COHESIVE SEDIMENT EROSION:
Part I Test Devices and Field Instrument Assemblies
Part II Relationship Between the Erosion Rate Constant
and Bed Shear Strength
COHESIVE SEDIMENT EROSION:
Test Devices and Field Instrument Assemblies
Relationship Between the Erosion Rate Constant and Bed Shear
The report consists of two parts: Part I deals with erosion test devices and field
instrument assemblies, while Part II is an analysis of published erosion rate (i)-bed shear
strength (Tr) data in an attempt to establish a functional link between the characteristic erosion
rate constant (defined as the slope of the erosion rate-bed shear stress plot), s, and 7,. A wide
range of erosion test devices has been used both in the laboratory and in the field in the effort
to further elucidate the phenomena of cohesive sediment erosion. They differ in configuration
and flow geometry, size, mechanisms of generating flow, operation protocol, and measurement
methodology. These devices are further complemented by various custom-designed field
instrument assemblies, which provide valuable prototype data for model calibration and
validation. The report reviews six categories of laboratory erosion test devices and eight
categories of field devices/measuring assemblies. For each category, a representative device
from the literature is described with a view to illustrating its salient features and typical test
results. Similarly, the field instrument assemblies are individually described, including a brief
discussion on typical field results. The description is confined to a documentation of the suite
of equipment with incidental comments on their merits without an effort to compare them on a
In Part II, altogether 152 pairs of measured erosion rate constant-bed shear strength
values have been gleaned from the literature for analysis. The data are limited to those that
exhibit a linear relationship linking the rate of erosion to the bed shear strength. A family of
curves of the functional form s = 200exp(-ar8) where a and b are empirical constants and the
relevant units are g/N-s and Pa for s and 7,, respectively, is apparent from the data spread.
Further examination of the trend of several important erosion resistance characterizing factors
reveals that the succession of curves behaves consistently with respect to the variation in the state
of sediment bed (undisturbed or remolded) and the total salt concentration in the pore fluid.
Based on these observations, a nomograph has been suggested for estimating the erosion rate
constant, given the bed shear strength, subject to the caveat that the nomograph be used only as
a last resort to provide general guidance, and is not intended to supplant erosion experiments,
which should always remain the first choice. It may be noted that the nomograph is applicable
only for the type of beds such as placed bed/undisturbed bed/remolded bed which has a uniform
shear strength with depth.
Table of Contents
Synopsis ................................................... i
List of Figures ........................... ..... .. ............ v
List of Tables ......................... .......... .......... ... vii
Part I: Test Devices and Field Instrument Assemblies . . . ..... ... 1
IA: Laboratory Devices . . .
Straight Flumes ................
Sample Results . . .
Other Similar Flumes . .
Laboratory Rotating Annular Flumes .
Introduction . . . .
Sample Results ............
Other Rotating Annular Flumes ..
Race-Track Flumes ............................
Sample Results ..........................
Other Similar Devices ......................
Rocking Flumes ........
Sample Results ..
Other Similar Devices
. .. ..... 8
. .. .. ..... 8
.. .. .. ..... 8
. . . . .
. . . . .
Closed-conduit Sediment-Water Tunnel
Sample Results ..........
Other Similar Devices ......
Rotating Cylinder Apparatus ......
Theory . . . .
Sample Results . . .
Other Rotating Devices . .
Drill-hole Test Apparatus ........
Sample Results ..........
Other Similar Devices ......
. . . . . . . . 10
Vertical Grid Oscillator .
Sample Results ....
Other Similar Devices
Propeller-based Apparatus ..
Sample Results ....
Other Similar Devices
Other Erosion Test Devices ....
Introduction . .....
Rotating Disk Device ..
Submerged Jet ........
. . . .
IB: Field Devices and Measuring Assemblies
Field Inverted Channels ...
Introduction . .
Sample Results ....
Other Similar Flumes
Field Rotating Annular Flumes .
Sample Results ......
Other Similar Flumes ..
. . . .. 22
ISIS Instrument for Shear Strength In-Situ ............................ 23
Sediment-water Interface Probe ..................................... 24
DAISY Tower Instrumentation Array ................................ 24
OSU Instrumentation .......................................... 25
COE Field Tower ............................................. 26
Instrumented Bed Frame ........................................ 26
Part I: Relationship Between the Erosion Rate Constant and the Bed Shear Strength ... 28
Introduction ................... .... .................... 28
Data Analysis ........................ .................. 28
References .................. ... ........................... 32
Tabulation of Test and Soil Conditions, Bed Shear
Strength and Characteristic Erosion Rate Constant ........... 39
Figures .................................................... 47
. .. 22
. . . .
List of Figures
1. Schematic of the Delft Tidal Flume (from Kuijper et al., 1989) ............ 47
2. A typical suspended sediment-time curve for an erosion experiment in the tidal
flume (from Kuijper et al., 1989) .............................. 47
3. Schematic of a laboratory rotating annular flume (from Sheng, 1989) ........ 48
4. A typical result of resuspension experiment using the annular flume for a
stratified bed at a bed shear stress (rb) of 0.21 Pa (from Mehta and Partheniades,
1979) ................ ..................................... 48
5. A typical result of resuspension experiment using the annular flume for a uniform
bed at a bed shear stress (rb) of 0.41 Pa (from Mehta and Partheniades, 1979) 49
6. Schematic of a race-track flume (from Mehta and Srinivas, 1993) .......... 49
7. Typical velocity profiles from a test run using a race-track flume with bottom
mud composed of kaolinite (from Mehta and Srinivas, 1993). t is the elapsed
time following test initiation, U is the characteristic mixed-layer velocity, and 6,
is the shear layer thickness. As an example, the dashed line indicates the location
of the bottom elevation of 56 for profile No. 4 (t = 54 min) ............. 50
8. Density profiles corresponding to the velocity profiles shown in Fig. 11 (from
Mehta and Srinivas, 1993). 5 is the density interfacial layer thickness and Pb is
the density value at the base of the corresponding 3. The initial fluid-bed
interface is indicated by the density profile at t = 0 .. ................ 50
9. a) Schematic of a rocking flume (from Villaret and Paulic, 1986)
b) Top constriction placed in the rocking flume (from Villaret and Paulic,
1986) ............... ....................... .. 51
10. A typical concentration-time curve for a deposited kaolinite bed using the rocking
flume (from Villaret and Paulic, 1986) ........................... 51
11. A typical concentration-time curve for deposited Cedar Key (Florida) mud bed
using the rocking flume (from Villaret and Paulic, 1986) . . ....... 52
12. Schematic of a closed-conduit sediment-water tunnel (from Teeter and Pankow,
1989) .......... ... ....... ................... ..... .52
13. Schematic of a rotating cylinder apparatus (from Rohan and Lefebvre, 1991) ... 53
14. A typical erosion rate versus shear stress curve using a rotating cylinder apparatus
for intact clay samples and triaxially prepared samples (from Chapius, 1986) 53
15. Schematic of a drill-hole test apparatus (from Rohan et al., 1986) .......... 54
16. A typical eroded sediment mass versus tractive force curve from a drill-hole test
apparatus (from Rohan et al., 1986) ............................ ... 54
17. Schematic of a vertical grid oscillator (from Tsai and Lick, 1986) .......... 55
18. Time histories of the shaker concentration at five stations (from Tsai and Lick,
1986) .............. ................. ............. 55
19. Schematic of the EROMES System (from Schunemann and Kuhl, 1993) ...... 56
20. Typical results from an erosion experiment using the EROMES System (from
Schunemann and Kuhl, 1993). Upper: suspended concentration; middle: bottom
shear stress; lower: erosion rate ............................... 56
21. Schematic of a rotating disk device (from Liou, 1970) ..... ........... ..57
22. Typical time-variations of erosion shear strength using a rotating disk device for
high water content bentonite samples without chemical additives, and with 0.1 N,
0.01 N NaCI and 0.01 N CaC12 additives (from Liou, 1970) ............. 57
23. Schematic of a submerged jet apparatus (from Dash, 1970) .............. 58
24. Typical time-variations of weight loss using a submerged jet apparatus for
consolidated samples with no sand (from Dash, 1970). ocma denotes maximum
consolidation pressure ..................................... 58
25. A typical time-variation of optical transmissometer outputs during a SEAFLUME
deployment (from Gust and Morris, 1989). Open and solid circles refer to the
front and back optical transmissometers, respectively .................. 59
26. Schematic of a field rotating annular flume (from Maa et al., 1993) ......... 59
27. Typical results of an incipient erosion experiment during the Sea Carousel
deployment (from Maa, 1991): ) ring speed versus time; b) OBS reading for
suspended sediment concentration (SSC). Numbers are shear stresses in
dynes/cm2 . . . . .............. . . ..... 60
28. Schematic of the Sediment-water Interface Probe (from Nichols et al., 1978) ... 60
29. A typical analog time-series record (from Nichols, 1989): A) current speed, u
and v, components at 0.2 time-constant, showing intermittent nature of
contributions at 30 cm above a fluid mud during ebb-tidal current, B) turbidity
showing small-scale fluctuations, station 19, upper Chesapeake Bay. Turbidity
units are in percent meter reading ................... .......... 61
30. A photograph of the DAISY Tower Instrumentation Array (from Bohlen, 1982) 61
31. Typical results from a field deployment of DAISY Tower (from Bohlen, 1982).
Upper: suspended sediment concentration; middle: near-bottom current speed;
lower: wind speed at 10 m elevation, Millstone Point, Long island Sound ..... 62
32. A typical time-trace of 15-min average horizontal velocity and suspended load for
continuous deployment of the OSU instrumentation (from Bedford et al., 1987) .62
33. Schematic of the COE Field Tower (from Mehta and Jiang, 1990) .......... 63
34. A typical computed horizontal mud acceleration spectrum from accelerometer data
collected during a deployment of the COE Field Tower (from Mehta and Jiang,
1990) ............................................. 63
35. Typical results of field measurement at Mersey site spring tide, 11/20/90: A)
concentration, B) bed level, C) shear stress (from Teisson et al., 1993) ....... 64
36. s- r curve for Group 1 ...................................65
37. s- T curve for Groups 2, 3,4 and 5 ........................... 66
38. s r, curve for Groups 6 and 7 ............................... 67
39. Nomograph for characteristic erosion rate constant ................... 68
List of Tables
1. List of erosion test devices and field measuring assemblies covered in the report .. 2
Straight flumes .............
Laboratory annular flumes ......
Race track flumes ...........
Closed-conduit devices . .
Rotating cylinder devices . .
Drill-hole test devices . . .
Vertical grid oscillators . .
Other erosion test devices . .
Field measuring assemblies ......
Field rotating annular flumes .....
Sediment data groups . . .
General characteristics of data groups
. . . . . . . .4
...................... .. ... 7
. . . . . . . 31
Part I: Test Devices and Field Instrument Assemblies
A wide array of erosion devices has been employed in the study of cohesive sediment
erosion both to elucidate the associated phenomena under controlled conditions, and to provide
the values of the relevant input parameters that appear in the erosion flux functions used in
analytical and numerical modeling. In order to obviate scale effects inherent in laboratory
investigations, some of the devices have also been applied to field situations. Earlier reviews
of laboratory equipment can be found in the Task Committee on Erosion of Cohesive Material
(1968) and Partheniades and Paaswell (1970). Later reviews appear to be more selective in
coverage (e.g., Berlamont et al., 1993), and do not include field equipment (e.g., van Leussen
and Winterwerp, 1990). In addition, ad hoc field instrument assemblies have also been used for
The report reviews selected laboratory and field-based devices/assemblies. They are
conveniently categorized according to their primary features of flow geometry as summarized
in Table 1. For each category (and sub-category where appropriate) a representative device or
assembly is described with respect to the basic features and operation of the equipment. Several
field measuring instrument assemblies listed in Table 1 are also included, and brief descriptions
of each follow those of the erosion test equipment. Each description is accompanied by typical
results. The report is confined to a documentation of erosion test devices and field measuring
assemblies with incidental comments on their merits; no effort has been made to compare them
on a relative basis.
IA: Laboratory Devices
This is by far the most common laboratory equipment used in studying erosion of
cohesive sediments. The number of such flumes is too numerous to enumerate, which include
recirculating, non-recirculating, and tilting types. One of the largest of its kind, the Delft tidal
flume, is briefly described herein based on the work of Kuijper et al. (1989).
The Delft tidal flume, schematically shown in Fig. 1, consists of a 130 m straight open
channel with a cross-section of 1 m x 1 m. The flume terminates in a sea basin where the water
level, salinity and sediment concentration can be controlled. A glass bottom and glass sidewalls
enable visual observations to be made from all sides. Tides with a maximum period of 1800 s
and flow rates up to 0.3 m3/s can be realized. At the upstream side, tap water with a selected
sediment content can be injected to simulate river inflow. Two return pipes allow the flume to
be used as a flow recirculating system.
Table 1. Erosion test devices and field instrument assemblies.
Category Sub-category Investigators
Laboratory Erosion Test Devices
Flumes Straight flume
Alishahi and Krone (1964); Dash (1968); Peirce et al. (1970);
Lonsdale and Southard (1974); Einsele et al. (1974); Thorn and
Parsons (1980); Mehta et al. (1982); Kamphuis and Hall (1983); Maa
and Mehta (1987); Shaikh et al. (1988); Kuijper et al. (1989);
Talebbeydokhti and Klingeman (1992); Tofts (1993); Mimura (1993)
Partheniades and Kennedy (1966); Mehta (1973); Fukuda (1978);
Kusuda et al. (1985); Burt and Game (1985); Tsai and Lick (1988);
Kuijper et al. (1989); Murakami et al. (1989); Li (1989); Lau and
Krishnappan (1991); Krishnappan (1991); Moller-Jensen (1993)
Watanabe et al. (1978); Ali et al. (1991); Mehta and Srinivas (1993)
Villaret and Paulic (1986)
Raudkivi and Hutchison (1974); Kelly and Gularte (1981); Teeter and
Moore and Masch (1962); Arulanandan et al. (1975); Croad (1981);
Chapuis and Gatien (1986)
Christensen and Das (1973); Rohan et al. (1986)
Tsai and Lick (1986); Wolanski et al. (1989); Teeter (1994)
Propeller-based Holland et al. (1970); Paulic et al. (1986); Schunemann and Kuhl
Others Rotating disk Liou (1970)
Submerged jet Dunn (1959); Moore and Masch (1962); Dash (1970)
Field Erosion Test Devices and Measuring Assemblies
Field Inverted channel Young (1977); Nowell et al. (1985); Gust and Morris (1989)
Field Rotating annular flume Peirce et al. (1970); Maa (1991); Amos et al. (1992)
ISIS Instrument for Shear Strength Hydraulics Research Limited (1994)
Sediment-water interface probe Nichols et al. (1978); Nichols (1989)
DAISY Tower Bohlen (1982)
OSU instrumentation Bedford et al. (1982, 1987)
COE field tower Mehta and Jiang (1990)
Instrumented bed frame Diserens et al. (1993), Teisson et al. (1993)
The flow is generated by a constant-head tidal tank and the ensuing fluid-sediment
interaction can be studied from the perspective of erosion and sedimentation fluxes under steady
and non-steady, and homogeneous and non-homogeneous conditions, the last including gravity
circulation and flocculation.
