55 Table 3.16: A correlation matrix, showing the correlation (expressed as ΦK, between 0.0 . and 1) between potential driving factors with NWP bush spread for the time frame 1993-2018. 69 Table 3.30: A correlation matrix, showing the correlation (expressed as ΦK, between 0.0 . and 1) between potential driving factors with the spread of shrubs in the Pilanesberg area for the time frame 1993-2018.
INTRODUCTION
Background and problem statement
The analyzes performed by Stafford et al. 2016), indicates that forest encroachment is most severe in the northern and western parts of the North West Province. Previous studies on the drivers of forest encroachment in the North West Province mainly focused only on fire suppression and herbivory (O'Connor et al., 2014).
Aims and objectives of the study
Hypothesis of the study
LITERATURE REVIEW
Introduction
Bush encroachment establishment models
- Walter’s two-layer model
- Moir’s one-layer model
- State-and-transition model
- Equilibrium rangeland model
- Non-equilibrium rangeland model
State-and-transition models are used to organize and communicate information regarding ecosystem change, particularly as a result of rangeland management actions (Bestelmeyer et al., 2017). Therefore, compared to dry savannas, mesic savannas are suggested to be controlled by non-equilibrium dynamics that affect the woody-grass ratio (Higgins et al., 2000).
Bush encroachment in the North West Province
The non-equilibrium model assumes that savannas are unstable mixtures of woody and herbaceous (mainly grass) vegetation that persist as a result of disturbance (Higgins et al., 2000). This unstable system can either tend towards open grassland or savanna forest (a higher ratio of trees than grass) depending on rainfall (Higgins et al., 2000).
The management and control of bush encroachment
- Biological clearing
- Manual clearing
- Mechanical clearing
- Chemical clearing
Since manual control is labor intensive, it creates employment opportunities, especially in communal areas (Nghikembua et al., 2020). Arboricides are applied to the leaves, trunks of woody plants or to the soil around plants (Turpie et al., 2019).
Driving factors of bush encroachment
- Climate
- Precipitation
- Temperature
- Atmospheric carbon dioxide (CO 2 )
- Topography
- Herbivory
- Fire
- The combined factor of herbivory and fire
- Geology
- Soil
- Soil texture
- Soil depth
- Soil structure
A downward slope would reduce the rate of fire spread and intensity (Trollope et al., 2002). A study by Buitenwerf et al. 2012), showed that BE occurs rapidly in granite-based (sandy) soils, but is mostly absent in basalt-based (clayey) soils.
GEOGRAPHIC INFORMATION SYSTEM (GIS)
Introduction
Materials and methods
- Summary
- Site description
- Obtaining and pre-processing GIS data
- Bush spread
- Potential driving factors of bush encroachment
- GIS analysis
- Bush spread
- Determining drivers of bush encroachment
- Significant areas
- Analysis of variance (ANOVA)
The same potential BE factors (from Table 3.2) were also used to analyze the significant areas. Ae Free-draining, red, eutrophic, apedal soils comprise >40% of the land (yellow soils comprise <10%).
Results and discussion
- Provincial scale
- Bush spread
- Determining drivers of bush encroachment
- Significant areas
- Bush spread
- Determining drivers of bush encroachment
- Precipitation data
- Analysis of variance (ANOVA)
BT showed slight correlations with all potential drivers, except MAP and geology, which showed insignificant correlations (< 0.20). BL showed slight correlations with all potential drivers, except for land types and land cover, which showed insignificant correlations (< 0.20). BT showed slight correlations with all potential drivers, except vegetation units, which showed insignificant correlations (< 0.20).
BT showed weak correlations with all the other potential driving factors, except MAP, which showed insignificant correlations (< 0.20). BT showed small correlations with all the potential driving factors, except for MAP, MAT and geology, which showed insignificant correlations (< 0.20) (Table 3.26). BL showed weak correlations with all the potential driving factors, except for MAT and vegetation units, which showed insignificant correlations (< 0.20) (Table 3.33).
Conclusions
Since the low MAP is likely to cause less grass to grow, and overgrazing has reduced the grass, which led to severe BE and BT in the Taung area. The low MAP, together with fire and herbivores is likely to reduce the bush in the Pilanesberg and Rustenburg areas. Correlation matrices showed that EU had significant correlations with soil types and vegetation units in the Taung area for the second and third time and.
Correlation matrices showed that BT had significant correlations with soil types in the Taung area for the second time. As the low MAP is likely to have caused less grass growth as well as overgrazing which led to severe BE and BT in the Taung area and the low MAP together with fire and herbivores may have reduced the shrubs in Pilanesberg and Rustenburg areas. Therefore, at the provincial scale MAP can be considered as the main driving factor of BE and BT in NWP from 1993 to 2018.
INVESTIGATING THE EFFECT OF SOIL PHYSICAL PROPERTIES ON
Introduction
Materials and methods
- Study area
- Transect layout
- Vegetation survey
- Soil analysis
- Determining the relationship between the woody species and soil variables
Pilanesberg is found on the Pilanesberg Mountain Bushveld, which is part of the Central Bushveld bioregion (Mucina and Rutherford, 2006). Legkraal is found on the Dwaalboom Thornveld, which is part of the Central Bushveld bioregion (Mucina and Rutherford, 2006). This study area is found on the Springbokvlakte Thornveld, which is part of the Central Bushveld bioregion (Mucina and Rutherford, 2006).
