Abstract: Background and Objective: Assessment of vegetation in catchment areas need to be assessed and understood in terms of plant diversity, ecological processes and functions that support appropriate ecosystem goods and services. The aim of this study was to assess plant species composition and diversity within the Tsitsa river catchment area in the Eastern Cape province, South Africa. Materials and Methods: Nineteen square plots measuring 5×5 m were established in Tsitsa river catchment area. Within each plot, environmental data and species present were recorded including Braun-Blanquet cover-abundance values for all species present in the plot. Vegetation and environmental data were analyzed using palaeontological statistics (PAST) version 3.06. Results: In total of 78 plant species were recorded belonging to 24 families and 57 genera. Among the documented species, 11.5% are exotic to South Africa. Plant families with the highest number of species were: Asteraceae with 15 species, followed by Poaceae with 14 species, Cyperaceae (10 species), Fabaceae and Rubiaceae (5 species each), Lobeliaceae (3 species), Acanthaceae, Asphodelaceae, Lamiaceae, Oxalidaceae, Polygalaceae, Scrophulariaceae, Verbenaceae and Vitaceae (2 species each). Six main floristic clusters were identified from the hierarchical cluster analysis (HCA) and detrended correspondence analysis (DCA). Results from canonical correspondence analysis (CCA) revealed that species composition was mainly influenced by calcium, carbon, erosion, magnesium, potassium and the slope of the landscape. Conclusion: The diverse species diversity and composition documented is due to several environmental factors particularly calcium, carbon, erosion, magnesium, potassium and the slope of the landscape.
INTRODUCTION
Vegetation in catchment areas is being lost at an alarming rate mainly due to land use changes1, as artificial alterations of the hydrological cycle and the drowning of alluvial riparian vegetation influences growth, survival and species composition resulting in major negative impact on their functions and conservation. Such land use changes result in rapid transformation in plant and animal communities, drastically altering not only the ecological processes and functions that maintain appropriate ecosystem goods and services but also adversely impacting the livelihood needs of local communities. Van Rooyen et al.2 argued that species habitats and their diversity need to be assessed and understood as knowledge on habitat types, environmental determinants, dynamics and biodiversity enable researchers to evaluate impacts of major developmental projects such as dam construction and plan for conservation of biodiversity in catchment areas. According to Kuma and Shibru3 plant diversity, regeneration status, floristic composition and vegetation structure are crucial elements to clearly visualize the anthropogenic activities as well as environmental factors affecting the vegetation of an area. Therefore, information on floral diversity is a fundamental requirement to understand ecosystem type, biodiversity composition and other ecological parameters pertaining to biodiversity management and conservation planning at local, regional and global levels. Other researchers such as Ssegawa and Nkuutu4 argued that botanical assessments are crucial in identifying plant diversity, protecting threatened and economic species, monitoring the status of reserves and understanding the extent of plant diversity in natural ecosystems. Hence, the vegetation analysis enables us to build a mental picture of an area under investigation, compare and ultimately classify communities of the vegetation and understand the relationship that exists within communities and among their environments5. It is within this background that an assessment of floral composition and diversity in the Tsitsa river catchment area, site of the proposed Ntabelanga dam in the Eastern Cape province, South Africa was carried out.
The Department of Water and Sanitation, South Africa commissioned the construction of Ntabelanga dam on the Tsitsa river catchment area in the Eastern Cape province, South Africa. The proposed Ntabelanga dam is an integrated multi-purpose dam aimed at rejuvenating domestic and industrial water supply, irrigation and hydroelectric power purposes, tourism, conservation and other related activities. According to the Department of Water and Sanitation (DoWS)6, the proposed Ntabelanga dam has a storage capacity of 490 million m3 and is estimated to supply potable water to 730,000 people by the year 2050. The dam will also provide water to irrigate approximately 2,900 ha of arable land and there will be a small hydropower plant at the dam to generate between 0.75 and 5 MW (average 2.1 MW)6. Research by Van Tol et al.7 revealed that large dams play an important role in rejuvenating economic and social development but are often associated with environmental degradation through permanent inundation of previously dry areas, alteration of stream flow regimes, reduction in natural flooding and fragmentation of river ecosystems, thereby reducing species diversity. Overall, there is a dearth of detailed vegetation phytosociological studies in Southern Africa showing plant diversity in catchment areas, floristic composition and pattern. Such vegetation studies can be used to understand the pattern and processes influencing vegetation occurrence in catchment areas and ecological impacts of dam construction in catchment areas. The present study therefore, was aimed at filling this knowledge gap by assessing plant species composition and diversity within the Tsitsa river catchment area in the Eastern Cape province, South Africa.
