JWSS - Journal of Water and Soil Science

Sediment selectivity during transport may provide basic information for evaluating on-site and off-site impacts of the soil erosion. Limited information is, however, available on the selectivity of sediments in rill erosion, particularly in the rainfed furrows. Toward this, the sediment selectivity was investigated in three soil textures (loam, loamy sand sand clay loam) under 10% slope using 90 mm.h-1 rainfall intensity for 40 min. Soil samples were passed from a 10 mm sieve and packed in to the erosion flume with 0.4m × 4 m in dimensions. Particles size distribution (PSD) was determined in the sediments (PSDs) and compared with the original soil PSD (PSDo). The proportion of PSDs and PSDo was stated as PSDs/PSDo to show the selectivity of soil particles by rill erosion. Based on the results, all three soils appeared as the coarse particles (coarse sand and very coarse sand) in sediments with the PSDs/PSDo>1, indicating the higher selectivity of these particles by rill erosion. Loamy sand was the most susceptible soil to rill erosion among the studied soils, which generated a higher runoff (0.0035 m².s^-1) and sediment load (0.1 kg.m-1.s-1) during rainfall. The PSDs of this soil were similar to those of the original soil PSD. This study revealed that the stability of aggregates could be regarded as the major soil factor controlling rill erosion rate and the sediment selectivity in the semi-arid soils. With an increase in the water-stable aggregates, soil infiltration rate and as a consequence, shear stress of flow could be decreased in the rills.

Few studies have been done regarding the role of the raindrop in the hydrodinamic mechanism of soil erosion. In this study, rainfall simulation experiments were conducted to evaluate the role of raindrop in runoff discharge, sediment concentration and hydraulic properties of flow under four slope gradients (5, 10, 15 and 20%) in a clay soil using a 90 mm.h^-1 rainfall intensity to reach the steady state flow. Soil sample was packed into the erosion flume with 0.3m× 0.4m × 4 m in dimensions and tested under two soil surface conditions:  one with raindrop impact and one without raindrop impact. The results showed that runoff discharge, sediment concentration, flow depth, shear stress, stream power, Reynolds number and runoff velocity under without raindrop impact condition were significantly lower than those in the condition  with the raindrop impact with a factor of 0.62 to 3.54, 0.08 to 11.83, 0.91 to 0.96, 0.26 to 3.25, 0.52 to 4.45, and 0.36 to 3.27, 0.23 to 0.79 times, respectively; on the other hand, the Darcy Wysbach, Chezy and Manning coefficients were increased significantly under without raindrop impact (P<0.01). Flow velocity was the key hydraulic parameter strongly affecting the hydraulic properties. These findings indicated the importance of raindrop impact in the detachment rate of soil particles through the change of the hydraulic characteristics. This study also revealed the key role of raindrop impact on the runoff hydraulic characteristics, as well as particle detachments rate in rills. Information about the role of raindrop impact is a substantial step in modeling the rill erosion. Therefore, elimination of raindrops impact, especially in the steep slopes, with the conservation of natural vegetation cover can sufficiently prevent runoff production as well as the particle detachment rate.