Geology Reference
In-Depth Information
plant height but in such a way that, for very tall
canopies, the energy can exceed that of the direct
throughfall, whereas ground-level vegetation will
protect the soil completely. The user also has
a choice between a model that simulates the
detachment and transport of soil particles as a
function of the shear stress exerted by the flow,
and one that describes the same processes using
unit stream power. The other fundamental differ-
ence between WEPP and EUROSEM is that the
former operates over a range of time steps from
individual storms to daily. Within each time
step, steady-state conditions are assumed, which
means that each time step is either one of erosion
or deposition of sediment. In contrast, EUROSEM
uses very short time steps (1-2 minutes) and
therefore continuously updates the sediment
concentration in the runoff and the transport
capacity. The latter is assumed to be the sediment
concentration at which erosion and deposition
are in balance. EUROSEM is therefore a dynamic
rather than a steady-state model, which implies
that there is a continuous exchange of soil
particles between the runoff and the soil surface
which controls the sediment concentration in
the flow.
There is no doubt that researchers with an
interest in modelling the processes of erosion
and deposition will develop even more complex
models than WEPP and EUROSEM as they work
towards the goal of a comprehensive description
of the processes and a dynamic simulation of
the factors affecting them. For practical purposes
the model user may well question whether
all this complexity is necessary. It is well known
that in terms of the amount of sediment reaching
the bottom of a hillside or discharging into a
water course, some erosion processes are more
important than others. Indeed, in terms of pro-
ducing a relatively simple and efficient model, it
is recommended that attention is focused on the
most important processes and that those con-
tributing little to the generation, transport and
deposition of sediment should be ignored
(Kirkby, 1980). The user therefore requires some
knowledge of the most important processes that
affect his or her problem so that a model which
emphasizes these can be selected. Even if a very
detailed model is not chosen, the user can gain
much by establishing the conceptual framework
of the problem as fully as possible. By under-
standing the processes involved and their con-
trolling factors, it is possible to decide which are
the most relevant and which models best match
what is required. Without a comprehensive
framework, there is a danger that something
important will go undetected.
2.5 Spatial Considerations
When selecting an erosion model it is necessary to
define the area over which it should operate. This
may vary from a small segment of a hillslope, to a
complete hillside, a small catchment (typically
0.01-0.5 km
2
in area but sometimes as large as
10 km
2
) encompassing hillslopes and a river chan-
nel, or a large catchment (typically 10-100 km
2
but sometimes as large as 100,000 km
2
). A deci-
sion is then needed on whether the area can be
treated as a single unit or whether it is necessary
to know what is taking place at different locations
within it. The first approach is suitable where only
knowledge of the amount of sediment leaving the
area is required. The second is essential where
knowledge of the source of the sediment is needed
so that the implementation of erosion protection
measures can be targeted. Generally the larger the
area, the greater the need for internal understand-
ing since deposition of sediment may occur at
several locations and sediment movement may be
concentrated along preferred flow paths, all of
which can influence the design of a system for
sediment management. These two approaches are
catered for respectively by the use of lumped and
distributed models.
The most commonly used lumped model in
erosion work is the USLE. As already noted, this
predicts mean annual soil loss from a single area,
in this case a field, within which rainfall, soil,
slope and land cover are either considered uni-
form or can be represented by coefficients which
express the average condition. Lumped models
are more commonly used in hydrology. They are
