Sediment transport capacity in erodible beds with reconstituted soils of different textures

Precise estimation of sediment transport capacity and wide applicability of a prediction equation could provide a positive response to increasing challenges of soil erosion. Currently, the models developed by many researchers are focused on a few or single soils. In this study, a prediction equation of sediment transport capacity was attempted to establish from the flume experiments performed on erodible beds under the experimental settings of two slopes (5.6%, 16.7%), four unit width discharges (6.67 x 10(-5), 1.33 x 10(-4), 2.00 x 10(-4) and 2.67 x 10(-4) m(2) s(-1)) and five reconstituted soils (sand contents of 0%, 30%, 50%, 70%, 100%). Here, the sediment transport capacity under clay-sand mixture of different soil texture were analyzed, and a prediction equation for sediment transport based on soil cohesion and hydraulic variables was established. The experiments confirmed that the sediment concentration and transport capacity increased with the unit width discharge and slope gradient as a power function, and was more dependent on slope gradient. Besides, the hydrological and erosive processes on the beds with reconstituted soils varied with soil texture, the beds with a high clay content had a stable channel of sediment transport. Comparatively, unit stream power was the most suitable hydraulic variable to describe the sediment transport capacity (r = 0.951, P < 0.01). The introduction of soil cohesion of bed materials further improved the reliability of the prediction equation by combining the soil cohesion and unit stream power as a composite predictor (R-2 = 0.940, N-SE = 0.943, P < 0.01). The new equation could provide dependable prediction for various erodible beds with packed soils of different textures. Soil cohesion could be considered as an indicator of soil properties in prediction equation of sediment transport capacity. Given the broad range of physical properties of reconstituted soils of different textures, the results are applicable to a wide variety of soil types.