4.

Modeling Water and Wind Erosion

Modeling Erosion
Modeling water and wind erosion is important to understanding the processes governing soil
erosion, predicting runoff and soil erosion rates, and identifying or choosing appropriate
measures of erosion control. Modeling permits the: (1) understanding of the driving processes,
(2) evaluation of on-site and off-site impacts on soil productivity and water and air pollution on
large scale, (3) identification of strategies for erosion control, and (4) assessment of the
performance of soil conservation practices for reducing water and wind erosion. Well-developed
and properly calibrated models provide good estimates of soil erosion risks. Soil erosion results
from a complex interaction of soil-plant-atmospheric forces. Thus, modeling soil erosion
requires a multidisciplinary approach among soil scientists, crop scientists, hydrologists,
sedimentologists, meteorologists, and others. Models must be able to integrate processes, factors
and causes at various spatial and temporal scales.
Numerous models of differing prediction capabilities and utilities have been developed. The
advent of technological tools such as remote sensing and GIS has significantly enhanced the
usefulness of soil erosion models. The coupling of GIS and remote sensing with empirical and
process-based soil erosion models has improved their predictive capability. The GIS stores the
essential database needed as input for modeling erosion and elaboration of maps of erosion-
affected areas. Remote sensing is, for example, useful to estimate land cover over large
geographic areas, which is a critical input for modeling erosion. Remote sensing and GIS tools
also allow the scaling up of modeled data from small plots (e.g., USLE) to large areas. Modeling
soil erosion involves integration of complex and variable hydrological processes across large
areas to understand the magnitude of soil erosion. There are empirical and process-based models
to estimate soil erosion at various scales (e.g., plot, watershed, and field).

Universal Soil Loss Equation (USLE)

The USLE developed in the USA is the most widely used empirical model worldwide for
estimating soil loss (Wischmeier and Smith, 1965). Information from the USLE is used in
planning and designing conservation practices. This model is not strictly based on hydraulic
principles and soil erosion theory. It thus simplifies the processes of soil erosion. The USLE was
specifically intended to predict soil loss from cultivated soils under specific characteristics. It has
1
sometimes been used inappropriately and applied to soil and land use conditions different from
those for which it was developed. It provides a long-term annual average estimate of soil loss
from small plots or field segments with defined dimensions. The USLE was developed from
measured data rather from physically-based modeling approaches. The limited consideration of
all the complex and interactive factors and processes of soil erosion with the USLE limits its
applicability of USLE to all conditions.
The USLE is, however, advantageous over sophisticated models because it is simple, easy to use,
and does not require numerous input parameters or extensive data sets for prediction. The
simplicity of the equation for its practical use has sacrificed accounting for all the details of soil
erosion. Parameters are estimated from simple graphs and equations.
The average annual soil loss is estimated as
A = R × K × LS × C × P
where A is average annual soil loss (Mg ha−1), R is rainfall and runoff erosivity index for the
location of interest, K is erodibility factor, LS is topographic factor, C is cover and management
factor, and P is support practice factor. The early versions of USLE were exclusively solved
using tables and figures (e.g., nomographs).
The continued improvement has resulted in MUSLE and Revised USLE (RUSLE 1 and 2).
Rainfall and Runoff Erosivity Index (EI)
The EI is computed as the product of total storm energy (E) times the maximum 30-min intensity
(I30) of the rain.
E I = E × I30
The USLE uses the annual EI which is computed by adding the EI values from individual storms
that occurred during the year. According to Wischmeier and Smith (1978), the EI corresponds
closely with the amount of soil loss from a field. The EI as used in the USLE overestimates the
EI for tropical regions with intensive rains. The USLE-computed EI is only valid for rain
intensities ≤63.5mmh−1. Modifications to EI have been proposed for tropical regions (Lal,
1976). The 30-minute intensity for a given storm and location is obtained from rain gauge charts
recording the 30-minute with the largest amount of rainfall. Data on R for different locations of
the continental USA and estimates for the world are available (Foster et al., 1981). In the USA,
about 4000 sites were analyzed for their rainfall intensities for a range of rain-return periods to

