M. A. Nearing

A Process-Based Soil Erosion Model for USDA-Water Erosion Prediction Project Technology

ABSTRACT Amodel was developed for estimating soil erosion by water on hillslopes for use in new USDA erosion prediction technology. Detachment, transport, and deposition processes were represented. The model uses a steady-state sediment continuity equation for predicting rill and interrill processes. Net detachment in rills is considered to occur when the hydraulic shear stress of flow exceeds the critical shear stress of the soil and when sediment load in a rill is less than the sediment transport capacity. Net deposition is calculated when the sediment load is greater than the transport capacity. Rill detachment rate is dependent upon the ratio of sediment load to transport capacity, rill erodibility, hydraulic shear stress, surface cover, below ground residue, and consolidation. Rill hydraulics are used to calculate shear stresses and a simplified transport equation, calibrated with the Yalin transport equation, is used to compute transport capacity in rills. Interrill erosion is represented as a function of rainfall intensity, residue cover, canopy cover, and interrill soil erodibility. The model has capabilities for estimating spatial distributions of net soil loss and is designed to accommodate spatial variability in topography, surface roughness, soil properties, hydrology, and land use conditions on hillslopes..

Hydraulics and erosion in eroding rills

Rills often act as sediment sources and the dominant sediment and water transport mechanism for hillslopes. Six experiments were conducted on two soils and a uniform sand using three experimental methodologies. The results of this study challenge the assumption often used in hydrologic and erosion models that relationships derived for sheet flow or larger channel flow are applicable to actively eroding rills. Velocity did not vary with slope, and Reynolds number was not a consistent predictor of hydraulic friction. This result was due to interactions of slope gradient, flow rate, erosion, and the formation of rill roughness, bed structures, and head cuts. A relationship for rill flow velocities was proposed. Stream power was found to be a consistent and appropriate predictor for unit sediment load for the entire data set, while other hydraulic variables were not. The data for stream power and sediment load fit the form of a logistic curve (r2 = 0.93), which is promising relative to recently proposed erosion models which are based on probabilistic particle threshold theory.

Slope Gradient Effects on Soil Loss for Steep Slopes

Data for assessing the effects of slope gradient on soil erosion for the case of steep slopes are limited. Widely used relationships are based primarily on data that were collected on slopes up to approximately 25%. These relationships show a reasonable degree of uniformity in soil loss estimates on slopes within that range, but are quite different when extrapolated beyond the range of the measured data. In this study, soil loss data from natural runoff plots at three locations on the loess plateau in China were used to assess the effect of slope gradient on soil loss for slopes ranging from 9 to 55% steepness. Plot size at each location was 5 m wide by 20 m long, and the soils were silt loams or silty-clay loam. The results indicated that for these plots, soil loss was linearly related to the sine of the slope angle according to the equation: S = 21.91 sinq – 0.96, where q is the slope angle and S is the slope steepness factor normalized to 9%. This relationship was assessed in terms of the limited existing experimental data for rainfall erosion on steep gradients and found to be reasonable for data collected on longer plots, but somewhat different than the data from shorter plot studies. The results of this study would indicate a lesser soil loss at high slopes than does the relationship used in the Universal Soil Loss Equation, but a greater soil loss than predicted by the Revised Universal Soil Loss Equation for steep slopes.

EVALUATION OF WEPP AND ITS COMPARISON WITH USLE AND RUSLE

The USDA-Water Erosion Prediction Project (WEPP) demonstrates a new generation of water erosion prediction technology for use in soil and water conservation planning and assessment. The WEPP computer model is based on various interacting natural processes in hydrology, plant sciences, soil physics, and erosion mechanics. The model offers several advantages over existing erosion prediction technology and has the capability of accommodating spatial and temporal variability in topography, soil properties, cropping and management, and sediment detachment and deposition. The model has a wider range of applicability as it accounts for most of the variables affecting runoff and erosion processes. However, considerable validation is required to assess the reliability of predictions obtained from the model. Sixteen-hundred plot years of natural runoff plot data were used for verification and validation of WEPP, including most of data used to develop Universal Soil Loss Equation (USLE). WEPP predictions of soil loss from natural runoff plots at 20 different locations were compared to measured data and existing technology (i.e., USLE and RUSLE). WEPP recorded a model efficiency of 0.71 compared to 0.80 and 0.72 for the USLE and RUSLE respectively. While the USLE and RUSLE did exhibit better model efficiency than WEPP this could be attributed to availability of more refined and site specific input parameters for the empirical models.

