Geology Reference
In-Depth Information
A third approach was therefore tried whereby, in
addition to determining the effective buffer area,
field evidence was used to define the actual
contributing area more precisely. At the first field
site, where no concentrated flow channels were
observed, the contributing area was defined by
the extent of the localized depression upslope
of the barrier, giving an area of 20 m by 20 m.
Since there were no channels, the flow depth used
was that recommended for unchannelled flow
(0.005 m). A further refinement with this approach
was to reduce the effective buffer area further to
consider only the area where deposition was
recorded. At this site, the effective buffer area
was considered equal to the 20 m of the contrib-
uting field area and a 6 m buffer length, reflecting
the recording of deposition in the front, middle
and back of the buffer.
At the second field site, it is clear that runoff
and associated sediment transport are concen-
trated in the gully. The contributing element is
therefore modelled based on the length (60 m) of
the gully and the width of the contributing gully
side slopes (10 m in total) instead of the area of
the whole field, that is, an area of only 600 m
2
instead of 97,000 m
2
. With sedimentation occur-
ring on mats at the front, middle and back of the
buffer, a buffer length of 8 m is used. Effective
buffer width is taken as equal to the width of the
contributing area (i.e. 10 m).
Table 13.6 gives the dimensions of the upslope
(element 1) and buffer (element 2) used in these
simulations. It can be seen from Table 13.7 that a
detailed understanding of the effective area of
the processes operating in the field, followed by
setting up the model properly to reflect these,
gives substantially better predictions.
In addition to the total amount of deposition,
the model predicts the particle size distribution
of the deposited material. Table 13.8 shows the
predicted and measured values. It can be seen
that the predictions are poor for these two sites.
The main reason for this is that the model
substantially underpredicts the sand fraction
deposited within the buffers. Whereas the field
data demonstrate that it is mostly the sand frac-
tion which is trapped by the buffers, the model
Table 13.6
Element dimensions (m) used with
configuration 3 based on effective contributing
area and effective buffer area.
Site
1
2
Length of contributing
element
20
60
Width of contributing
element
20
10
Length of buffer
element
6
8
Width of buffer
element
20
10
Table 13.7
Model results for element configuration 3.
Measured deposition
in buffer (kg m
−2
)
Predicted deposition
in buffer (kg m
−2
)
Site
1
6.21
6.35
2
4.85
4.29
Note
: The measured values differ from those in Table 13.5 because
they have been corrected for the effective buffer widths used in
configuration 3.
Table 13.8
Model results for the particle size
distribution of sediment deposited in the buffer.
Site
Particle size
Measured
Predicted
1
% clay
3
24
% silt
7
76
% sand
90
0
2
% clay
4
17
% silt
16
83
% sand
80
0
output shows that no sand particles are delivered
to the buffer. Instead, most are deposited close to
the point of detachment and only small amounts
are picked up and transported by the runoff. All of
the latter are then deposited in the upslope ele-
ment and do not reach the buffer. To some extent
this reflects reality in that sandy material is
deposited on the lower slope as evidenced by the
