Suspended-sediment concentration and calibre in relation to surface-flow structure in
Squamish River estuary, southwestern British Columbia
KENNETHM. ROOD
K. Rood & Associates, 3484 Oxford St., Vancouver, B. C . , Canada V5K IN9
AND
EDWARD
J. HICKIN
Department of Geography, Simon Fraser University, Burnaby, B. C., Canada V5A IS6
Received November 24, 1988
Revision accepted April 26, 1989
Surface grabs of suspended sediments from boils and the background flow in Squamish Estuary revealed strong differences
between the two environments. Boils exhibit higher sediment concentrations and larger particle sizes than the background
flow. The higher sediment concentrations and larger grain sizes in boils are related to tidal phase, with maximum concentrations and sediment sizes occurring in the early decelerating phase of the tidal flow. Sediment concentrations and grain sizes
in the background flow are comparatively invariant with tidal phase. Boils appear to play an important, if not dominant, role
in suspending and transporting sand-sized sediments in Squamish Estuary.
Des prCltvements de sMiments en suspension B la surface de remous et de zones h tcoulement moyen dans l'estuaire
Squamish ont mis en Cvidence des diffkrences considtrables entre ces deux milieux. Les Cchantillons provenant des remous
prtsentent des concentrations de sCdiments plus grandes et avec une granulomktrie plus grosshe que ceux recueillis dans
les zones B Ccoulement moyen. La concentration en sMiments la plus ClevCe et la granularit6 la plus grossitre dans les remous
sont relikes h une des phases de la marte, les maxima de concentration et de grosseur des sediments apparaissent durant la
phase de dCcClCration du flot de marCe. D'autre part, dam les zones B Ccoulement moyen la concentration en sCdiments et
la grosseur des grains ne varient pas avec les phases de la marCe. Les remous semblent jouer un rBle important, sinon
dominant, dans la mise en suspension et dans le transport des sCdiments de la grosseur des sables dans l'estuaire Squamish.
[Traduit par la revue]
Can. J. Earth
Sci. 26, 2172-2176
(1989)
Introduction
Boils and other surface disturbances commonly are observed
in rivers and estuaries, but their role in the suspension of bed
materials has not been directly studied. Within estuaries, the
general character of the flow is strongly influenced by tidal
phase. Laboratory investigations (Anwar 1981; Anwar and
Atkins 1980) as well as field experiments (Anwar 1983;
Bowden and Howe 1963; Gordon 1975) indicate that maximum turbulent stresses (Reynolds stresses) occur during
decelerating flows and are lagged relative to the mean column
velocities, producing a hysteresis between velocity and turbulent stress and, by inference, between velocity and suspendedsediment concentration. Thorn (1975) was able to measure the
hysteresis of sediment concentration in a deep tidal flow.
The Reynolds stresses, as calculated from velocity records,
generally exhibit intermittency and "peakiness" (Anwar
1981; Gordon 1975). Records are dominated by intermittent
periods of high-momentum transport that are generally assumed
to reflect the small-scale bursting cycle described by Kline
et al. (1967) and Kim et al. (1971). This assumption is particularly appealing in the estuarine environment because the
intensity and frequency of bursting increase under adverse
pressure gradients (Kline et al. 1967), as are experienced in
decelerating estuarine flows.
There is certainly some question as to whether the large
intermittent events observed on the surface of geophysical
flows are actually a product of the small-scale bursting cycle.
Other mechanisms also produce intermittent and large contributions to momentum transport; specifically, shedding of
roller eddies from the lee of sand waves (Lane 1944;
Korchokha 1968; Nordin et al. 1972). This seems a more suitPrinted in Canada / Imprime au Canada
able mechanism in Squamish estuary where boils are associated with sand-wave fields and, during certain tidal phases,
are clearly distinguished from the background flow by their
energy and sediment content.
The purpose of this note is to further elucidate the role of
boils or surface disturbances, over a tidal cycle, in suspending
and transporting bed sediments.
Field site and methods
Field observations were collected in Squamish River estuary, roughly 1 krn upstream of the delta front (Fig. 1). On the
day of sampling (July 14, 1987), discharge in the Squamish
River was 560 m3/s (Squamish River near Brackendale
gauge; 08GA022), in excess of the long-term mean July discharge. Average suspended-sediment concentration entering
the tidal reach was estimated as 180 mg/L from a rating curve
developed by Hickin (1989). The Squamish is a sand-bed river
within the sampling reach. During the freshet, sand waves
develop on the bed of the channel whose wavelength and
amplitude vary over the tidal cycle (Fig. 2).
Sediment concentrations, both within boils and in the
general background (interboil) flow, were sampled from a
moored boat during a falling tide (tidal range 3.3 m; Point
Atkinson). Samples were surface grabs consisting of 6-9 L
of water and sediment. Paired samples, one within the surface
expression of a recently emerged boil and one in the general
background, were collected at intervals ranging from 10 to 60
min. It was occasionally necessary to wait several tens of
minutes before a boil broke the surface of the river near
enough to the boat to be sampled. Velocities 1 m below the
water surface were measured by a cable-suspended Model
2173
NOTES
FIG. 1. Location map of the Squamish River estuary.
