Restoring degraded patches in a semi-arid rangeland of South Africa

It is known that, depending on the degree of degradation, the vegetation recovery of severely degraded areas by means of natural succession processes is very slow, if not impossible, and that active intervention in the form of restoration technologies has to be applied. Combinations of different restoration technologies, including ripping, over-sowing, brushpacking and the addition of organic material, were evaluated in a study site (characterized by denuded saline patches) in the semi-arid areas of South Africa. Quantitative vegetation sampling procedures, which included the monitoring of vegetation cover, frequency and density measurements were carried out over a period of 2 years (2000–2002) to determine the establishment and dynamics of the over-sown species used in the restoration treatments. The germination capacity and viability of the over-sown species were also tested in the laboratory, using methods prescribed by the International Seed Testing Association (ISTA). The over-sown species were positively correlated in the over-sowing treatments, while only annual pioneer species which were still represented in the soil seed bank in the degraded patches, correlated with the control and cultivation-only treatments. Of the over-sown species Digitaria eriantha and Chloris gayana had the highest frequencies. The restoration treatment that included a combination of ripping, over-sowing, brushpacking and organic material showed the highest establishment rate of all over-sown species. The results of the seed purity analysis indicated that seed obtained from a registered seed merchant showed higher purity percentages than seed collected locally by the land user. Species with high frequency and density values under natural conditions in the field trials also had a high purity and germination percentage in the laboratory experiments.