The sediment layer in the flume is first formed by deposition in still water after some
period of mixing. Subsequently the layer is allowed to consolidate for a longer period of time.
The bed is then eroded in various stages during which the fluid-induced bed shear stress is
successively increased. The bed shear stress is calculated assuming a logarithmic velocity
profile. The sediment concentration is measured with an optical instrument consisting of six
probes. From various water depths, water is pumped through the probes and led through small
cells and the concentration level determined via light absorption. The dry density of the mud
is determined using a conductivity probe. The same probe is also used for detecting the surface
of the mud layer.
Figure 2 shows a typical suspended sediment concentration-time curve under steady flow
testing condition using China Clay (80% kaolinite and 15% illite) with a mean diameter of 3
microns. In this particular case, the consolidation period was 8 days. From Fig. 2, it is seen
than for the first two lower stress applications, i.e., 0.11 Pa and 0.17 Pa, the erosion rate
reaches zero after about 10 to 15 hours following test initiation, while the erosion rate eventually
becomes constant at the application of the third shear stress (0.22 Pa).
Other Similar Flumes
As mentioned at the outset, the laboratory straight flume is the most common equipment
used in the study of cohesive sediment erosion both under steady and oscillatory flow conditions.
Table 2 is a non-exhaustive list of flumes in this category. Straight closed-end flumes, equipped
with wave makers, are also commonly used in wave-sediment interaction studies (e.g., Maa and
Mehta, 1987; Mimura, 1993). Typically, the waves are mechanically generated using a plunger
or a piston-type wave maker, and the horizontal sediment bed is accommodated in a drop section
or trench in the middle of the false bottom construction. Typical measurement parameters
include suspended sediment concentration, wave height envelope, density and pore pressure
variation with depth, and mud mass transport. In some flumes, e.g., Alishahi and Krone (1964),
waves are wind-generated. Further details on some representative flumes in this category are
included in Table 2.
Laboratory Rotating Annular Flumes
Annular flumes have been used in laboratory investigation by several investigators.
Those adapted for field use are noted under a separate heading. The following description
pertains to the flume used by Mehta (1973).
Table 2. Straight flumes.
Investigators) Sub-category (length x width x depth) Remarks
Dash (1968) Recirculating,
Peirce et al. Non-recirculat-
(1970) ing, tilting
18.3 x 0.30 x 0.39
9.1 x 0.76 x 0.46
3.7 x 0.14 x 0.15
Wind-generated waves using centrifugal fan.
0.32 m x 0.32 m square sample housing unit
with a movable base; filter frame covered with
wire mesh and filter paper at downstream end.
Upstream supply via surge chamber and
downstream gated control outlet to waste.
Einsele et al.
Mehta et al.
Shaikh et al. Recirculating,
Kuijper et al.
10.0 x 0.57 x 0.25"
17.4 x 0.30 x 0.20
10.0 x 0.6 x 0.9
9.1 x 0.15 x 0.30
20.0 x 0.46 x 0.45
2.5 x 0.16 x 0.11
130.0 x 1 x 1
7.9 x 0.46 x 0.61
20.6 cm x 13.4 cm x 5.4 cm sample holder.
Closed rectangular section.
Underflow gated control at downstream end.
0.61 m x 0.15 m x 0.15 m drop section to hold
Plunger-type wave maker.
15.2 cm x 10.5 cm x 2.25 cm sample container
for unsaturated compacted soils.
Steady, non-steady, homogeneous and non-
homogeneous (gravity circulation, etc.) testing
Sediment layer thickness 4 cm.
Tofts (1993) Recirculating,
9.0 x 0.4 x 0.4
13.0 x 0.3
Sediment layer thickness 8 cm.
Bottom-hinged flap-type wave maker.
The apparatus, shown schematically in Fig. 3, consists of a system of an annular channel
and ring that rotate in opposite directions. The channel is made of 10 mm thick fiberglass and
the annular space is flanked by two concentric cylinders with a mean diameter of 1.50 m, giving
a rectangular cross-section of width 0.20 m and depth 0.46 m. The cylinders are mounted on
a 13 mm thick steel turntable of diameter 0.61 m. Four units of 76 mm x 51 mm plexiglass
windows are provided every 900 in the lower part of its outer wall for visual observation. An
annular false bottom made of 3 mm thick plexiglass and having the same dimensions as the
annular ring is used to facilitate direct measurement of the bed shear stress using temperature-
compensated strain gages.
The annular 6 mm thick plexiglass ring, which has the same mean diameter as the
channel, but with a slightly smaller width of 0.19 mm, is positioned within the channel. The
ring is suspended from four flexible stainless steel blades of dimensions 0.6 mm thick, 0.08 m
wide and 0.52 m long. The width of the ring is 6 mm less than that of the channel, yielding a
radial gap of 3 mm between each edge and the channel wall. The speeds of the ring and the
channel are controlled independently through two variable speed driving motors using a
concentric shaft assembly.
The fluid confined within the annular space is driven by the differential motion of the
ring in contact with the fluid and the channel. The rotational motion of the channel and the ring
sets up a flow field that is mainly in the azimuthal direction, thereby eliminating undesirable end
effects associated with a straight flume assembly. The essentially axi-symmetric flow that ensues
is then analyzed based on one-dimensional (azimuthal) flow condition. However, a secondary
flow in the radial direction is also induced due to the varying radial pressure gradient across any
cylindrical plane, and may violate the assumptions of little variation in both the azimuthal and
radial directions inherent in one-dimensional flow analyses.
Partheniades and Kennedy (1966) argued that the desired near-uniform flow field in the
radial direction can be achieved by rotating the channel and the ring in opposite directions, the
latter at a greater speed. At a sufficiently large differential speed, the larger vertical momentum
of the downward flow induced by the faster rotating ring then balances the radial pressure
gradient near the channel bottom, and hence reduces the secondary flow there. Their contention
was supported by flow experiments using plastic beads and small threads attached to the channel
bottom, which indicated that the secondary current velocities were of the order of 10 to 20 %
of the corresponding azimuthal velocities when the operating curves for the channel and ring
speeds were followed.
More recently, Sheng (1988) calculated the strength of secondary flow within rotating
annuli using an integral boundary layer model and measured azimuthal velocities. When applied
to a rotating-ring-fixed-channel assembly with specified configuration [see Table 3 under Fukuda
(1978) for dimensions], it was shown that the average radial velocity would be approximately
20% of the mean azimuthal velocity outside the boundary layer, but 50% of the azimuthal
velocity at 0.1 cm above the bottom.
Using the Reynolds equations in a simplified form in conjunction with the continuity
equation in cylindrical coordinates, Maa (1993) showed by numerical experiments that near the
channel bottom [see Table 3 under Maa (1993) for dimensions] the maximum radial velocity
would be about 10% of the nearby azimuthal velocity, which is reasonably constant except close
to the two covers for a ring rotating speed of 8 rpm. In addition, the radial bed shear stresses
would be about 20% of the azimuthal bed shear stresses, and the total stresses would be close
to the azimuthal bed shear stresses. These calculations indicate that the effects of secondary
flow can be reduced to an insignificant level by a judicious choice of flume dimensions and the
Water and sediment are first introduced into the annular space through the open top of
the channel with the ring raised. The ring is then positioned in the channel such that it just
touches the surface of the water. For each channel speed, the ring is then rotated at a
predetermined speed based on operating curves established to accomplish uniform deposition of
sediment across the channel width.
Two typical results of resuspension experiments using the apparatus (Mehta and
Partheniades, 1979) are shown in Fig. 4 (density-stratified bed) and Fig. 5 (uniform bed). The
density-stratified bed was obtained from gradual sediment deposition from suspension while the
channel was in slow motion and the bed shear stress was kept slightly less than the shear stress
at which the entire sediment deposits. The uniform bed was prepared outside the channel at a
mean density close to that of the deposited sediment and subsequently placed and leveled into
the channel. The sediment used in both cases was a commercial kaolinite.
In the case of the stratified bed, which emulates a typical situation with respect to the
surficial sediment layers found in estuaries, it is seen that the slope of the concentration versus
time curve, hence the erosion rate, decreases with time as the strength of the bed increases with
depth. In contrast, Fig. 5 exhibits a constant slope, i.e., constant rate of erosion, after a
relatively short transient period. More results are available in Parchure (1984).
Other Rotating Annular Flumes
Variously dimensioned rotating annular flumes can be found in the literature. They all
follow the same general principles described above and differ mainly in instrumentation for
measurement. Table 3 lists several such flumes.
Several flumes of this type have been used. The following description pertains to that
of Mehta and Srinivas (1993). The apparatus is similar in concept to its counterpart in the study
of salt stratified flows (Narimousa and Fernando, 1987).
The flume, shown schematically in Fig. 6, consists of a horizontal recirculating open
channel with a disk pump system to generate fluid shear in the study of interfacial instability and
resultant fluid mud entrainment. The flume is made of variable thickness plexiglass (3.2 mm
at the curved portions and 12.7 mm at the thickened linear test section and pump section). The
width of the flume narrows linearly from 48 cm at the base of a bi-triangular section containing
the pump and flow separators to 10 cm at the junction with the curved portion (a tapering
distance of 79 cm) and thereafter. The flume is symmetrical about the above pump section with
a 200 cm long linear section for testing purposes.
Table 3. Laboratory annular flumes.
Investigators) d d h Operating Mode
di d w h
Partheniades and Kennedy (1966) 0.72 0.91 0.10 0.30 Counter-rotating
Mehta (1973) 1.12 1.52 0.20 0.46 Counter-rotating
Fukuda (1978) 1.02 1.32 0.15 0.30 Rotating ring
Fukuda (1978) 0.92 1.52 0.30 0.30 Rotating ring
Kusuda et al. (1985) 0.80 1.20 0.20 0.20 Counter-rotating
Burt and Game (1985) 5.20 6.00 0.40 0.35 Rotating ring
Tsai and Lick (1988) 1.70 2.00 0.15 0.20 Rotating ring
Kuijper et al. (1989) 1.90 2.30 0.20 0.30 Counter-rotating
Murakami et al. (1989) 1.25 1.55 0.15 0.20 Counter-rotating
Li (1989) 1.08 1.50 0.21 0.41 Counter-rotating
Lau and Krishnappan (1991) 1.60 2.00 0.20 Rotating ring
Krishnappan (1991) 4.70 5.30 0.30 0.30 Counter-rotating
Moller-Jensen (1993) 1.50 1.90 0.20 0.26 Rotating ring
'di = 2r = inner diameter;d = 2r outer diameter; w =r r- = width; h = height.
The disc pump consists of two vertical, motor-driven shafts, which rotate in opposite
directions. Each shaft is stacked with a number of thin discs of sand-blasted plexiglass. These
are of two diameters, 4 and 13 cm, stacked alternately on each shaft, and so arranged that a
small disc on one shaft meshes with the larger of the other, thus almost sealing the center of the
pump, while the fluid is thrown as a series of horizontal jets around the outside of the smaller
discs and between the larger discs. In order to prevent the discs from "sucking up" the density
interface, a horizontal "splitter plate", which extends downstream into the curved segment, is
placed at a height of 31 cm above the flume floor to effect horizontal partitioning.
The disk pump, which is designed to minimize any intrusive effect of pumping on the
interfacial dynamics, drives the initially clear water layer relative to the higher density fluid
below. The ovate geometry guides the flow gradually into the linear test segment without undue
interference. In addition, the effect of helical secondary flows on the processes occurring within
the linear test segment is minimized by the installation of the splitter plate. Hence, the
incorporation of the straight test segment allows secondary flows developed in the bends to
decay, yielding an experimental environment with a generally unidirectional flow.
When the pump is turned on, the energy of the generated turbulent shear flow leads to
the development of substantial instabilities at the density interface. The cumulative effect of the
movement of the density interface and the increasing dissipation of turbulent kinetic energy on
the entrained sediment aggregates manifests in a change in the mixed (upper) layer mean
velocity, which is monitored.
During test, the flume is first filled to the requisite pre-selected height of water. Pre-
mixed fluid mud is then introduced through the intake at the flume bottom. The initial height
of the fluid mud layer is kept at just below the elevation of the splitter plate. With time, as the
fluid mud is entrained, the movement of the density interface is tracked and the velocity
variation in the mixed layer monitored.
Figure 7 illustrates typical velocity profiles when kaolinite was used as fluid mud
sediment. Time t is the elapsed time during which the characteristic velocity is U, and the shear
layer thickness is 8,. With increasing velocity, an outward "bulge" developed in the velocity
profile due to the combined effect of the disk pump, flow inertia below the disks, and the
confining effect of the interface. The corresponding variation in density (Fig. 8) attests to the
fact that above the lutocline (mud-water interface) the sediment was consistently well-mixed.
Below the lutocline, hindered settling and self-weight dewatering caused the density to become
increasingly non-uniform with depth.
Other Similar Devices
In mud studies, an apparatus of similar setup has been used by Ali et al. (1992). It
differs essentially from the one just described in that the flow is driven by a toothed belt in
contact with the water surface in the rear straight section. Further details are summarized in
Table 4, which also includes an earlier version used by Watanabe et al. (1978).
Thus far only one such apparatus has been reported in the literature. The following
description is based on the work of Villaret and Paulic (1986).
The rocking flume, shown schematically in Fig. 9a, is constructed of 12.5 mm thick
plexiglass. It measures 2.4 m in length and 0.36 m in height with an inner width of 0.15 m.
A 7 cm high false bottom is built into the flume within which a deepened section (54 cm long
x 5 cm deep) is located to function as the test section, hence yielding an actual depth of 0.29 m.
The entire assembly is mounted on a table via a pivot, giving a bottom clearance of 16 cm above
the table top that permits rocking motion.
Table 4. Race-track flumes.
Investigators Dimensions Instrumentation Use
Watanabe et Two semicircular bends of internal radius 50 Rotating paddle wheel Study of
al. (1978) cm joined by two straight sections each 5 m capable of generating both concentration
long. The channel is 0.15 m wide and 0.5 m steady and sinusoidal profiles.
Ali et al. Two semicircular bends of internal radius 75 Toothed belt in the rear Study of fluid
(1992) cm joined by two straight sections 4 m long straight section and laser- mud formation
each. The channel is 30.5 cm wide and about doppler anemometer for and transport.
1 m high (with a maximum working depth of velocity measurement.
