developed in response to the limited applicability of models such as CREAMS for
analysing the effects of soil management practices such as tillage or fallow management
strategies (Littleboy et al. 1996). Models such as CREAMS calculate runoff as a function
of rainfall and soil water content, excluding surface and crop cover changes resulting
from tillage practices. PERFECT was designed to predict runoff, erosion and crop yield
for some management options in dryland cropping areas of Australia, including sequences
of plantings, harvests and stubble management during fallows (Littleboy et al. 1996).
The model comprises six modules: data input, water balance, crop growth, crop residue,
erosion and model output. These modules are arranged in a framework that allows
alternative modules to be used as required for the potential range of applications. The
modules draw on other models such as MUSLE and CREAMS. Erosion is simulated in
the model using MUSLE, while the mineral nitrogen removed from the topsoil by
erosion is simulated using the following relationship taken from CREAMS:
SEDN = SOIL
where SEDN is the mineral nitrogen lost in the sediment, SOIL is the daily erosion (kg
), MNIT is the mineral nitrogen in the topsoil (kg kg
) and ENR is the enrichment
ratio (Littleboy et al. 1992b).
The inputs to the models are daily climate data, soil parameters, cropping sequence
criteria (i.e. crop type and length of fallow), crop growth parameters and fallow
management (tillage) options. The climate data requirements include daily rainfall, pan
evaporation, temperature and evaporation.
Littleboy et al. (1992b) found that PERFECT was more reliable than CREAMS in
predicting runoff, accounting for 77%–89% of the variation in daily runoff volume. This,
in addition to the consideration of crop cover and surface runoff on infiltration and soil
evaporation, indicates that PERFECT is a more appropriate model to analyse runoff from
cropping systems with complex crop/fallow rotations than the CREAMS model.
Although PERFECT was not developed specifically as a water quality model, the
incorporation of a runoff component in addition to the large crop component of the
model may provide an advantage over models such as CREAMS, where the major
emphasis is placed on surface hydrology, sediment and pesticide movement, and nutrient
models with little or no accounting for land management practices.
The major disadvantage with PERFECT in terms of water quality and erosion modelling
is that it does not contain a sediment transport or nutrient component. However, the
structure of the model is such that a hydrological component of the model may be
incorporated. Additionally, the erosion component of the model does not account for
rainfall intensity, thus raising the possibility for overestimation or underestimation of
erosion depending on the rainfall event. Although the model structure is generally robust,
Littleboy et al. (1992a) noted that the model was not designed for application beyond
those environments typical of north-eastern Australia and recommended that the model
be calibrated against suitable field data before use in any other environment.
In summary, PERFECT provides a potentially valuable tool for assessing conservation
cropping options by simulating the water balance, crop yield and erosion for
combinations of soil type, climate, fallow management strategy and cropping sequence.
The incorporation of a sediment transport and nutrient component would be required for
the model to be useful in water quality modelling. If this were to occur, the detail of the
crop cover and management components may provide an advantage over other models,
where these processes are considered important.
Examples of Model Users:
Queensland Department of Primary Industries
IBM PC, UNIX
The model, including source code, is available free of charge.
For further information contact:
Dr. Mark Littleboy
Phone: (07) 3896 9593
3.2.15 STARS and IHACRES
The Solute Transport with Advection, Resuspension and Settling (STARS) model was
developed at the Integrated Catchment Assessment and Management Centre at the
Australian National University. It is a one-dimensional model of advective transport
between two gauging stations or nodes given flow at both nodes (Green et al. 1999;
Dietrich et al. 1999).
The STARS model is conceptually based, and requires upstream and downstream
concentration over some period (including a few events) for calibration of the model
parameters. The model has only five parameters and is thus less likely to experience
problems with model identifiability than more complex conceptual and physics-based
The model simulates processes such as particle settling, deposition and resuspension of
sediment as well as lateral sources of sediment from bank erosion and sediment inputs
associated with local rainfall. The model compensates for differences in flow between
upstream and downstream nodes by computing an average flow rate over the reach, Q
The scaled equation for downstream suspended sediment concentration, c
, as a function
of time is:
is the concentration upstream,
is the effective water parcel travel time
estimated from the data, and
Deposition is controlled by the particle settling velocity via
, lateral sources are given
and resuspension is determined by a combination of
. These are the five
parameters requiring calibration in the model.
When it is necessary to model streamflow, the IHACRES model is used (Jakeman et al.