Soil erosion by surface water flow on a stony, semiarid hillslope

Soil erosion on hillslopes occurs by processes of soil splash from raindrop impacts and sediment entrainment by surface water flows. This study investigates the process of soil erosion by surface water flow on a stony soil in a semiarid environment. A field experimental method was developed whereby erosion by concentrated flow could be measured in predefined flow areas without disturbing the soil surface. The method allowed for measurements in this study of flow erosion at a much wider range of slopes (2·6 to 30·1 per cent) and unit discharge rates (0·0007 to 0·007 m2 s−1) than have been previously feasible. Flow velocities were correlated to discharge and hydraulic radius, but not to slope. The lack of correlation between velocity and slope might have been due to the greater rock cover on the steeper slopes which caused the surface to be hydraulically rougher and thus counteract the expected effect of slope on flow velocity. The detachment data illustrated limitations in applying a linear hydraulic shear stress model over the entire range of the data collected. Flow detachment rates were better correlated to a power function of either shear stress (r2 = 0·51) or stream power (r2 = 0·59). Published in 1999 by John Wiley & Sons, Ltd.

The WEPP Watershed Model : I. Hydrology and Erosion

The Water Erosion Prediction Project (WEPP) watershed scale model is a continuous simulation tool that extends the capability of the WEPP hillslope model to provide erosion prediction technology for small cropland and rangeland watersheds. The model is based on fundamentals of erosion theory, soil and plant science, channel flow hydraulics, and rainfall-runoff relationships, and contains hillslopes, channels, and impoundments as the primary components. The hillslope and channel components can be further divided into hydrology and erosion components. Channel infiltration is calculated by a Green-Ampt Mein-Larson infiltration equation. A continuous channel water balance is maintained, including calculation of evapotranspiration, soil water percolation, canopy rainfall interception, and surface depressional storage. The channel peak runoff rate is calculated using either a modified Rational equation or the equation used in the CREAMS model. Flow depth and hydraulic shear stress along the channel are computed by regression equations based on a numerical solution of the steady state spatially varied flow equations. Detachment, transport, and deposition within constructed channels or concentrated flow gullies are calculated by a steady state solution to the sediment continuity equation. The impoundment component routes runoff and sediment through several types of impoundment structures, including farm ponds, culverts, filter fences, and check dams. The purpose of this article is to provide an overview of the model conceptual framework and structure. In addition, detailed mathematical representations of the processes simulated by the channel hydrology and erosion components are presented. The processes simulated by the impoundment component are not described in this article, but it does include impoundment effects on watershed model channel peak discharge and time of concentration calculations.

Evaluation of WEPP Runoff And Soil Loss Predictions Using Natural Runoff Plot Data

Model testing and evaluation are critical to the acceptance of any new prediction tool. This study was conducted to evaluate the overall performance of the Water Erosion Prediction Project (WEPP) hillslope model in predicting runoff and soil loss under cropped conditions. Natural runoff plot data, including 4,124 selected events, 556 plot years, and 34 cropping scenarios, from eight locations were selected. The average length of record for the cropping scenarios was about nine years. Several common crops and tillage systems were included. The WEPP input files for soil, slope, climate, and crop management were compiled based on measured data. The coefficient of determination (r2) between model-predicted and measured-runoff volumes for optimized Green and Ampt hydraulic conductivity (Kb) was 0.77 for selected events, 0.76 for annual values, and 0.87 for average annual values; the r2 between predicted and measured soil losses (excluding fallow and corn plots at Bethany, Mo.) was 0.36, 0.60, and 0.85, respectively. Similar predictions of runoff and soil loss were also obtained with WEPP internally estimated Kb values. Runoff and soil loss were slightly overpredicted for small storms and for years with low runoff and soil loss rates, and were underpredicted for large storms and for years with high runoff and soil loss rates. However, average runoff and soil loss rates for different cropping and management systems were adequately predicted. The accuracy and reliability of the predictions were shown to improve from an event to annual to average annual basis. Results of this study show that the WEPP model is a useful tool for predicting runoff and soil loss rates under cropped conditions.