Mehta and Two semicircular bends of internal radius 46 Disk pump with a horizontal Study of fluid
Srinivas cm joined by a straight section 2 m long. The splitter plate, a vertically mud-water
(1993) rear section is bi-triangular in shape to adjustable electro-magnetic interfacial
accommodate the disc pump and flow separator current meter and a flow- dynamics.
assembly. The channel is 10 cm wide and 61 through densimeter
The flume is operated by a hydraulic transmission attached to a 0.75 hp motor. A metal
shaft (rocking arm) at one end of the flume connects a rotating plate driven by the motor through
a motor shaft to the flume via a hub. In operation, the rotating shaft moves the rocking arm up
and down, thereby causing the flume to rock back and forth. Both the period and the magnitude
of the rocking motion can be adjusted by varying the motor speed and the eccentricity of the
rocking arm/rotating plate connection, respectively.
A plexiglass top constriction of height 19 cm and length 54 cm (Fig. 9b) is placed in the
water column over the sediment bed to increase flow velocity. Its ends are sloped to reduce
turbulence at the flow entrance and its height above the bed can be varied. Hence, the free
surface flow in the flume is transformed into flow in a "tunnel" in the central portion of the
Due to the controlled rocking motion imparted to the flume, the current generated at the
sediment surface has a sinusoidal velocity variation. The flume is calibrated to produce a
maximum shear stress up to 0.8 Pa. The rocking flume data were compared with those from
the annular flume as follows. The maximum velocity was first determined by using direct
measurement of the displacement of the water level, by an electromagnetic current meter, and
by considering flow continuity. These velocity data were then used to compute the maximum
applied shear stress using the relationships established by Jonsson (1966), which was formulated
for progressive waves. To adapt to the case of standing waves, the computed maximum shear
stress was integrated over one-half a wave period to obtain the mean shear stress, which is one-
half the maximum shear stress. This is the shear stress that was used in comparing the results
of erosion experiments obtained from an annular flume (steady current) and the rocking flume
oscillatoryy current). In general, the bed shear strength was lower under oscillatory currents
than under steady currents. The discrepancy was attributed to the process of bed softening under
oscillatory currents, implying a degradation of bed shear strength due to a breakdown of the
structure of the deposited aggregates.
Two types of bed can be tested: placed bed wherein a thick pre-mixed slurry is placed
in the flume to uniform depth, and deposited bed wherein a dilute suspension of sediment is
allowed to settle out of water column and consolidate into a bed. The bed is then subjected to
selected shear stresses. During the test, samples at various depths in the water column are
collected periodically for concentration determination.
Figures 10 and 11 show typical concentration-time curves for two deposited beds
composed of different sediments. The kaolinite was of a commercial grade while the estuarine
mud was collected from a tidal flat near Cedar key, Florida, which had a principally
montmorillonitic content (73%). Both curves exhibit a series of steady states (characterized by
constant final concentrations). Higher concentrations were observed for the kaolinite bed for
the same applied shear stress, due to its greater erodibility than the estuarine mud.
Other Similar Devices
In mud studies, this device appears to be the only one of its kind. Tilting flumes are
generally used to study the effects of gravity-induced instability, and are thus different materially
from the rocking flume arrangement (e.g., Ali and Georgiadis, 1991).
Closed-conduit Sediment-water Tunnel
Closed conduits in which the fluid recirculates in either the vertical or the horizontal
plane have been used to study cohesive sediment erosion. The following description pertains
to that of Teeter and Pankow (1989).
The device, shown schematically in Fig. 12, is a closed-conduit sediment-water tunnel
designed to safely test contaminated sediments. Hence, it is open to air only at a small
expansion chamber. The water tunnel, which is made of 12.5 mm thick clear acrylic, has a
uniform cross-section, which changes from rectangular (38 cm wide x 9 cm deep) in the
horizontal deposition/resuspension sections to circular (20.3 cm internal diameter) in the vertical
settling and pumping sections. The volume of the water tunnel is 0.28 m3, and the available
surface area for sediment deposition and resuspension is 1.64 m2. Flow in the tunnel is driven
by a tandem pair of two-bladed, skewed propellers and by a variable-induction motor. Settling
tests can be performed in the descending tube while the deposition and resuspension tests are
performed in the rectangular conduit sections by sampling.
The water tunnel has been calibrated so that the propeller speed is related to average
velocity and bed shear stresses using a tachometer, a flowmeter, and a hot-film shear stress
sensor. The effects of secondary flow and other flow irregularities have been examined using
flow visualization techniques and judged to be small based on the overall smooth flow pattern
and a uniform turbulence structure. The propeller wakes are confined within the ascending tube.
In a particular test involving composite dredged material from New Bedford Harbor,
Massachusetts, reconstituted sea water was used in the tunnel, with the addition of formaldehyde
solution to inhibit microbial growth. An initial water tunnel sediment bed was established by
injection of sediment during deposition periods. Subsequent tests then alternated between
resuspension and deposition episodes accompanied by addition of sediment to the water tunnel
without removal of material from previous tests. Hence, sediments resuspended from the bed
of the tunnel at the beginning of the each test became incorporated into the test material.
Strict safety procedures such as full length disposable suits, neoprene gloves, respirators,
etc., required of all personnel are implemented when the sediment contains hazardous
In the above test, no resuspension results in graphical form were presented, but the
equilibrium resuspension concentration at each applied shear stress was tabulated. It was
observed that erosion was rapid during the first few minutes after the application or increase in
bed shear stress. Erosion decreased rapidly as the tests progressed, and the suspended
concentration reached an equilibrium value.
Other Similar Devices
A similar device, but of simpler construction, was used by Raudkivi and Hutchison
(1974). The details are given in Table 5, which also includes another device (Kelly and Gularte,
1981) where the fluid recirculates in the horizontal plane.
Table 5. Closed-conduit devices.
Investigators Dimensions Instrumentation Remarks
Raudkivi and 51 mm x 51 mm rectangular Circulating pump, Made from stainless steel, perspex
Hutchison cross-section with a sediment flowmeter, preston tubes and rigid PVC for studying the
(1974) holder (51 mm x 38 mm and heat exchanger unit dependence of sediment erodibility
wide) placed flushed with on temperature, salinity, zeta
the conduit floor potential and ion exchange
Kelly and 15.2 cm x 15.2 cm Four-blade ducted
Gularte rectangular cross-section propeller; refrigerated
(1981) water tunnel
Teeter and 38 cm x 9 cm rectangular Two-bladed skewed Made from 12.5 mm thick clear
Pankow cross-section propeller, flowmeter, hot- acrylic to test contaminated
(1989) film shear stress probes sediments.
Rotating Cylinder Apparatus
Several devices of this general type have been used. The following description pertains
to that of Chapuis and Gatien (1986).
The device, shown schematically in Fig. 13, consists of a cylinder of cohesive soil
mounted coaxially inside a slightly larger transparent plexiglass cylinder which rotates during
test. The annular space between the cylindrical soil sample and the outer rotating cylinder is
filled with the desired eroding fluid to transmit shear from the rotating cylinder to the surface
of the soil sample. In this particular configuration, the prepared cylindrical soil sample (75 mm
diameter x 89 mm long) is placed between two metallic short cylinders (base and head), both
guided in rotation by ball bearings. There is no shaft within the sample, which distinguishes it
from that of Moore and Masch (1962) in that in the latter device the clay sample was
reconstituted around a metallic shaft to which lower and upper plates were connected for support
and trimming of the sample. The base rotates freely relative to the bottom of the outer cylinder
in order to measure the torque transmitted by the eroding fluid to the soil cylinder by means of
an upper shaft connected to the head. The transparent cylinder has an inside diameter of 102
mm and can be rotated at regulated speed up to 1,750 rpm.
The apparatus makes use of annular water flow between two concentric cylinders due to
the motion of the outer cylinder while the inner one remains stationary. When the outer cylinder
is rotated, rotation is imparted to the fluid, which in turn transmits a shear to the surface of the
soil sample. The shear stress at the clay-fluid interface is computed from the torque required
to hold the sample stationary, which is directly measured using a pulley and weight system with
masses ranging from 0 to 40 g with a 0.1 g precision, as opposed to the use of a calibration
curve between the applied shear stress and rpm of the outer cylinder by Moore and Masch
Moore and Masch (1962) reasoned that because of the stabilizing inertial forces resulting
from a velocity distribution that increases in the radial direction, the turbulence level at the soil
surface should be relatively low, resulting in a small variation in the instantaneous value of shear
stress. Hence, the cylindrical form of soil samples and the action of the end pieces (mounted
immediately above and below the soil sample) should result in a uniform value of the shear
stress over the entire surface of the sample. However, in analyzing the hydrodynamic conditions
in a rotating cylinder setup, Rohan and Lefebvre (1991) showed that the flow between the two
cylinders is always turbulent due to the relatively high rotational velocity attainable (1,750 rpm,
maximum in the device of Chapius and Gatien, 1986). The turbulent flow structure in an
annulus departs from that of a rectilinear flow due to the presence of a centrifugal force. The
latter flow regime mimics more closely field situations. Hence, Rohen and Lefebvre (1991)
caution that the real shear stress may be underestimated by an order of magnitude due to
streamline curvature and the fluctuations in the radial component of velocity.
First the sample is either cut from intact blocks using a template and a steel wire, or
reconstituted and reconsolidated in a triaxial cell after physicochemical or mechanical treatment.
The clay cylinder is then mounted on a pivoting base and emplaced. After the eroding fluid is
added into the annular space, the test commences with several stages, each at a constant rpm.
Each stage is held for 10-30 minutes and the shear-stress-induced couple continuously recorded.
At the end of each stage, the fluid is removed and the cell cleaned with fresh fluid. All eroded
particles thus recovered are oven-dried and weighed to obtain the eroded mass. This procedure
obviates repetitive manipulations of the sample, which may contribute to sample remolding.
Typical results of tests conducted on undisturbed natural clays from Northern Quebec are
shown in Fig. 14, which illustrates the influence of sample preparation method (Chapuis, 1986).
The consistently lower curve for the minimum registered value of the hydraulic shear stress at
a given speed exhibited by the triaxially prepared samples is attributed to their smoother surface
compared to the cut samples. The test results indicate that the hydraulic shear stress depends
on surface roughness, which in turn is influenced by the erosion process.
Other Similar Devices
The device described is a modified version of the one originally developed by Moore and
Masch (1962); the changes being in the use of intact samples, rotation guidance, alignment and
influence of end supports on the annular flow regime. A similar device has also been used by
Arulanandan et al. (1975). Further details are summarized in Table 6.
Drill-hole Test Apparatus
The following description pertains to the apparatus of Rohan et al. (1986).
The device, shown schematically in Fig. 15, consists of a stainless steel tube serving as
the sediment compartment through which the eroding fluid is forced down an axial circular hole
created earlier by drilling. The sample tube measures 44.5 mm in outer diameter and 35.4 mm
in inner diameter and 100 mm in length. The hydraulic head for the flow is provided by a
constant head reservoir capable of generating a flow velocity in the drill hole up to 10 m/s
corresponding to an applied shear stress of 450 Pa.
First the relative change of the diameter of the drill hole caused by erosion is calculated
based on the dry weight of the eroded sediment collected in the sedimentation basin at the
outflow end of the specimen. The flow rate is measured by a flowmeter, and the pressure
difference caused by frictional losses in the specimen is then measured by differential
manometer, the various losses being accounted for with the use of the Moody diagram. The
evaluation of the applied shear stress is then based on the control-volume momentum equation.
Table 6. Rotating cylinder devices.
Investigators Dimensions Instrumentation Remarks
Moore and Cylindrical clay sample of Torque is derived from a Only remolded
Masch (1962) dimensions 76 mm outer diameter x calibration curve obtained using a samples can be
76 mm long with a central metallic pulley-weight system tested.
shaft as the core in a slightly larger
cylinder capable of rotating up to
Arulanandan Cylindrical clay sample of Torque is obtained from a Only remolded
et al. (1975) dimensions 76 mm outer diameter x calibration curve obtained using a samples can be
82 mm long with a center metallic pulley-weight system and wet tested.
shaft as core in outer cylinder of sample is periodically removed
inner diameter 102 mm capable of for determination of eroded
rotating up to 1,100 rpm. sediment mass.
Croad (1981) Cylindrical sample suspended Rotating outer drum drives the
concentrically in the flow flow.
Chapius and A wholly cylindrical clay sample of Pulley-weight assembly to Both reconstituted
Gatien (1986) dimensions 75 mm diameter x 89 measure torque, eroding fluid and intact samples
mm long in plexiglass outer withdrawn for determination of can be tested.
cylinder of inner diameter 102 mm eroded sediment mass.
capable of rotating up to 1,750 rpm.
The device is designed for studying the erodibility of intact undisturbed clays, and hence
the specimens are trimmed from block samples to minimize sample disturbance. The specimen
is first trimmed to a diameter of about 37 mm and lowered into the tube whose other end is
fixed by a circular cutting tool. A mechanical lathe is used to drill the hole in the rotating soil
cylinder, giving an initial hole diameter of 6.4 mm and presenting a total surface of 20.0 cm2
Typical results of two Canadian clays are shown in Fig. 16. As noted by Rohan et al.
(1986), at the end of the test period, the St. Bamabe clay had not reached its critical shear stress
at an applied shear stress of 400 Pa, as evidenced by the rather constant amount of eroded
material. On the other hand, a critical shear stress of 170 Pa can be defined for the Grande
Baleine clay as characterized by the net increase in the rate of erosion.
Other Similar Devices
A somewhat similar device was used earlier by Christensen and Das (1973), in which the
sediment sample is prepared in the annular space between a brass tube and an inner molding
tube. Further details are given in Table 7.
Table 7. Drill-hole test devices.
Investigators Dimensions Instrumentation Remarks
Christensen Molded 3.2 mm thick Constant-head tank, Made from brass tube to test
and Das smooth clay lining of 1.9 cm flowmeter and shear saturated, compacted samples.
(1973) inside diameter, 10.2 cm stress computed using
long approximate friction
Rohan et al. Drilled cylindrical erosion Constant-head tank, Made from stainless steel tube to
(1986) surface of 6.4 mm diameter flowmeter, differential test undisturbed sediments.
x 100 mm long manometer and shear
stress computed using
Vertical Grid Oscillator
This is a portable device designed for rapid erodibility assessment. The following
description relates to the device of Tsai and Lick (1986).
The apparatus, shown schematically in Fig. 17, consists of a cylindrical chamber inside
of which a horizontal grid oscillates vertically. The cylindrical chamber, which is made of cast
acrylic tubing, is 27.9 cm high with an outside diameter of 12.7 cm and inside diameter of 11.7
cm. The grid element is a 6 mm-thick plexiglass disc of diameter 11 cm, which is perforated
with holes of 12 mm diameter at 15 mm centers, giving a porosity of 42.8%. The grid is
oscillated by a 1/8 hp permanent magnet DC motor via drive rod and linkage bar connection.