1994a, 1994b, 1990; Evans and Jakeman 1997) for predicting discharge at catchment
outlets, and a simple discharge routing model is used for instream sections. The
IHACRES rainfall-runoff model is a hybrid metric–conceptual model based on the
instantaneous unit hydrograph. It was developed by the Centre for Resource and
Environmental Studies with the Institute of Hydrology. This model accounts for the
effects of evapotranspiration, drainage and precipitation on rainfall-runoff. Rainfall is
modified using temperature data to reflect the effects of drainage, evapotranspiration and
antecedent weather conditions to become effective rainfall. This effective rainfall is then
modelled as passing through one or two internal reservoirs or storages. The exact number
of storages used is determined by the calibration data. IHACRES has been widely applied
within Australia and overseas in a range of climatic conditions. It has been shown to
predict runoff as effectively as other models but has the advantage of containing only 5–7
parameters. It has been augmented with power law relations between sediment
concentrations and discharge (and between phosphorus and sediment concentrations) to
predict water quality concentrations. This has been successfully prototyped in several
catchments of the Namoi Basin (Jakeman et al. 1999).
The STARS and IHACRES models have the advantage of requiring relatively little input
data, as the conceptual nature of the models means that spatially distributed input data on
catchment characteristics is not required for model calibration. The small number of
model parameters also means that the models are less likely to suffer from problems of
identifiability than more complex models.
STARS and IHACRES were developed in Australia, and as such are applicable to
Australian conditions. STARS has been applied to catchments in the Namoi,
Murrumbidgee and Murray River Basins (Green et al. 1997; Dietrich and Jakeman 1997).
Examples of Model Users:
Centre for Resource and Environmental Studies (CRES), Integrated Catchment
Assessment and Management Centre (ICAM).
The STARS model is not commercially available. IHACRES is available for
approximately £300 from:
Institute of Hydrology, Maclean Building
Wallingford OX10 8BB UK
Phone: +44 1492 83 8800
For further information contact:
Professor Tony Jakeman
(02) 6249 4742
THALES is a hydrological model that uses TAPES-C, a set of computer programs that
allypartitions automatically subdivides the model area into interconnected irregular-
shaped elements and calculates a number of topographic attributes for each element
(Moore and Grayson 1993). The THALES model is event-based and models runoff and
subsurface flow (Hatton et al. 1998). The model has the potential to be incorporated into
a sediment and nutrient transport model where the simulated flow characteristics of the
catchment would be used to calculate soil movement or nutrient transport (Grayson and
THALES is a relatively simple physics-based model that enables a wide range of hydrologic
processes to be represented through the incorporation of the Hortonian mechanism of surface
runoff as well as a representation of variable-source-area runoff and exfiltration of
subsurface flow (Grayson et al. 1992a). The elemental structure of THALES allows each
element to have different infiltration, surface flow and subsurface flow parameters, although
parameters are generally measured for each soil type or region of different surface conditions
and it is assumed that these do not vary within each region or soil type. Grayson et al.
(1992b) note that assumptions underlying models such as THALES are extensive and occur
at all levels from the overall model structure to the constituent algorithms. As a
consequence, there is a danger in using this model out of context.
THALES has been developed as an investigative tool and has proved useful in the
analysis of catchment response. Moore et al. (1991) stated that a common deficiency of
many hydrologic or water quality models is their inability to represent the effects of
three-dimensional terrain on flow processes without a large number of often unrealistic
assumptions. In this way, THALES, with the use of TAPES-C, provides an advantage
over models that do not account for three-dimensional terrain. At present, THALES has
been used mainly in research and will require further development to incorporate a water
quality component prior to use by catchment managers.
Examples of Model Users:
Department of Civil Engineering, University of Melbourne
DEM component (TAPES) requires UNIX with X Windows graphics (written in Fortran-
77 and C). Not compiled under DOS or Windows.
TAPES-C, THALES, and TAPES-G (grid-based version of TAPES) available from
For further information contact:
Dr. Rodger Grayson
Phone: (03) 9344 7305
3.2.17 USLE and modifications
The Universal Soil Loss Equation (USLE) is a soil erosion prediction model used widely
within the USA and worldwide, either on its own or incorporated into such models as
AGNPS. Developed in the 1970s by the US Department of Agriculture, the model has
undergone much research and a number of modifications (e.g. MUSLE, USLE-M). The
model has also been upgraded to take into account additional information that has
become available since the development of the USLE (RUSLE). The basic USLE is an
empirical overland flow or sheet–rill erosion regression equation based primarily on
observations (Zhang et al. 1995). Although USLE is an empirical model, it has some
conceptual components. The model relates sediment delivery to slope, slope length,
rainfall, erosivity and soil erodibility, of which the latter two are predicted both
empirically and conceptually.
