Lagrangian Modelling Techniques Simulating Wave and Sediment Dynamics ...
7
Backscatterance Sensor (OBS) at 7 cm and (iii) electromagnetic current meters
at 48 and 97 cm height above the bed. All instruments sampled at 5 Hz and
4400 samples were stored each burst. Data treatment is described by Green et
al. (1998).
Sediment dynamics on the intertidal flat is controlled by waves, which occur intermittently (Green et al.1998; Bell et al.1998). However, when waves are present,
there is a characteristic continuous variation over the tidal cycle in the controlling
processes. Both the surface wave heights and the bed orbital currents vary considerably over the tidal cycle in response to changing sand bank emergence and water
depth respectively. Wave heights are largest at high tide. However, as depth profoundly influences the penetration of orbital motion to the bed, sse is much lower
around high tide than at lower tide levels, due to this factor alone.
Green et al. (1998) also noted a change in "kind", which occurred with the arrival at the measurement site of the "turbid fringe", the narrow, highly turbid
edge of the estuarine water body. The turbid fringe initially travels with the rising tide across the sand flat, spreads more uniformly with time and ultimately
leads to an overall increase in turbidity throughout the estuary as storm waveenergy develops. The turbid fringe exhibited concentrations which were larger
than expected for the measured orbital currents. The precise mechanisms for
this were not isolated, although factors such as turbulence penetration due to
wave breaking in the following winds were considered.
RALPH was situated on the crest of a subtle linear bar, in a field of bars which
emerge from the water at spring low tide. The crests are 0.5-1.0 km long with average spacing of 50 m and amplitude of approximately 0.05 m. Surface sediments
on the intertidal flat were found by Dolphin (1992) to comprise 96% sand and
4% mud, while anaerobic sediments are typically encountered 1-2 cm below the
surface. The low mud content and constant stirring of the bed allows the bed to
remain non-cohesive.
Sediments at the RALPH site have mean grain size of 0.1 mm which compares
to median grain sizes of 0.2 mm in shallower water further up the intertidal flat
(Dolphin 1992). Muds were observed to intermittently accumulate as a fine veneer, but do not persist (Green et al. 1998). Symmetrical wave-generated ripples
(height 1-2 cm; length 10-50 cm) were frequently observed. The sediments can
therefore be described as predominantly noncohesive fine sands mixed with
small amounts of surficial and interstitial muds. They experience a depth of disturbance of around 1-2 cm (Dolphin 1992); similar to the ripple height, although
greater disturbance depths (up to 8 cm) have been noted in some parts of the intertidal flat (Dolphin 1992).
To examine all relevant processes, a storm was chosen for model simulation.
This was the period consisting of hourly data bursts 2152-2188 from 8:00 on May
9 to 03:00 on May 10, 1995. (Note that a burst number increment of 1 corresponds to a half-hour time interval). To test the model, measurements are compared with model predictions from two adjacent model cells at the RALPH site.
These are cells I,J=(119,95) and (118,96), where I is positive to the east and J is
positive to the north (Fig. 1).
7
Backscatterance Sensor (OBS) at 7 cm and (iii) electromagnetic current meters
at 48 and 97 cm height above the bed. All instruments sampled at 5 Hz and
4400 samples were stored each burst. Data treatment is described by Green et
al. (1998).
Sediment dynamics on the intertidal flat is controlled by waves, which occur intermittently (Green et al.1998; Bell et al.1998). However, when waves are present,
there is a characteristic continuous variation over the tidal cycle in the controlling
processes. Both the surface wave heights and the bed orbital currents vary considerably over the tidal cycle in response to changing sand bank emergence and water
depth respectively. Wave heights are largest at high tide. However, as depth profoundly influences the penetration of orbital motion to the bed, sse is much lower
around high tide than at lower tide levels, due to this factor alone.
Green et al. (1998) also noted a change in "kind", which occurred with the arrival at the measurement site of the "turbid fringe", the narrow, highly turbid
edge of the estuarine water body. The turbid fringe initially travels with the rising tide across the sand flat, spreads more uniformly with time and ultimately
leads to an overall increase in turbidity throughout the estuary as storm waveenergy develops. The turbid fringe exhibited concentrations which were larger
than expected for the measured orbital currents. The precise mechanisms for
this were not isolated, although factors such as turbulence penetration due to
wave breaking in the following winds were considered.
RALPH was situated on the crest of a subtle linear bar, in a field of bars which
emerge from the water at spring low tide. The crests are 0.5-1.0 km long with average spacing of 50 m and amplitude of approximately 0.05 m. Surface sediments
on the intertidal flat were found by Dolphin (1992) to comprise 96% sand and
4% mud, while anaerobic sediments are typically encountered 1-2 cm below the
surface. The low mud content and constant stirring of the bed allows the bed to
remain non-cohesive.
Sediments at the RALPH site have mean grain size of 0.1 mm which compares
to median grain sizes of 0.2 mm in shallower water further up the intertidal flat
(Dolphin 1992). Muds were observed to intermittently accumulate as a fine veneer, but do not persist (Green et al. 1998). Symmetrical wave-generated ripples
(height 1-2 cm; length 10-50 cm) were frequently observed. The sediments can
therefore be described as predominantly noncohesive fine sands mixed with
small amounts of surficial and interstitial muds. They experience a depth of disturbance of around 1-2 cm (Dolphin 1992); similar to the ripple height, although
greater disturbance depths (up to 8 cm) have been noted in some parts of the intertidal flat (Dolphin 1992).
To examine all relevant processes, a storm was chosen for model simulation.
This was the period consisting of hourly data bursts 2152-2188 from 8:00 on May
9 to 03:00 on May 10, 1995. (Note that a burst number increment of 1 corresponds to a half-hour time interval). To test the model, measurements are compared with model predictions from two adjacent model cells at the RALPH site.
These are cells I,J=(119,95) and (118,96), where I is positive to the east and J is
positive to the north (Fig. 1).