Professional Documents
Culture Documents
Soil Erosion: A Review of Models and Applications
Soil Erosion: A Review of Models and Applications
Abstract— Soil erosion is a global environmental agriculture and also the natural environment (Shougang,
problem influenced by both natural and human factors. Na and Ruishe, 2014). Soil erosion has become one of the
Modeling provides a quantitative and consistent global environmental hazards that limits today’s human
approach to estimate soil erosion and sediment yield survival and restricts global socio-economic sustainable
under a wide range of conditions, and is needed to guide development (Han, Ren, Zhang and Li, 2016). Land
the comprehensive control of soil erosion. Over the years degradation due to erosion processes incurs substantial
various soil erosion models have been developed. The costs both for individual farmers and for society as a
application of these models is dependent on the soil type whole (Phai, Orange, Migraine, Toan and Vinh, 2006).
and climate of the given area because models differ in With growing pressure on natural resources and
complexity and input requirements. This paper reviews landscapes, there is an increasing need to predict the
various soil erosion models and their applications, consequences of any changes to the environment
focusing more on the most widely applied models which (Shougang et al, 2014). They further stated that modelling
are: Universal Soil Loss Equation (USLE), Revised plays an important role in this by helping our
Universal Soil Loss Equation (RUSLE) and Water understanding of the environment and by forecasting
Erosion Prediction Project (WEPP). The method used for likely impacts. Soil erosion models are useful to estimate
this research is a review of academic articles, bulletins, soil loss and runoff rates from agricultural land, to plan
conference papers, textbooks, research reports and land use strategies, to provide relative soil loss indices
publicly available materials on soil erosion models and and to guide government policy and strategy on soil and
their applications. The results of this study revealed that water conservation (Smith, 1999). Effective modelling
most soil erosion models have been developed for the can provide information about current erosion, its trends
assessment of rill and interill erosion at plot or catchment and scenario analysis (Ganasri and Ramesh, 2016).
scale on agricultural lands and watersheds in terms of Soil erosion prediction technology began over 70 years
estimating mostly soil loss, sediment yield, erodibility (K) ago, but it was in 1965 that the work expanded into the
values, rainfall factor (R) factors, runoff rates and Universal Soil Loss Equation (USLE) developed by
forecasts of likely impacts. Again, the study indicated that Wischmeier and Smith, perhaps the foremost achievement
most previous authors on soil erosion assessment used the in soil erosion prediction (Laflen and Flanagan, 2013).
empirical models due to their limited data and parameter Since then several models have been developed to
inputs. Recommendations of this study include: (1) simulate soil erosion prediction process. They all consider
expansion of the USLE and RUSLE models for the slope steepness, slope length, vegetative cover, rainfall,
simulation of gully erosion and sediment processes; (2) soil properties and erosion control methods as parameters
researchers should be encouraged through grants to which influence erosion (Smith, 1999). Erosion models
develop empirical models (that make use of limited data) utilize the various factors that affect erosion to simulate
based on rainfall (R) factor and erodibility (K) factor that erosion processes in order to predict the levels of erosion
provide two opposing forces in soil erosion processes; in a region (Anejionu, Nwilo and Ebinne, 2013). They
and (3) management of soil erosion based on the opined that insights could be drawn from present and
indigenous knowledge of the affected people and land future trends of erosion impacts in a region with these
holders. models. Various studies on erosion models have clearly
Keywords— Applications, Environmental Sustainability, demonstrated that the dominant factor contributing to
Models, Review, Soil, Erosion, USLE. sediment discharge is the erosive power of rainfall (Phai
et al, 2006).
I. INTRODUCTION Soil erosion models fall into three main categories,
Soil erosion is one of the most serious environmental depending on the physical processes simulated by the
problems in the world today because it threatens model, the model algorithms describing these processes
4.1 Universal Soil Loss Equation (USLE) the USLE. As with most empirical models, the USLE is
The USLE is an empirical soil model developed by not event responsive, providing only an annual estimate
Wischmeier and Smith, (1978). Originally, USLE was of soil loss as it ignores the processes of rainfall, runoff,
developed mainly for soil erosion estimation in croplands and how these processes affect erosion, as well as the
or gently sloping topography (Ganasri and Ramesh, heterogeneities in inputs such as vegetation cover and soil
2016). The USLE quantifies soil erosion as the product of types (Merritt et al, 2003). They asserted that the model
six factors representing rainfall and runoff erosivity (R), is not event-based and as such cannot identify those
soil erodibility (K), slope length (L), slope steepness (S), events most likely to result in large-scale erosion.
