Field rainfall simulator studies on two clay soils of the Darling Downs, Queensland. III. An evaluation of current methods for deriving soil erodbilities (K factors)

Abstract

Simulated rain has been widely used to derive soil erodibility (K) values for the Universal Soil Loss Equation (USLE). Because of concern that recent work using smaller plots may not give realistic results, this paper considers the effects of plot length and erosion process on values of K derived from rainfall simulator studies. It also highlights problems in the calculation of K from rainfall simulator data, using the factors of the USLE. Rainulator data on slope length/erosion process interactions were used to calculate soil losses and K for plot lengths of 10.7 and 22.5 m tilled up and down the slope, on two soils, both on 4% slope. K showed up to threefold variation with changes in plot length, because different erosion processes contributed to soil loss. The results also showed major differences between single-event and annual average responses of erosion to slope length, leading to the conclusion that the annual average factors of the USLE cannot be used to analyse single-event rainfall simulator data. Instead, rainfall simulator data must be converted to average annual soil losses, which can then be validly analysed, using the factors of the USLE, to derive K. The procedures presently used to calculate annual average soil losses from rainfall simulator data do not take into account erosion process/runoff rate interactions, and are therefore unsatisfactory. Single-event soil loss models may provide a means for producing better estimates of annual average soil losses suitable for the derivation of K.