The sediment whose erodibility is to be determined is placed at the bottom and overlain
by water. The grid oscillates in the water and creates turbulence which penetrates down to the
sediment-water interface and causes resuspension. It is recognized that while the turbulence
generated by an oscillating grid is different from that operating in the field, i.e., due to currents
and waves, the issue is circumvented by calibrating resuspension results from the grid oscillator
to those obtained in a conventional flume experiment. The basic premise is that when the flume
(with a given applied bed shear stress) and oscillator (with a given frequency) produce the same
concentration of resuspended sediment under the same environmental conditions, the stresses
required are considered equivalent.
The grid oscillating device can be used in both the laboratory or field setting. In field
operation, the cylindrical chamber consists of three parts: an open-ended cylindrical tube, a
bottom disc, and a top plate. The diver first pushes the tube down into the sediment. The
bottom disc is then forced into the sediment from the side of the tube and slid to the location of
the tube. By pushing the bottom disc up, the sediment core is firmly confined in the tube.
Finally, the top plate is placed on the top of the tube and the whole assembly brought up to the
ship. After the thickness of the sediment core is measured, the overlying water is adjusted to
the height used in the calibration experiment (12.7 cm) and the test started with periodic
suspension concentration measurement by sampling.
A typical result from a shallow water field test carried out in Lake St. Clair near Detroit,
Michigan is shown in Fig. 18 based on two different equivalent shear stresses. At each station,
the concentration at the higher shear stress is observed to be higher. In all cases, the
concentration rises very rapidly initially and then levels off until it reaches a quasi-steady state.
The decrease of the concentration noted in most cases thereafter is believed to be the result of
sediment compaction created by the fluctuating grid-generated pressure, which could change the
entrainment process. Hence, it is advised that the device not be used for multiple shear tests.
Other Similar Devices
Other similar devices used in sediment entrainment studies are given in Table 8.
Table 8. Vertical grid oscillators.
Investigators di dd hRemarks
Tsai and Lick (1986) 11.7 11.0 27.9 6 mm thick plexiglass horizontal perforated grid in cast
acrylic tube cylinder.
Wolanski et al. (1989) 9.5 8.5 50.0 15 circular flat rings 7 mm wide x 5 mm thick, 2 cm
apart vertically constitute the grids.
Teeter (1994) 11.5 11.0 12.7 6 mm thick plexiglass horizontal perforated grid in cast
acrylic tube cylinder. The height indicated is the depth
adi = inner diameter; dd = disc diameter; h = height of cylinder.
Several apparatuses in this category have been reported in the literature. The following
description pertains to that of Schunemann and Kuhl (1993).
The device, code-named the EROMES System and shown schematically in Fig. 19,
consists of a tube containing the sample and a large container to store the eroded material during
an experiment in a turbulent suspension. It is another experimental approach to investigate the
erosion of muddy sediments with possible surface coatings intact and sedimentary textures
preserved. Contrary to the more conventional approaches whereby clearly defined parameters
from a wide-ranging suite of environmental factors are singled out for examination, the emphasis
here is to investigate what happens in a simulated environment. The sample tube is the actual
tube used in collecting field samples. The submerged sample is mounted in a vice with a
calibrated propeller in water and located 3 cm above the sample surface. The propeller consists
of six evenly spaced lamellae as baffles. The sample tube, which is made of perspex with a
diameter of 10 cm, is connected by pipes via a dispersing machine and a pump to a large
container for storing the eroded material in a turbulent suspension. The sample as placed has
an exposed surface area of 78.5 cm2 for testing purposes. A calibrated gamma probe measures
the density profile of the sediment bed sample based on attenuation of emitted rays, while an
attenuationmeter continuously monitors the concentration in the container by measuring the
attenuation of a beam of transmitted light of constant intensity.
The apparatus uses turbulence artificially induced by the propeller to erode the samples
and to keep the eroded material in suspension. The eroded material is stored in the large
container with a different and larger set of propeller, which runs at a higher speed in order to
induce a homogeneous distribution of suspended matter in the container. The suspension in the
sample tube is continuously pumped into the storage container and a second connection, the
return pipe, carries the suspension back to the sample tube. Thus, there is an exchange of
suspension between the storage container and the sample tube to simulate natural conditions in
the sample tube. The attenuationmeter is calibrated by sampling the concentration of the
suspension and using filtering technique to relate the attenuation factor to the suspension
concentration. In this way, the erosion rate can be computed from the calibrated curve that
converts the attenuation value to g/1 of suspended matter.
The propeller initially turns at 50 rpm. The revolution is then increased in steps of 20
rpm in 5 minute intervals until clearly perceptible erosion starts. Then the propeller revolution
is kept constant for 40 minutes to record the erosional behavior under constant conditions.
Finally, the revolution is reduced below the level of the critical bottom shear stress and the
suspension sampled for the determination of concentration. After 10 minutes of stabilization the
next increment is initiated.
Figure 20 shows the results from a typical erosion experiment. The three subplots are
from the same experiment. The top diagram displays the time variation of concentration in the
storage container. The center diagram depicts the time series of bottom shear stress applied to
the sample. The bottom diagram shows the change in the erosion rate with time. It is seen that
during increasing bottom shear stress episodes, the erosion rate generally increases as well,
while during the 40 minutes of constant applied shear stress it decreases, which reflects
increasing erosion resistance of the sediment bed with depth.
Other Similar Devices
Simpler versions of the above concept have been used to study microbially enhanced
stability in the laboratory setting. The apparatus essentially consists of a flask containing the
cultured bed agitated by a mechanical stirrer. The relative erosion of different sediments and
under different biotic control is measured as the dry mass of the sediment in suspension after
stirring at a known rpm for 5 minutes (Holland et al., 1974; Paulic et al., 1986).
Other Erosion Test Devices
The devices included here are those that were used in the 1960's through early 1970's
and have since seldom been used in studying cohesive sediment dynamics. They are included
here for completeness.
Rotating Disk Device
The rotating disk device of Liou (1970), schematically shown in Fig. 21, was used to
study the variation of the critical erosion shear stress of clay samples with different chemical
additives and subjected to different ambient temperatures. It consists of a rotating disk (0.33
m diameter x 1.3 cm thick) in a steel cylindrical container (0.34 m diameter x 0.2 m high), the
bottom of which is a circular steel soil sample pan (0.29 m diameter x 5 cm high). The disk,
rotating in a fluid that is at rest and confined, essentially generates an axi-symmetric boundary
layer structure. At sufficiently high disk Reynolds number, separate boundary layers form on
the surfaces of the disk and the fixed bottom. The flow condition near the soil sample consists
of an inward radial velocity, tangential velocity and axial upward velocity. By solving the
Navier-Stokes equation in cylindrical coordinates with simplifying assumptions, it is shown that
the fluid shear stress above the soil sample depends on the rotational speed in the core region,
I.e. the fluid region between the top and bottom boundary layers. There is a dynamic
equilibrium between the centrifugal forces and the radial pressure gradient. Hence, for a given
rotational speed, the shear stress is linearly proportional to the distance from the center, which
is calibrated with the disk positioned 2.5 cm above the surface of the sample and covered in 6.4
cm of eroding fluid.
The test commences by turning on the rotating disk at the selected rotational speed. The
turbid fluid is periodically drained and replaced with fresh fluid. The point at which the outer
edge of the sample begins to erode is found by the use of a point gage.
A typical test result on bentonite samples in terms of the erosion shear stress-time
variation with different chemical additives is shown in Fig. 22. The constant value reached by
the erosion shear stress after a period of time is identified as the critical erosion shear stress.
Hence, the critical erosion shear stress is seen to be the highest for the case of 0. IN NaCl as
This is essentially a technique used to empirically relate the scour resistance of a cohesive
soil sample to a vertically directed water jet to bulk soil properties such as plasticity and clay
mineralogy. The description here is based on the work of Dash (1970).
The jet apparatus, schematically shown in Fig. 23, consists of a 0.76 m long plexiglass
tube of 2.5 cm internal diameter fitted with a 3.2 mm diameter jet nozzle at the lower end.
Water is supplied from a constant-head tank. The cylindrical sample tube is 7.6 cm diameter
x 32 cm with the soil sample seating snugly at its bottom.
During test, the jet tip is placed 2.5 cm above the sediment surface. Erosion losses are
determined based on the change in the weight of the sample. Fig. 24 shows a typical test result
conducted on kaolin samples. In the figure, cmax refers to the maximum consolidation pressure
the sample has been subjected to prior to testing. Generally it is seen that a lower consolidation
pressure and a higher jet head both lead to higher erosion. Table 9 compares the submerged jet
devices used by several investigators.
Table 9. Submerged jet devices.
Investigators Description Primary Soil Primary Major
Variables Hydraulic Dependent
Dunn (1959) A nozzle produces a submerged vertical jet Plastic Head Critical
of water directed perpendicularly at the properties, tractive stress
surface of the soil sample placed at the vane shear
bottom of a lucite cylinder. No dimensions strength
Moore and 2.5 cm diameter tube fitted with variable Shear strength Jet elevation Depth of
Masch size nozzles; 12.7 cm diameter x 10.2 cm erosion,
(1962) cylindrical sample holder in 0.91 m square location of
x 0.46 m lucite tank incipient scour
Dash (1970) 0.76 m long plexiglass tube of 2.5 cm Clay Head Eroded
internal diameter fitted with a 3.2 mm mineralogy, sediment mass
diameter nozzle; 7.6 cm diameter x 32 cm percent clay,
cylindrical sample tube tensile strength
IB: Field Erosion Test Devices and Instrument Assemblies
Both Field erosion test devices and instrument assemblies have been used, the latter
specifically to collect erosion data in near bottom areas composed of cohesive beds. Following
the descriptions of the field erosion test devices, several field instrument assemblies as
summarized in Table 10 are briefly described.
Field Inverted Channels
This is an application of the commonly used straight flumes to field situations for in-situ
determination of erosion threshold and entrainment rates of undisturbed deep sea bottom
sediments. Several investigators have used this device; the following description pertains to that
of Gust and Morris (1993).
Table 10. Field instrument assemblies.
Investigators) Assembly/ Construction S C
ArraTU CM PT DP SA CA WS
Nichols et al.
Steel tripod frame 3.7
m high with a circular
base of 1.8 m diameter
equipped with lead
ballast and steel pads
for stability; central
hydraulic cylinder with
1.44 m vertical free
DAISY Aluminum tripod
Tower frame 3 m high with a
circular base of 2 m
diameter in-filled with
lead and concrete for
/ / /
/J J /
Bedford et al. OSU
(1982; 1987) Instru-
Mehta and COE field
Jiang (1990) tower0
Mounted on a host
tower (e.g., DAISY
Aluminum tower 2.45
m high with a
rectangular base 1.5 m
x 1.0 m with
protruding pins from
the leg supports for
anchoring; a central
4.2 cm diameter
aluminum shaft within
a concentric pipe of
5.8 cm outer diameter
serves as sensor
Diserens et al. HR Triangular framed / / /
(1993); Instrument open structure with
Teisson et al. -ed bed protruding legs for
(1993) frame anchoring
TU =- turbidimeter/transmissometer; CM = current meter; PT pressure transducer; DP density probe; TE -
thermistor; SA = salinity sensor; CA = camera; WS = water sampler.
includes an acoustic sediment concentration profiler.
includes a biaxial accelerometer.
The apparatus consists of a 2 m x 0.4 m x 0.2 m deep inverted channel, which forms a
duct upon contact with the sediment surface. The design is based on the original equipment of
Young and Southard (1978) used in the study of the erosion of in-situ fine sands, but with
modifications to the instrument package and superstructure that permits free-falling deployments
and recovery at water depths less than 4,000 m.
The flume is located within a triangular outer tripod with a base length 3.5 m and height
3 m, which carries the flotation-anchoring assembly and the hydraulic piston from which the
flume is suspended. The flume section is equipped with a pump, discharge control, and sensor
and power supply packages. The channel is made from 6 mm thick aluminum sheet and is fitted
with a 40 cm flared entrance to reduce flow disturbance there. The duct outlet is bolted to a
gated control box and a DC-driven axial pump that sucks the water through the duct. The speed
of the flow is determined by the cross-section of the duct, the position of the gate and the power
delivered to the pump.
The straight flume operates as a flow-through system. The rectilinear flow generated by
the sucking action of the axial pump produces an applied bed shear stress to the sediment
surface, which is measured by flush-mounted friction velocity sensors. An additional array of
hot-film anemometers measures the velocity profile. The suspended sediment load that results
is measured by fore and aft optical attenuation meters, and the remolding of the sediment surface
is monitored by a camera. Natural flows superimposed on the pump-generated flow do not
invalidate the measurements, since the friction velocity probes always measure the resulting wall
shear stresses exerted on the sediment surface.
For deployment, the flume section is first retracted above the tripod base by the bleeding
action of the suspending piston, and the entire assembly is lowered into water. Once the
superstructure makes contact with the sea bottom, the suspended flume is made to settle onto the
sea bottom whereby the sidewalls penetrate to a depth of about 5 cm, resulting in a flow cross-
section of 630 cm2. A weight then triggers off a succession of automated operations involving
the recorder, probes, camera, discharge control gate and pump motor. Data are stored in a
digital multi-channel recorder. The sensors are calibrated in the laboratory for the anticipated
ambient temperature range, and checked with a portable calibration device prior to deployment.
At the completion of field operation, a spring-loaded pulley jettisons the anchor stones
and the assembly returns to the surface where its retrieval is facilitated by the emission of radio
and strobe signals.
A typical result from a field deployment in Puget Sound, Washington where the sediment
surface consisted of a layer of consolidated cohesive mud is shown in Fig. 25 in terms of the
time-variation of the optical transmissometer output voltages, which in turn is used to compute
the mean concentration of suspended sediment in the flume duct. In the figure, open circles
denote voltage from the front (inlet) transmissometer and closed circles, the back (outlet)
transmissometer. The time series of entrainment rates are then calculated using these
concentration data, leading eventually to the determination of the functional relation between
erosion rate and applied shear stress.
Other Similar Flumes
Two other similar flumes that have been reported earlier in the literature. They are the
SEAFLUME (Young, 1977), and the SEADUCT deployed at the HEBBLE (the High Energy
Benthic Boundary Layer Experiment) site (Nowell et al., 1985) to measure in-situ erosion rates
of sea bed.
Field Rotating Annular Flumes
This is an application of laboratory-developed rotating annular flumes to field scale for
in-situ experiments to study the complex erosion processes of natural sea beds. While the theory
of generating hydrodynamic forcing is the same as their laboratory counterparts, these devices
differ essentially in that they are necessarily bottomless, and employ a high level of
sophistication in deployment requirement and instrumentation packages in order to operate in the
field. Several investigators have used this device; the following description pertains to that used
by Maa (1991).
The apparatus, shown schematically in Fig. 26, is functionally similar to laboratory
annular flumes. Two 0.2 m deep cylinders of diameters 2 m and 2.3 m, respectively, form the
inner and outer walls of the annular space, yielding a width of 0.3 m. During operation, the
flume is lowered from a boat to penetrate the sea bed at a predetermined elevation, giving a
designed flume depth of 0.1 m. The flume has no bottom, and a ring at the top of the flume
rotates at selected speeds to generate flow in the flume. The ring speed, which has a maximum
at 14 rpm, is regulated by a shipboard personal computer and a 1 hp DC motor controller.
