of bush clumps to capture and retain soil resources) have been crossed. Furthermore, excessively high concentrations of base cations in these sampling zones, relative to the untransformed reference site, affirm the crossing of a degradation threshold. My results on the changes in the variation of vegetation production with distance from water using the Normalized Difference Vegetation Index, support the growing body of evidence in the literature that suggests that increases in the variance of ecosystem processes and services are important indicators of impending ecological thresholds. In general, my findings support the prediction in the literature that domestic meso-herbivore induced transformation of Arid Mosaic Thicket will follow a state-andtransition model.
- Format
- vi, 144 leaves
- Format
- Publisher
- Nelson Mandela Metropolitan University
- Publisher
- Faculty of Science
- Language
- English
- Rights
- Nelson Mandela Metropolitan University
- Hits: 763
- Visitors: 1006
- Downloads: 390
Collections
NMU School of Natural Resource Management
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | SOURCE1 | Adobe Acrobat PDF | 8 MB | Adobe Acrobat PDF | View Details Download |