Improvement in soil loss estimation in USLE type experiments

Abstract

Soil erosion experimentation around the world commonly uses field runoff plots from which runoff and soil loss are collected. The volume of the collected water-sediment mixture is usually so great that sub-sampling techniques are required to estimate total soil loss in any erosion event. It has been shown that the commonly used method involving a collection box or tank, agitation of the water-sediment mixture and then sub-sampling can lead to serious under-estimation of total soil loss. The various possible sources of this error are investigated. For the error due to the unavoidable short time delay between completion of sediment stirring and sample collection, a simple practical method of correction based on settling theory is described and illustrated for four different soil types. Application of this correction requires measurement or estimation of the time delay, and of the soil's settling velocity characteristics.