The USLE estimates the average annual soil loss from:
A = R K L C S P
where A is the soil loss averaged over slope length, R is the combined erosivity of rainfall
and runoff, K is the soil erodibility, L is the factor dependent on slope length, S is the
factor dependent on slope gradient, C is dependent on vegetative cover and management
and P is dependent on conservation practices (Zhang et al. 1995). The simplicity of this
equation and the availability of parameter values, at least in the USA, has made this
model relatively easy to use (Loch and Rosewell 1992).
There are a number of limitations to the USLE equation. The model is not event-based
and as such cannot identify those events most likely to result in large-scale erosion. Gully
erosion and mass movement are not considered in the erosion process, and the deposition
of the sediment is not considered to occur within the area under consideration (Zhang et
al. 1995). Runoff leaving a field generally concentrates in a few major channels, the
profile of which is often concave, such that ephemeral gully erosion can occur along the
upper reach of the channel and deposition occurs in the lower reaches of the channels.
This gully erosion can be as extensive as sheet and rill erosion (Lane et al. 1992).
Additionally, unlike in the USA, the use of USLE in Australia has been limited by the
perceived lack of data for the parameters required to run the model under Australian
conditions (Loch and Rosewell 1992). Nearing et al. (1994) noted that the adaptation of
USLE to a new environment requires a large investment of time and resources to develop
the database required to run the model. Evans et al. (1992) identified that due to rainfall
variability, data must be collected for at least 10 years and this, combined with the lack
of data for overburden spoil and replaced spoils to be applied to USLE, was a
disadvantage for the use of this model in spoil pile erosion prediction.
Due to the identified limitations of USLE, a number of modifications to the basic format
for have been proposed in the literature. These include the Modified USLE, the Revised
USLE (Renard and Ferreira 1993; Renard et al. 1994), the USLE-M (Kinnell 1998a;
Kinnell 1998b; Kinnell and Risse 1998) and SOILOSS (Rosewell 1995; Rosewell and
Lang 1996). These continue to improve components of the model, tending to make it
more process-based. RUSLE maintains the basic form of the USLE, although all
equations used to arrive at the factor values have been modified (Lane et al. 1992).
Changes to the form of the LS factor in RUSLE enables the prediction of soil loss due to
Hortonian overland flow in three-dimensional terrain with convergent and divergent
slopes (Ryan and McKenzie 1997). USLE-M, for example, provides a more complex
representation of processes than the USLE as it more directly considers the effect of
runoff on erosion with changes to the R factor (Kinnell 1998b). Consequently, USLE-M
has a greater ability to account for the more frequent small to medium erosion losses.
Kinnell (1998a) noted that ‘USLE technology will form the basis of modelling the spatial
and temporal variation in soil erosion within catchments in the future and as such there
are benefits in continued improvements in the model’.
The SOILOSS computer program is a local adaptation of RUSLE, being adapted to NSW
conditions through the estimation of the rainfall erosivity and soil erodibility factors from
local rainfall erosivity maps or calculated from rainfall intensity data and soil landscape
maps respectively (Rosewell and Lang 1996). The map required for the estimation of
rainfall erosivity factors can be obtained from Rosewell and Turner (1992). The program
applies the USLE and is used to assist in the selection of land and crop management
practices to decrease erosion (Rosewell 1995). The SOILOSS program has been used
extensively by the Soil Conservation Service, now the NSW Department of Land and
Water Conservation, to estimate water pollution hazard for Water Pollution
Requirements (Rosewell and Lang 1996). This was achieved by combining the site-
specific factors of R, K and S for a fixed slope length of 20 m and a P factor of 1. The
cover management factor, C, is calculated based upon average soil loss levels following
specific logging operations (Rosewell and Lang 1996). Factor C measures the combined
effect of all the interrelated cover and management variables and is defined as the ratio of
soil loss from land managed under specified conditions to the corresponding loss from
clean-tilled continuous fallow (Rosewell 1997).
