Field-scale run-off and erosion in relation to topsoil aggregate stability in three tropical regions (Benin, Cameroon, Mexico)

The influence of aggregate stability on run-off and erosion has often been reported from experiments on microplots (about 1 m(2)) under simulated rain. Our objective was to compare the aggregate stability of topsoil (0-10 cm) with run-off and erosion from experiments on run-off plots (about 100-1000 m(2)) under natural rain. Run-off and soil losses were measured over three years on 14 plots in Benin, Cameroon and Mexico. All plots were under herbaceous vegetation and had moderate slope length and slope declivity, but differed in climate (400-1600mm annual rainfall), soil type (sandy clay loam Nitosol, loamy sand Ferralsol, loamy Regosol), and management (from savanna to long-duration mouldboard ploughing). The stability of aggregates was determined by immersing and wet-sieving 2-mm sieved air-dried samples into water. Mean annual run-off rate and soil losses generally increased, and the proportion of stable macroaggregates (>0.2mm) decreased, with increasing duration and intensity of tillage, and with decreasing cover on the soil surface. For all 14 plots, run-off and soil losses were closely correlated with aggregate stability; correlations were improved when slope gradient and climate aggressivity were considered in addition to aggregation. Slaking, the main mechanism of aggregate breakdown which occurs when dry soil is immersed, accounted well for run-off and erosion. The stability of topsoil aggregates seems therefore to be a valuable indicator of field-assessed run-off and erosion for plots on moderate slopes with herbaceous vegetation.