Soil erodibility factors for water erosion risk estimation in south-central Chile

Abstract

When estimating the maximum potential average annual soil loss due to sheet and rill erosion, the Revised Universal Soil Loss Equation (RUSLE) is usually selected because it is a simple model to operate if the various parameters are known. In RUSLE, the soil erodibility factor K is traditionally estimated from the nomogram developed by Wischmeier and Smith (1978), which is based on soil properties (texture, organic matter, permeability, and structure). Unfortunately it has been found that the utility of this nomogram is largely limited when soil organic matter content is higher than 4%, which is common condition in many soils in Chile. For those soils where the organic matter content is higher than 4%, different methods need to be used. This paper describes a theoretical study conducted in order to generate a soil erodibility map for most of the cultivated land in Chile. The study area was approximately 840 km long by 100 km wide, and included more than 500 soil series. Soil erodibilities were calculated based on soil survey data using two different methods: the Wischmeier and Smith’s nomogram and a method based only on the soil particle size distribution. The erodibilities estimated with both methods were compared and when considerable differences were found, the estimates were investigated based on soil properties, and spatial distribution. When erodibility values were correlated with the soil organic matter content, a gradual transition between the estimates of both methods was observed for soils with around 4% organic matter. This transition suggests that it is suitable to use Wischmeier and Smith’s nomogram for soils with less than 4% organic matter and the particle size distribution method for soils with more than 4%.