lagrangian Modelling Techniques Simulating Wave and Sediment Dynamics ...
13
address spatial variability in SSC, advection and changing sea bed conditions.
Several additional critical factors determining SSC have been identified.
First, the advection of less turbid channel water onto the intertidal zone during the flood evidently explains anomalously low values of SSC during peak
flood flow. Second, it appears that changes in grain size at the bed, when fines
are suspended, lead to a small increase in the wave friction factor fW' This causes
the Shields stress and entrainment to increase and a feedback to fw ensues.
Coarsening of the grain size and a corresponding lowering of the mean bed level
is shown in Fig. 3 in a period when fines have been lifted into the water column.
Third, the concept of layer thickness defining sediment availability was of
paramount importance to the success of the simulations. Although the waves are
small, Shields stress is more than adequate to lift fine material into the water column such that entrainment could have continued if sediment were available. If
suspension was allowed to continue up to the maximally possible sediment
transport capacity of the fluid, the predicted concentrations were much larger
than those measured. The settlement time scales of the fines are long, much
longer than the time scales for wave generation and development of critical bed
shear stress. By imposing a layer thickness of 0.1 mm for the finest sediment (Table 1), their entrainment was limited by availability a short time after waves first
developed. The model showed that SSC would have been much higher if more
mud were available at the bed and suspension had been allowed to continue
without imposing availability limitations.
This concept of availability-limited entrainment over beds of mixed sediment
grain sizes is fundamental to the methodology employed in numerical models
and to understanding estuarine sediment dynamics. In the field, it means that
bed sediments must be carefully sampled over the depth of disturbance, while
retaining both the surficial and interstitial muds, if model accuracy is to be improved.
The pick-up function which allows the process-s of entrainment and settlement to be separated (Black 1994) makes it possible in a Lagrangian model to
Bed mean gfaln size 811 • 4 8 hours
Bed level ., 1 = '.8 hou's
11)4
11)4
x 10
102
102
•
100
~
-2
9B
.,
100
98
-6
96
96
9'
94
·8
· 10
92
92
=
115
120
125
05~m
=
115
05km
120
125
Fig.3 Mean sediment grain size at the bed and bed erosion/accretion at 4.8 h after 8 am on
May 9. The legend on the left shows mean grain size, while the legend to the right shows bed
level change (xlO- 4 )
13
address spatial variability in SSC, advection and changing sea bed conditions.
Several additional critical factors determining SSC have been identified.
First, the advection of less turbid channel water onto the intertidal zone during the flood evidently explains anomalously low values of SSC during peak
flood flow. Second, it appears that changes in grain size at the bed, when fines
are suspended, lead to a small increase in the wave friction factor fW' This causes
the Shields stress and entrainment to increase and a feedback to fw ensues.
Coarsening of the grain size and a corresponding lowering of the mean bed level
is shown in Fig. 3 in a period when fines have been lifted into the water column.
Third, the concept of layer thickness defining sediment availability was of
paramount importance to the success of the simulations. Although the waves are
small, Shields stress is more than adequate to lift fine material into the water column such that entrainment could have continued if sediment were available. If
suspension was allowed to continue up to the maximally possible sediment
transport capacity of the fluid, the predicted concentrations were much larger
than those measured. The settlement time scales of the fines are long, much
longer than the time scales for wave generation and development of critical bed
shear stress. By imposing a layer thickness of 0.1 mm for the finest sediment (Table 1), their entrainment was limited by availability a short time after waves first
developed. The model showed that SSC would have been much higher if more
mud were available at the bed and suspension had been allowed to continue
without imposing availability limitations.
This concept of availability-limited entrainment over beds of mixed sediment
grain sizes is fundamental to the methodology employed in numerical models
and to understanding estuarine sediment dynamics. In the field, it means that
bed sediments must be carefully sampled over the depth of disturbance, while
retaining both the surficial and interstitial muds, if model accuracy is to be improved.
The pick-up function which allows the process-s of entrainment and settlement to be separated (Black 1994) makes it possible in a Lagrangian model to
Bed mean gfaln size 811 • 4 8 hours
Bed level ., 1 = '.8 hou's
11)4
11)4
x 10
102
102
•
100
~
-2
9B
.,
100
98
-6
96
96
9'
94
·8
· 10
92
92
=
115
120
125
05~m
=
115
05km
120
125
Fig.3 Mean sediment grain size at the bed and bed erosion/accretion at 4.8 h after 8 am on
May 9. The legend on the left shows mean grain size, while the legend to the right shows bed
level change (xlO- 4 )