Conservation practices to mitigate and adapt to climate change

Climate change, in combination with the expanding human population, presents a formidable food security challenge: how will we feed a world population that is expected to grow by an additional 2.4 billion people by 2050? Population growth and the dynamics of climate change will also exacerbate other issues, such as desertification, deforestation, erosion, degradation of water quality, and depletion of water resources, further complicating the challenge of food security. These factors, together with the fact that energy prices may increase in the future, which will increase the cost of agricultural inputs, such as fertilizer and fuel, make the future of food security a major concern. Additionally, it has been reported that climate change can increase potential erosion rates, which can lower agricultural productivity by 10% to 20% (or more in extreme cases). Climate change could contribute to higher temperatures and evapotranspiration and lower precipitation across some regions. This will add additional pressure to draw irrigation water from some already overexploited aquifers, where the rate of water recharge is lower than the withdrawal rates. These and other water issues exacerbated by climate change present a serious concern because, on average, irrigated system yields are frequently double those of nonirrigated systems. The…

Detachment of Undisturbed Soil by Shallow Flow

portance of the roles played by raindrop impact and overland flow (Gilley and Finkner, 1985; Bradford et Quantification of soil detachment rates is necessary to establish a al., 1987), the effect of flow depth and sediment load basic understanding of soil erosion processes and to develop fundamental-based erosion models. Many studies have been conducted on on splash (Hirschi and Barfield, 1988; Kemper et al., the detachment rates of disturbed soils, but very little has been done 1985), and transport capacity (Guy et al., 1987; Kinnell, to quantify the rates of detachment for natural soil conditions. This 1993) have been simulated and analyzed. The relationstudy was conducted to evaluate the influence of flow discharge, slope ship between soil detachment by raindrop impact, raingradient, flow velocity, shear stress, stream power, and unit stream drop size and mass, drop velocity, kinetic energy, soil power on detachment rates of natural, undisturbed, mixed mesic typistrength, water drop impact angle, and surface sealing cal Udorthent soil. Flow rates ranged from 0.25 to 2.0 L s 1 and slope have also been investigated (Nearing and Bradford, gradient ranged from 8.8 to 46.6%. This study was compared with a 1985; Bradford et al., 1987; Sharma et al., 1991; Sharma previous study that used disturbed soil prepared by static compression. et al., 1993; Cruse et al., 2000). These experiments have The results indicated that the detachment rates of disturbed soil were contributed to the better understanding of the mecha1 to 23 times greater than the ones of natural undisturbed soil. It was necessary to use natural undisturbed soil samples to simulate the nism of soil detachment by raindrop impact and prodetachment process and to evaluate the influence of hydraulic paramevided a basis for models for interrill areas (Gilley and ter on detachment rate. Along with flow rate increasing, detachment Finkner, 1985; Sharma et al., 1991; Sharma et al., 1995). rate increased as a linear function. Detachment rate also increased Detachment of cohesive soils by shallow clear-water with slope gradient, but the functional relationship between the two flow under laboratory conditions has received less attenvariables depended on flow rate. Stepwise regression analysis indition (Nearing et al., 1991). Detachment by overland cated that detachment rate could be well predicted by a power function flow occurs when the stress or energy applied by the of flow rate and slope gradient (R2 0.96). Mean flow velocity was overland flow is great enough to pull the soil particles closely correlated to detachment rate (r2 0.91). Flow detachment away from the bulk material. Shear stress ( ), stream rate was better correlated to a power function of stream power (r2 power ( ), and unit stream power (P) are normally 0.95) than to functions of either shear stress or unit stream power. used hydraulic parameters to simulate detachment rate in