Quantifying Sediment Sources to the Suspended Load of an Agricultural Stream Using Radioisotopes

Abstract

The goal of this study was to understand better the delivery of sediment to streams in small, intensively agricultural watersheds of the U.S. Midwest by determining the amount of sediment coming from the fields and stream banks during three consecutive runoff events. The natural activities of Beryllium-7 and Lead-210 in different source soils were compared with the corresponding activities of the suspended sediment collected in the stream during these events. Both a simple two end-member mixing model and a Bayesian model were used to determine the relative contributions from the source areas to the suspended load of each event. The two end-member approach suggested that ~60% of the sediment carried in the stream during the first event was eroded upland soils and was attributed to a “first flush” of readily available material from past events. For the second and third events, the amounts of eroded upland soils were ~34% and ~26%, respectively, because less material was readily available for mobilization. The two end-member model results compared favorably with the Bayesian model, which also incorporated Cesium-137 as a third tracer. Additionally, these results were confirmed with the clockwise hysteresis observed in the different events. During the third event, a flash flood, stream bank collapse was observed and bank retreat estimates from multiple methods compared favorably with the partitioning results. Quantifying sediment sources in watersheds will allow land managers to target more accurately areas where Best Management Practices (BMPs) are most needed to control sediment-related problems. INTRODUCTION Quantifying erosion and sediment delivery at the watershed scale has proven difficult due to high temporal and spatial variability of the different erosion processes occurring over a landscape (Church 2006), despite recent progress in understanding the mechanisms of erosion (e.g., Romkens et al. 2002; Govers et al. 2007). This problem is further exacerbated in agricultural areas where anthropogenic activities, including tillage (e.g., Van Oost et al. 2006) and channel straightening (e.g., Urban and Rhoads 2003) are added stressors. The need still exists to identify and quantify 1243 World Environmental and Water Resources Congress 2014: Water without Borders © ASCE 2014