MATERIALS AND METHODS
Study area: The study was conducted in the Tsitsa river catchment area within Elundi Local Municipality in the Joe Gqabi District Municipality of the Eastern Cape province, South Africa. The Eastern Cape province is characterized by landlessness, pervasive chronic poverty, low levels of education, economic activity, vulnerability, lack of basic services, a dearth of employment opportunities and high levels of dependency on welfare8. Tsitsa river catchment area is located on 31°7 35.9 S and 28°40 30.6 E. The study area receives an annual rainfall of about 749 mm, with most of it falling in December and January, with the lowest (15 mm) average rainfall received in June and the highest (108 mm) in January9. The study area is underlain by sedimentary rocks of the Tarkastad subgroup and Beaufort karoo supergroup with post karoo doleritic intrusions9. Tsitsa river catchment area is characterized by highly unstable soils that are prone to erosion as evidenced by extensive areas of severe gully erosion on the inter-fluvial areas adjacent to stream channels and these erosional and piping characteristics are suggestive of the presence of dispersive soils9. Mucina and Rutherford10 described the vegetation of the study area as sub-escarpment grassland and sub-escarpment savanna bioregions dominated by moist grasslands and Acacia spp. This vegetation type occurs at an altitude of 880-1860 m above sea level with the landscape characterized by moderately rolling hills10. Households in Tsitsa river catchment area have small permanent arable land between 0.1 and 0.5 ha of the 1 ha homestead land allocated to them by the tribal authorities to subsistence agriculture7. The arable lands are typically consolidated rainfed farming areas, which can be made up of several plots (1-3 ha or more)7. With high levels of poverty, low levels of economic activity and the poor quality of land allocated to Tsitsa river catchment area residents, non-farm activities are potentially an important source of livelihood for the residents.
Sampling data collection: Fieldwork was conducted between March and November, 2016 in Tsitsa river catchment area. Nineteen plots measuring 5×5 m, based on the results of a species-area curve11 determined prior to the sampling process were used to assess plant species composition and abundance via the Braun-Blanquet survey technique11. The exact locality of each plot was recorded using Global Positioning System (GPS). Within each sample plot, the habitat information and species present were recorded. A cover-abundance value was assigned to each species present in a sample plot according to the Braun-Blanquet cover-abundance scale11-14 as shown in Table 1. Plant species were identified in the field and the taxon names conform to those of Germishuizen et al.15. Unknown plant species were collected, pressed, dried and identified by taxonomists of the Giffen Herbarium (UFH), University of Fort Hare. The following environmental data were collected from every quadrat following methods outlined by Omar et al.16: C (%), Ca (cmol kg–1), clay (%), erosion (%), herb height (cm), K (cmol kg–1), litter cover (%), Mg (cmol kg–1), Na (cmol kg–1), NH4-N (mg L–1), NO3-N (mg L–1), pH, rock cover (%), sand (%), silt (%), slope (%), total vegetation cover (%) and tree height (cm). Multivariate data analysis was performed on the vegetation data to explore the floristic variation, to detect and visualize similarities in the plots16.
Data analysis: Canonical correspondence analysis (CCA) was performed using Palaeontological Statistics17, version 3.06. Patterns of plant species composition in relation to the measured environmental factors were analyzed using CCA. Detrended correspondence analysis (DCA) was performed on the same data set using Palaeontological Statistics17. According to Legendre and Legendre18, CCA and DCA are direct gradient analysis techniques that relate species composition and abundance to environmental variation enabling the significant relationship between plant species and environmental variables to be determined. Factors hypothesized to influence vegetation composition and abundance in this study were captured in a spreadsheet as environmental variables.
