Search

This paper presents a rainfall simulation experiment carried out on three 50 m2 plots in the Senegalese groundnut belt. One plot was not cultivated. Groundnut and millet had previously been grown in the other two. The experiment consisted of three rain events applied over 5 days at the end of the dry season. Erosion was monitored inside the plots by the use of a relief meter and, at their outlets, by sampling the discharge. The number of indigenous nematodes, and an exotic species introduced before the first rain event, was monitored in the soil and in the discharge. This experiment allows, for the first time, a set of simple hypotheses to be proposed to explain the spread of nematodes by the runoff: raindrop impacts on the soil surface set them in suspension; then, their low bulk density and their relatively large size do not allow them to settle when the raindrops shake the water surface. Thus, nematodes follow the flow path where they are as far as its velocity remains significant. The biological aspects are decisive in the mobility of nematodes, which can vary by a factor of 100 depending on the trophic groups. A very high raindrop erosion occurred during the experiment, up to 60 tons per hectare for the first rain event after hoeing. This represents more than 40 per cent of the volume of soil previously moved by soil work. The geometric properties of the plough, and their hydraulic consequences, appear very ephemeral. And yet these large movements of soil inside the plots are little related to the sediment load at the outlet, which follows its own rules. Analysis of the results indicates that the carrying capacity of the runoff at the scale of 10 m2, on gentle slopes ploughed perpendicular to the slope, could not be directly calculable from the discharge. It could depend on the history of past discharges because the shape of the flow paths, which condition their carrying capacity, permanently interacts with the discharge. These interactions could explain the great difficulties encountered by the erosion models in the case of low discharges on non-cohesive soils. Copyright © 2000 John Wiley & Sons, Ltd.