PREDICTION OF M A X I M U M SCOUR D E P T H

AT SUBMARINE PIPELINES
By Yee-Meng Chiew, 1 M e m b e r , A S C E
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

ABSTRACT: This paper examines and reviews published results relating to local
scour around submarine pipelines. It highlights the limitations of existing methods
for estimating scour depth at the pipeline. Based on experimental results, the study
proposes an empirical function relating the amount of gap flow through the scour
hole for given flow conditions. With the aid of this function, it is possible to
predict the maximum Scour depth at submarine pipelines for given flow and geo-
metric boundary conditions. Published results suggest that the maximum equilib-
rium scour depth occurs when the pipeline is just lying on a plane bed and sub-
jected to a pure unidirectional current. The undisturbed bed shear stress is equal
to the critical shear stress for sediment entrainment. This condition implies that
there is no general sediment transport away from the pipeline. The predicted max-
imum scour depth using the iterative method proposed in this study compares well
with experimental results.

INTRODUCTION

The rapid development of offshore oil fields has increased the construction
of submarine pipelines for transport of crude oil to onshore refineries. In-
teractions between the pipelines and an erodible bed under current and/or
wave conditions tend to cause scouring around the pipelines. Scour under-
neath the pipeline may expose part of the pipe causing it to suspend in water.
If the free span of the pipe is long enough, the pipe may experience resonant
flow-induced oscillations, leading to structural failure. Accurate estimates of
the scour depth is important in the design of submarine pipelines. At present,
several empirical methods, based on various research findings, can be found
in literatures for estimating the equilibrium scour depth under unidirectional
current. The present study summarizes and critically reviews these research
works, and proposes a semiempirical method for estimation of maximum
scour depth based on understanding of the physics of flow and sediment
transport around the pipeline.

REVIEW OF PREVIOUS STUDIES

The number and variety of formulas proposed for prediction of local scour
depth around bridge piers is numerous, but comparatively few are found for
estimating scour depth at submarine pipelines. This study scrutinizes five
research investigations and their findings reported in published literatures.
The first of these five investigations was conducted by Kjeldsen et al. (1973).
The data were based on flume tests conducted at the Technical University
of Norway. Ibrahim and Nalluri (1986) proposed two empirical equations
relating scour depths to the flow parameters. These equations were derived
'Sr. Lect., School of Civ. and Struct. Engrg., Nanyang Tech. Inst., Nanyang Ave-
nue, Singapore 2263.
Note. Discussion open until September 1, 1991. To extend the closing date one
month, a written request must be filed with the ASCE Manager of Journals. The
manuscript for this paper was submitted for review and possible publication on April
27, 1990. This paper is part of the Journal of Hydraulic Engineering, Vol. 117,
No. 4, April, 1991. ©ASCE, ISSN 0733-9429/91/0004-0452/$ 1.00 + $.15 per page.
Paper No. 25672.
452

J. Hydraul. Eng. 1991.117:452-466.

 TABLE 1. Boundary Conditions of Tests Conducted by Previous Investigators

Pipe
Water depth Velocity Grain size diameter
Reference (m) (m/s) (mm) (mm)
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

d) (2) (3) (4) (5)

60, 110,
Kjeldsen et al. (1973) 0.43, 1.43 0.20-0.52 0.074 225, 500
Ibrahim (1986) 0.032-0.30 0.19-0.48 0.425, 0.8, 1.5 25, 50, 75
Mao (1986) 0.25, 0.35 — 0.36 50, 100

from experimental data (Ibrahim 1986) collected at the University of New-

castle upon Tyne in England. Chao and Hennessy (1972) proposed an an-
alytical method for estimating maximum scour depth under pure current con-
dition. Herbich (1981, 1985) and Herbich et al. (1984) advocated use of this
method and reproduced it with little change in several publications. The fourth
investigation contains several research projects (Bijker 1986; Bijker and
Leeuwestein 1984; Hulsbergen 1984; Leeuwestein et al. 1985) conducted at
the Delft University of Technology in the Netherlands. The last research
work in this list was conducted at the Technical University of Denmark
(Freds0e and Hansen 1987; Hansen et al. 1986; Jensen et al. 1988; Mao
1986—1988). The following section discusses findings from each of these
investigations mentioned above.
There are other relevant information on this subject, notably the field tests
conducted in the Hydraulic Research Station in Wallingford, England (Lit-
tlejohns 1977) and by Carstens (1983) and Dahlberg (1983).

