Raindrop‐impact‐induced erosion processes and prediction: a review

Raindrop‐impact‐induced erosion is initiated when detachment of soil particles from the surface of the soil results from an expenditure of raindrop energy. Once detachment by raindrop impact has taken place, particles are transported away from the site of the impact by one or more of the following transport processes: drop splash, raindrop‐induced flow transport, or transport by flow without stimulation by drop impact. These transport processes exhibit varying efficiencies. Particles that fall back to the surface as a result of gravity produce a layer of pre‐detached particles that provides a degree of protection against the detachment of particles from the underlying soil. This, in turn, influences the erodibility of the eroding surface. Good understanding of rainfall erosion processes is necessary if the results of erosion experiments are to be properly interpreted. Current process‐based erosion prediction models do not deal with the issue of temporal variations in erodibility during a rainfall event or variabilities in erodibility associated with spatial changes in dominance of the transport processes that follow detachment by drop impact. Although more complex erosion models may deal with issues like this, their complexity and high data requirement may make them unsuitable for use as general prediction tools. Copyright © 2005 John Wiley & Sons, Ltd.     

Simulations demonstrating interaction between coarse and fine sediment loads in rain‐impacted flow

Rain‐impacted flows dominate sheet and interrill erosion and are important in eroding soil rich in nutrients and other chemicals which may have deleterious effects on water quality. Erosion in rain‐impacted flow is associated with raindrop detachment followed by transport either by the combination of flow velocity and raindrop impact (raindrop‐induced flow transport, RIFT) or the inherent capacity of the flow to transport detached material. Coarse particles tend to be transported by RIFT, while fine particles tend to be transported without any assistance from raindrop impact. Because the transport process associated with coarse particles is not 100 per cent efficient, it generates a layer of loose particles on the soil surface and this layer protects the underlying soil from detachment. Simulations were performed by modelling the uplift and downstream movement of both fine and coarse particles detached from the soil surface by individual raindrop impacts starting with a surface where no loose material was present. The simulations produced a flush of fine material followed by a decline in the discharge of fine material as the amount of loose material built up on the bed. The decline in the discharge of fine material was accompanied by an increase in the discharge of coarse material. The relative amounts of coarse and fine material discharged in the flow varied with flow velocity and cohesion in the surface of the soil matrix. The results indicate that the discharge of various sized sediments is highly dependent on local soil, rain and flow conditions and that extrapolating the results from one situation to another may not be appropriate. Copyright © 2006 John Wiley & Sons, Ltd.     

Sediment delivery from hillslopes and the Universal Soil Loss Equation: some perceptions and misconceptions

The Universal Soil Loss Equation (USLE) or the revised USLE (RUSLE) are often used together with sediment delivery ratios in order to predict sediment delivery from hillslopes. In using sediment delivery ratios for this purpose, it is assumed that the sediment delivery ratio for a given hillslope does not vary with the amount of erosion occurring in the upslope area. This assumption is false. There is a perception that hillslope erosion is calculated on the basis that hillslopes are, in effect, simply divided into 22·1 m long segments. This perception fails to recognize the fact the inclusion of the 22·1 m length in the calculation has no physical significance but simply produces a value of 1·0 for the slope length factor when slopes have a length equal to that of the unit plot. There is a perception that the slope length factor is inappropriate because not all the dislodged sediment is discharged. This perception fails to recognize that the USLE and the RUSLE actually predict sediment yield from planar surfaces, not the total amount of soil material dislocated and removed some distance by erosion within an area. The application of the USLE/RUSLE to hillslopes also needs to take into account the fact that runoff may not be generated uniformly over that hillslope. This can be achieved by an equation for the slope length factor that takes account of spatial variations in upslope runoff on soil loss from a segment or grid cell. Several alternatives to the USLE event erosivity index have been proposed in order to predict event erosion better than can be achieved using the EI₃₀ index. Most ignore the consequences of changing the event erosivity index on the values for the soil, crop and soil conservation protection factors because there is a misconception that these factors are independent of one another. Copyright © 2007 John Wiley & Sons, Ltd.     