2
develop an iso-erodent map (Fig. 4.1). Values of EI30 below 50 correspond to dry regions (e.g.,
Great Plains) and those above 500 correspond to humid regions.
Soil Erodibility Factor (K)
Soil erodibility refers to soil’s susceptibility to erosion. It is affected by the inherent soil
properties. The K values for the development of USLE were obtained by direct measurements of
soil erosion from fallow and row-crop plots across a number of sites in the USA primarily under
simulated rainfall. The K values are now typically obtained from a nomograph (Foster et al.,
1981) or the following equation:
K = 0.00021 × M1.14 × (12 − a) + 3.25 × (b − 2) + 3.3 × 10−3(c − 3)
100
where M is particle-size parameter, a is % of soil organic matter content, b is soil structure code
(1 = very fine granular; 2 = fine granular; 3 = medium or coarse granular; 4 = blocky, platy, or
massive), and c profile permeability (saturated hydraulic conductivity) class [1 = rapid
(150mmh−1); 2 = moderate to rapid (50–150mmh−1); 3 = moderate (12–50mmh−1); 4 = slow to
moderate (5–15mmh−1); 5 = slow (1–5mmh−1); 6 = very slow (<1mmh−1)]. The size of soil
particles for very fine sand fraction ranges between 0.05 and 0.10 mm, for silt content between
0.002 and 0.05, and clay <0.002 mm. The soil organic matter content is computed as the product
of percent organic C and 1.72.

Topographic Factor (LS)

The USLE computes the LS factor as a ratio of soil loss from a soil of interest to that from a
standard USLE plot of 22.1m in length with 9% slope as follows:

where S is field slope (%) and θ is field slope steepness in degrees

Cover-Management Factor (C)

The C-factor is based on the concept that soil loss changes in response to the vegetative crop
cover during the five crop stage periods: rough fallow, seedling, establishment, growing, and
maturing crop, and residue or stubble. It is computed as the soil loss ratio from a field under a
given crop stage period compared to the loss from a field under continuous and bare fallow
conditions with up- and down-slope tillage (Wischmeier and Smith, 1978). Crop type and tillage

3
method, the two sub-factors defining the C, are multiplied to compute the C-values. Estimates of
C values for selected vegetation types are shown in Table 4.1. Detailed calculations of C values
are presented by Wischmeier and Smith (1978).
Table. C values for some tillage and cropping systems (After Wischmeier and Smith, 1978)
Vegetation Description C values
Grain corn Moldboard plow, no residues, plowed during:
– fall 0.40
– spring 0.36
Mulch tillage 0.24
Chisel plow, >50% residue cover, spring plowing 0.20
Ridge tillage 0.14
No-till with 100% residue cover 0.05
Corn silage and beans Moldboard plow, no residues, plowed during:
– fall 0.50
– spring 0.45
Mulch tillage 0.30
Ridge tillage 0.17
No-till with 100% residue cover 0.10
Cereals Fall plowed 0.35
Spring plowed 0.32
Mulch tillage 0.21
Ridge tillage 0.12
No-till with 100% residue cover 0.08
Corn-soybean rotation Moldboard plow, no residues, fall plowing 0.50
Chisel plow, >50% residue cover, spring plowing 0.23
No-till with 100% residue cover 0.05
Corn-soybean rotation Moldboard plow, no residues, fall plowing 0.20
Chisel plow, >50% residue cover, spring plowing 0.14
No-till with 100% residue cover 0.05
Hay and pasture Dense stand of sod-like grass 0.02
Forest >90% canopy cover and 100% litter cover 0.001
Short and managed trees At least 75% of canopy cover without litter cover 0.35
without understory
vegetation (fruit trees)
At least 75% of canopy cover with about 30% 0.08
litter cover
Support Practice Factor (P)
The P-factor refers to the practices that are used to control erosion. It is defined as the ratio of
soil lost from a field with support practices to that lost from a field under

Modified USLE (MUSLE)

The MUSLE is a modified form of USLE. While USLE predicts sediment yield based on
rainfall, MUSLE predicts it by using runoff factor, which accounts for the antecedent soil water
content. This modification allows the use of USLE for predicting sediment loss on a storm event
basis.

