A research report submitted in partial fulfilment of the requirements for the degree of Master of Science in Environmental Science
(Coursework and Research Report) to the Faculty of Science, University of the Witwatersrand,
Johannesburg, 2017. / Various studies have been conducted on the determinants of herbaceous vegetation composition and structure in savannas, but there is still no consensus on the extent of the role played by each. This is particularly so for disturbed savannas in communal rangelands. This study aimed to investigate the relative influence of rainfall, catenal position and distance from village on composition and structure of herbaceous vegetation in communal rangelands of Bushbuckridge, Mpumalanga Province in South Africa. The study used pre-existing data collected in 2012 in 56 plots located across nine villages in three rainfall zones: Low (<600 mm), Medium (600-700 mm) and High (>700 mm). In each zone, the communal rangelands of three villages were sampled in upslope and downslope positions and across three distance (disturbance) categories of, 0-599 m, 600-1799 m and >1800 m relative to nearest villages. The composition-related measures investigated were: absolute and relative abundance of species present, species richness, Simpson’s Diversity Index and relative abundance of perennial and annual grasses, as well as categories of forbs whilst the herbaceous structure measures used were distance in meters measured along transect line to perennial grass tuft and grass basal cover. Both individual and interactive effects of the rainfall zone, landscape position, and distance from village on herbaceous composition and structure were investigated using bivariate and multivariate statistics.
Both grass species richness and Simpson’s Diversity Index were significantly higher in the high rainfall zone than in the low and medium rainfall zones. Perennial grasses in all rainfall zones heavily dominated the herbaceous layer, with the mean percentage perennial grass contribution being lowest in the high rainfall zone whilst the low rainfall zone had the highest. The percentage composition of annual grasses was highest in the medium rainfall zone and lowest in the low rainfall zone whilst it was intermediate in the high rainfall zone. The mean distance to perennial grass, which is an index of density of perennial grass tufts, was significantly higher in the high rainfall zone than in the low rainfall zones. The mean percentage grass basal cover was higher in low rainfall zone than in high rainfall zone.
The mean species richness in the upslope and downslope catenal positions was not significantly different which was contrary to what was expected. Mean Simpson’s Diversity Index was higher on the downslope position than on the upslope catenal position. The relative abundance of perennial grasses between the upslope and downslope catenal position was not significantly different. The proportion of perennial grasses was more than that of annual grasses and other life forms sampled.
Neither mean distance to perennial grass nor basal cover differed significantly between catenal positions. It was established in this study that closest plots to villages had 22% more species than the plots between medium and far plots from the villages. Near the villages (0-599 m), the species richness was found to be higher than in the medium (600-1799 m) and (>1800 m) distance categories. The mean Simpson‘s Diversity Index in all distance (disturbance) categories were found to be not significantly different. There was no significant difference in the percentage of perennial grass among the disturbance gradient categories. The perennial grasses were consistently dominant over the annual grasses along the disturbance gradients. The distance to perennial grass indicated no difference across all distance gradients. Percentage basal cover decreased with increasing distance from settlements.
The composition and occurrence of grass species were associated with different environmental gradients studied. There was significant interactive effect due to a combination effect of rainfall, topographical gradients and distance gradient on the distance to perennial grass. However, the mean distance to perennial was lower at distance and rainfall combination, though was not significant. The interactive effect on basal cover due to a combination effect of rainfall, catenal position and distance gradient was found to be not significant.
Overall, the herbaceous composition and structure was more strongly impacted by rainfall zone than by catenal position. The herbaceous composition and structure was affected by disturbance gradient specifically on species richness and grass basal cover only, while there was no effect on Simpson’s Diversity Index, perennial grass percentage, and distance to perennial grass as measured along transect line. It was also established that there was association of species with environmental gradients. It is recommended that in future a multi-year study on the same variables that have been studied here be undertaken in order to establish long-term trends on the effect of the gradients on herbaceous vegetation. It must be ensured also that there be representativity of disturbance gradients samples when designing sampling programme. It will also be beneficial to establish the density per village or stocking rates of different villages and the management aspects of the stock and how it is related to herbaceous composition and structure. Such studies will provide further knowledge on the extent of human induced disturbance like grazing in the communal rangelands given a set of environment gradients. / LG2018
Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:wits/oai:wiredspace.wits.ac.za:10539/25030 |
Date | January 2017 |
Creators | Seabi, William Maropeng |
Source Sets | South African National ETD Portal |
Language | English |
Detected Language | English |
Type | Thesis |
Format | Online resource (ix, 79 leaves), application/pdf |
Page generated in 0.1913 seconds