Geoscience Reference
In-Depth Information
Boundary conditions
Boundary conditions include inflow water discharge, water stage (at inlet or outlet
depending on the flow regime), and inflow sediment discharge and size composi-
tion. The chosen time series of flow and sediment data should represent the average
hydrological cycle in the study domain. Usually, the time series should be long
enough and include high, intermediate, and low water years, with various recurrence
frequencies.
Historical data
Historical measurement data of flow properties, sediment discharges, channel mor-
phological changes, etc., should be collected and analyzed for better understanding of
the study problem and calibration of the numerical model.
5.6 MODEL SENSITIVITY TO INPUT PARAMETERS
Out of all the model parameters, the adaptation length (coefficient) and mixing layer
thickness are least understood and must be prescribed empirically in the sediment
transport models described in Sections 5.3.2 and 5.3.3. Therefore, the concern here is
to analyze the influence of these parameters on the model results. This analysis was
performed in three typical cases by Wu and Vieira (2002), using the semi-coupled
model described in Section 5.3.3.
Case 1. Channel degradation
The experiments performed by Ashida and Michiue (1971) for bed degradation and
armoring processes due to clear water flow downstream of a dam were simulated. The
experimental flume was 20 m long and 0.8 m wide. The flume bed was filled with
non-uniform sediment with a median size of 1.5 mm and a standard deviation of 3.47.
Clear water was pumped into the entrance of the flume at a constant discharge. In
simulated experimental run 6, the flow discharge was 0.0314 m
3
s
−
1
, and the initial
bed slope was 0.01. The computational grid consisted of 40 elements with an equal
spacing of 0.5 m, and the time step was 10 s. The experiments started from a flat bed.
In order to account for the development of bed forms in the simulation, the bed form
height was assumed to vary linearly with time. The Manning roughness coefficient for
the fully developed bed was about 0.023. The bed-material porosity was calculated
using the Komura (1963) formula. The sediment transport capacity was calculated
using the Wu
et al
. (2000b) formula.
The sensitivity of the model results to the adaptation length was investigated using
various functions
L
t
=
7.3
h
,
L
t
=
t
, and
L
t
=
+
0.5
t
while keeping the mixing layer
thickness constant as
d
50
, the median size of the parent sediment mixture. Here,
h
is
the flow depth in meters, and
t
is the time in hours. Fig. 5.18 compares the measured
and calculated bed scour depths at 7, 10, and 13 m upstream of the weir. The trends of
intensive scour in the initial period and weak scour in the final equilibrium stage were
reproduced well. The function
L
t
1
7.3
h
provides the best results for the bed scouring
process, especially regarding the time to reach the equilibrium state. The results for
=