RESULTS
In total 78 plant species were recorded from the study conducted within the Tsitsa river catchment area and these plants are grouped into 24 families and 57 genera (Table 2). Among recorded plant species, 11.5% are exotic to South Africa and the rest are indigenous to the country. Plant families with the highest number of species were: Asteraceae with 15 species, followed by Poaceae with 14 species, Cyperaceae with 10 species, Fabaceae and Rubiaceae with 5 species each, Lobeliaceae with 3 species and Acanthaceae, Asphodelaceae, Lamiaceae, Oxalidaceae, Polygalaceae, Scrophulariaceae, Verbenaceae and Vitaceae with 2 species each.
| Table 1: | Braun-Blanquet cover-abundance codes, values and median values11-14 |
| Table 2: | List of plant species recorded from eight sites in Tsitsa river catchment area. Species marked with an asterisk (*) are exotic to South Africa |
| Fig. 1: | Hierarchical cluster analysis dendogram classification of vegetation plots based on weighted species presence |
| Fig. 2: | Detrended correspondence analysis (DCA) ordination diagram showing the grouping of six vegetation types (A-F) identified in Tsitsa river catchment area |
The rest of the plant families were represented by a single species each (Table 2). The most common genera in descending order of frequency were Cyperus with four species followed by Bulbostylis, Helichrysum, Lobelia and Senecio with three species each and Berkheya, Eragrostis, Finicia, Hypparrhenia, Oxalis, Richardia and Sporobolus with two species each (Table 2).
Six main floristic clusters were derived from the hierarchical cluster analysis (HCA) out of nineteen sampled plots (Fig. 1). The analysis presented about 35% similarity among the six clusters (Fig. 1). Similar results were obtained by detrended correspondence analysis (DCA) which separated 19 plots into six main clusters (Fig. 2). Results from canonical correspondence analysis (CCA) revealed that species composition was mainly influenced by calcium, carbon, erosion, magnesium, potassium and the slope of the landscape (Fig. 3).
The floristic and environmental characteristics of the six clusters are summarised below:
| Cluster A: | A total of 19 species were recorded on this cluster, ruled by way of perennial herbs and grasses namely Bulbine abyssinica, Crabbea hirsuta, Cynodon dactylon, Helichrysum glomeratum and Taraxanum officinale with Bulbine abyssinica recorded in this cluster scarcely (Table 3). Some of the plots were rocky, with an overage rock cover of approximately 8.5%. This cluster had the least carbon and sodium content of 0.29% and 23.43 cmol kg–1 respectively (Table 3) |
| Fig. 3: | Canonical correspondence analysis (CCA) ordination scatter plot indicating the influence of environmental variables on species composition in Tsitsa river catchment area |
| Cluster B: | An overall of 33 species have been recorded on this cluster, dominated by annual and perennial herbs, shrubs and grasses specifically namely Aristida congesta spp. barbicollis, Ficinia deusta, Helichrysum glomeratum, Hyparrhenia hirta, Richardia humistrata, Sporobolus africanus and Taraxanum officinale (Table 3). Unique plant species recorded in this cluster only include Chamaecrista capensis, Euphorbia inaqualatera var. inaqualatera, Cineraria spp., Conostomium spp., Digitaria ternata, Schkuhria pinnata and Selago spp. (Table 3). This cluster had the highest average vegetation cover, carbon content and were commonly flat in comparison with other clusters |