Kjeldsen et al. (1973)

Kjeldsen et al. (1973) conducted flume experiments to investigate local
scour around submarine pipelines under unidirectional current. The boundary
conditions of the tests are summarized in Table 1.
They proposed an empirical equation relating the scour depth, ds, mea-
sured from the bottom of the pipeline, to the velocity head, Vl/2g, and the
pipe diameter, D:
/v2\02
ds = 0.9721 — I D 0 8 (1)
Eq. 1 applies to conditions where the pipeline is initially resting on the bed
level. The equation implies that the scour depth only depends on flow ve-
locity and pipe diameter, but excludes the effect of flow depth and grain
size. By considering the size of the sediment and the velocity used in the
experiments, a live-bed condition clearly exists in the tests conducted by
Kjeldsen et al. (1973). This condition means that there is always sediment
replenishment into the scour hole from upstream because of the occurrence
of general sediment transport on the bed. When studying local scour at bridge
piers under a live-bed condition, Chiew and Melville (1987) showed that the
equilibrium live-bed scour depth may be substantially less than a clear-water
scour depth. This is because in live-bed conditions, general sediment trans-
port occurs, and this transport causes the development of a dynamic equi-
librium between the scouring process and influx of sediment into the scour
453

J. Hydraul. Eng. 1991.117:452-466.

 Chao 6 Hennessy (1972)
+ Ibrahim £ Nalluri (1986)
Q Kjelsen et al. (1973)
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

0 Dutch

Measured scour depth, mm

FIG. 1. Comparison of Measured and Predicted Scour Depth

hole, resulting in a lesser scour depth. This effect is the most likely cause
of the consistently smaller computed scour depths as compared with the ex-
perimental results obtained in this study (see Fig. 1).

Ibrahim and Nalluri (1986)

Ibrahim and Nalluri (1986) proposed two equations based on an extensive
experimental program on local scour around submarine pipeline conducted
at the University of Newcastle upon Tyne (Ibrahim 1986). Table 1 contains
the boundary conditions for the flume tests conducted under the influence
of unidirectional flow only. The two empirical equations are:

- = 4.706^ + 0.06 (2a)

D \VC, gy„
and

* = 0.084^ + 1.33 (2b)

gy<
where ds = scour depth; D = pipe diameter; V0 = undisturbed mean flow
velocity; Vc = critical velocity for sediment entrainment; g = gravitational
acceleration; and ya = flow depth.
Eqs. 2a and 2b apply to clear-water and live-bed conditions, respectively.
Similar to Eq. 1, proposed by Kjeldsen et al. (1973), Ibrahim and Nalluri
(1986) also stated that the scour depth is related to velocity and pipe di-
ameter. In addition, they also included the influence of flow depth in their
analysis. However, instead of clarifying how flow depth affects the equilib-
rium scour depth, their two empirical equations compounded the problem.
This is because Eq. 2a implies that the scour depth is directly and inversely
proportional to flow velocity and flow depth, respectively, whereas Eq. 2b

454

J. Hydraul. Eng. 1991.117:452-466.

 implies the exact opposite. The paradox arose because the equations were
derived purely from curve fitting technique and no due considerations ap-
peared to have been given to the physics describing the scouring process.
Furthermore, the constant of 1.33 in Eq. 2b is extremely large and com-
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

