An examination of the relationship between erodibility parameters and soil strength

Erosion rate of soil by the impact of raindrops and overland flow of water is often considered to be affected by the shear strength of surface soil. Physically based erosion models indicate a link between defined erodibility parameters and soil strength. The objectives of this paper are to determine erodibility parameters with the process-based erosion model GUEST for a. krasnozem soil of two contrasting strengths, and to examine the influence of soil strength on erodibility parameters. Soil beds of width 1 m and length 5.8 m, with and without compaction, were exposed to simulated, constant rate rainfall. A range of slopes was used. Detachment trays of width 300 mm and downslope length 200 mm containing soils of identical strength were placed at the same slope and exposed to the same rain in order to determine the effects of rainfall-driven processes alone on erosion. Soil strength was measured with a hand vane tester and a pocket penetrometer to determine whether compaction was effective in modifying soil strength. Temporal variation in sediment concentrations (c) for the large soil beds and detachment trays was measured for each slope and soil strength. The settling velocity characteristic of soil, with and without exposure to rain, was determined with the modified bottom withdrawal tube technique. Values of c decreased with increase in soil strength. The relationship between c and slope was influenced by soil strength in a manner consistent with the theoretical expectation of the role of soil strength in controlling erosion. Rilling during erosion was absent only when the soil was compacted. The average settling velocity of the soil exposed to rain (i.e. its depositability) was significantly lower than for the same soil not subjected to rain, indicating a breakdown of soil aggregates as a result of raindrop impact. Rainfall detachability parameters (estimated with GUEST) Were lower when soil strength was high. Runoff-driven erodibility parameters, namely the specific energy of entrainment (J), increased and the approximate erodibility parameter (â) decreased with increase in soil strength. The Variation in these erodibility parameters with soil strength was consistent with the theory implemented in GUEST. Detailed analysis of the relative contribution of rainfall- and runoff-driven processes to c at varying stream powers and soil strengths indicated that, at high soil strength, uncertainty in the values of J and â is high because of the higher contribution to c of rainfall-driven rather than runoff-driven processes. The adequacy of in situ measurement of soil strength as an indicator of soil erodibility is discussed in relation to the results presented.