Comparing grain size composition of inter-rill and rill-eroded sediment from cultivated hillslope soils using caesium-134 and cobalt-60 as tracers

Understanding the extent to which textural characteristics of inter-rill and rill-eroded sediment differ across a range of agricultural soils is important in order to more accurately predict sediment transport capacities and thus increase the reliability of erosion models. However, a lack of consensus still exists over how much influence these two mechanisms exert on the physical characteristics of sediment during an erosion event. This communication reports results from a prototype tracing technique using two different artificial radionuclides to compare the grain size composition of inter-rill and rill-eroded sediment from eight different cultivated hillslope soils from southern England. A 1.0 m * 0.2 m * 0.11 m soil box was filled with soil labelled with caesium-134 (134Cs) to a depth of 0.1 m. A second layer of soil labelled with cobalt-60 (60Co) was distributed across the first soil layer, to a depth of 0.01 m, and lightly textured to replicate a fine tilth. The soil box was pitched at a ca. 17% slope (9.5°) and each soil was subjected to ca. 35 mm hr−1 of simulated rainfall until surface runoff was initiated and a rill developed. All runoff-eroded sediment was collected over 90 s intervals, dried and radiometrically assayed. Changes in radiometric signals were used to identify the transition from inter-rill to rill erosion, demarking a shift in sediment-source. Samples recording highest activity of either radionuclide were deemed to contain largest contributions of sediment mobilised by each respective erosion process and were selected for grain size analysis. Grain sizes, measured at 10 percentile intervals, were subjected to a paired-sample statistical analysis to determine whether size-difference was significant. Rill-eroded sediment was significantly coarser than inter-rill eroded sediment for four soils, whereas inter-rill eroded sediment was significantly coarser than rill-eroded sediment for one soil. There was no significant size-difference for two soils, and one soil was omitted from statistical analysis due to excessive contamination by the sub-dominant erosion process.