Soil erodibility, Soil erosion, Soil texture, Natural run off, Universal soil loss equation

Abstract

Soil erodibility is regarded as a key parameter for evaluating the soil’s susceptibility to erosion and is essential for predicting soil loss. To date, the available soil erodibility cal-culation techniques have been widely used around the world. However, it is not well known which of the erodibility calculation techniques is best for a given group of soil at specific geographical location. Likewise, very few investigations have been done so far for the spe-cific situation of soils in Ethiopia. The limited availability of an appropriate method for the estimation of soil erodibility factor-K is the main bottleneck for prediction of a reliable sediment yield. Hence, this study was carried out to investigate the predictive capability of the K-values as estimated by different techniques; to study the interrelationship between different techniques of estimating soil erodibility, and to validate and suggest the best technique(s) for practical applications. Accordingly, different soil erodibility estimation techniques such as Wischmeier and Smith (178), Soil Erodibility Nomograph described by Wischmeier et al. (1971), Shirazi and Boersma (1984), Williams et al. (1984, and based on soil color and class were employed to estimate soil erodibility. The actual soil loss was collected from the runoff test plots and compared to the values of different techniques un-der investigation. The findings of the study show that the determined K value of soil of the study area, i.e., Cambisols found to be 0.11 t h (MJ mm)-1,whereas estimated K values us-ing different soil erodibility estimation techniques range from 0.13 to 0.34 t h (MJ mm)-1. The K value estimated using Shirazi & Boersma (1984) is relatively high, while those of estimated based on soil type and Wischmeier & Smith (1978) are relatively low. Compared to those K values estimated using different soil erodibility estimation techniques, the K value estimated by soil type was generally found to be closure to the measured K value un-der natural run off plots. Therefore, under unavailability of soil physicochemical proper-ties of the soil that can be used in different soilerodibility estimators, K value estimated based on soil typeis more reliable to be used in RUSLE.The study has shown that testing and validation of the estimated soil erodibility using measured soil loss from the standard RULSE runoff plots are critical for unbiased assessment of the magnitude and spatial dis-tribution of soil erosion at the large-scale. However, as this experiment was conducted only for a single season, it is not likely to give definite recommendation; thus, similar experi-ment has to be repeated under the same condition.For all major soils in Ethiopia, different soil erodibility estimator techniqueshave to be validated for estimationof K values to be used in RUSLE model for soil erosion prediction.