Improved Soil Mixing and Delivery System for a Storm Runoff Simulator

Abstract

An earlier version of a storm runoff simulator to test conservation buffers reproduced target hydrographs and sedigraphs using uniform, fine sand; however, it was unable to uniformly mix and deliver native sediment. The objectives of this work reported were to create a method to process native agricultural sediment, mix a uniform sediment slurry at a target concentration, and create a control system that will deliver the slurry in varying flow rates corresponding to a target sedigraph. Eroded silty clay (14% sand) was scraped, dried, and processed with a hammer mill. A sand (93% sand) and loam (44% sand) were dried and screened for organic debris and large clods. Each soil type was mixed by an axial flow impeller in an 1890-L cone-bottom tank. Recirculation through a trash pump was used to further break down aggregates and maintain a uniform sediment concentration in the tank. A V-Port ball valve under pressurized flow was used to achieve outflow control. The sediment mixing system was capable of producing concentrations within 3.3% of the target concentration with a maximum test concentration of 0.294 kg L -1 with the silty clay soil. Simulated hydrographs had a Nash Sutcliffe Efficiency of 0.998, a Root Mean Square Error of 0.06 L s -1 , and a peak flow rate within 1% of the target flow. Simulated sedigraphs with silty clay had similar performance. Neither the sandy soil, nor the loam, were successfully delivered through the system to match target sedigraphs. The sand could not be uniformly mixed in the tank, but the loam was uniformly mixed to the target concentration.