The flume is sealed between the rotating ring and side walls, which prevents sediment
from escaping or entering the annulus. Therefore, measurement of an increase in suspended
sediment within the flume determines the extent of bottom erosion. Instruments mounted over
the sea-bed flume include an Optical Backscatter Sensor (OBS), two miniature Optical
Transmissometers (OTS), a counter for ring speed, a water sampling system, and a data logger.
On the ring, there is a plexiglass window to mount an underwater camera for photography of
the sea floor around the channel to provide a view of the surface roughness.
The flume is lowered into place from a boat. After the flume has penetrated to the
predetermined elevation, photographs of the bed condition are first taken and the ring rotation
started. The ring speed is increased in a step-wise fashion. At each stage, there is a short
period (first 2 minutes) of increasing speed followed by a longer period of constant speed. The
OBS and the two OTS's then measure the suspended sediment concentration. The data logger
serves as the analog-to-digital data converter and feeds the personal computer onboard for data
A typical result of an incipient erosion experiment conducted at Wolftrap shoal in the
lower Chesapeake Bay is shown in Fig. 27 in terms of the time-change of the measured
suspended sediment concentration. The numbers shown above each step in the step-wise
increase in ring speed are numerically computed bed shear stresses (dynes/cm2). The mean
water depth at this test site was 11.6 m, and the in-situ sediments consisted of a mixture of fine
sand (74%), silt (14%) and clay (12%) with a mean diameter of 0.03 mm. Biological activity
at the site was judged to be quite significant.
Figure 27 indicates a critical erosion threshold of 1.28 dynes/cm2, corresponding to the
point of departure from the trend of constant concentration. The observed decreasing
concentration level at constant ring speed is ascribed by the author to a combination of flume
leakage, influence of suspended sediment on the bed shear stress and redeposition of large
sediment particles in low shear stress area of the flume.
Other Similar Flumes
Two other similar flumes have been reported in the literature (Peirce et al., 1970; Amos
et al., 1992). They differ largely in the driving system, size, and sealing device as summarized
in Table 11.
Table 11. Field rotating Annular flumes.
Investigators Dimensions Instrumentationa Driving Mechanism
Peirce et al. 0.42 m outer diameter x Pipette for sampling Battery-operated paddle wheel.
(1970) 0.05 m water depth suspension
Maa (1991) 2.15 m mean diameter x One OBS, two Flow is generated by a rotating
0.20 m high x 0.30 m wide OTS, data logger lid controlled by DC motor.
Amos et al. 2.00 m mean diameter x Three OBS, data Flow is driven by a rotating lid
(1992) 0.30 m high x 0.15 m wide logger equipped with paddles.
"OBS = optical backscatter sensor; OTS = optical transmissometer.
ISIS Instrument for Shear Strength In-Situ (Hydraulics Research Ltd., 1994)
The instrument, which was designed to measure the erosion shear strength of muddy
sediments on inter-tidal mud flats, consists of a circular, inverted bell-shaped funnel of 84 mm
diameter placed inside a cylindrical perspex column of 90 mm diameter. Other attachments
include pump and flow controller, batteries, reservoir and flow meter. The whole assembly is
mounted on a baseplate measuring 1 m x 0.35 m x 0.7 m. In operation, the bell head is
positioned at a typical distance of 4 8 mm above the mud bed. Water is drawn up through the
center of the bell via pumping into a reservoir, which is then recirculated to replace the pumped
water via the sides of the bell. The induced radial flow toward the bell center exerts an
approximately even shear stress across the bed, the magnitude of which is controlled by the
height between the bell head and the bed surface, and the pumped discharge through the system.
In addition, turbidity is measured in the reservoir by an optical backscatter probe.
Sediment-Water Interface Probe (Nichols et al., 1978; Nichols, 1989)
This is a dedicated bottom-mounted field instrument array designed for measuring fluid
mud transport and behavior in the bottom boundary layer. As shown schematically in Fig. 28,
it consists of a central tripod support frame that houses the various measuring sensors and
samplers. These include an optical turbidimeter, electromagnetic current meters, a nuclear
transmission density probe, a pressure transducer and suspended sediment pump samplers.
These instruments are mounted independently and are connected by signal cables to shipboard
signal processors and a data acquisition system.
The steel frame stands 3.7 m tall with a circular base 1.8 m in diameter. It is fitted with
lead ballast and steel pads to ensure stability when in position. A hydraulic cylinder, which is
housed in a central shaft and has a vertical free play of 1.44 m, thrusts the sensors through the
lower water column and drives the density probe through the sediment-water interface into the
bed. The sensors provide simultaneous and continuous measurement of sediment density,
turbidity and current meter over the water depth. The suspended sediment sampler consists of
a nozzle, which is connected to a submerged pump via a flexible tubing. The pump delivers
water to deck overline and thence into sample bottles and filtration units.
The instrument system is designed to operate in two modes: 1) vertical profiling through
the water column into the bed, and 2) temporal monitoring at one or several depth measurement
points. Since the system is operated from a boat, its use is limited to normal weather conditions.
Figure 29 shows a portion of a typical analog time-series record during a deployment at
the upper Chesapeake Bay at 30 cm above a fluid mud be during ebb tidal current. The
turbidity units are in per cent reading. These high frequency records display small-scale
fluctuations, suggesting that the mud participates in intermittent bursts and sweeps associated
with local accelerations and decelerations.
DAISY Tower Instrumentation Array (Bohlen, 1982)
The system is designed to provide reasonably long-term in-situ observations of near-
bottom suspended material conditions as well as the prevailing hydrodynamic condition that
induces it. As photographed in Fig. 30, it consists of an aluminum frame approximately 3 m
in height with a circular base of 2 m diameter. The array consists of a control unit and four
basic subsystems comprising a pump, an optical array, a current meter and camera setups,
together with supplementary temperature and conductivity sensors.
The control unit consists of a digital data logger, which supplies power to the instrument
array and records the output signals. Hence, it is self-contained with sufficient data storage
capacity to permit sampling of all instruments four times an hour over 36 days. The individual
instruments comprising the primary instrument package include transmissometers, savonius rotor
current meter, super 8-mm movie camera with strobed light unit, and variable-volume
displacement pump and associated filtration units. Thermistors and flow-through cells for
conductivity measurement make up the supplementary package.
Figure 31 shows the result of a field deployment in the vicinity of the New London
dredged material disposal grounds, eastern Long Island Sound in approximately 20 m of water
in January 1980 (13 days). The current was measured near the bottom. The suspended
material concentrations display relatively low variability. The anomalously high spike near 100
hours has been attributed to aberrant fouling or proximity to disposal of dredged material.
OSU Instrumentation (Bedford et al., 1982; 1987)
This is an acoustics-based system that uses ultrasonic transducer technology to measure
detailed vertical profiles of sediment concentration. The instrumentation is attached to a small
mobile support system configured such that the instruments can either be deployed on a host
tower or as a stand-alone unit. For example, it was part of the OSU C-DART Data Acquisition
Tower deployed at the Old Woman Creek Estuarine Sanctuary, Lake Erie to record the high-
density multi-faceted data needed to couple wind wave conditions to lake bottom turbulent
responses (Bedford et al., 1982). The array of instruments includes an acoustic sediment
concentration profiler, two-axis electromagnetic current meters, transmissometer, pressure
transducer and thermistor.
Figure 32 shows a typical temporal variation pattern of the horizontal velocity and
suspended load during a deployment in Central Long Island Sound in 20 m of water. In this
particular deployment, the instrument array was attached to one leg of the DAISY Tower
described above. The acoustic profiler was positioned 1 m off the bottom and measured the
backscatter intensity in 1 cm intervals throughout the one meter column. The 100 point
sediment concentration profile was sampled at 32 Hz and ensemble averaged into a stored 1 Hz
signal. The signals from the current meter, positioned 68 cm above the bottom, were sampled
at 4 Hz and smoothed to an effective temporal resolution of 1 Hz. An in-situ computer operated
The values shown in Fig. 32 are 15-minute averages. The data pertain to a continuous
deployment (CC) whereby the instruments sampled data at the selected frequencies until data
storage capacity was reached, a total duration of 3.5 hr. Another mode of deployment is the
discrete or block-sampled deployment (CD) during which the devices sample at the same
frequencies for 15 minutes every 3 hours. While the CD deployment is designed to sample
through at least two tidal cycles, the CC deployment permits an estimate of the "sub-grid scale"
activity. In this deployment, the effect of tower tilt, which can cause a misalignment between
the current meter axis and the horizontal streamwise component leading to very large errors in
Reynolds stress calculation, was corrected by an axis rotation of the data.
The velocity shows a gradual decline from a maximum flood velocity toward slack tide,
which implies a corresponding decline in the total average kinetic energy. The suspended load
also decreases, but starts to increase toward the end of the measurement period. It was found
that the time histories of suspended load does not correlate with the total energy distribution, but
does correlate very closely with the time traces of the turbulent and wave energy. These
measured data were then used to quantify the associated bottom sediment entrainment, deposition
and vertical net fluxes based on a turbulent control volume approach.
COE Field Tower (Mehta and Jiang, 1990)
The field tower, shown schematically in Fig. 33, consists of an aluminum frame 2.45 m
high with a rectangular base 1.5 m x 1.0 m, which tapers to 0.25 m by 0.15 m at the top.
Slanted bracing members provide adequate strength against buckling and torsion during
installation and retrieval operations. The tips (pins) of the four legs are conical in shape and
anchored into the ground to provide stability. A 0.8 m square wooden plank firmly fixed at the
top of the tower functions as a base for the mounting of the data acquisition system. A central
shaft 4.2 cm in diameter held within a concentric pipe of 5.8 cm outer diameter acts as a holder
for an accelerometer mounted in a plexiglass "boat", which consists essentially of a horizontal
oval disc with two vertical guide vanes. Other instruments include a pressure transducer and
a current meter.
Figure 34 shows a typical horizontal mud acceleration spectrum computed from the
accelerometer data collected during a deployment in Lake Okeechobee, Florida. The water
depth was 1.43 m, the mud thickness was 0.54 m, and the accelerometer was embedded 20 cm
into the mud. The spectrum shows a marked peak at a very low frequency corresponding to a
long period oscillation distinct from direct wind forcing. A plausible explanation for the low
frequency peak is a second-order effect resulting from wind-induced wave forcing leading to surf
beat-like response of the fluid-like mud bottom to wave forcing.
HR Instrumented Bed Frame (Diserens et al., 1993; Teisson et al., 1993)
The bed frame is designed (by Hydraulics Research Ltd of Wallingford, England) for
continuous measurement of time-varying hydrodynamic and sediment-related parameters due to
both tides and waves. It consists of a triangular frame with protruding legs that anchor it to the
bottom. The instrument package includes current meters (Braystoke for tidal current and
electromagnetic for wave-induced flow), turbidity sensors, and a pressure sensor. In addition,
ultrasonic probes mounted on smaller frames 3 m away from the main frame enable the bed
level change to be monitored. The sensors are connected to shore units (control and data
logging) housed in a van by cable links.
Figure 35 shows some typical results of field measurement obtained during a deployment
in an intertidal area at Eastham Dock, Mersey estuary, U.K. The bed frame was positioned
approximately 1.5 m above the Mean Low Water. A vertical array of current meters and
turbidity sensors sampled the data at 5 hz to enable recording of high frequency wave and
turbulence fluctuations. The data pertain to a spring-tide episode during which the water level
rose to a maximum water depth of 5.5 m. The bed shear stress was calculated from the tidal
mean velocities measured at three heights based on a logarithmic velocity distribution. The
wave heights during the recorded deployments were very small and had no significant influence
on bed shear stress.
The concentration and shear stress values shown are mean values averaged over 10
minute intervals. The maximum suspended sediment at 0.1 m (0.9 kg/m3) was higher than the
corresponding maximum at 1.0 m above the bed (0.55 kg/m3). Assuming that the sediment is
generated locally and there is little spatial variation over the immediate area, the rise in near-bed
suspension concentration was interpreted as suspended sediment descending through the water
column. The measured deposition of mud was shown to correspond with periods of low bed
shear stress, and more deposition occurred during periods of sustained high concentration levels.
Part H: Relationship Between the Erosion Rate Constant
and the Bed Shear Strength
A review of laboratory data on the erosion of cohesive sediment beds reveals that the
resulting empirical relationship between the rate of erosion, and the applied bed shear stress,
Tb, generally has one of the following two functional forms (Mehta, 1988):
3 b -Ts.'1
e = _b__ (1)
'3[rb 7,s)1/2 (2)
= u2e ;(z) I
where r7 is the erosion bed shear strength for a placed bed, rs(z) is the bed shear strength as
a function of depth, z, below the bed surface for a deposited bed, and ac, oa2 and 13 are
As discussed elsewhere (Hayter, 1983), Eq. 2 can be approximated by Eq. 1 in certain
cases. In any event, the focus here is on the linear functional form, Eq. 1, with a view to
establishing a possible link between the characteristic erosion rate constant, s (= a1/7r), and rs,
by reviewing the relevant laboratory data on placed bed erosion. It is clear from Eq. 1 that s
is the slope of the erosion line expressed as the rate of change of erosion rate with respect to the
excess shear stress, rb- s, and is consistent with the definition used by Arulanandan et al.
(1980). In this context, 152 data points have been gleaned from the literature. Further details
on the data are given in Appendix A.
An earlier effort in this direction has been made by Arulanandan et al. (1980) for natural
soil samples using distilled water. Their efforts yielded the following expression for estimating
the rate of change of erosion rate for a natural undisturbed soil subjected to hydraulic shear
stress from river (eroding) water:
s = 223exp(-0.13re) (3)
applicable over the range 3 < r, < 20 where s is in g/dynes-min and r7 is the critical shear
stress in dynes/cm2. After examining the distribution of the 151 data points in a semi-log plot,
it was noted that an equation of the form of Eq. 3, which is a straight line on a semi-log plot,
is unable to account for the observed flattening out tendency of the curve toward larger values
of 7,. Hence, the following functional form was selected to regress the data:
s = smaxexp(-ar) (4)
Based on an inspection of the data trend and spread, sma was fixed at 200 g/N-s, and a family
of seven curves could be distinguished, each surrounded by a separate cluster of data points.
The resulting two-parameter fits to the data were obtained by the method of least squares,
leading to the results given in Table 12. The respective curves are shown in Fig. 36-38. The
figures are separated into groups with different abscissa values in order to show the different
extents of the accompanying data points and yet with enough clarity at the low r, region to
distinguish them. Hence, for example, Group 1 curve is shown alone in Fig. 36 since the
farthest data point is located close to 70 Pa. It is noted that the curves overlap in the small r,
region. Also, the range of the reported values of erosion rate constant spreads over seven orders
Table 12. Sediment Data Groups.