where sed is sediment yield on a storm event basis (Mg), Q is surface runoff volume (mm), qp is
peak runoff (m3 s−1), A is area of the hydrologic response unit (HRU) (ha), and CFRG is coarse
fragment factor, which is estimated as
4
where rock is % rock in the uppermost soil layer.
Revised USLE (RUSLE)
This model is more comprehensive and detailed than USLE and is based on empirical- and
process-based approaches (Renard et al., 1997). As compared to USLE, it includes more EI
values for the western U.S. in addition to those in the eastern U.S. It incorporates soil processes
(e.g., freezing and thawing) and changes in water content into the USLE. It uses computer tools
to calculate complex LS interactions based on rill and interrill erosion relationships and
incorporates information on canopy and surface residue cover and the effects of temperature and
soil water on above- and below-ground residue decomposition at short time (1/2 month)
intervals. In USLE, the C values are calculated from tables with data from field experiments, but
RUSLE computes these values from four sub-factors, which are the following:
 prior land use (PLU) factor which accounts for the amount and biomass and tillage
practices from previous years,
 the canopy (CC) factor accounting for the vegetative cover,
 the surface cover (SC) factor that reflects the amount of residue mulch left on the soil
surface, and
 surface roughness (SR) factor.
The RUSLE accounts for the influence of farming across slopes as well as stripcropping and
buffer strips within the P factor. The P values are estimated based on slope length and steepness,
ridge height, soil deposition, soil infiltration, and the cover and roughness conditions. Friendly-
user online assessments of soil loss for RUSLE are available to estimate soil loss by simply
entering the county name, slope, length, and soil series name, and crop rotation of the cropped
field. RUSLE1 and RUSLE2 compute transport capacity (Tc) as

where kt is transport capacity that depends on the hydraulic resistance of soil surface roughness
and vegetative cover, and qp is runoff rate, and θ is angle of the slope. Sediment deposition (D)
is estimated as

where is Vf is fall velocity of the sediment and g is sediment load.

5
Modeling Wind Erosion
Similar to those for water erosion, a number of empirical and physically-based models exist for
predicting wind erosion. Models vary in rigor and strictness with which factors and processes are
considered. The available models are under continuous refinement to incorporate the complex
and variable parameters that govern wind erosion. Most of the current knowledge on the
dynamics and mechanics of wind behavior for the development of models comes from the work
done in the deserts of North Africa by Bagnold (1935). The first investigations dealt with why
and how sand particles accumulated in dunes and what interactive mechanisms occurred between
wind blowing and flying soil particles. Climate and soil surface characteristics were recognized
as the first drivers of wind erosion early in research. Chepil and Milne (1941) and Chepil (1945a,
b, c) expanded the theoretical basis on the mechanics of wind erosion. Chepil (1959) proposed a
generalized equation to estimate wind erosion as

where I is soil cloddiness factor, R is crop residue factor, K is ridge roughness equivalent factor,
F is soil abradability factor, B is wind barrier factor, W is width of field factor, and D is wind
direction factor.
One of the simplest empirical equations was developed by Pas´ak (1973) as
E = 22.02 − 0.72P − 1.69V + 2.64Rr (4.29)
where E is erodibility (kg ha−1), P is percent of non-erodible fraction of soil, V is relative soil
moisture, and Rr is wind velocity (km h−1). This model has limited use because does not
incorporate variables for vegetative surface cover and soil roughness.

Limitations of Water and Wind Models

While the available models have advanced the understanding of soil erosion processes and
estimation of erosion rates, their applicability to conditions different from those for which they
were developed remains limited. The large and detailed data required as input for most of the
current models are seldom available. Performance of current models is highly variable and site-
specific. Further model development of process-based models for a wide range of soil,

6
management and climate scenarios is warranted. Model domains must incorporate the temporal
and spatial variability of conditions.
Example 1. A 130m long field with 5% slope is under continuous corn managed with chisel
plowing in eastern Ohio. The soil is silt loam (10% coarse and medium sand, 10% very fine
sand, 20% clay, and 60% silt) with 2.5% of soil organic matter content. The structure is fine
granular and the saturated hydraulic conductivity is 40mmh −1. Estimate the average annual soil
loss if the field is contoured and strip cropped with no terraces.
Rainfall Erosivity, R = 2100, a = 2.5, b = 3 and c = 3,