Geology Reference
In-Depth Information
80
60
Fig. 11.1
Sediment concentration
(
c
) calculated as a function of
streampower (
) for particular
values of parameter
J
and
b
defined in the text. The set of
curves for
J
> 0 and
b
< 1 are
calculated assuming that the
sediment concentration
remains at the source limit
corresponding to the chosen
values of
J
. Reproduced with
permission from Rose (1993),
p. 324 in
Hydrology and Water
Management in the Humid
Tropics
by M. Bonell
et al
.,
1993, Cambridge University
Press.
Ω
J = 1
00
40
20
0
0
0.1
0.2
0.3
0.4
0.5
0.6
(W/m
2
)
Ω
JN
the flow-driven erosion of bare soil, is invaluable,
not only in itself, but also in providing justification
for the form in which the empirical soil erodibi-
lity parameter
b
was introduced. As shown in
Rose (1993) any decrease in
b
below its theoreti-
cal upper limit (of
b
the soil erodibility parameter
b
has a physical
meaning which can be related to
J
. In such analy-
sis the possible contribution of other erosion
processes to soil loss, such as rainfall detachment,
rolling, or mass movement, is recognized as
increasing the value of
b
obtained from the analy-
sis, even possibly to a value greater than unity
(which is the upper limit for flow-driven erosion
alone). In some experiments, evidence of the
expected additional erosive role of rainfall has
also been found (Yu
et al
., 1999).
Monitored erosion events in which
b
was
found to modestly exceed unity were such that
soil loss by processes that add to flow-driven ero-
sion could be expected to occur, or where mass
movement was observed in some steep slope ero-
sion events under intense tropical rainfall
(Presbitero
et al
., 2005).
Explicit recognition of the separate physical
role of settling velocity, and of the effects of rilling
(if it occurs), both increase the physical relevance
of erodibility
b
.
The information on runoff rate required
to operate GUEST can be measured directly
1) affects sediment concen-
tration variation with the driving variable
stream-power in a manner very similar indeed to
the effect of increasing
J
above its lower limit
of
J
=
0 (at the transport limit). This is illustrated
in Fig. 11.1, and this similarity relieves the empir-
icism of
b
by giving more confidence to its phy-
sical meaning when flow-driven erosion is
dominant.
It is the Type B version of GUEST employing
b
which has been widely used in extensive multi-
country experiments in the tropics, subtropics
and southeast Asia referred to earlier. Hereafter
in this chapter this version of the program will be
simply referred to as GUEST. In these multi-
country studies it was experimentally established
that flow-driven erosion was usually the major
process involved for steeper slopes and significant
erosion events (Rose, 1995), and in such contexts
=