pletely overshadows both the influence of velocity and flow depth. Whether
this value means that the bed level is degrading or a contribution from the
height of the propagating bed features is not clear. The equation suggests a
scour depth of 1.33 times the pipe diameter when the flow velocity is zero.
This inference is rather puzzling. Based on Eqs. 2a and 2b, Ibrahim and
Nalluri (1986) stated that grain size has no influence on the scour depth apart
from the indirect influence on the value of Vc.
Comparison between results computed using the clear-water scour depth
equation (Eq. 2a) with the experimental data obtained in this study (Fig. 1)
shows that most of the computed data is less than the experimental data.
This is probably because Ibrahim and Nalluri (1986) assume the formation
of an equilibrium scour depth in three to four hours after commencement of
experiment. This test duration is far too short in scour experiments based on
the writer's experience in clear-water scour at both bridge piers and pipe-
lines. Normally, it takes up to two to three days to reach equilibrium. Based
on results obtained from this study, only 5 0 - 7 0 % of the equilibrium clear-
water scour depth is reached in three to four hours of testing. The duration
of experiment is the most probable cause of the reduced scour depth obtained
from Ibrahim and Nalluri's data set (1986).

Chao and Hennessy (1972)

The method proposed by Chao and Hennessy (1972) to evaluate maximum
scour depth around submarine pipelines is applicable to pure current con-
dition only. It includes the following steps:
1. For a given incoming flow condition, calculate the discharge through and
jet velocity in the scour hole.
2. Scouring takes place until the jet velocity equals the critical shear stress,
T„ of the bed sediment in the hole. The jet velocity reduces as the scour hole
enlarges. It must be pointed out that this approach of computing the scour depth
is only applicable to clear-water conditions, where there is no sediment influx
to the scour hole. Where there is general transport of sediment upstream from
the pipeline, a dynamic equilibrium exists when the rate of sediment transport
into the scour hole equals that leaving.
To examine this procedure in evaluating the maximum scour depth, one
needs to look into these two steps. The analytical model is based on the two-
dimensional potential flow theory and several assumptions made in their original
paper. The discharge per unit width through the hole, qbot, is

A* = VJH - * ) for H > R (3)

and the average jet velocity, wavg, is

JHV in
gb
\R \R
yb = °' = V forH>R (4)
VM Vc
(H-R) (H\ 2
(H
2
[R)-\R] + 1

455

J. Hydraul. Eng. 1991.117:452-466.

 H
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

I H - R ) = moximum scour
i depth

FIG. 2. Definition Sketch of Variables Used by Chao and Hennessy (1972)

0.12
(PQ)

Z 0-10
ex

c_
00

t 0-06
CJ

0-04
"5
u

^ 0-02

0.01 0.05 0.1 . , . 0-5 1 5

u5o- (mm)

FIG. 3. Critical Shear Stress versus Grain Size According to Herbich (1981); Data
Computed Using Shields Entrainment Function (Raudklvi and Sutherland 1981)
Are Superimposed

where R = radius of pipe; and H = scour depth from the center of pipe.
Fig. 2 contains the definition sketch of the variables used in Eqs. 3 and
4. The main drawback of the method is the use of a potential flow theory
in deriving the solution. In real flow, the fluid is not inviscid, and separation
occurs at the rear of the pipe. This phenomenon affects the flow conditions.
This method of computation is likely to overestimate the discharge through
and jet velocity in the scour hole.
This method of computation also requires estimation of the critical shear
stress of the bed sediment in the scour hole. In their paper, Chao and Hen-
nessy (1972) did not explicitly state how the critical shear stress was eval-
uated. Herbich (1981), when reproducing this method of computation, pro-
posed a plot relating the critical shear stress versus grain size. Fig. 3 shows
the curve (the critical shear stress computed using the Shields entrainment
is superimposed on the curve for comparison) and overestimates the critical
shear stress by up to five times for grain size ranging from 0.1 mm to 1
mm when comparing with the customarily used Shields entrainment func-
tion. No reason was given for the use of such a curve. The overestimation
will reduce the predicted scour depth. Fig. 1, which contains the comparison
between experimental data obtained in this study and data computed using
456

J. Hydraul. Eng. 1991.117:452-466.

 this method, clearly shows that the computed value is consistently lower than
the experimental results obtained from this study.

