British Academy: The UK's National Academy for the Humanities and Social Sciences

The social and ecological implications of landscape change since ca. 1800 in the South African Karoo

Professor Ian Foster

This field-based research project was funded by the British Academy from 2006 to 2008 and has involved a team of researchers from both UK and South African Universities. It has focused on widespread problems of land degradation in the South African Karoo (Eastern Cape) over the last ca. 200 years. Over two field seasons we have combined information from several sources in order to understand the relationship between agricultural change, climate change and land degradation. First, samples of sediment have been collected from a number of peat bogs (locally known as vleis) and from sediments accumulating in two small farm reservoirs. These sediments preserve archives of information that we can use to reconstruct environmental change over different timescales. Pollen analysis of the vlei sediments, for example, tells us how local vegetation has changed as a result of cultivation and selective stock grazing, while changes in sedimentation rates in the reservoirs tells us when the landscape began to degrade and whether recently adopted farming practices are slowing or even reversing the trend in landscape degradation. Secondly, weather records kept by local farmers have been collected and coded and these data allow us to analyse rainfall characteristics in order to establish the extent to which climate change has already affected the region. Thirdly, the analysis of historical farm diaries, supplemented by interviews with local farmers whose families have lived in the area for up to four generations, allows us to explore when, why and how changes in farm management have occurred over the last 200 years.

While the fieldwork has been completed, analysis of sediment samples, rainfall records and interviews is ongoing. The following section outlines some of our key findings so far.

The two reservoirs cored as part of this research programme (Ganora, near Nieu Bethesda, and Cranemere, near Pearston) have both accumulated over 4 m of sediment since they were built in 1910 and 1843 respectively. While we know the age of the reservoirs themselves, we have used two methods to establish the time when different sedimentary layers were deposited. First, we have used an atmospheric pollutant (Cs-137) that has only existed in the environment since the beginning of atmospheric thermonuclear weapons testing in the 1950s. It has a known fallout history globally, with a marked peak in the southern hemisphere in 1965, and produces relatively high concentrations in sedimentary sequences that can be correlated with the known atmospheric fallout record. Secondly we have used a natural radionuclide (Pb-210) that comes from the decay of radon gas in the atmosphere. This falls onto reservoir surfaces and is buried with the sediment. Once buried, it decays slowly through time. Since we know its decay rate, it is possible to calculate the age of the sediment into which it is incorporated.

By combining the analysis of Cs-137 and Pb-210 it is possible to calculate changes in sedimentation rate for the lifetime of the reservoir as shown in Figure 1. Here, at Ganora, the early 20^th century was a period of relatively slow sediment accumulation but sediment accumulation, and therefore erosion rates in the catchment, increased significantly from the late 1960s onwards.

Figure 1. The combined Cs-137 and Pb-210 chronology for Ganora Reservoir (A); changing patterns of sediment accumulation through time (B).

A

B

In addition to establishing rates of sedimentation, our analysis has included measurements of sediment properties that help us to understand whether sources of sediment in the catchment have changed through time. Here we compare a range of characteristics of the sediments in the reservoir with those of possible sources in the catchment. Figure 2 shows the downcore change in magnetic susceptibility.

Figure 2. Changes in magnetic susceptibility in the Ganora Reservoir sediments

These preliminary results suggest that erosion rates have increased significantly in the last ~40 years (Fig. 1) and that sources of sediment accumulating in farm reservoirs has changed from an early (1900-1960) phase dominated by channel erosion, with occasional inputs from gully systems, to a more recent phase (post 1990) when localized degraded badlands appear to have become the most important sediment source (Fig. 2). However, a significant change in sediment source also occurred in the 1930s. We know from the weather records that the 1930s was a period of extreme drought in the Karoo that was punctuated by occasional extreme rainfalls. We also know from farm records that the 1930s witnessed the highest stocking densities ever seen in the Karoo. Grazing removes much of the ground cover and reduces the resistance of soil to water erosion. Coupled with extreme drought, which reduces plant growth and the protective plant cover, these are ideal conditions under which to destabilize landscapes and initiate high rates of erosion. This erosion often gives rise to localized features known as badlands. Once these start to form, they tend to develop until almost all of the soil has been removed.

Land use histories are complex and often farm specific. General trends in stocking density show a peak in the 1930s of ~14 large livestock units per hectare of agricultural land. This has now reduced to between 2-4 large livestock units per hectare. In catchments that experienced rainfed or irrigated cultivation (including Cranemere), peak activity occurred in the late 1950s. However, many farmers have diversified their enterprises and now provide a range of tourist activities and accommodation. The number of game farms in the region has also increased significantly in the last 10 years. Stock and rangeland management practices have also been variable, with some farmers preferring high intensity grazing in small camps for short periods before moving stock to new camps on a rotational system. Historically, controlled burning was used to improve grass growth and, while this practice has declined in recent years, natural fires still occur extensively throughout the region.

South Africa will face a crisis in water management in the very near future if recent global climate change predictions prove to be correct. In 2008, it was estimated that the country already allocated between 95 and 98% of its available water resources. Much of this water is stored in reservoirs and is moved to areas of need through inter-basin transfers. Climate change is not the only threat to sustainable water supply. Our results suggest that rates of erosion have increased substantially in recent decades and, of course, much of the sediment will be deposited in major water supply reservoirs. Understanding how, why and by how much erosion rates have increased will allow us to quantify the future sustainability of this vital resource and engage with water resource managers in identifying schemes targeted at erosion reduction.

To date we have presented the results of research funded by the British Academy at several conferences (British Society for Geomorphology, Exeter, July 2008; the Southern African Association of Geomorphologists annual conference, Pretoria, September 2008; and the International Association of Hydrological Sciences conference, Christchurch NZ, December 2008). We have secured additional funding from the South African National Research Foundation to continue our collaboration. As part of this initiative, we will embark on a scheme of knowledge transfer by running a workshop at Rhodes University in September 2010.

March 2009