The transects at the Pilanesberg site were laid out between Tshwene Drive and Tilodi Drive of the Pilanesberg National Park (Figure 4.8). The transects at the Legkraal site were laid out in the middle of the Legkraal site area (Figure 4.9). The transects of the Kgomo-Kgomo site were also laid out in the middle of the site area (Figure 4.10).
Results and discussion
- Vegetation survey
- Species composition and abundance
- Vegetation structure at the three study sites
- Single or multi-stem of woody species
- Soil analysis
- Soil classification
- Soil texture and soil particle size distribution
- pH and electrical conductivity (EC)
- Hydraulic soil parameters
The bottoms of the three transects of the Kgomo-Kgomo study site had orthic topsoil horizons (300 mm deep) above the subsurface horizons of the red apes, which were 1 200 mm deep, except for Transect 3 where the red apes horizon extended down to only 600 mm, above a neocutanic subsurface horizon (1 500 mm deep). However, there was no clear relationship between soil pH and EC and the TSA of all woody species from the three study sites. D. However, there was no clear relationship between groundwater retention and TSA of all woody species from the three study sites. study locations.
From the results, there was no clear relationship between soil moisture retention and TSA of all wood species of the three study areas.
Relationships between woody species and soil variables
A large part of the water retained at field capacity (0.33 bar) is available for uptake by plants through roots (Brady & Weil, 2017). Since there was no clear relationship between groundwater retention and TSA of all woody species or dominant species affecting the three study sites, this indicates that soil moisture retention did not affect BE at the three study sites. The soil of all three transects of the Pilanesberg and Legkraal study sites contained ideal Pbs for plant growth (USDA, 2018), but the soil of the Kgomo-Kgomo study site contained Pb that affected or limited root growth.
Pb ol’aanaan iddoo qorannoo Kgomo-Kgomo guddina hundee daangeffameef sababa ta’e, kunis gosoota weerartootaa irratti dhiibbaa uume (Turpie et al., 2019), kan akka G. RDA ordination indicates that a positive correlation exists between the three transects of baay’ina biqiltoota mukaa dhuunfaa, Pb fi giddugaleessa pH iddoo qorannichaa (Transect 1, Transect 2 fi Transect 3). Akkasumas walitti dhufeenyi gaariin TSA tiraanseektoota hunda iddoo qorannoo Legkraal fi Pilanesberg Transect 2 fi porosity gidduutti ture.
Conclusion
No clear correlation was found between the particle size distribution of clay (%) and TSA for the woody species found at the three study sites, however D. There was no clear correlation between soil water retention and TSA for all woody species or dominant invasive species of the three study sites, indicating that soil moisture did not affect BE at the three study sites. Soils in all three transects of the Pilanesberg and Legkraal study sites had ideal Pbs for plant growth.
The Kgomo-Kgomo study site had the highest Pb compared to the Pilanesberg and Legkraal study sites. There was no clear correlation between Pb and TSA for all the woody species of the three study sites. The RDA ordination indicates that a positive correlation existed between TSA for all tree species of the three transects of the Kgomo-Kgomo study site (Kgomo-Kgomo Transect 1, Kgomo-Kgomo Transect 2 and Kgomo-Kgomo Transect 3) with the number of individual woody plants, bulk density and average pH.
CONCLUSIONS AND RECOMMENDATIONS
Conclusions
This led to the spread of woody plant seedlings and the encroachment of shrubs, particularly in the Taung area of NWP. Soil types and properties did not have a significant impact on all woody species identified at each study site, but rather on specific encroachment species causing BE in the NWP. The results showed that the higher clay content in the Kgomo-Kgomo study area had a definite relationship with D.
There was no clear relationship between the soil water retention and the TSA of all the woody species or specific woody species of the three study sites, indicating that soil moisture retention did not affect BE at the three study sites. The RDA ordination also further substantiated the correlation between the total woody species found and Pb at the Kgomo-Kgomo study site. As there was almost no grass cover within the three transects of the Kgomo-Kgomo study site, trampling and overgrazing may have contributed to the high mass density (Pb) values, resulting in limited root growth.
Recommendations
- Recommendations for implementations
- Recommendations for future research
166 Appendix D: A correlation matrix, indicating the correlation (expressed as ΦK, between 0.0 and 1) between the potential drivers of BE with bush. 167 Appendix E: A correlation matrix, which indicates the correlation (expressed as ΦK, between 0.0 and 1) between the potential drivers of BE with bush. 167 Appendix F: A correlation matrix, which indicates the correlation (expressed as ΦK, between 0.0 and 1) between the potential drivers of BE with bush.
172 Appendix P: Correlation matrix showing the correlation (expressed as ΦK, between 0.0 . and 1) between potential BE factors with shrubs. 173 Appendix Q: Correlation matrix showing the correlation (expressed as ΦK, between 0.0 . and 1) between potential BE factors with shrubs. Appendix R: Correlation matrix indicating the correlation (expressed as ΦK, between 0.0 and 1) between potential BE factors with shrubs.
Appendix KK: Correlation matrix indicating the correlation (expressed as ΦK, between 0.0 and 1) between possible BE factors with shrubs. Appendix CCC: Correlation matrix indicating the correlation (expressed as ΦK, between 0.0 and 1) between the possible factors BE and bush.
The soil pH and EC from the three study sites