| Cluster C: | A total of 33 plant species have been recorded on this cluster was dominated by perennial herbs and grasses namely Helichrysum glomeratum, Hyparrhenia hirta, Richardia humistrata, Sporobolus africanus, Sporobolus fimbriatus and Zornia capensis (Table 3). This cluster had the lowest sand content and maximum silt and ammonium nitrogen contents (Table 3) |
| Cluster D: | This cluster consisted of a single plot characterised through a total of 24 plant species. Thirteen plant species were recorded in this cluster only including annual plants such as Bulbostylis densa ssp. afromontana and Zinnia peruviana. The majority of the common plant life in this cluster were either perennial herbs, shrubs or trees, which included Aloe ferox, Asparagus laricinus, Berkheya discolor, Cussonia paniculata ssp. paniculata, Cyphostemma spp., Helichrysum cerastioides var. cerastioides, Kalanchoe rotundifolia, Lantana rugosa, Senecio inaequidens, Sutera cooperi and Teucrium trifidum (Table 3). This cluster had the lowest litter cover and maximum slope, tree height, pH, calcium, nitrate nitrogen and clay content in comparison with other clusters (Table 3) |
| Cluster E: | This cluster consisted of a single plot characterised by a total of 12 plant species. Plant species common on this cluster and recorded in this cluster only included annual or perennial herbs such as Cyperus spp., Lobelia flaccida ssp. flaccida and Verbena spp. (Table 3). This cluster had the highest sand content and the lowest pH, potassium, calcium, magnesium, nitrate nitrogen, clay and silt contents in comparison with other clusters (Table 3) |
| Cluster F: | This cluster consisted of a single plot characterised by a total of 13 plant species. Plant species not usual in this cluster and recorded in this cluster only included Acacia karroo, perennial herbs such as Commelina africana var. africana and Richardia brasiliensis (Table 3). This cluster had the lowest vegetation cover and the highest litter cover, potassium, sodium, magnesium and ammonium nitrogen contents in contrast with other clusters (Table 3) |
| Table 3: | Summary of floristic associations with environmental variables assessed in eight sites within the Tsitsa river catchment area |
DISCUSSION
Results of this study provided baseline data such as species diversity and composition in Tsitsa river catchment area. With the exception of Verbenaceae and Vitaceae, the rest of major plant families recorded in this study Acanthaceae, Asphodelaceae, Asteraceae, Cyperaceae, Fabaceae, Lamiaceae, Lobeliaceae, Oxalidaceae, Poaceae, Polygalaceae, Rubiaceae and Scrophulariaceae are among the largest families in South Africa characterized by at least 100 species each15. Among the common species recorded in 30% of the plots included eight indigenous species namely, Aristida congesta spp. barbicollis, Cynodon dactylon, Eragrostis chloromelas, Ficinia deusta, Helichrysum glomeratum, Hyparrhenia hirta, Sporobolus fimbriatus and Sporobolus africanus. Exotic species recorded in 30% of the plots included Richardia humistrata and Taraxacum officinale. These two exotic species together with five additional exotic species recorded in this study namely Conyza bonariensis, Oenothera rosea, Richardia brasiliensis, Richardia humistrata, Schkuhria pinnata, Taraxacum officinale and Zinnia peruviana have been recorded in other provinces of South Africa19. All these exotic plant species recorded in this study are also recognized weeds in several countries throughout the word and are listed in the global compendium of weeds20. Documentation of exotic plants and weeds is important as such records can contribute to the global knowledge of invasive alien plants in South Africa19 because one of the useful predictors of invasiveness is whether a species is invasive elsewhere in the world or not21.
