Developing empirical relationship between interrill erosion, rainfall intensity, slope gradient and soil types in highlands of Salale, Ethiopia
In order to develop an empirical relationship for interrill erosion based on rainfall intensity, slope steepness and soil types, an interrill erosion experiment was conducted using laboratory rainfall simulator on three soil types (Vertisols, Cambisols and Leptosols) for the highlands of North Shewa Zone of Oromia Region. From simulation work done using 450 x 320 x 100 mm erosion test pan; splashed soils, runoff and washed soils were collected at 5 minute interval for each 15 minutes simulation run at various combinations of design rainfall intensities of 25.67, 52.14, 73.50, and 99.20 mmhr-1, slope steepness of 5, 20, 35, and 50% and the three soil types. Runoff rate, splash and wash loss were measured to obtain a total soil loss from the test pan. Different models relating interrill erosion with rainfall intensity and runoff rate were considered and their coefficients and parameters were estimated. From the models, those with highest prediction potential (R2) were selected to be incorporated into models consisting soil properties and used to form six basic models of which three of them had shown average R2 values of more than 0.95. Accordingly, five slope factor equations were combined with the three selected models and thus 15 models were formed. Out of these models, three of them had shown R2 > 0.90 and were further compared with each other using the graph plotted for observed versus predicted interrill erosion. This showed that model incorporating rainfall intensity, runoff rate, median soil particle diameter, % clay and quadratic form of slope factor carried higher prediction potential than the other two models and selected as the final model for predicting interrill erosion. As this model incorporates more factors than those previously developed models, it ensures more accurate estimation of interrill erosion.
Keywords: Interrill erosion Empirical equations Erosion models
Copyrights belongs to the Science, Technology and Arts Research Journal