DNC-3 (NBA Controls Ltd.) direct-reading current meter at
the time of sediment sampling.
Suspended-sediment samples were filtered in the field, and
concentrations calculated from the dry weight of sediment and
the total sample volume. Calibre of the suspended sediments
was measured in a visual accumulation tube. Percent finer
measurements from the accumulation tube were graphed, and
the resulting curve was used to estimate the Dlo, DS0,and DgO
sizes of the suspended-sediment samples. Because of the large
sample volumes, the weight of suspended sediments ranged
from 0.2 to 5 g, more than adequate for accurate grain-size
analysis.
A limited field program was undertaken on June 28, 1988
(discharge 510 m3/s), to measure the change in sediment concentration in the boils as they evolve and coarser material
drops out over time. Boils in the lower estuary were grab
sampled from a boat floating with the mean surface current as
they broke the surface and every 5 s as they evolved downstream. Pure chance determines whether a boil surfaces in an
appropriate location, and as a result, only two sets of samples
were collected during the falling tide.
Field observations
Figure 3 plots tide heights (Point Atkinson gauge), surface
-
CAN. J . EARTH SCI. VOL.26,1989
flow direction
A
flow direction
4
metres below
water surface
I
,
l
l
~
4
-m
FIG.2. Fathometer profiles of the sand-wave fields in the Squamish River estuary at 0700 July 14, 1987 (A), and at 1400 July 14, 1987 (B).
O F l o w velocity (V: m l s )
@Tide heiaht (h: m)
Time (hours)
Q
,
.Bo~i
'
.
CO (rng/L x 0 0 1 )
.
.
.
.
.
.
. . 1.5. .
I n t e r b o l l Co trng/L x 0.0 I )
.
.
.
.
*
1.4
k
1,'6
1:7
1.'8
.
1.9
2
Flow velocity (V: rnls)
.
2.1
-
2.b
-2:3
2.'4
FIG.4. Suspended-sediment concentrations (Co) and surface velocities (V); July 14, 1987. Horizontal lines through points indicate
measurements during decelerating flows.
7
8
9
-
10-11
Tlrne (hours)
-
12
-
13
14
FIG. 3. (A) Tide heights (h) and surface velocities (V); July 14,
1987. (B) Boil and interboil suspended-sediment concentrations (Co);
July 14, 1987.
velocities, and boil and interboil suspended-sediment concentrations. The boil concentrations clearly vary with the tidal
phase, and both the boil and interboil sediment concentrations
show a dependence on velocity (Fig. 4). Hysteresis of the sediment concentrations is weakly apparent for the boil samples
and obscured for the background, or interboil, samples. This
may be due, in part, to the use of surface rather than mean
velocity, variation induced by the grab-sampling technique, or
NOTES
TABLE1. Overall statistics of the boil and interboil suspended-sediment samples (N = 25)
Variable
Mean
Concentration (mg/L)
Boil
Interboil
Boillinterboil
Sediment size
Boil D,, (mm)
Dso (mm)
D9o (mm)
Interboil D,, (mm)
D50 (m)
090(mm)
Boil D5, 1 interboil D,,
304
r c o . m g / ~x 0.01 (boil
mco, mg/L x
I)
370
67
8.5
0.11
0.22
0.37
0.06
0.14
0.23
1.7
Minimum
87
15
1.6
790
240
36.6
29
22
16
20
32
23
36
.Boll D50
Time. s
the difficulty of identifying "true" interboil periods, i.e.,
those unaffected by dissipating boils.
Of more importance is the clear distinction between the concentrations in the boil and interboil environments (Table 1).
Over all the paired samples the boil concentrations averaged
eight times greater than the background concentration. The
ratio of boil to interboil concentration varied widely for the
individual samples: interboil concentrations were affected by
the unavoidable sampling of dissipating boils. The observed
boil concentrations (Fig. 3) varied with the energy of the roller
or boil carrying the sediments and the time between the boil
breaking the surface and sampling. Dissipation within the
boils also introduces sampling variability. One sequence of
samples (Fig. 5) shows the rapid decline in sediment concentration associated with evolution of the boils after they break
the water surface. Results of the dissipation measurements are
also variable. Continued upwelling, after the boil breaks the
surface, often briefly increases sediment concentrations. Small
boils, expanding within the dissipating main feature, also
affect the results.
Sediment sizes are less affected by factors such as varying
boil strengths than the sediment concentrations. The calibre of
sediment in the boils is tidal-phase dependent: hysteresis
clearly is apparent for the boils but not apparent in the interboil
environment (Fig. 6). Coarse sediment in the interboil samples
is derived, in part, from dissipation of upstream boils. The
observed sediment sizes appear to vary in response to
upstream boil activity. The boil and interboil paired samples
58
81
96
0.16
0.28
0.46
0.09
0.22
0.36
3.3
0.06
0.14
0.28
0.05
0.08
0.12
1 .O
0 0 l (boil 2 )
FIG. 5. The change in suspended-sediment concentrations (Co) in
evolving boils; June 28, 1988.