Applying the QREI30 index within the USLE modelling environment

The USLE/RUSLE model was designed to predict long‐term (~20 years) average annual soil loss by accounting for the effects of climate, soil, topography and crops. The USLE/RUSLE model operates mathematically in two steps. The first step involves the prediction of soil loss from the ‘unit’ plot, a bare fallow area 22.1 m long on a 9% slope gradient with cultivation up and down the slope. Appropriate values of the factors accounting for slope length, gradient, crops and crop management and soil conservation practice are then used to adjust that soil loss to predict soil loss from areas that have conditions that are different from the unit plot. Replacing EI₃₀, the USLE/RUSLE event erosivity index, by the product of the runoff ratio (Q_R) and EI₃₀, can enhance the capacity of the model to predict short‐term soil loss from the unit plot if appropriate data on runoff is available. Replacing the EI₃₀ index by another index has consequences on other factors in the model. The USLE/RUSLE soil erodibility factor cannot be used when the erosivity factor is based on Q_REI₃₀. Also, the USLE/RUSLE factors for slope length, slope gradient crops and crop management, and soil conservation practice cannot be used when runoff from other than the unit plot is used to calculate Q_R. Here, equations are provided to convert the USLE/RUSLE factors to values suitable for use when the erosivity factor is based on the Q_REI₃₀ index under these circumstances. At some geographic locations, non linear relationships exist between soil loss from bare fallow areas and the Q_REI₃₀ index. The effect of this on the slope length factor associated with the Q_REI₃₀ index is demonstrated using data from runoff and soil loss plots located at the Sparacia site, Sicily. Copyright © 2012 John Wiley & Sons, Ltd.     

Accounting for the influence of runoff on event soil erodibilities associated with the EI30 index in RUSLE2

Soil erodibilities (K) associated with the EI₃₀ index vary not only with soil properties but also with soil moisture as it varies in time and space. In Revised Universal Soil Loss Equation Version 2 (RUSLE2), temporal variations in soil erodibility in the USA are calculated using monthly precipitation and temperature as independent variables. K_UM, the soil erodibility factor associated with the Q_REI₃₀ index, varies independently of runoff and the product of K_UM and the runoff ratio for the unit plot (Q_R1) provides an alternative to the temporally varying Ks currently used in predicting storm soil loss in RUSLE2. Comparisons were made between the product of Q_R1 and K_UM and RUSLE2 Ks for representative storms at four locations representing the north to south variation in climate in the USA. Peak erosion associated with the current approach used in RUSLE2 was slightly higher at two locations and slightly lower at the other two locations. One other location, Morris, MN, provided an exception with the peak loss predicted by using the product of Q_R1 and K_UM being 1.7 times that obtained using RUSLE2 Ks. In theory, average annual K_UM values should be better related to soil properties than the average annual values of K frequently used when the average annual values of EI₃₀ are used to predict soil loss. However, work has yet to be performed to determine how K_UM varies directly with soil properties and in space and time. Copyright © 2014 John Wiley & Sons, Ltd.     

Evidence of iceberg-ploughing in a subaqueous ice-contact fan, glacial Lake Rinteln, NW Germany

The Coppenbrügge subaqueous ice‐contact fan complex of early Saalian age is about 10 km long and up to 10 km wide and is composed of offset‐stacked fan clinothems that are transgressive‐regressive sequences formed during an overall lake level rise. The individual fan bodies consist of coarse gravel in the ice‐proximal part, passing distally into sandy facies and showing large‐scale foreset bedding. The iceberg scour recognized in an open‐pit outcrop is up to 1.5m deep, up to 2m wide and cut in undisturbed mid‐fan deposits. The scour‐fill can be traced laterally for about 15m and consists of sheared sand and, in the frontal zone, of downbent overlying strata surrounded by a zone of deformed sediments. The deformed sediment produced by the iceberg keel's shearing of the trough walls is a sand mass containing angular soft‐sediment clasts that show internal folds and fractures. The basal surface of the deformed sediment is a nearly horizontal shear plane, steepening up laterally as a discrete thrust and showing a flat‐ramp‐flat geometry. The scour was formed by the iceberg keel's ploughing the substrate and pushing the sediment sideways and frontally, forming a ridge of deformed sediments at the trough end. This ridge was concurrently eroded by an accompanying meltwater underflow which apparently developed a horseshoe system of scouring vortices around the grounded iceberg. The current's scour was filled with massive, non‐stratified sand deposited rapidly from turbulent suspension. The iceberg eventually broke up and its keel part was buried. As these ice fragments gradually melted, the space was closed by normal faulting and downbending of overlying strata. The collapsing scour‐fill became partly liquified, and the resulting water‐escape structures cut the normal faults and the overlying deposits. Though produced chiefly by tangential shear strain, iceberg‐ploughing features are readily distinguishable from other glaciotectonic deformations. They can serve as a diagnostic criterion for glaciolacustrine or glaciomarine environments and the distinguishing of ice‐contact subaqueous fans from ice‐contact deltas in the stratigraphic record.     