RESEARCH CONDUCTED AT DELFT UNIVERSITY OF TECHNOLOGY

Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

Besides merely presenting a method to compute scour depth at submarine

pipelines, researchers at the Delft University of Technology first attempted
to study the mechanics of flow around submarine pipelines. They investi-
gated changes to the flow pattern around the pipe and the subsequent re-
sponse of the sediment. Bijker and Leeuwestein (1984) stated that the scour
depth depends on the undisturbed flow velocity, pipe diameter, flow depth,
height of pipe above bed level, and grain size. They stated that the principal
cause of erosion is a local increase in transport capacity of the water passing
a pipeline, while deposition occurs where this capacity decreases. This state-
ment is merely a corollary of their observation rather than an explanation.
Chiew (1990) identified that piping plays a dominant role in initiating scour
at submarine pipelines. When the pressure gradient, induced by the high
stagnation pressure upstream and the comparatively low wake pressure
downstream from the pipe, exceeds the flotation gradient of the bed sedi-
ment, piping occurs. This combines with the upstream stagnation eddy to
cause the onset of scour.
Using results from a series of model tests, including those conducted by
Kjeldsen et al. (1973), the Dutch research group proposed a slightly different
empirical equation for computing scour depth at submarine pipelines [the
boundary conditions for these model tests can be found in Bijker and
Leeuwestein (1984)]:
0.26

d, = 0.929( — J x D 078 x rfjo004 (5)

Comparing Eqs. 5 with that proposed by Kjeldsen et al. (1973), the Dutch
equation includes a moderate effect of grain size on the scour depth. They
also proposed a method for predicting prototype scour depth through a scale
series. In other words, results obtained from model tests can be extrapolated
and used to predict scour depth associated with a much larger pipe diameter
and velocity. Verification of their method was done in the large-scale fa-
cilities at the Delft Hydraulics Laboratory and using a computational model.
Their results showed that both Eqs. 1 and 5 underestimated the maximum
scour depth.
One important conclusion drawn from the Dutch study is the inference
that the scour depth under unidirectional current is always higher than that
under pure wave action or the combined effect of wave and current with the
same bottom shear stress.
Apart from investigating local scour at submarine pipelines per se, the
Dutch group also studied self-burial of submarine pipelines (Hulsbergen 1984;
Leeuwestein et al. 1985). They reported that pipelines can bury themselves
down to three diameters under certain circumstances, and they proposed means
to stimulate this process. Fins were attached to the pipelines with the inten-
tion of increasing the rate of scouring, thereby encouraging self-burial.

RESEARCH CONDUCTED AT TECHNICAL UNIVERSITY OF DENMARK

Several papers (Freds0e and Hansen 1987; Hansen et al. 1986; Jensen et
al. 1988; Mao 1986-1988) have been published based on an extensive re-
457

J. Hydraul. Eng. 1991.117:452-466.

 search program conducted at the Technical University of Denmark. One of
the main objectives of the research program was to develop a mathematical
model for the scour process below submarine pipelines. The Danish group
studied both two- and three-dimensional scour, but the present review con-
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

centrates on the two-dimensional case.

Hansen et al. (1986) obtained a relationship between the bed velocity in
the scour hole and the two geometric quantities, G/D and ds/D, where G
is the vertical distance between the underside of the cylinder and the original
bed level. They stated that the width of the flume would also influence this
relationship. For a given G/D, Vbot/V0 (where Vbot is the mean flow velocity
in the scour hole) decreases with increasing scour depth; for a fixed scour
depth, Vboi/V0 increases with decreasing gap ratio. In principle, their math-
ematical model is based on this empirical relationship and an equilibrium
scour depth is reached when the computed bed shear stress is equal to the
critical shear stress for sediment transport. The model uses the sediment
transport formula of Meyer-Peter and Muller (1948). They reported good
comparison between results from their mathematical model and experiments.
It must be pointed out that the grain size used in the experiment for veri-
fication has a mean value of 0.36 mm. The applied shear stresses used sug-
gested occurrence of a live-bed condition for all the tests.
Apart from merely studying the scouring process, the Danish group de-
voted much of their time into investigating the flow and forces around the
pipeline as the scour develops from an initially flat bed to the equilibrium
stage. Excellent data and analysis can be found in Jensen et al. (1988) and
Freds0e and Hansen (1987). Mao (1987) also investigated flow-induced vi-
bration during the sagging process of the pipeline.

