The effect of tillage direction on soil redistribution by mouldboard ploughing on complex slopes

Abstract

Seven mouldboard ploughing experiments were conducted to systematically investigate the effect of different tillage directions on soil redistribution on hillslopes. The present study included tillage directions other than parallel to the gradient or along the contour, that is, in our experiments the slope gradient changed simultaneously in tillage and in turning direction. Using physical tracers we developed a model of the two-dimensional tracer displacement as a function of topography and tillage variables. The displacements in tillage and in turning direction were separately described as 2nd degree polynomials in both tillage and in turning directions. This model fully accounted for the directionality of tillage. Displacement in turning direction additionally depended on tillage depth, while that in tillage direction was affected by tillage speed and soil bulk density. We found a large effect of tillage directionality on soil redistribution, and tillage at 45° to the gradient turning soil upslope was the least erosive tillage direction. We obtained non-linear relationships between soil redistribution and profile curvature, instead of the linear relationships reported previously. Consequently tillage erosivity varied in tillage direction and a unique tillage transport coefficient could not be obtained for all tillage directions.

Introduction

Tillage erosion is a major process of soil redistribution on sloping land in tillage-intensive agriculture (Govers et al., 1999, Lobb et al., 1999, Van Oost et al., 2003a). Topography is the dominant control of the process leading to a characteristic pattern of soil loss on convexities and soil aggradation on concavities (Govers et al., 1994). Analysis of the spatial distribution of caesium-137 in arable soils suggested annual erosion rates frequently exceeding 20 t ha⁻¹ on eroding sites (Quine et al., 1997). Such extent of soil loss on crests and shoulder slopes has major implications for soil quality, such as organic matter and nutrient dynamics, and soil productivity (e.g. Verity and Anderson, 1990, Schumacher et al., 1999, Quine and Zhang, 2002). In Mediterranean environments soil productivity has been directly related to soil redistribution due to tillage (Kosmas et al., 2001). Additionally, the tillage-induced pattern of soil properties within hummocky fields may also increase the risk of nutrient loss by water erosion (Van Oost et al., 2000).

Researchers have used chemical or physical tracers in controlled tillage experiments to study the effect of topography, and tillage speed and depth on the soil redistribution pattern of different tillage implements. Lindstrom et al. (1990) were among the first to report linear relationships between soil displacement and slope gradient for contour and up- and downslope tillage. Based on similar experiments, Govers et al. (1994) derived a general expression for the unit soil transport rate due to tillage as a linear, univariate function of slope gradient. Hence, for a tillage process the change in soil transport rate along a hillslope, i.e. the occurrence of erosion or deposition, becomes proportional to the change in slope gradient (Govers et al., 1999). The proportionality factor (or tillage transport coefficient) is an expression of tillage erosivity and permits prediction of tillage translocation rates (e.g. Lobb et al., 1995, Quine et al., 1997). Apart from tillage speed and depth, tillage erosivity is a function of tillage direction. This is evident from the different tillage transport coefficients obtained for up-downslope tillage along the direction of maximum slope and for contour tillage (Van Muysen et al., 2002). However, these two tillage directions represent a special set of the topographic variables controlling tillage translocation, as the slope gradient only varies in a single direction, either in tillage or perpendicular to tillage direction, i.e. turning direction, while it is zero in each of the other. Soil movement by tillage essentially is a two-dimensional process characterized by a displacement vector in tillage and one perpendicular to tillage direction. Each of these can be affected by the slope gradient in the corresponding direction on a more complex topography (De Alba, 2001). Hence, the simultaneous change of slope gradients in both tillage and turning direction may exert an important influence on tillage erosivity. To our knowledge hardly anyone has investigated soil displacement due to tillage in a direction other than parallel to the gradient or along the contour. Under normal agricultural practice on hummocky terrain simultaneously changing slope gradients in tillage and turning direction will be rather common as field geometry more than topography determines the tillage direction. Therefore, there exists an important need for experimental data and a characterisation of tillage erosivity that account for this directionality, if reliable predictions of soil translocation by tillage are to be obtained on complex topography (Quine, 1999, Van Muysen et al., 2002).

The mouldboard plough still is the standard implement for primary tillage in mechanized agriculture and it is associated with high tillage erosivity (e.g. Gerontidis et al., 2001, Van Muysen et al., 2002). The general objective of our tillage experiments was, therefore, to investigate the relationship between slope gradient and soil movement for mouldboard ploughing on light textured soils. Specifically we investigated the effect of simultaneously changing slope gradients in tillage and in turning direction by ploughing slantwise on a hillslope. The experimental data were used to develop a general, slope-dependent, two-dimensional model of soil displacement due to tillage for hummocky terrain and to derive an expression of tillage erosivity from it.