Modelling event soil losses using the QREI30 index within RUSLE2

Recently, a USDA Curve Number‐based method for obtaining estimates of event runoff has been developed for use in enhancing the capacity of Revised Universal Soil Loss Equation (RUSLE2) to deal with runoff‐driven phenomena. However, RUSLE2 still uses the EI₃₀ index as the basis for determining the erosivity of the rainfall for sets of runoff producing storms at a location even though the product of the runoff ratio (Q_R) and EI₃₀ index is better at prediction event erosion when runoff is known or predicted well. This paper reports the results of applying the Q_REI₃₀ index using data available from tables within RUSLE2 to predict storm event soil losses from bare fallow areas and areas with continuous corn at Holly Springs, MS, and Morris, MN. In RUSLE2, all rainfall during a calendar year is considered to detach soil material that is flushed from the area if and when runoff occurs. However, the Q_REI₃₀ index is calculated using the EI₃₀ value for the amount of rain in the storm that produces runoff. Consequently, changes were made to the timing of events during the calendar year in order to meet the criteria for using the Q_REI₃₀ index. As a general rule, the peak event soil loss produced using the Q_REI₃₀ index were higher than produced by RUSLE2, and the peak event soil loss for the bare fallow occurred later than for the continuous corn. The results of the work reported here show that the Q_REI₃₀ index may be used to model event erosion produced by a set of storms within RUSLE2 provided that the appropriate mathematical rules upon which the USLE was developed are adhered to. Copyright © 2013 John Wiley & Sons, Ltd.     

Runoff dependent erosivity and slope length factors suitable for modelling annual erosion using the Universal Soil Loss Equation

Despite revisions and refinements, the Revised Universal Soil Loss Equation (RUSLE), which is the revised version of the Universal Soil Loss Equation (USLE), over predicts small annual soil losses and under predicts large annual soil losses. To some large extent, this results from the equation over estimating small event soil losses and under estimating large event soil losses. Replacing the USLE/RUSLE event erosivity index (EI₃₀) by the product of EI₃₀ and the runoff ratio (Q_R) significantly reduces the errors in estimating event erosion when runoff is measured, but the USLE‐M, the USLE variant that uses the Q_REI₃₀ index, requires crop and support practice factors that differ from those used in the RUSLE. The theory which enables the Q_REI₃₀ index to be used in association with the RUSLE crop and support practice factors is presented. In addition, the USLE/RUSLE approach was developed for conditions where runoff is produced uniformly over a hill slope. A runoff dependent slope length factor that takes account of runoff variations over a hill slope is presented and demonstrated for the situation where runoff from a low runoff producing area passes onto an area where runoff is produced more readily. Copyright © 2007 John Wiley & Sons, Ltd.     

Modeling of the effect of flow depth on sediment discharged by rain‐impacted flows from sheet and interrill erosion areas: a review

Sediment, nutrients and pollutants discharged from sheet and interrill erosion areas by rain‐impacted flows may influence water quality in streams and rivers. The depth of water on the soil surface influences the capacity of raindrop impacts to detach soil material underlying rain‐impacted flows, and a number of so‐called process‐based and mechanistic models erroneously use equations on the basis of the effect of water depth on splash erosion to account for this effect. Also, a number of these models require complex mathematical solutions to make them operate and can only predict sediment composition and discharges well if many of their parameters are calibrated specifically to the situations where they are being applied. Experiments with rain‐impacted flows, where flow depth and velocity over eroding surfaces have been controlled, have been reported in the literature and provide more appropriate equations to account for the drop size – flow depth interactions that affect detachment and transport of particles in rain‐impacted flows. There is a need to develop modeling approaches that rely on relevant data obtained under well‐controlled flow conditions where flow depths and velocities are known. Copyright © 2012 John Wiley & Sons, Ltd.