Patterns of plant species richness and diversity across two habitat types in the Upper Karoo, South Africa

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

The Nama-Karoo biome is relatively understudied in terms of its baseline biodiversity. Apart from its rich agricultural land-use history, the region is also under pressure from the development of the Square Kilometre Array, an increased demand for cleaner energy from shale gas fracking and/or other renewable energy installations, and the overarching impacts of global climate change. A reliable baseline inventory of biodiversity for the region is essential if these impacts are to be monitored and managed effectively. The main aim of this study was to relate fine-scale patterns of plant diversity and community structure to broader-scale vegetation mapping in the Karoo regions. It also investigated the role of several environmental and climatic variables as drivers of species richness, relative cover, and growth form diversity in two habitat types (i.e. plains and rocky dolerite hillslopes), and along a longitudinal environmental gradient. A pairwise floristic survey approach was implemented, using modified Whittaker plots in each habitat type at 30 sites within the shale gas exploration area in the Upper Karoo bioregion. Data were collected on species richness, relative cover, and growth form diversity of the observed vegetation. Soil samples were collected from each Whittaker plot, and climate data were obtained by point sampling from raster layers using GIS. The results showed that mean species richness was significantly higher (p < 0.01) in slope habitats than in plains habitats across the environmental gradient. Trees and woody shrubs had significantly higher species richness (p < 0.001) and relative cover (p < 0.01) in slope habitats. Low woody shrubs comprised the greatest percentage of growth form diversity in both habitat types in terms of species richness and relative cover, and were the dominant growth form across the longitudinal gradient. At the arid western extent, leaf-succulent shrubs had slightly higher relative cover (~ 25%) in slope habitats compared to other growth forms, apart from low woody shrubs. In the central regions, perennial grasses were more abundant (25 – 60%) in slope habitats, while annual grasses were more abundant (25%) in plains habitats. At the more mesic eastern end of the longitudinal gradient, perennial grasses were dominant (> 50%) in plains habitats. Cluster analysis, based on species presence data in each habitat type, showed relatively strong correspondence between plant associations in slope habitats and their respective vegetation types as currently defined. These plant associations were spatially aggregated according to their position along the environmental gradient. Little to no correspondence was found between plant associations in plains habitats and their respective vegetation types. These plant associations were interspersed with each other when plotted spatially, and occurred in a repeating pattern in plains habitats across the study area. Multiple regression models indicated that a combination of climatic and environmental variables, and soil properties significantly predicted overall species richness and relative cover of the five dominant growth forms. Habitat type commonly emerged as a significant predictor for overall species richness and relative cover. Overall dissimilarity, and dissimilarity in the relative cover of five plant functional types between plains and slope habitats, were also predicted by a combination of climatic and environmental variables, and soil properties. These predictors varied greatly between the different response variables, suggesting that different plant functional types are influenced by different drivers, depending on the habitat in which they occur. Habitat heterogeneity, coupled with local and regional variation in prevailing climate and soils, has consistently emerged as an important driver of plant species richness and relative cover in global drylands research. Measuring richness and diversity within habitats requires a high sampling resolution in both plot size and number. This study has shown that sampling at the 0.1 ha scale (or larger) captures a representative sample of richness and diversity within a given habitat type in the Nama-Karoo biome, where low woody shrubs (< 60 cm tall) are the dominant growth form. In the current vegetation map of South Africa, the vegetation types for the Upper Karoo bioregion are coarse, and hence give the impression of homogeneity in what is in reality a relatively heterogenous landscape. Fine-scale baseline biodiversity data such as are presented in this study may improve the resolution of the existing vegetation map, as well as inform better conservation and management practices in economically important and biologically diverse rangelands in the Nama-Karoo biome, prior to future developments in the region.