CHANGES IN GRASS SPECIES COMPOSITION AND PRODUCTION IN TWO RANGELAND MANAGEMENT SYSTEMS: A LANDCARE INITIATIVE IN THE NORTH WEST PROVINCE, SOUTH AFRICA

Abstract

Land degradation is worldwide in geographic extent, global in its environmental and socio-economic impacts and it leaves no continent unaffected. It is thus a matter of urgent priority that ways be found to prevent and even reverse, the degradation of the natural resources. This is also true for the North West Province of South Africa. An enormous challenge faces this Province: that of protecting and sustainably managing the natural resource base, even under the present condition of population growth and rising unemployment. It is the intention of the Province to ensure that the natural resources are wisely and carefully managed, utilized and improved. Various LandCare programs, initiatives of the South African National Government, has been implemented in order to encourage and support sustainable land use practices, raising awareness and promoting a resource conservation ethic. LandCare is a sustainable resource utilization program, based in the community and driven by both the public and private sectors through partnerships and co-operation. The purpose of this project was to develop a baseline study in order to establish and implement monitoring and evaluation systems to monitor progress and assess emerging and changing needs on an ongoing basis. Determining the changes in the grass species composition and production at eighteen different benchmark sites, in three magisterial districts, namely Ganyesa, Kudumane and Taung, of the western region in the North West Province, was used to achieve this goal. The changes in the grass species composition and production were monitored in the communal grazing (Austrey, Maketlele and lpelegeng study sites) and Morafe ranch management systems (Waterfouche, Heuningvlei and Orange Grove study sites). The Morafe ranch management system and some strategies to evaluate the effect of resting in communally managed systems were implemented at these study sites since January 2001. The study sites are situated in different Agricultural Development Centers (ADC's), which in turn are divided into different Field Service Units (FSU's). Benchmark sites representing different conditional states, were identified in selected areas in both the Morafe ranch and communal managed systems in collaboration with extension officers of the Department of Agriculture, Conservation and Environment. Exclosures, used to exclude any grazing by cattle and goats, were erected inside the benchmark sites. The main purpose of erecting the exclosures was to demonstrate the advantages of resting the rangeland. In order to achieve the aims of the project, the grass species composition and production of the rangeland were determined with the help of the agricultural extension officers and the local land users at each of the study sites. The grass species composition was determined inside and outside the exclosures, by means of the line transect method, where the nearest perennial and annual grass species was recorded in a radius of 30 cm at 1 meter intervals along two transects of 1 OOm each. The grass production was only determined inside the exclosures by means of the quadrat method. Two replications of twenty-five lm2 quadrats were alternately placed at four meter intervals, along the two 1 00m transects. The grass was cut per species at a height of 3cm above the ground, dried (24 hours at 60°C) and weighed. The classification of the grass species into highly desirable, desirable, less desirable and non desirable classes for grazing, according to Van Oudtshoorn ( 1999) and the expert knowledge of technicians and researchers of the Department of Agriculture, Conservation and Environment, were used for the purpose of this study. Soil samples were taken on the inside and outside of each exclosure. Although results from the soil analyses can be used to explain changes in the grass species composition and production, drastic changes in the soil characteristics usually only take place over the long-term. These results thus mainly serve as baseline data for the monitoring of these benchmark sites in future. The frequency at which the different grass species, forbs and bare ground occurred inside and outside the exclosures at the different benchmark sites, were calculated by determining the percentage that each species composed of the total amount of species, recorded during a specific species compositional survey. The Z-index (%) of agreement, or similarity (Du Toit, 1998), was applied in order to determine if changes in the grass species composition between the inside and outside of the exclosures did occur during the study period of two years (during the years 2001 and 2002). The concept of a large stock unit (LSU), (Tainton, 1999), was used to establish the current seasonal grazing capacity of the rangeland by means of the formula proposed by Moore et al. (1985). By appointing different utilization factors (Smit, 2002) to the proportional production (kg/ha) of the grass species, divided into the different desirability classes, the seasonal grazing capacities of the different benchmark sites were determined. Three different approaches in calculating the seasonal grazing capacities were used and compared with each other. The high Z-index values at all the benchmark sites, indicates that the similarity between the grass species composition inside and outside the exclosure is still very high. This signifies that the study period was too short for definite changes to occur between the grass species composition inside and outside the benchmark sites, due to the rest (applied inside the benchmark sites) and different management systems applied. The slight changes that occurred could mainly be attributed to the rainfall variability at each specific study site. O'Connor (1994) also stated that population dynamics of important savanna grass species appears to be more dependent upon rainfall variability than upon grazing. However, the availability of seed in the soil seed bank is an important constraint on seedling recruitment in years of suitable rainfall. Consequently, it is very important that rangelands should be very carefully managed, especially during periods of low rainfall. The rangeland at each benchmark site could be classified into a good, moderately good, moderate, or poor rangeland condition mainly according to its grass species composition, which in turn has a decisive influence on the productivity and eventually on the seasonal grazing capacity expected from grasslands (Snyman, 2000). The grass species, Schmidtia pappophoroides, Eragrostis lehmanniana and Aristida stipitata, were identified as good indicators of the rangeland condition in the Ganyesa and Kudumane magisterial districts, indicating a good, moderately good and moderate rangeland condition, respectively. The grass species composition at the benchmark sites situated in the Taung magisterial district however differed from those in the Ganyesa and Kudumane districts, due to differences in the topography, soil and geology at the three magisterial districts. Even though some benchmark sites in the same study area were characterized by the same soil type, and were subjected to the same rainfall pattern and management system, their responses to a higher rainfall, in terms of grass species composition and production changes, differed to a greater extent. This could mainly be attributed to the fact that these sites were subjected to different grazing pressures during the past. The historical grazing regime that was followed over many years in the past, will still have a great impact on the results which are and will be monitored in future. A plant's particular strategy, or life history traits, thus determines its success at a particular region of any environmental gradient (McCook, 1994). Even though it was not the objective of this study to compare the Morafe ranch and communal management systems, some differences were observed. It was evident that the benchmark sites managed by the Morafe ranch system (the Waterfouche, Heuningvlei and Orange Grove benchmark sites), were generally in a better condition than the communally managed benchmark sites (the Austrey, Maketlele and Ipelegeng benchmark sites). This was deduced from the fact that they consisted of a more desirable grass species composition, higher total production figures and better seasonal grazing capacities than most of the communally managed benchmark sites. Seeing that the Morafe ranch management system was only recently implemented and could therefore not have led to such an improvement in the rangeland condition, the better condition could mainly be attributed to the limited availability of watering points at these sites in the past. It can therefore be concluded, that these benchmark sites were not subjected to overgrazing in the past, due to the low livestock densities and water resource constraints. In conclusion it can be said that although the timespan of two years is too short for vegetation dynamic studies to detect species compositional changes, these preliminary results can already be used as baseline data to establish and implement a long-term monitoring system. The results will also assist to further improve the environmental awareness of land users and agricultural extension officers, which was already initiated by means of this project, with regard to sustainable management practices. As a consequence a very positive collaboration between the academic institution (Potchefstroom University for CHE) and the agricultural extension officers and land users were created.