Search

Ephemeral gullies (EGs) are important erosion features that could account for as much as 70% of total erosion on the Loess Plateau of China. However, few quantitative studies have been conducted in the field on the impact and mechanism of upslope sediment runoff in EG erosion processes. A series of field rainfall simulation experiments were conducted on an artificial EG to investigate the impacts of upslope inflow with two flow rates (15 and 30 L min−1) and five sediment concentrations (20, 40, 80, 120, and 160 g L−1) on runoff and erosion processes within the EG. The results showed that the lower flow rate of 15 L min−1 increased soil loss by 26% compared with the higher inflow rate (30 L min−1). The lower inflow rate triggered greater sheet or rill erosion, whereas the higher rate mainly strengthened the incision of the main EG channel and led to increasing gully erosion. The channel shoulders expanded laterally very little, especially under the higher flow rate. Inflow sediment concentration had a major influence on EG erosion processes. With increasing sediment concentration, EG erosion and total slope erosion first increased, and then decreased. When sediment concentration exceeded 120 g L−1, the inflow sediment began to be deposited, and 120 g L−1 can be regarded as a critical value of EG erosion. These results are beneficial to establish a soil erosion model and to control ephemeral gully erosion in loess areas.

The effects of vegetation cover on overland flow and erosion processes on hillslopes vary with vegetation type and spatial distribution and the different vegetation components, including the above- and below-ground biomass. However, few attempts have been made to quantify how these factors affect erosion processes. Field experimental plots (5 m × 2 m) with a slope of approximately 25° were constructed and simulated rainfall (60 mm hr−1) (Rainfall) and simulated rainfall combined with upslope overland flow (20 L min−1) (Rainfall + Flow) were applied. Three grass species were planted, specifically Astragalus adsurgens (A. adsurgens), Medicago sativa (M. sativa) and Cosmos bipinnatus (C. bipinnatus). To isolate and quantify the relative contributions of the above-ground grass parts (stems, litter cover and leaves) and the roots to reducing surface runoff and erosion, each of the three grass species was subjected to three treatments: intact grass control (IG), no litter or leaves (only the grass stems and roots were reserved) (NLL), and only roots remaining (OR). The results showed that planting grass significantly reduced overland flow rate and velocity and sediment yield, and the mean reductions were 21.8%, 29.1% and 67.1%, respectively. M. sativa performed the best in controlling water and soil losses due to its thick canopy and dense, fine roots. Grasses reduced soil erosion mainly during the early stage of overland flow generation. The above-ground grass parts primarily contributed to reducing overland flow rate and velocity, with mean relative contributions of 64% and 86%, respectively. The roots played a predominant role in reducing soil erosion, with mean contribution of 84%. Due to the impact of upslope inflow, overland flow rate and velocity and sediment yield increased under the Rainfall + Flow conditions. The results suggest that grass species on downslope parts of semi-arid hillslopes performed better in reducing water and soil losses. This study is beneficial for forage selection, allocation and management practices, such as forage harvesting, when implementing restoration strategies to control soil and water losses.

Limited information has isolated the impacts of rainfall on rill formation and erosion on steep hillslopes where upslope inflow simultaneously exists. Field simulation experiments were conducted on steep hillslopes (26°) under rainfall (60 mm h −1 ), inflow (6, 12, 18, 24, 30, 36 L min −1 m −1 ), and combination of rainfall and inflow to explore the impacts of rainfall on rill formation, and the interaction between rainfall and inflow on soil erosion. Rainfall decreased soil infiltration rate (10%–26%) mainly due to soil crust by raindrop impact. Rainfall strengthened rill formation, which behaved in the increment in rill width (5%–26%), length (4%–22%), and depth (3%–22%), but this increment decreased as inflow rates increased. Additionally, the contribution of rainfall on rill formation was most significant at the initial stage, followed by the final stage and active period of rill development. Rainfall increased rill erosion (8%–80%) and interrill erosion (36%–64%), but it played a dominant role in increasing interrill erosion under relatively high inflow rates. The most sensitive hydrodynamic parameter to soil erosion was shear stress and stream power under inflow and ‘inflow + rainfall’ conditions, respectively. For the lowest inflow rate, the reduction in soil loss by interaction between rainfall and inflow accounted for 20% of total soil loss, indicating a negative interaction. However, such interaction became positive with increasing inflow rates. The contribution rate to rill erosion by the interaction was greater than that of interrill erosion under relatively low inflow rates. Our results provide a better understanding of hillslope soil erosion mechanism.

Summary Sedimentation resulting from soil erosion degrades the surface water environment. This study determines the sediment trapping capacity (STC) of grass filter strips and how slope gradient and vegetation cover affect it. STC is defined as the maximum sediment trapped within vegetative filter strips (VFS) under given conditions. An exponential model was derived to mirror the relationship between instantaneous sediment trapping efficiency and runoff duration, with findings that STC is closely related to the initial efficiency and attenuation coefficients. The model effectively describes the sediment trapping process, and demonstrates that STC decreases with increasing slope and increases with vegetation cover. Grass strips have a small sediment trapping thickness measured in mm or cm, and grass stems and leaves have little influence on STC, which indicates that reasonable forage cutting may have little influence on sediment removal. STC is an intrinsic characteristic of VFS, and it can be recommended to assess the VFS performance in trapping sediment in severe soil erosion areas. A sediment trapping modulus at the watershed scale based on STC, can offer help to effectively evaluate soil and water conservation engineering and transformation of cropland into forest and grassland in China.

Evaluation of grass influence on soil erosion process can provide important information in soil and water conservation. The laboratory experiment was conducted to study runoff and sediment producing processes and runoff hydraulics in the grassplots with different covers (35%, 45%, 65% and 90%) and bare soil plot (control) at a slope of 15. The results showed that grass significantly reduced runoff and sediment. Compared with bare soil plot, the grassplots had a 14-25% less runoff and an 81-95% less sediment, and played a more important role in reducing sediment at the final stage of rainfall. There was a significantly negative logarithmic relationship between sediment yield rate (SDR) and cover (C): SDR = 1.077-2.911 ln(C) (R-2 = 0.999'"). Sediment yield rate of grassplots decreased with rainfall duration, and decreased linearly as runoff rate increased. Overland flow velocities deceased with increase in grass cover, and the cover had greater effect on lower slope velocity than upper one. Froude numbers decreased with increase in cover, and flow regimes of all treatments were laminar and tranquil. Darcy-Weisbach and Manning friction coefficients of grassplots increased as ground cover increased. Therefore, increase in grass coverage can efficiently reduce soil loss and improve ecological environments. (c) 2006 Elsevier B.V. All rights reserved.