cover and management practices (C), and supporting Applying the equation to purposes for which it was not
conservation practices (P) (Renard and Freimund, 1994). intended, however, cannot be recommended (Wischmeier
This empirical equation is based on the statistical analysis 1978). Since it was designed for interrill and rill erosion,
of more than 10,000 plot-years of data of sheet and rill it should not be used to estimate sediment yield from
erosion on plots and small watersheds (Roose, 1977). The drainage basins or to predict gully or stream-bank erosion
equation is: (Morgan, 2005). He reported that care should be taken in
A = RKSLCP using it to estimate the contribution of hill slope erosion
in which erosion (A) is the estimated soil loss per unit to basin sediment yield because it does not estimate
area, R is the rainfall-runoff erosivity ∑factor, K is the deposition of material or incorporate a sediment delivery
soil erodibility factor, L is the slope length factor, S is the ratio. In his opinion, since the equation was developed to
slope steepness factor, C is the cover management factor, estimate long-term mean annual soil loss, it cannot be
and P is the supporting practices factor (Wischmeier and used to predict erosion from an individual storm.
Smith, 1978).
The model predicts rainfall based on rainfall erosivity (R 4.2 The Revised Universal Soil Loss Equation
factor) and soil erodibilitty (K factor). Bols (1978) (RUSLE)
proposed a formular for calculating the R factor in The RUSLE has been revised to more accurately estimate
Indonesia in a model: soil loss from both crop and rangeland areas (McCool,
2.5𝑃2 Foster, Renard, Yoder, and Weesies, 1995). The RUSLE
R=
100(0.073𝑃+0.73)
maintains the basic structure of the USLE but is a
where P = Annual precipitation in millimetres and R is in computerized version that incorporates the results of
MJmmha-1hr-1yr-1 additional research and experience obtained since the
The soil erodibility index is calculated with the following 1978 publication of USLE by Wischmeier and Smith
equation (Roose, 1977): (Renard and Friedmund, 1994). The equation is:
A
K= A = R.K.L.S.C.P
R x SL x 2.24
where A is the erosion in tons per hectare, R is the rainfall where A is the computed soil loss, R is the rainfall-runoff
erosivity index, SL is the topographic factor, and 2.24 the erosivity factor plus a factor for any significant runoff
coefficient necessary to go from metric units (t/ha) to from snow melt expressed in MJ mm ha-1h-1yr-1; K is the
English units (t/acre). soil erodibility factor – the soil-loss rate per erosion index
Although the simplicity of this equation and the unit for a specified soil as measured on a standard plot
availability of parameter values have made this model which is defined as a 72.6-ft (22.1m) length of uniform
relatively easy to use, there are a number of limitations to 9% slope in continuous clean-tilled fallow expressed in t
V. RESULTS AND DISCUSSION Renard and Frerimund (1994) agreed that the most
As a result of problems arising from soil erosion and land accurate estimate of R-values for both USLE and RUSLE
degradation, various models have been developed for can only be obtained from long-term rainfall intensity
estimation and simulation of soil erosion. Based on the data. In the opinion of Ganasri and Ramesh (2016),
review of previous studies on soil erosion modes and RUSLE is sensitive to land use - land cover as result
applications, the following results were obtained as obtained matched reasonably with observed data.
summarised in Table 2. Although simple to apply, USLE In their application of the WEPP model, Han et al. (2016)
and RUSLE are empirical models and therefore cannot be observed that the WEPP-simulated runoff and sediment
used to simulate erosion on an event basis. Roose (1977) yield predictions were relatively consistent with the
reported that USLE predicts sheet and rill erosion on hilly measured values at slope scale but at watershed scale both
slopes and approaches neither the problem of flow nor the simulated values of runoff and erosion were higher
that of transport in solution and neglects the qualitative than the measured. Pieri et al. (2006) held the same view
aspect of eroded materials. Present USLE soil loss that WEPP could adequately simulate the water balance
equations are substantially less accurate for prediction of for the model plot but further stated that comparison
specific events than for the prediction of long time between WEPP-simulated and field measured sediment
averages (Wischmeier and Smith, 1978). Similarly, yields suggested that WEPP tends to under-predict