rills, which given the functions as follow: S erosion has been defined as the process of deghS [1] tachment and transportation of soil material by erowhere (Pa) is shear stress, (kg m 3) is water mass sive agents (Ellison, 1947). Soil detachment is the subdensity, g (m s 2) is the gravity constant, h (m) is the process of dislodgment of soil particles from the soil depth of flow, and S (fraction) is the tangent value of mass at a particular location on the soil surface. The bed slope degree. dislodgment is caused by the forces applied on the soil particles by the erosive agents, which are mainly rainV ghSV [2] drops and overland flow (Owoputi and Stolte, 1995). In process-based soil erosion models, the sediment source where (kg m 3) is stream power, V (m s 1) is mean is conceptually separated into that from interrill and rill flow velocity. areas. In interrill areas, dominant processes are detachP VS [3] ment by raindrop impact and transport by raindropimpacted shallow flow. In rills, dominant processes are where P (m s 1) is unit stream power. It is clear that detachment and transport by concentrated flow (Huang shear stress, stream power, and unit stream power are et al., 1996). Therefore, understanding of the detachfunctions of flow depth, velocity, and slope gradient. ment mechanisms for both interrill and rill areas is necTherefore, through combinations of different slope graessary for the development of process-based erosion dients, flow rates, and flow depths, the relationship bemodel. tween soil detachment rate and these hydraulic parameDetachment by raindrop impact has been studied in ters can be derived based on the data from hydraulic detail during the past several decades. The relative imflume studies. Lyle and Smerdon (1965) were among the first to use a hydraulic flume to investigate the relaG. Zhang, B. Liu, and X. He, Dep. of Resources and Environmental tionship between soil erosion and flow shear stress unSciences, Beijing Normal Univ., Beijing, 100875, China; M.A. Nearing, der constant slope. The results revealed a unique relaUSDA-ARS Southwest Watershed Research Center, Tucson, AZ tionship for a given soil type. 85719; G. Liu, Institute of Soil and Water Conservation, Chinese Academy of Sciences and Ministry of Water Resource, Yangling, Nearing et al. (1991) conducted a series of experiShaanxi, 712100, China. G. Zhang, currently at: CREST, Japan Science ments in a hydraulic flume with varying bed slope to and Technology Corp., Japan. Received 6 Sept. 2002. *Corresponding investigate the relationship between soil detachment by author (zhang@urban.env.nagoya-u.ac.jp). shallow flow, flow depth, bed slope, and mean weight diameter of the aggregates with small, statically comPublished in Soil Sci. Soc. Am. J. 67:713–719 (2003).

Rill erosion and morphological evolution: A simulation model

A mathematical model is advanced to simulate dynamically and spatially varied shallow water flow and soil detachment, transport, and deposition in rills. The model mimics the dynamic process of rill evolution, including variable rates of sediment redistribution along the bed and changes in local bed morphology. The sediment source term in the model uses a point scale, probabilistic relationship based on turbulent flow mechanics and a recently developed sediment transport relationship for rills based on stream power. The interdependent feedback loops between channel bed morphology, local flow hydraulics, and local scour and deposition, within the framework of the full hydrodynamic equations with inertial terms, constitute a mathematical model with the capacity to represent spatial variability and temporal evolution of the rill. Finite elements were applied to numerically solve the hydrodynamic and sediment continuity equations. A series of laboratory flume experiments were performed to evaluate the model. Initial bed slopes were 3, 5, and 7% with step increases of water inflow rates of 7.6, 11.4, and 15.2 L min−1. The soil material used in the flume was a kaolinitic, sandy-clay loam. The rill model equations were solved for increasingly complex cases of spatial and temporal variabilities. The model followed measured patterns of morphological changes as the rill evolved, which suggests that the feedback loops in the model between erosion, bed morphological changes, and hydraulics were adequate to capture the essence of rill evolution.