Coefficients in Eq. 4
Group a b Number of data points
1 1.345 0.368 7
2 2.892 0.372 16
3 3.905 0.356 34
4 4.938 0.355 20
5 6.594 0.382 26
6 9.011 0.386 23
7 10.582 0.252 26
For each group, the characteristics of the relevant data points were further examined
under five important erosion resistance characterizing factors, namely, bulk density, clay
content, total salt concentration in the pore fluid, clay cation exchange capacity, and state of the
sediment bed (undisturbed or remolded), in an attempt to establish a rational means of data
division. The factors are summarized in Table 13. The grouping based on bulk density, clay
content, and cation exchange capacity does not seem to indicate any distinct influence of these
parameters. On the other hand, the grouping does correlate with state of sediment bed and total
salt concentration in the pore fluid. In general, undisturbed sediment beds and those with higher
total salt concentration exhibit higher bed shear strength (and correspondingly higher
characteristic erosion rate constant) compared with their remolded counterparts and those with
lower total salt concentration.
A nomograph shown in Fig. 39 is then proposed for estimating the characteristic erosion
rate constant, given the bed shear strength, based on the state of sediment bed and the total salt
concentration in the pore fluid. However, it is cautioned that the grouping has been premised
on the assumption that the results of erosion experiments used in this study are not dependent
on the type of erosion apparatus employed. The soundness of this assumption has not been
proven although it is inherent as a basis for reporting the results by the various investigators.
Therefore, given the five decadal range of the reported s values that were used in establishing
the individual curves, the nomograph is to be used as a last resort for guidance purpose and
should not in any way supplant the need for erosion experiments when the necessary facilities
Table 13. General characteristics of data groups.
Bulk Density Clay Content Total Salt Concentration
(kg/m3) (%) (meq/1)
Number Range Mean SD Range Mean SD Range Mean SD I
1 (7) 1,440-2,750 1,910 460 5-53 24 16 1.1-205.0 54.1 87.3 9
(6) (6) (5) (
2 (16) 1,420-2,080 1,730 170 12-46 27 12 2.2-145 26.9 41.7 7
(11) (12) (13) (
3(34) 1,480-1,860 1,670 160 6-50 28 10 377-39.5 19.0 6.7
(5) (32) (32)
4(20) 1,270-1,990 1,820 210 12-42 23 10 1.3-205.0 30.5 64.1
(15) (19) (18) (
H-5 (26) 1,350-2,090 1,820 200 11-53 27 11 1.7-205.0 21.7 42.0
(16) (22) (22) (
6 (23) 1.070-2,240 1,740 310 6-80 33 26 2.2-32.5 6.9 9.0
(20) (19) (13)
7(26) 1,100-2,400 1,730 380 6-94 34 19 1.1-6.2 3.2 1.8
(20) (22) (6)(
The number within parentheses refer to the number of data points.
SD = standard deviation; meq/1 = milliequivalents per liter; meq/100 g = milliequivalents per 100 g.
Cation Exchange Capacity
Range Mean SD
P.2-19.8 12.5 3.9
7.3-27.5 17.5 6.4
.5-22.9 15.1 5.5
.9-30.1 15.0 5.4
1.6-26.1 14.6 5.3
1.6-25.0 16.1 6.4
t.8-100.0 22.8 22.9
Ali, K.H.M. and Georgiadis, K. 1991. Laminar motion of fluid mud. Proceedings of the
Institution of Civil Engineers, Part 2, London, 795-821.
Ali, K.H.M., O'Connor, B.A. and Nicholson, J. 1992. Theoretical and experimental study of
fluid mud. Proceedings of the 5th International Symposium on River Sedimentation, Karlsruhe,
Alishahi, M.R and Krone, R.B. 1964. Suspension of cohesive sediments by wind-generated
waves. Technical Report HEL-2-9, Hydraulic Engineering Laboratory, University of California,
Berkeley, 24 p.
Amos, C.L., Christian, H.A., Grant, J. and Paterson, D.M. 1992. Sea-carousel a benthic,
annular flume. Estuarine Coastal ShelfScience, 34(6), 557-577.
Arulanandan, K., A. Sargunam, P. Loganathan and R.B. Krone. 1973. Application of chemical
and electrical parameters to prediction of erodibility. In: Soil erosion: causes and mechanisms,
prevention and control, Special Report 135, Highway research Board, Washington, DC, 42-51.
Arulanandan, K., Loganathan, P. and Krone, R.B. 1975. Pore and eroding fluid influences on
surface erosion of soil. Journal of Geotechnical Engineering Division, ASCE, 101(1), 51-66.
Arulanandan, K., Loganathan, P. and R.B. Krone. 1975. Pore and eroding fluid influences on
surface erosion of soil. Journal of Geotechnical Engineering Division, ASCE, 101(1), 51-66.
Arulanandan, K., Gillogley, E. and Tully, R. 1980. Development of a quantitative method to
predict critical shear stress and rate of erosion of natural undisturbed cohesive soils. Technical
Report GL-80-5, Geotechnical Laboratory, US Army Engineer Waterways Experiment Station,
Vicksburg, MS, 41p.
Bedford, K., Van Evra III R.E. and Valizadeh-Alavi, H. 1982. Ultrasonic measurement of
sediment resuspension. Proceedings of the Conference Applying Research to Hydraulic Practice,
P.E. Smith (ed.), ASCE, New York, 575-583.
Bedford, K., Wai., O., Libicki, C. and Van Evra, R. 1987. Sediment entrainment and
deposition measurements in Long Island Sound. Journal of Hydraulic Engineering, 113(10),
Berlamont, J., Ockenden, M., Toorman, E. and Winterwerp, J. 1993. The characterization of
cohesive sediment properties. Coastal Engineering, 21, 105-128.
Bohlen, F. 1982. In-situ monitoring of sediment resuspension in vicinity of active dredge spoils
disposal area. Proceedings of Oceans '82, IEEE Press, New York, 1028-1033.
Burt, T.N. and Game, A.C. 1985. The Carousel. Report No. SR33, Hydraulics Research,
Wallingford, U.K., 15p.
Chapuis, R.P. 1986. Quantitative measurement of the scour resistance of natural solid clays.
Canadian Geotechnical Journal, 23(2), 132-141.
Chapuis, R.P. and Gatien, T. 1986. An improved rotating cylinder technique for quantitative
measurement of the scour resistance of clays. Canadian Geotechnical Journal, 23(1), 83-87.
Christensen, R.W. and Das, B.M. 1973. Hydraulic erosion of remolded cohesive soils.
Highway Research Board Special Report No. 135, Washington DC, 8-19.
Croad, R.N. 1981. Physics of erosion of cohesive soils. Ph.D. dissertation, University of
Auckland, New Zealand.
Dash, U. 1968. Erosive behavior of cohesive soils. Ph.D. dissertation, Purdue University,
Diserens, A., Ockenden, M.C. and Delo, E.A. 1993. Application of a mathematical model to
investigate sedimentation at Eastham Dock, Mersey Estuary. In: Nearshore and Estuarine
Cohesive Sediment Transport, A.J. Mehta (ed.), American Geophysical Union, Washington DC,
Dunn, I.S. 1959. Tractive resistance of cohesive channels. Journal of Soil Mechanics and
Foundation Engineering Division, ASCE, 85(3), 1-24.
Einsele, G., Overbeck, R., Schwarz, H.U. and Unsold, G. 1974. Mass physical properties,
sliding and erodibility of experimentally depositional and differently consolidated clayey muds.
Sedimentology, 21(3), 339-372.
Espey, W.H. 1963. A new test to measure the scour of cohesive sediment. Technical Report
UYD-01-6301, Hydraulic Engineering Laboratory, University of Texas, Austin.
Fukuda, M.K. 1978. The entrainment of cohesive sediments in freshwater. Ph.D. dissertation,
Case Western Reserve University, Cleveland, Ohio, 210p.
Gularte, R.C., Kelly, W.E. and Nacci, V.A. 1977. The threshold erosional velocities and rates
of erosion for redeposited estuarine dredged material. Proceedings of the 2nd International
Symposium on Dredging Technology, Paper H3, British Hydraulic Research Association,
Gust, G. and Morris, M.J. 1989. Erosion thresholds and entrainment rates of undisturbed in situ
sediments. Journal of Coastal Research, SI5, 87-99.
Holland, A.F., Zingmark, R.G. and Dean, J.M. 1974. Quantitative evidence concerning the
stabilization of sediments by marine benthic diatoms. Marine Biology, 27(3), 191-196.
Hwang, K.N. and A.J. Mehta. 1989. Fine sediment erodibility in Lake Okeechobee, Florida.
Report UFL/COEL-89/019, University of Florida, Gainesville, 140p.
Hydraulics Research Limited. 1994. ISIS Instrument for shear strength In-Situ. Information
Pamphlet, Wallingford, England, Ip.
Jonsson, I.G. 1966. Wave boundary layers and friction factors. Proceedings of the 10th
Coastal Engineering Conference, Vol. 1, ASCE, New York, 127-148.
Kamphuis, J.W. and Hall, K.R. 1983. Cohesive material erosion by unidirectional current.
Journal of Hydraulic Engineering, 109(1), 49-61.
Kandiah, A. 1974. Fundamental aspects of surface erosion of cohesive soils. Ph.D.
dissertation, University of California, Davis, 236p.
Kelly, W.E. and Gularte, R.C. 1981. Erosion resistance of cohesive soils. Journal of
Hydraulics Division, ASCE, 107(10), 1211-1224.
Krishnappan, B.G. 1991. A rotating flume for cohesive sediment transport research. NWRI
Contribution No. 91-, National Water Research Institute, Burlington, Ontario, 28p.
Kuijper, C. Cornelisse, J.M. and Winterwerp, J.C. 1989. Research on erosive properties of
cohesive sediments. Journal of Geophysical Research, 94(C10), 14341-14350.
Kusuda, T., Umita, T., Koga, K., Futawari, T. and Awaya, Y. 1985. Erosional process of
cohesive sediments. Water Science Technology, 17(6/7), 891-901.
Lau, Y.L. and Krishnappan, B.G. 1991. Size distribution and settling velocity of cohesive
sediments during settling. Journal of Hydraulic Research, 30(5), 673-684.
Li, Y. 1989. Experiment and analysis on the transport and deposition of clay sediment. M.Sc.
thesis, Hohai University, Nanjing, The People's Republic of China, 120p (in Chinese with
Liou, Y.-D. 1970. Hydraulic erodibility of two pure clay systems. Ph.D. dissertation,
Colorado State University, 134 p.
Lonsdale, P. and Southard, J.B. 1974. Experimental Erosion of North Pacific red clay. Marine
Geology, 17(4), M51-M60.
Maa, J.P.-Y. 1991. In-situ measurements of the critical bed shear stress for erosion. In:
Environmental Hydraulics, Lee and Cheung (ed.), Balkema, Rotterdam, 627-632.
Maa, J.P.-Y. 1993. VIMS Sea Carousel: its hydrodynamic characteristics. In: Nearshore and
Estuarine Cohesive Sediment Transport, A.J. Mehta (ed.), American Geophysical Union,
Washington DC, 265-280.
Maa, P.-Y. and Mehta, A.J. 1987. Mud erosion by waves: a laboratory study. Continental
ShelfResearch, 7(11/12), 1269-1284.
Maa, J.P.-Y., Wright, L.D., Lee, C.-H., and Shannon, T.W. 1993. VIMS Sea Carousel: a
field instrument for studying sediment transport. Marine Geology, 115, 271-287.
MAST-G6M. 1992. On the methodology and accuracy of measuring physico-chemical
properties to characterize cohesive sediment. G6 Coastal Morphodynamics, under preparation.
Mehta, A.J. 1973. Deposition behavior of cohesive sediments. Ph.D. dissertation, University
of Florida, Gainesville, 296p.
Mehta, A.J. 1988. Laboratory studies on cohesive sediment deposition and erosion. In:
Physical Processes in Estuaries, J. Dronkers and W. van Leussen (eds.), Springer-Verlag,
Mehta A.J. and Partheniades, E. 1979. Kaolinite resuspension properties. Journal ofHydraulics
Division, ASCE, 105(4), 411-416.
Mehta, A.J., Parchure, T.M., Dixit, J.G. and Ariathurai, R. 1982. Resuspension potential of
deposited cohesive sediment beds. In: Estuarine Comparison, V.S. Kennedy (ed.), Academic
Press, New York, 591-609.
Mehta, A.J. and Jiang, F. 1990. Some field observations on bottom mud motion due to waves.
Report No. UFL/COEL-90/008, Coastal and Oceanographic Engineering Department, University
of Florida, Gainesville, 76p.
Mehta, A.J. and Srinivas, R. 1993. Observation on the environment of fluid mud by shear flow.
In: Nearshore and Estuarine Cohesive Sediment Transport, A.J. Mehta (ed.), American
Geophysical Union, Washington DC, 224-246.
Mimura, N. 1993. Rates of erosion and deposition of cohesive sediments under waves. In:
Nearshore and Estuarine Cohesive Sediment Transport, A.J. Mehta (ed.), American Geophysical
Union, Washington, DC, 247-264.
Moller-Jensen, P. 1993. Wadden Sea mud; methods for estimation of transport, erosion and
consolidation of marine cohesive sediments. Ph.D. dissertation, Department of Civil
Engineering, Aalborg University, Denmark, 145p.
Moore, W.L. and Masch, F.D. Jr. 1962. Experiments on the scour resistance of cohesive
sediments. Journal of Geophysical Research, 67(4), 1437-1449.
Murakami, K., Suganuma, F. and Sasaki, H. 1989. Experimental investigation on erosion and
deposition of fine cohesive sediments in an annular rotating channel. Report of the Port and
Harbor Research Institute, 28(1), 43-76 (in Japanese with English abstract).
Narimousa, S. and Fernando, H.J. S. 1987. On the sheared interface of an entraining stratified
fluid. Journal of Fluid Mechanics, 174, 1-22.
Nichols, M.M. 1989. Field instrument system for fluid mud transport and behavior. Journal
of Coastal Research, SI5, 31-38.
Nichols, M.M., Thompson, G.S. and Faas, R.W. 1978. A field study offluid mud dredged
material: its physical nature and dispersal. Technical Report D-78-40, U.S. Army Engineer
Waterways Experiment Station, 74p.
Nowell, A.R.M., McCave, I.N. and Hollister, C.D. 1985. Contributions of HEBBLE to
understanding marine sedimentation. Marine Geology, 66(1/4), 397-409.
Parchure, T.M. 1984. Erosional behavior of deposited cohesive sediments. Ph.D. dissertation,
University of Florida, Gainesville, 321p.
Parchure, T.M. and Mehta, A.J. 1985. Erosion of soft cohesive sediment deposits. Journal
of Hydraulic Engineering, 111(10), 1308-1326.
Partheniades, E. 1965. Erosion and deposition of cohesive soils. Journal of the Hydraulics
Division, ASCE, 91(1), 105-138.
