Aggregate stability and erosion response to antecedent water content of a loess soil

Soil erosion processes are affected by the erodibility of the soil and by the erosivity of the rain. The effects of rain characteristics and invariant soil properties such as texture and organic matter content on soil erosion processes are well documented. However, the effect of antecedent soil-water content (U+03B8a) on aggregate breakdown, seal formation and subsequent soil erosion is much more disputable as opposing effects have been reported. We conducted lab experiments with a rainfall simulator on a Belgian silt loam soil. The objectives were to determine the effect of U+03B8a on seal formation, runoff and soil loss and to evaluate its effect on an empirical sediment transport equation. Air-dried soil aggregates were subjected to antecedent soil-water contents of 0.04 (air-dry aggregates), 0.12 and 0.19 m3 m-3. No runoff occurred on the soils with highest antecedent soil-water content, highest total runoff values were observed for the intermediate U+03B8a, while intermediate amounts of total runoff were noticed for the air-dry aggregates. Soil loss, however, showed a different trend: highest values were found for the lowest U+03B8a, intermediate values for the intermediate U+03B8a and no soil loss for the highest U+03B8a. We further observed that U+03B8a had no influence on the final runoff rates and on the final infiltration rate through the soil surface. In using a water discharge and stream power equation to predict sediment transport, the intercept and exponent of the regression equations were found to be lower for U+03B8a of 0.19 m3 m-3 compared to U+03B8a of 0.12 m3 m-3, indicating a decreasing erodibility with increasing U+03B8a. We therefore suggest including U+03B8a as an additional variable to assess soil erodibility in deterministic event-based water erosion models.