Geoscience Reference
In-Depth Information
of a turbulent eddy). This parameter is used by the
turbulence model to limit the turbulent dissipation
so that the turbulent viscosity does not become
excessively large. If the mixing length is set to a
high value, the turbulent viscosity (and conse-
quently deposition) will be over-predicted.
Fig. 9 demonstrates the effect of TLEN on the
geometry of the deposit. A TLEN of 1 mm renders
the flow unable to sustain the sediment in suspen-
sion and results in premature deposition in the
feeder channel (Fig. 9C). The amount of sediment
reaching the expansion table is thus significantly
reduced and the resulting lobe shape is inconsist-
ent with the one obtained from the physical exper-
iment. An appropriate mixing length for the
particular turbidity current was found to be between
0.15 mm and 0.2 mm. This turbulent mixing length,
along with the default erosion and bedload coeffi-
cient, provides the most genuine flow structure and
gives a realistic distribution of sediment on the
expansion table (Fig. 9C). Fig. 10 displays some of
the results obtained for Run 4 in a similar way to the
ones obtained for Run 2. Longitudinal and trans-
verse sections of the resulting deposit for Run 2
show a good match with the measurements reported
from laboratory experiments (Fig. 11). The numer-
ical results match very well to the physical data,
both on the longitudinal and the transversal cross
sections, slightly underestimating the thickness of
the deposits in the distal part of the lobe but getting
the shapes and general architecture of the deposit
right. It is possible to appreciate how the software
is capable of reproducing both the depositional
areas and the erosional zones, such as the one at
the channel mouth in the laboratory experiments.
This case study suggests that the numerical model
is capable of producing realistic depositional results.
The parameters discussed above need to be adjusted
to particular conditions and often vary from our the-
oretical assumptions from literature. The geometry
of the simulated deposits in the present case do not
seem to be affected by the discrepancies in the sedi-
ment transport mechanism.
(A)
0.040
Baas
et al
., 2004
TLEN 0.20 mm
RUN 2
Mesh A
y = 0m
0.035
0.030
0.025
0.020
0.015
0.010
0.005
0
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
x (m)
(B)
0.040
RUN 2
Mesh A
x = 0.34 m
0.035
0.030
0.025
0.020
0.015
0.010
0.005
0
−1.5
−1.0
−0.5
0
0.5
1.0
1
y (m)
(C)
0.040
RUN 2
Mesh A
x = 1.41m
0.035
0.030
0.025
0.020
0.015
0.010
0.005
0
−1.5
−1.0
−0.5
0
0.5
1.0
1.5
y (m)
Fig. 11.
Comparison between two-dimensional cross-
sections from Baas
et al
. (2004) (blue line) and MassFLOW-
3DTM results (red line). A) longitudinal sections through
the centre line of the lobe. B) cross section shape of the
lobe at 0.34 m from the channel mouth. C) cross section
shape of the lobe at 1.41 m from the channel mouth. Note
that the z axis is exaggerated.
Case study III: Flow structure of turbidity
currents in sinuous channels
Deep-water sinuous channels show great variation
in their transverse and plan-view geometry, infill
sand net/gross, turbidite facies range and deposi-
tional architecture (Abreu
et al
., 2003; Hubbard
et al
., 2009; Kneller, 2003; Wynn
et al
., 2007).
However, the dynamics of their formative flows
and causes of variation are poorly understood, as
the results of laboratory and numerical experiments
are contradictory and inconclusive (Corney
et al
.,
2006; Imran
et al
., 2007; Peakall
et al
., 2007).
Further work is thus needed in order to establish a
set of well-accepted ground rules for sedimentation
Search WWH ::
Custom Search