Plant species identified in this study as components of the six clusters shown in Fig. 2-3, as well as Table 2 and 3 corroborate earlier findings by Mucina and Rutherford10 that the area is characterized by about six vegetation types. These vegetation types which are characteristic of Tsitsa river catchment area and its surrounding areas include the Bisho thornveld, Drakensberg foothill moist grasslands, East Griqualand grassland, Eastern valley bushveld, Mabela sandy grassland and Mthata moist grassland10. The following species which are key elements of the Bisho thornveld have been documented in this study (Table 2): Cynodon dactylon, Eragrostis chloromelas, Hyparrhenia hirta, Hypoxis argentea, Kyllinga alata, Microchloa caffra and Sporobolus africanus10. According to Mucina and Rutherford10, Drakensberg foothill moist grasslands which was represented by Eragrostis chloromelas, Helichrysum odoratissimum, Hyparrhenia hirta, Microchloa caffra, Senecio retrorsus and Sporobolus africanus in the current study is moderately rolling and mountainous, incised by river gorges of drier vegetation. The following species which are key elements of East Griqualand grassland have been documented in this study: Acacia karroo, Aristida congesta, Digitaria ternata, Eragrostis chloromelas, Hyparrhenia hirta, Microchloa caffra, Senecio retrorsus and Sporobolus africanus10. In the current study, Eastern valley bushveld vegetation type was represented by Aristida congesta, Hyparrhenia hirta, Melinis repens and Sporobolus fimbriatus. Mabela sandy grassland vegetation type was represented by Aristida congesta, Cynodon dactylon, Digitaria ternata, Eragrostis chloromelas, Eragrostis gummiflua, Hyparrhenia hirta, Microchloa caffra, Paspalum distichum and Setaria sphacelata. Mthata moist grassland was represented by Cynodon dactylon, Hermannia parviflora, Hyparrhenia hirta, Microchloa caffra, Richardia humistrata, Senecio retrorsus and Sporobolus africanus. Previous studies by Mucina and Rutherford10 revealed that alien species such as Richardia humistrata is widespread in Mthata moist grassland and regarded as a conservation concern.
The present study revealed that species composition is often influenced by environmental factors such as calcium, carbon, erosion, magnesium, potassium and the slope of the landscape. Similar results were obtained by Fonge et al.22 showed that plant diversity, distribution and abundance in Lewoh-Lebang in the Lebialem highlands of southwestern Cameroon was affected by organic carbon, nitrogen, calcium and cation exchange capacity of the soil. Similarly, Ahmed et al.23 found that calcium carbonate, magnesium, organic matter, pH, salinity (electrical conductivity) and sodium contributed to the distribution of plant species and plant communities in Omayed Biosphere reserve in Egypt. Many types of human-influenced disturbances such as dam construction in catchment areas are known to influence riparian ecosystems. Research by Ceschin et al.24 revealed that severe alterations in the catchment’s hydromorphological and biological features are associated with dam construction which usually lead to the development of riparian and aquatic communities. Such riparian and aquatic developments include an increase in the number of vegetation types in comparison with the natural aquatic environment pre-existing the dam construction24. Results obtained in this study are important for monitoring any changes in the vegetation and other environmental factors of Tsitsa river catchment area due to land use changes. Such baseline data will be used to assess any changes in the environmental factors, floristic diversity, composition and structure caused by the planned construction of Ntabelanga dam within Tsitsa river catchment area. After the dam has been constructed, vegetation succession is expected in response to several environmental factors and the identification of the principle environmental factors and dominant vegetation types are viewed as major challenges in trying to understand ecological processes in a transformed landscape.
CONCLUSION
The vegetation of Tsitsa river catchment area is fairly diverse, characterised by 78 plant species and at least six vegetation types. The identified vegetation types included the Bisho thornveld, Drakensberg foothill moist grasslands, East Griqualand grassland, Eastern valley bushveld, Mabela sandy grassland and Mthata moist grassland. Such diverse vegetation types may imply that the environmental conditions in Tsitsa river catchment are favourable for a wide range of plant species. Each of these vegetation types reflects the homogeneity of the plant communities in terms of plant species composition and dominance. Therefore, the uniqueness of each vegetation type documented in Tsitsa river catchment area may be due to a combination of various environmental factors such as such as calcium, carbon, erosion, magnesium, potassium and the slope of the landscape.
SIGNIFICANCE STATEMENTS
This study showed that catchment areas are characterized by diverse plant species and vegetation types. Key environmental factors found to influence plant species diversity and composition included calcium, carbon, erosion, magnesium, potassium and the slope of the landscape. Results of this study will help researchers to uncover critical environmental factors likely to have an effect on species diversity and composition. Thus, insights into vegetation ecosystem structure and composition should also evaluate environmental variables.
ACKNOWLEDGMENTS
The authors would like to express their gratitude to the Water Research Commission (WRC, grant number K8/1027), National Research Foundation (NRF grant number T398) and Govan Mbeki Research and Development Centre (GMRDC, grant number C169), University of Fort Hare for financial support to conduct this study.