Partheniades, E., Kennedy, J.F., Etter, R.J. and Hoyer, R.P. 1966. Investigations of the
depositional behavior of fine cohesive sediments in an annular rotating channel. Report No. 96,
Hydrodynamics Laboratory, Massachusetts Institute of Technology, Cambridge, 81p.
Partheniades, E. and Paaswell, R.E. 1970. Erodibility of channels with cohesive boundary.
Journal of Hydraulic Division, ASCE, 96(3), 755-771.
Paulic, M., Montague, C.L. and Mehta, A.J. 1986. Influence of light on sediment erodibility.
Proceedings of the 3rd International Symposium on River Sedimentation, S.Y. Wang et al. (ed.),
Jackson, MS, 1759-1764.
Peirce, T.J., Jarman, R.T. and de Turville, C.M. 1970. An experimental study of silt scouring.
Proceedings of the Institution of Civil Engineers, London, 231-243.
Raudkivi, A.J. and Hutchison, D.L. 1974. Erosion of kaolinite clay by flowing water.
Proceedings of the Royal Society of London, A337, 537-554.
Rohan, K., Lefebvre, G., Douville, S. and Milette, J.-P. 1986. A new technique to evaluate
erodibility of cohesive material. Geotechnical Testing Journal, 9(2), 87-92.
Rohan, K. and Lefebvre, G. 1991. Hydrodynamic aspects in the rotating cylinder erosivity test.
Geotechnical Testing Journal, 14(2), 166-170.
Schunemann, S. and Kuhl, H. 1993. Experimental investigations of the erosional behavior of
naturally formed mud from the Elbe Estuary and adjacent Wadden Sea, Germany. In: Nearshore
and Estuarine Cohesive Sediment Transport, A.J. Mehta (ed.), American Geophysical Union,
Washington DC, 314-330.
Shaikh, A., Ruff, J.F. and Abt, S.R. 1988. Erosion rate of composite NA-montmorillonite
soils. Journal of Hydraulic Engineering, 114(3), 296-305.
Sheng, Y.P. 1989. Consideration of flow in rotating annuli for sediment erosion and deposition
studies. Journal of Coastal Research, SI5, 207-216.
Talebbeydokhti, N. and Klingeman, P.C. 1992. Suspension of fine-grained sediment. Iranian
Journal of Science and Technology, 16(1), 29-42.
Task Committee on Erosion of Cohesive Materials. 1968. Erosion of cohesive sediments.
Journal of Hydraulic Division, ASCE, 94(4), 1017-1049.
Teeter, A.M. 1994. A Particle entrainment simulator to gage cohesive sediment erodibility.
Dredging Research Technical Note DRP 01-xx, U.S. Army Engineer Waterways Experiment
Station, Vicksburg, MS (under review).
Teeter, A. M. and Pankow, W. 1989. Deposition and erosion testing on the composite dredged
material sediment sample from New Bedford Harbor, Massachusetts. Technical Report HL-89-
11, Hydraulics Laboratory, Waterway Experiment Station, Vicksburg, MS, 36p.
Teisson, C., Ockenden, M., Le Hir, P., Kranenburg, C. and Hamm, L. 1993. Cohesive
sediment transport processes. Coastal Engineering, 21(1-3), 129-162.
Thorn, M.F.C. and Parsons, J.G. 1980. Erosion of cohesive sediments in estuaries: an
engineering guide. Proceedings of the 3rd International Symposium on Dredging Technology,
Bedford, England, 349-358.
Tofts, H. 1993. Erosional behavior of partly cohesive, partly non-cohesive sediments in uniform
flow. Proceedings of the 6th International Conference on Urban Storm Drainage, Niagara Falls,
Tsai, C.-H. and Lick, W. 1986. A portable device for measuring sediment resuspension.
Journal of Great Lakes Research, 12(4), 314-321.
Tsai, C.H. and Lick, W. 1988. Resuspension of sediments from Long Island Sound. Water
Science Technology, 20(6/7), 155-164.
van Leussen, W. and Winterwerp, J.C. 1990. Laboratory experiments on sedimentation of fine-
grained sediments: a state-of-the-art review in the light of experiments with the Delta Tidal
Flume. In: Residual Currents and Long-term Transport, R.T. Cheng (ed.), American
Geophysical Union, Washington DC, 241-259.
Villaret, C. and Paulic, M. 1986. Experiments on the erosion of deposited and placed cohesive
sediments in an annular flume and a rocking flume. Report No. UFL/COEL-86/007, Coastal and
Oceanographic Engineering Department University of Florida, Gainesville, 61p.
Watanabe, A., Thimakorn, P. and Das Gupta, A. (1978). Concentration of suspended clay in
periodic flow. Proceedings of the 16th Coastal Engineering Conference, Vol. 2, ASCE, New
Wolanski, E., Asaeda, T. and Imberger, J. 1989. Mixing across a lutocline. Limnology and
Oceanography, 34(5), 931-938.
Young, R.A. 1977. SEAFLUME: a device for in-situ studies of threshold erosion velocity and
erosional behavior of undisturbed marine muds. Marine Geology, 23, M11-M18.
Tabulation of Test and Soil Conditions, Bed Shear Strength (rs) and
Characteristic Erosion Rate Constant (s)
Investigators) Test Conditiona Soil Conditionb s
Espey (1963) Rotating Cylinder Taylor marl 60.91 0.51
0/0.47/-/2,750 50% < 5 pm
Partheniades Straight Flume (Series 1) Mare Island Strait, San 0.437 0.01
(1965) dense, uniform bed Franscisco Bay: princiapally M
33/1.1/-/2,240 60% < 2 pm
Straight Flume (Series I) Same as above 1.206 0.01
Christensen Drill-hole Apparatus K 0.495 0.55
and Das 0/0.35/13/- 53% < 2 pm
0/0.33/13/- K 0.776 0.06
62% < 2 pm
Raudkivi and Closed Conduit K (CEC= 4.8) 0.227 0.13
Hutchison 0.01M NaNO3/0.51/21/- 2 5 pm
(1974) 0.01M NaNO3/0.51/32/- Same as above 0.247 0.08
Kandiah Rotating Cylinder I(30%)+SF(70%) 3.43 0.87
(1974) Homoionic (MgCl2 pore fluid)
TS = 20.0 meq/1
homoionic (CaC12 pore fluid) Same as above 1.34 1.1
TS = 9.8 meq/1
TS = 21.0 meq/1 Same as above 3.03 0.66
TS = 39.5 meq/1 Same as above 4.73 0.48
TS = 80.0 meq/1 same as above 8.12 0.45
TS = 20.0 meq/1 Yolo Loam Silt + K(50%) 0.73 8.8
SAR = 48
K(40%) 0.73 8.2
K(30%) 0.723 4.6
Investigators) Test Conditiona Soil Conditionb I s s
TS = 20.0 meq/1
Yolo Loam Silt + K(30%) 1.23 2.0
SAR = 2.5
K(40%) 1.43 1.9
K(50%) 1.63 1.6
TS = 20.0 meq/1 Yolo Loam Silt + I(40%) 0.35 4.8
SAR = 48
1(30%) 0.45 2.5
1(20%) 0.45 2.2
TS = 20.0 meq/1 Yolo Loam Silt + 1(20%) 1.93 0.97
SAR = 2.5
1(30%) 2.03 0.90
1(40%) 2.13 0.80
TS = 20.0 meq/1 Yolo Loam Silt + M(30%) 0.15 6.9
SAR = 48
M(20%) 0.155 5.6
M(10%) 0.256 2.3
TS = 20.0 meq/1 Yolo Loam Silt + M(10%) 2.0 1.0
SAR = 2.5
M(20%) 2.43 0.83
M(30%) 2.73 0.51
TS = 20.0 meq/1 Remolded bay mud:CC=11% 0.25 8.6
SAR = 46.5
CEC = 15.0 meq/100g
Remolded natural soil 0.34 10.5
SAR = 46.5
CEC = 12.0
Remolded Yolo Loam: 0.54 5.6
SAR = 46.5
CEC = 10.4
TS = 20.0 meq/1
Remolded bay mud: CC=11%
SAR = 4
CEC = 15.0 meq/100g
Investigators) Test Condition" Soil Conditionb 7
TS = 20.0 meq/1
Remolded Yolo Loam: 1.6- 2.5
SAR = 4
CEC = 10.4 meq/100g
Remolded natural soil 1.73 2.0
SAR = 4
CEC = 12.0 meq/100g
TS = 20.0 meq/l Remolded natural soil 2.33 1.8
SAR = 4
CEC = 21.2 meq/100g
TS = 20.0 meq/1 Remolded natural soil (Meyers 2.43 0.53
SAR = 4
CEC = 22.5 meq/100g
Remolded natural soil 2.84 0.16
CEC = 30.1 meq/100g
TS = 20.0 meq/1 1(30%) + SF(70%) 0.305 3.2
OM = 2.8%
OM = 5.7% 0.40 3.4
SAR = 3.0 1.803 1.3
OM = 0.85% 2.403 0.87
OM = 2.7% 4.30 0.47
OM = 5.6% 3.602 1.2
TS = 20.0 meq/1
1(30%) + SF(70%)
SAR = 3.0
Arulanandan Rotating Cylinder Yolo loam 0.415 1.92
et al. (1973) -/0.28/-/- SAR = 23.2 _
Investigators) Test Conditiona Soil Conditionb s
Arulanandan Rotating Cylinder SAR = 12.4 3.042 1.92
et al. (1973) -/0.28/-/-
(continued) SAR = 9.4 4.773 0.31
SAR = 1.1 19.42, 0.15
-/0.40/-/- 0.001N NaCI pore fluid 2.142 2.82
SAR = 35
0.005 N NaCI pore fluid 4.0 0.63
Arulanandan Rotating Cylinder Yolo Loam: 0.005N NaCI 0.106 5.25
et al. (1975) 0/0.28/-/- CEC = 19.8; 19% < 2 itm
pore fluid with SAR = 1.1
0/0.28/-/- Pore fluid SAR = 1.6 0.064 21.20
0/0.30-/-/ Pore fluid SAR = 10.7 0.011 147.3
Gularte et al. Closed Conduit Thames River spoil: 0.277 0.11
(1977) 31/1.13-2.22/-/2,400 OM = 6.1 12.6
10% < 2 pm
Fukuda Annular Flume: Type I Western Basin sediment, Lake 0.167 0.75
(1978) 0/0.76/-/- Erie
0/0.79/-/- dso = 3 jim 0.077 0.44
Thorn and Straight Flume Grangemouth Mud 0.027 2.62
Parsons 10 min erosion test K(17%)+I(17%)+CH(17%)
(1980) 28/-/-/- CEC = 20
OM = 10%
20 min erosion test (p = 2,080) 0 1.
30 min erosion test (p = 2,080) 0.08 1.53
10 in erosion test (p = 1,510) Brisbane mud 0.047 1.76
M(30%) +K(15%) +I1(5%)
20 min erosion test (PB = 1.510) CEC = 35 0.047 1.41
30 min erosion test (B = 1.510)0.057 1.18
10 min erosion test (B = 1,720) Belawan Mud 0.216 2.89
M(20%)+K(30%) +1(30 %)
20 min erosion test (pB = 1,720) CEC = 25 0.22 1.79
30 min erosion test (PB = 1,720) 0.186 1.39
Investigators) Test Conditiona Soil Conditionb s
et al. (1980)
I -70--0- -. 1 .
205/0.36/-/1,680 9.0/7.0/21.7/11.3/22% 3.81 19.2
145/0.42/0/1,550 0./7.8/23.1/7.3/13% 0.72 17.8
2.7/0.29/-/1,810 26.6/11.0/0.1/20.4/40% 0.32 26.7
34.0/0.24/-/1,420 14.3/10.4/0.4/11.6/25% 2.22 2.7
55.0/0.26/-/1,980 49.7/7.1/14.8/9.8/53% 3.81 21.3
2.5/0.36/-/1,740 16.5/5.0/0.1/14.5/30% 0.24 13.7
5.6/0.29/-/1,810 1.9/4.0/0.2/9.0/17% 0.43 15.3
29.0/0.38/-1,810 23.7/5.2/4.7/20/30% 0.15 23.3
9.6/0.23/-/2,090 11.0/4.2/1.9/12.1/25% 0.35 3.5
10.8/0.30/-/2,080 36.1/7.5/9.2/27.5/42% 0.42 21.3
2.95/0.41/0/1,850 31.3/5.5/0.5/26.1/46% 1.92 6.1
4.7/-/-/1,820 34.1/4/0.9/18.8/25% 3.32 2.2
6.0/0.46/-/1,610 25.3/8.6/1.3/24.1/42% 1.12 7.0
28.0/0.5/-/1,610 29.0/11.6/0.6/22.9/40% 1.53 2.5
2.4/0.39/-/1,630 0./5.1/0.2/7.6/12% 0.26 1.5
2.2/0.43/-/1,270 8.8/5.5/0.3/20/14% 0.54 7.1
3.8/0.44/-1,690 8.6/5.7/0.4/13.3/12% 0.32 49.2
8.0/0.35/-/1,950 6.8/3/0.7/9.2/18% 0.41 125.8
6.1/0.38/-/1,760 21.3/9.1/0.3/17.5/34% 0.54 6.5
1.5/0.34/-/1,440 0./4.9/0.1/11.3/5% 0.31 75.0
3.7/0.31/-/1,480 3./5./0.4/8.5/6% 1.13 4.5
1.1/0.30/-/1,670 10.5/5.1/0.2/13.5/29% 3.91 9.2
4.8/0.43/-/1,730 15.3/6.4/0.6/16/28% 0.32 36.7
4.0/0.3/-/1,610 14.3/5.7/0.5/13.9/29% 1.63 3.0
Investigators) Test Conditions Soil Conditionb s
205(0.)/0.29/-/1,950 9.0/7.0/21.7/11.3/22% 1.05 0.22
205(13.25)/0.30/-/1,930 4.24 0.15
205(26.5)/0.29/-/1,930 3.44 0.06
2.7(0.)/0.34/-/1,850 26.6/11/0.1/20.4/40% 2.05 0.03
2.9(0)/0.32/-/1,860 11.1/4.7/0.1/15.5/33% 0.66 0.16
2.9(1.45)/0.33/-1,850 11.1/4.7/0.1/15.5/33% 4.24 0.05
2.5(0.9)/0.32/-/1,890 16.5/5/0.1/14.5/30% 0.47 0.04
5.6(1.25)/0.28/-/1,950 1.9/4/0.2/9/17% 0.95 1.0
5.6(2.5)/0.28/-/1,930 1.9/4/0.2/9/17% 0.85 0.22
1.3(0.)/0.24/-/1,960 3.2/3.3/0.1/7.9/17% 2.24 0.27
1.3(0.65)/0.24/-/1,990 3.2/3.3/0.1/7.9/17% 2.84 0.09
1.7(0.33)/0.38/-/1,740 18.8/5.8/0.1/20.3/37% 0.27 0.09
1.7(0.66)/0.38/-/1,820 18.8/5.8/0.1/20.3/37% 1.35 0.08
2.95(0.)/0.54/-/1,630 31.3/5.5/0.5/26.146% 0.65 0.57
4.7(0.)/0.31/-/1,830 34.1/4/0.9/18.8/25% 0.36 0.72
2.8(0.)/0.44/-/1,740 29/11.6/0.6/22.9/40% 0.56 0.09
2.4(0.)/0.34/-/1,930 0./5.1/0.2/7.6/12% 0.355 1.1
2.4(0.235)/0.31/-/1,950 0./5.1/0.2/7.6/12% 0.36 0.62
2.4(0.705)/0.31/-/1,930 0./5.1/0.2/7.6/12% 0.356 0.57
2.2(0.69)/0.30/-/1,940 8.8/5.5/0.3/20/14% 0.56 0.38
2.2(1.38)/0.30/-/1,940 8.8/5.5/0.3/20/14% 1.64 0.43
3.8(0.69)/0.31/-/1,960 8.6/5.7/0.4/13.3/12% 0.17 0.38
3.8(1.38)/0.27/-/1,950 8.6/5.7/0.4/13.3/12% 0.56 0.11
8.0(0.565)/0.31/-1,900 6.8/3/0.7/9.2/18% 1.55 0.06
6.(./03// ,802././.31./4 .
