Causes of landscape mega-ripples: The kommetjies of South Africa

Abstract

A variety of explanations have been invoked for landscape patterning, such as regularly spaced mounds and hollows that occur globally. The regularity of the patterning has, however, commonly resulted in biotic agents being suggested as causal, despite the fact that abiotic factors also commonly induce patterning (e.g. water/wind ripples in sand). We studied the ‘kommetjie’ landscapes of South Africa, which comprise striking ripples that have been suggested to be produced by earthworms transferring soil from hollows to adjacent mounds. This landscape is superficially similar to the spatially extensive ‘surales’ of South America and pit-and-ridge landscapes in Papua New Guinea that have also been attributed to earthworms. We hypothesised that kommetjies were formed as a result of emergent patterning induced by water erosion exacerbated by earthworm activity under a special combination of environmental circumstances. To test this, we modelled the geographic distribution of kommetjies, produced a drainage map and examined the soil stratigraphy and elemental composition. We found that the kommetjies distribution was limited to an area with relatively high annual rainfall (>600 mm) and moderate annual average temperatures (<20 °C). The ripples are inter-connected into basins of meandering flow networks that commonly traverse the landscape slope. The soils are underlain by a plinthic horizon that prevents subsurface infiltration, resulting in ponding of water in hollows and surface runoff. The amount and distribution of pisolith gravel in the hollows and hummocks were consistent with erosional deflation of the hollows. Although giant earthworms are common in the area, they also occur in areas without the ripples. The ripples are likely produced by water erosion of gently sloping terrain that depends on restriction of sub-surface drainage exacerbated by earthworm bioturbation and cast production. We thus concluded that the ripples are an emergent product of an interaction between abiotic erosion and earthworm activity.