EXPERIMENTAL SETUP

Based on results from both studies of scour at bridge piers and submarine
pipelines to date (Chiew and Melville 1987; Bijker 1986), it is clear that
maximum scour occurs when the structure is subjected to unidirectional cur-
rent alone and that the undisturbed bed shear stresses, T, equals the critical
shear stress, TC, for sediment entrainment. Thus, when one plans to inves-
tigate the maximum scour depth at a structure, it is necessary to carry out
experiments under unidirectional current with the flow condition where T =
TC. Besides this, the duration of the experiment is important for determination
of the maximum scour depth. In clear-water conditions, the equilibrium scour
depth is approached asymptotically, and it takes a long time, even in flume
tests, to reach equilibrium. Published results often contradict each other when
one compares results obtained with the same flow and sediment boundary
conditions but a different time frame.
The experiments in this study were conducted in a flume 8 m long, 0.3
m wide and 0.45 m deep and had glass sides. Three differently sized pipes
32, 50, and 63 mm in diameter were used. For each test, the cylinder that
extended the entire width of the flume was used to represent the pipe. The
cylinder, fixed at both ends, was just lying on the sand bed. The scour depth,
measured from the underside of the cylinder, was measured at regular in-
tervals to monitor the scour development. When equilibrium was reached,
the flow was stopped and the equilibrium scour profile was measured. Equi-
librium is assumed to have achieved when less than 1 mm in changes to the

458

J. Hydraul. Eng. 1991.117:452-466.

 TABLE 2. Properties of Sediments Used in Experiments
Mean particle Specific Critical shear
size, d50 (mm) gravity velocity (m/s)
(1) (2) (3)
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

0.33 2.65 0.0139

0.61 2.65 0.0180
0.80 2.65 • 0.0210
1.70 2.65 0.0320

scour depth occurs within an eight-hour duration.

Four different types of uniform, cohesionless sediments were selected for
the experiment. Table 2 contains the properties of the sediments used. Five
different approach flow depths, at 50, 70, 100, 150, and 180 mm, were
used in the experiments. To simulate a submarine pipeline, the flow depth
is always greater than the pipe diameter. For all flows, the shear velocity,
u*, was kept constant at 0.95 times the critical shear velocity of the bed
sediments. The latter was computed using the Shields entrainment function
given in Raudkivi and Sutherland (1981). The bed slopes for all approach
flow depths were adjusted to provide uniform flow conditions.
The velocity profiles of the undisturbed approach flow for all the tests
were measured using a minipropeller at the centerline of the flume. These
profiles enabled the determination of both the shear velocity and the ap-
proach-unit flow rate, q„. The latter was obtained from integration through
depth of the velocity profile. The undisturbed mean flow velocity was then
determined by dividing q„ by the depth of flow. Besides this, the velocity
profiles have the pipeline for several tests were also measured after the scour
hole has reached the equilibrium condition. These profiles allow evaluation
of the unit flow rate over the pipeline, qtop. Because of the continuity of
flow, the difference between q„ and qtop is the amount of flow through the
scour hole, q^. These values were used to assess the quantity of flow through
the scour hole for given flow and geometric conditions.

PREDICTIONS OF MAXIMUM SCOUR DEPTH

As discussed earlier, for a given flow depth, pipe diameter, and grain size,
the maximum equilibrium scour depth occurs when the following two con-
ditions are satisfied:

1. A clear-water condition, where there is no general sediment transport up-

stream of the scour area, exists, and the undisturbed bed shear stress equals the
critical shear stress for sediment entrainment.
2. Scouring occurs under unidirectional current for the given shear stress.

