Geology Reference
In-Depth Information
in cropland area so that all fields on slopes above
a certain slope degree should be changed to other
uses. The decrease in cropland should be accom-
panied by an intensification of the remaining
cropland and by an increase in orchards (cash
trees). The idea is that in the long term the
income of the farmers should increase once they
get better yields from the remaining cropland as
well as income from fruit trees and other cash
trees. Since it takes time before the new land use
can start to benefit the farmers, the government
pays compensation to make the change economi-
cally feasible for them.
The Loess Plateau is likely to remain an area
of considerable erosion despite all efforts to
reduce erosion rates. It will remain a high-relief,
low vegetation-cover area with heavy storms on
erodible soils. Since the gully erosion has very
markedly increased local relief, it is unrealistic
to think that proper conservation methods will
reduce erosion rates to pre-cultivation levels.
Nevertheless, such conservation methods could
achieve large reductions in current erosion
rates.
Soil erosion modelling is potentially a power-
ful tool for combating soil erosion. It helps us to
understand erosion better, to locate erosion
hotspots, to predict erosion and to evaluate the
effects of different soil and water conservation
measures before implementing them. Even
though research on the Loess Plateau has been
intense for the past 50 years, process-based ero-
sion models have not often been applied. Instead,
more attention has been given to monitoring.
For erosion modelling on the Loess Plateau,
several characteristics of the Plateau need to be
addressed:
●
Slopes in the erodible loess can be very steep,
which may have consequences for flow velocity
and transport capacity of the flow.
●
Sediment concentrations in runoff may be
extremely high. At such concentrations the fluid
properties might differ from those of clear water.
●
The area is heavily dissected by gullies. Thus,
erosion models should be able to cope with gully
erosion, or at least with gullies as a source of
sediment.
The LISEM (Limburg Soil Erosion Model) model
is chosen for this exercise because it is, in
principle, suitable for simulating erosion on the
Loess Plateau for several reasons. Firstly, LISEM
is storm-based, so that it should be able to handle
the storm-dominated water erosion of the Plateau.
Secondly, LISEM is a distributed model, so that
spatial predictions of erosion inside a catchment
are possible. Thirdly, LISEM is a process-based
model. As the source code was available to us,
this meant that process descriptions in the model
could be adapted if the specific characteristics of
the Loess Plateau required this. Fourthly, LISEM
is integrated with the geographical information
system (GIS) PCRaster (Wesseling
et al
., 1996),
and reads GIS maps as input and produces GIS
maps as output. Finally, LISEM was developed for
loess soils. This combination of characteristics
made LISEM more suitable than other well-
known erosion models, such as WEPP, EUROSEM
and KINEROS. The principles of LISEM have
been described in several papers (De Roo
et al
.,
1994, 1996a,b; Jetten & De Roo, 2001).
The aims of this study were:
(1)
to evaluate the effects of the particular char-
acteristics of the Loess Plateau on soil erosion
processes;
(2)
to adapt the LISEM model to Loess Plateau
conditions if this proves necessary;
(3)
to calibrate and validate the LISEM model
for a small catchment on the Chinese Loess
Plateau; and
(4)
to simulate the effects that different soil and
water conservation methods have on soil erosion.
12.2 Study Area
The study area was the Danangou catchment, a
typical small (3.5 km
2
) Loess Plateau catchment in
Shaanxi Province, northern China, with steep
slopes and a loess thickness of more than 100
metres (Plate 9). The soils are mainly erodible silt
loams that classify as Calcaric Regosols/Cambisols
in the FAO-system (Messing
et al
., 2003a). Median
grain size of the loess is about 35
m. The climate
is semi-arid, with occasional heavy thunderstorms
μ