et al. (1980)
Investigators) Test Conditiona Soil Conditionb s
Arulanandan 6.1(0.572)/0.36/-/1,820 21.3/9.1/0.3/17.5/34% 1.75 0.04
et al. (1980)
remoldedd 1.5(3.81)/0.27/-11,910 0./4.9/0.1/11.3/5% 0.94 1.3
bed) 3.7(1.3)/0.27/-/1,990 3/5/0.4/8.5/6% 0.156 1.4
3.7(2.6)/0.27-/1,970 3/5/0.4/8.5/6% 0.127 0.4
1.1(0.2)/0.38/-/1,810 10.5/5.1/0.2/13.5/29% 0.107 1.2
1.1(0.4)/0.39/-/1,800 10.5/5.1/0.2/13.5/29% 0.505 0.5
4.8(0.2)/0.35/-/1,860 15.3/6.4/0.6/16/28% 2.50 0.68
4.8(0.4)/0.35/-/1,850 15.3/6.4/0.6/16/28% 2.304 0.17
4.0(0.952)/0.31/-/1,900 14.3/5.7/0.5/13.9/29% 0.606 0.12
6.2(0.)/0.28/-/1,970 15.5/6.2/0.6/14.2/30% 0.107 0.83
6.2(0.537)/0.28/-/1,950 15.5/6.2/0.6/14.2/30% 0.406 0.32
6.2(1.61)/0.28/-/1,930 15.5/6.2/0.6/14.2/30% 2.005 0.03
6.2(3.22)/0.29/-/1,910 15.5/6.2/0.6/14.2/30% 4.804 0.02
Gularte et al. Closed Conduit Grundite: I(50%)+Silt(50%) 0.067 0.965
10/0.6/-/- Same as above 0.257 0.314
Villaret and Annular Flume Placed K: CEC = 6 0.257 0.20
Paulic (1986) 10/-/24-27/1,630
Same as above Placed Cedar Key mud: 0.207 0.08
CEC = 100; OM = 11%
Hwang (1989) Annular Flume: Type I Lake Okeechobee mud 0.437 0.018
placed bed, lake water principally K + M + S
bulk density = 1,100 kg/m3 OM = 40%
bulk density = 1,190 kg/m3 0.647 0.01
bulk density = 1,070 kg/m3 0.346 0.47
bulk density = 1,090 kg/m3 0.556 0.074
MAST G6M bulk density = 1,140 kg/m3 0.097 0.30
Investigators) Test Conditiona Soil Conditionb s
MAST G6M bulk density = 1,210 kg/m3 0.247 0.34
(continued) bulk density = 1,270 kg/m3 0.506 0.31
bulk density = 1,310 kg/m3 0.526 0.23
bulk density = 1,350 kg/m3 1.225 0.28
bulk density = 1,380 kg/m3 1.735 0.15
bulk density = 1,410 kg/m3 2.474 0.15
TS = total salt concentration in pore fluid (milliequivalents/liter); K = kaolinite; M = montmorillonite; I = illite; SF
= silica fl6ur; SAR = sodium adsorption ratio; CEC = cation exchange capacity (milliequivalents/100 g); CH =
chlorite; S = sepiolite; OM = organic matter in percent; CC = clay content in percent.
a) The four values in 1/2/3/4 refer to 1 = salinity of pore fluid in ppt except for Arulandandan et al.
(1980) where it refers to total salt concentration in soil pore space (milliequivalents/liter), the value
in parentheses, where it appears, refers to total salt concentration (meq/l) in the eroding fluid, 2 =
percent water content by weight, 3 = temperature in oC, 4 = bulk density, p (kg/m3). A dash
denotes that the information is not available.
b) The five values in 1/2/3/4/5 refer to 1 = plasticity index, 2 = percent organic content by weight, 3
= SAR and 4 = CEC as defined above, 5 = clay content in percent. A dash denotes that the
information is not available.
c) The superscript on the values listed in column under T, refers to the data group number.
SItMJCtlON VIL tPiu
_I~_ _. --~--I -- -
ICMI 0 tUOIWl
OW104M to U.95COV*1
Schematic of the Delft Tidal Flume (from Kuijper
et al., 1989).
.0 Jau BED SHEAR STRESS: 0.t Po
2.0 BED SHEAR STRESS: 0.17 Po
ru..U BED SHEAR STRESS. 0.22 Pa
50 100 10o
TIME in hours
A typical suspended sediment-time curve for an
erosion experiment in the tidal flume (from Kuijper
et al., 1989).
"-'--- -- mi
6ll V Bottom
a laboratory rotating
a < 3 ~ 5 -o 1- is .3 A is r P Is -20 23-- -
".- .-_ I-- k.
A typical result of resuspension experiment using
the annular flume for a stratified bed at a bed shear
stress (rb) of 0.21 Pa (from Mehta and
7/7///77777//// ////// '
KI *-**** **** R t
JII : : ; -- :- : t;** *i ll
I- I I ''. .......... ..,i
_111~.- ,. ..... 1 ..
l '^ ---,-"------ !-- ---- !. i I III
SO L 4 2 4 I 61 20 II 24 a a 2 J 6 3 M M 4V0 4244 4
A typical result of resuspension experiment using
the annular flume for a uniform bed at a bed shear
stress (rb) of 0.41 Pa (from Mehta and
schematic of a race-track
flume (from Mehta and
VELOCITY, u (cm/s)
Fig. 7. Typical velocity profiles from a test run using a
race-track flume with bottom mud composed of
kaolinite (from Mehta and Srinivas, 1993). t is the
elapsed time following test initiation, U is the
characteristic mixed-layer velocity, and 6, is the
shear layer thickness. As an example, the dashed
line indicates the location of the bottom elevation of
5, for profile No. 4 (t = 54 min).
sSymbol I6 Pbl 1
S(mini) (cm) (glan3)
A 35 1.8 1.025
0 40 45 1.7 1.028
0 x 54 2.0 1.02s
1 60 1.5 1.03
.00 1.02 1.04 1.06
DENSITY, p (g/cm3)
Fig. 8. Density profiles corresponding to the velocity
profiles shown in Fig. 11 (from Mehta and
Srinivas, 1993). 5 is the density interfacial layer
thickness and Pb is the density value at the base of
the corresponding 5. The initial fluid-bed interface
is indicated by the density profile at t = 0.
hFnl .taf Hydul Troauic n / TnL s3o
arnd Soee COntroller
a) Schematic of a rocking flume; b) Top
constriction placed in the rocking flume (from
Villaret and Paulic, 1986)
A typical concentration-time curve for a deposited
kaolinite bed using the rocking flume (from Villaret
and Paulic, 1986).
A typical concentration-time curve for deposited
Cedar Key (Florida) mud bed using the rocking
flume (from Villaret and Paulic, 1986).
NOTE. NOT TO CAL.
CONTRUCTEO OOF 1.2Sc
THICK CLEAR ACRYLIC.
9em x 3Jcm x 2114cm DUCT
S_ SAMPLING TUBE
4 REINFORCING FLANGES
FLOWMTErR ASCENDINGTUBE '
2 03em 10 ANO PROPELLERS
CONNECTION FLANGES SACCESSAND
9cm x 3cm a 2114cm DUCT ELBOW
.- FLUID RESERVOIR
Schematic of a closed-conduit sediment water tunnel
(from Teeter and Pankow, 1989).
Schematic of a rotating cylinder apparatus (from
Rohan and Lefebvre, 1991).
Y. O ,
S NtM TRIAIIS.LY RECONSOUDATC
a INTACT. CUT IN A LOCK
CONTACT $HEAR STRESS (Pa)
A typical erosion rate versus shear stress curve
using a rotating cylinder apparatus for intact clay
samples and triaxially prepared samples (from
samples and triaxially prepared samples (from
I-- ,.. -
!OOk -i 130
- -I~--. .~ -~
PIEOMETER RING PRESSURE MEASURIG OCVICE
SMOOTH WALLS TUBE
STAINLESS STEEL TUBE de 44 4Smm d, a 3 3m.mm
SPORTS OF STAINLESS STEtL
1 CLAY SAMPLE
Schematic of a drill-hole test apparatus (from Rohan
et al., 1986).
SVELOCITY (m/stc) I0 029
12 3 4 T ? -
I 7 I I
VELOCITY (M/seC) f .0.0O
4 S 6
'0 .1 I I -- 1 --- -- -- T -- -- I -- i ,Ii
GO4 GRANDC BtAINC)
z50 ST I (ST-BARNABE)
s0 00 SO 200 250 300 350
TRACTIVE FOACC (Pt)
400 430 500
A typical eroded sediment mass versus tractive
force curve from a drill-hole test apparatus (from
Rohan et al., 1986).
S-: /Y i
A4~. A.l ~I )
I m n t a T i m o r e
Schematic of a vertical grid
and Lick, 1986).
oscillator (from Tsai
Fig. 18. Time histories of the shaker concentration at five
stations (from Tsai and Lick, 1986).
Schematic of the EROMES System (from
Schunemann and Kuhl, 1993).
0.e -il Ili iI.
0 I 2 3 4 1 6 7 8 9
____ I II
I I I 'II
0 I 2 3. 4 5 6 7 6 9
Typical results from an erosion experiment using
the EROMES System (from Schunemann and Kuhl,
1993). Upper: suspended concentration; middle:
bottom shear stress; lower: erosion rate.
ISe- a MIser
*T it -- 8 11 .. .. -- ,
T- ; i I r
Fig. 21. Schematic of a rotating disk device (from Liou,
010.1 1 NWC
0~ N CeC M .
0.01 M Na'c
Eeon TUe (Ik)
Fig. 22. Typical time-variations of erosion shear strength
using a rotating disk device for high water content
bentonite samples without chemical additives, and
with 0.1 N, 0.01 N NaCI and 0.01 N CaCI,
additives (from Liou, 1970).
Schematic of a submerged jet apparatus (from Dash,
1M I MNUTII
CLAV -.w T7o
* ta l.aJ OtI P. wuD. *J PT.
* KRAIl*%.4 '0A4. ~*Am 4.7sr
* I ma4PSI. MCAD* 4.75 PT
* ItKSa4,80 *Pa. A *n 1i, t
"~r K31ql lCY' P"i. 14gA)
Typical time-variations of weight loss using a
submerged jet apparatus for consolidated samples
with no sand (from Dash, 1970). ao denotes
maximum consolidation pressure.
,1*o 1o g2 (10 <0
A typical time-variation of optical transmissometer
outputs during a SEAFLUME deployment (from
Gust and Morris, 1989). Open and solid circles
refer to the front and back optical transmissometers,
0 FLUME PUMP OFF
aS 0 *
*. 0 0 0
0 0 .
Schematic of a field rotating annular flume (from
Maa et al., 1993).
1 3 4 51 I, r 211,1I.-L .,n...,7,I .l .,H l.. l ,71.77.7 1. 7 l' .<.
T I n E (minute)
Typical results of an incipient erosion motion
experiment during the Sea Carousel deployment
(from Maa, 1991): a) ring speed versus time; b)
OBS reading for suspended sediment concentration
(SSC). Numbers are shear stresses in dynes/cm2.
Schematic of the Sediment-water Interface Probe
(from Nichols et al., 1978).
a INCIPIENT MOTION EXPERIMENT .
LO WOLFTRAP SHOAL
4 ring speed-J-
A typical analogue time-series record (from
Nichols, 1989): A) current speed, u and v,
components at 0.2 time-constant, showing
intermittent nature of contributions at 30 cm above
a fluid mud during ebb-tidal current, B) turbidity
showing small-scale fluctuations, station 19, upper
Chesapeake Bay. Turbidity units are in percent
A photograph of the DAISY Tower Instrumentation
Array (from Bohlen, 1982).
NeIW .ONOO SirtE
* I ij i i
i "f "I ,
Typical results from a field deployment of DAISY
Tower (from Bohlen, 1982). Upper: suspended
sediment concentration; middle: near bottom current
speed; lower: wind speed at 10 m elevation,
Millstone Point, Long Island Sound.
- -0--w anuonwV n aMc.r aiunie. rop)
a- ----S w usnded Lfl6Miu* w Aerol)o)
- \ 0.
0 0 0
-~ ~ \'
S 20 40 60 80 100 120 140 160 18
A typical time-trace of 15-min average horizontal
velocity and suspended load for continuous
deployment of the OSU instrumentation (from
Bedford et al., 1987).
for Data Acquisition
Z5 \ -T-
tS \ n
of the COE Field Tower (from Mehta
0.4 0.6 0.8
A typical computed horizontal mud acceleration
spectrum from accelerometer data collected during
a deployment of the COE Field Tower (from Mehta
and Jiang, 1990).
- 0.014 -
Time after high water (mins)
Bed level 20.11.90 pm
L -- ------. 4 .
.300 -100 100 -
Time after high water (mins)
Time after high water (mins)
Typical results of field measurement at Mersey Site
spring tide, 11/20/90: A) concentration, B) bed
level, C) shear stress (from Teisson et al., 1993).
r- Group 1
L 1 0.
0 10 20 30 40 50
Bed shear strength (Pa)
Fig. 36. s 7, curve for Group 1.
Ld 0.1 ,
Bed shear strength (Pa)
s 7, curve for Groups 2, 3, 4 and 5.
Bed shear strength
s 7, curve for Groups 6 and 7.
5 6 7 8 9
Bed shear strength
Nomograph for characteristic erosion rate constant.
Note: Number refers to total salt concentration In
the pore fluid (milliequlvalents/liter)
2 Mostly remolded/30