The main advantage of RUSLE, on which SOILOSS is based, over the USLE is that the
RUSLE has the capacity to estimate the C factor from information on vegetation form,
decay and tillage practices rather than from experimental plot data as used in the USLE.
Another advantage of SOILOSS over other USLE-based alternatives is that it is
applicable to Australian conditions and should thus be more reliable in erosion
predictions. On the other hand, the SOILOSS program is still a non-event-based
prediction equation (Kinnell 1996), and as discussed previously may not be as useful a
management tool as an event-based predictive model. However, Rosewell (1995) noted
that a combination of SOILOSS with AGNPS is capable of indicating the relative
differences in nutrient generation between alternative land and crop management
practices. The incorporation of SOILOSS into models in place of USLE would be
expected to improve the validity of the model predictions under Australian conditions.
Examples of Model Users (SOILOSS):
NSW Department of Land and Water Conservation; NSW Department of Housing; NSW
Environment Protection Authority; State Forests of NSW; NSW Agriculture
Available from Publication Sales, Department of Land and Water Conservation, GPO
Box 39, Sydney, NSW 2001. (Cost in 1997 $100 for single user and $200 for multiple
For further SOILOSS information contact:
Mr. C.J. Rosewell
Phone: (02) 6742 9505
The Watershed Erosion Prediction Project (WEPP) is a physics-based, hillslope-scale
model developed in the USA in an initiative between the Agricultural Research Service,
the Soil Conservation Service, the Forest Service in the Department of Agriculture and
the Bureau of Land Management in the US Department of the Interior (Laflen et al.
1991). The model has been applied widely to hillslopes in the US (e.g. Laflen et al. 1991)
and worldwide, including Australia (e.g. Fogarty 1997). The model was intended to
determine and/or assess the essential mechanisms controlling erosion by water, including
anthropogenic impacts (Zhang et al. 1995; Liu et al. 1997).
Like many physics-based models, WEPP is based on a mass balance formulation, one of
the standard equations used in physics-based models:
is the sediment rate per unit width of rill channel, D
is the rill net
detachment or deposition rate, and D
is the interrill net detachment or deposition rate
(Zhang et al. 1995). Being a physics-based model, the computational requirements of
WEPP are high, with a large number of inputs required. The processes represented by
WEPP can be broadly characterised as erosional processes, hydrological processes, plant
growth and residue processes, water use processes, hydraulic processes and soil
processes (Laflen et al. 1991). Erosional processes are limited to sheet and rill erosion
and erosion occurring in channels where detachment is due to hydraulic shear. Through
the erosional components of the model, the three stages of erosion (detachment, transport
and deposition) are quantified using the rill–interrill concept of describing sediment
detachment (Laflen et al. 1991; Lane et al. 1995), which is the detachment and transport
of sediment through raindrop impact and shallow flows.
Originally, interrill detachment was modelled in WEPP as:
is the interrill detachment rate, K
is the interrill erodibility constant and I is the
rainfall intensity. Following work by Kinnell (1993a, b), the I
term was replaced by the
product of runoff and intensity in the 1995 release of WEPP.
Rill detachment is modelled using the relationship:
is the detachment capacity of clear water, K
is the rill erodibility of soil due to
is the shear below which there is no detachment and
is the hydraulic
shear of flowing water, where
is the density of water, r
is the hydraulic radius and s is the hydraulic gradient,
which is approximately equal to the slope of the rill bottom.
The erosional processes result from the forces and energies developed in hydrologic
processes (Laflen et al. 1991). The components of the hydrologic processes are climate,
infiltration and a winter component that accounts for snow accumulation and melt.
Knowledge of plant growth and residue components is required to make an accurate
assessment of the plant and residue characteristics above and below the soil. These
include canopy cover and height, above- and below-ground biomass of living and dead
plant material, leaf area index and basal area, and are estimated on a daily basis (Laflen et
al. 1991). As such, information regarding dates and management practices are essential
inputs to the model. The plant characteristics are of utmost importance to describe
adequately as they will have a large impact on the soil erosion and hydrological processes
within the site.
The soil water status is updated on a daily basis and is required to obtain infiltration and
surface runoff volumes—the driving force in the detachment by flowing water in rills and
channels (Laflen et al. 1991). The water balance component uses information about
climate, plant growth and infiltration to estimate daily potential evapotranspiration and
soil and plant evaporation.
Documents you may be interested
Documents you may be interested