Coefficient of
variation (%)
Maximum
Olnterboil D50
. . . . . . . . . . . . . , . . . . .
1.4
1.5
1.6
1.7
1.8
1:9
2
Flow veloclty (V; m h )
2.1
2.2
2.3
2.4
FIG.6. Suspended-sediment calibre and surface velocities; July 14,
1987. Horizontal lines through points indicate measurements during
decelerating flows.
are clearly distinguished on the basis of grain size (Table 1).
The largest discrepancies occur during the decelerating flow
when the ratio of boil to interboil DS0 averages nearly two.
Discussion
Despite the limitations imposed by sampling techniques and
the general variability of the estuarine environment, it is
apparent that boils play an important, if not dominant, role in
suspending and transporting sand-sized bed sediments in
Squamish River estuary. Certainly, the boils seem responsible
for carrying the coarsest bed sediments to the water surface
and distributing them throughout the water column. The intensity of the boil activity, measured as either concentration or
calibre relative to the background flow, is strongly dependent
on tidal phase. The background sediment concentration is
comparatively invariant and much less affected by the
measured velocities or tidal phase.
The role of the boils in increasing total suspended transport
past a section depends on the spatial density of boils and their
size and relative sediment concentration. The first factor
varies with tidal phase and discharge but can be simply
resolved. The second factor is more difficult to specify and
ultimately may involve measurement of correlations of vertical velocities with an array of suitable current meters.
Finally, these observations of spatial inhomogeneities in
2176
CAN. J. EARTH SCI. VOL. 26. 1989
suspended transport have implications for suspended-sediment
measurements. If the period between boils, T, scales approximately as UT/d = 5 (Jackson 1976), where U is a mean velocity and d is a mean depth, then the typical period between boils
in the Squarnish is approximately 15 s. This scaling appears
to apply over a broad water-surface area, and individual boils
may not impinge on a sensor. An accurate average sediment
concentration, sampled over several boils, may require more
than a minute of sampling. This exceeds typical filling times
for the 1 L containers used in many suspended-sediment
samplers (Interagency Committee 1952).
ANWAR,
H. 0 . 1981. A study of the turbulent structure in a tidal flow.
Estuarine, Coastal and Shelf Science, 13: 373 -387.
1983. Turbulence measurements in stratified and well-mixed
estuarine flows. Estuarine, Coastal and Shelf Science, 17:
243 -260.
ANWAR,H. O., and ATKINS,R. 1980. Turbulence measurements in
simulated tidal flow. ASCE Journal of the Hydraulics Division,
106(HY-8): 1273- 1289.
BOWDEN,
K. F., and HOWE,M. R. 1963. Observations of turbulence
in a tidal current. Journal of Fluid Mechanics, 17: 27 1 -284.
GORDON,
C. M. 1975. Sediment entrainment and suspension in a turbulent tidal flow. Marine Geology, 18: M57-M64.
HICKIN,E. J. 1989. Contemporary Squamish River sediment flux to
Howe Sound, British Columbia. Canadian Journal of Earth
Sciences, 26: 1953- 1963.
INTERAGENCY
COMMITTEE.
1952. The design of improved types of
suspended sediment samplers. Report No. 6. Subcommittee on
Sedimentation, Federal Interagency Basin Committee, Hydraulic
Laboratory of the Iowa Institute of Hydraulic Research. Iowa
City, IA.
JACKSON,
R. G. 1976. Sedimentological and fluid-dynamic implications of the turbulent bursting phenomenon in geophysical flows.
Journal of Fluid Mechanics, 77: 531 -560.
KIM,H. T., KLINE,S. J., and REYNOLDS,
W. C. 1971. The production
of turbulence near a smooth wall in a turbulent boundary layer.
Journal of Fluid Mechanics, 50: 133- 160.
KLINE,S. J., REYNOLDS,
W. C., SCHRAUB,
F. A., and RUNDSTADLER,
P. W. 1967. The structure of turbulent boundary layers. Journal of
Fluid Mechanics, 30: 741 -773.
KORCHOKHA,
Y. 1968. Investigations of the dune movement of sediment on the Polomet River. Soviet Hydrology, 11: 541 -559.
LANE,E. W. 1944. A new method of sediment-transportation. Transactions, American Geophysical Union, 25: 900.
NORDIN,C. F., MCQUIVEY,
R. S., and MEJIA,J. M. 1972. Hurst
phenomenon in turbulence. Water Resources Research, 8:
1480- 1486.
THORN,M. F. C. 1975. Deep tidal flow over a fine sand bed. Proceedings of the 16th congress of the international association of
hydraulic research, Sao Paulo, Brazil, pp. 217-223.