Fig. 4 contains a typical equilibrium scour profile obtained under this flow
condition. Under clear-water conditions, equilibrium occurs when T through
the gap beneath the pipeline is equal to the critical shear stress in the scour
hole. To evaluate this condition, one needs to answer two important ques-
tions:
459

J. Hydraul. Eng. 1991.117:452-466.

 Scour depth (mm)
20

Original Bed Level \^ J

0
^^^ Direction of Flow
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

r""
-20

-40

-60

-80

\Pipe
-100 Axis
-120
-800 -600 -400 -200 0 200

Distance from vertical pipe axis (mm)

FIG. 4. Typical Equilibrium Scour Profile

1. How much of the incoming flow is deflected up over the pipe and how
much goes through the scour hole beneath the pipe? The amount of flow through
the scour hole, or gap flow, will control the local bed shear stresses, and hence
the size of the scour hole.
2. What is the critical shear stress on the upstream slope of the scour hole?
Gap Flow
When a scour hole is present between a circular cylinder and the bed (see
Fig. 4), the oncoming flow divides itself into two components as it ap-
proaches the cylinder. Based on an extensive experimental investigation of
gap flow in shallow open-channel flows, Chiew (1991) found that the amount
of gap flow depends on the undisturbed flow depth, y0, pipe diameter, and
size of the gap, G. In general, for a given gap size, the larger the flow depth
ratio, y0/D, the smaller will the amount of gap flow be. This phenomenon
is not surprising because, with a large flow depth, most of the incoming
flow will be deflected over rather than under the pipe. Thus, less flow is
drawn down through the gap (or the scour hole under the pipe). Fig. 5 con-
tains the empirical function obtained by Chiew (1991). The important dif-
ference between the curves in Fig. 5 and the case of a scour hole around
submarine pipelines is the difference in the geometric boundary conditions.
In the former, the boundary is plane and the pipe is located well above the
plane bed, whereas in the latter, the pipeline is lying on an initially undis-
turbed bed level and the gap (or scour hole) lies beneath the pipe. Thus a
separate function needs to be evaluated in order to predict the amount of
gap flow for given flow and geometric parameters. Fig. 6 contains the em-
pirical function relating the flow depth parameter, y0/D, and the ratio of gap
over total flow rate, q' (qbM/<lo)' Unlike Fig. 5, the effect of the gap size
(or scour depth measured directly below the cylinder) does not appear to
affect the function. This is probably because the scour depth ratio, ds/D,
fluctuates within a small range of 1-2 and the scatter of experimental results
does not permit such fine precision. As in Fig. 5, the curve in Fig. 6 suggests
a reduction in gap flow when flow depth increases. By using the curve in
Fig. 6, it is possible to determine the gap flow rate and hence, to evaluate

460

J. Hydraul. Eng. 1991.117:452-466.

Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

FIG. 5. yJD versus q' with G/D as Third Parameter (Chiew 1991)

the bed shear stress, T ^ , in the scour hole. The method for determining T ^
is discussed in a later section.

Critical Shear Stress in Scour Hole

A typical equilibrium scoure hole does not have a symmetrical shape (see
Fig. 4), but rather has a steep upstream and much gentler downstream slope.
Results from this study showed that the values of the upstream slope ranged
from 20° to 28°. The value of this angle is important because it may dictate
the actual critical shear stress in the scour hole. Most engineers today accept
that the Shields entrainment function can adequately describe the initiation
of sediment transport. However, the function only applies to a horizontal or
near horizontal longitudinal bed slope. With a steep slope, gravity is likely
to affect the mobility of the sand grains significantly. For example, when
the bed slope approaches the angle of repose of the sediment, the bed shear
stress necessary to initiate sediment movement should approach zero.
Analysis of the results obtained in this study, however, did not advocate
the argument presented in the previous paragraph. The critical shear stresses
of the bed sediment in the scour hole were found to be very close to that
predicted using the Shields entrainment function despite the steep slope. This
is probably caused by the reversed eddies that exist behind the pipeline. Flow
separation occurs at the pipeline and reversed eddies are formed in the wake
behind the pipe. This reversed flow appears to have affected the critical shear
stresses of the bed sediment lying on the steep slope.
461

J. Hydraul. Eng. 1991.117:452-466.

 10

e
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

6 \
\
AY«.
4
\

k
2

0-2 0A q. 0.6 1-0

FIG. 6. y„/D versus <?'

Method for Predicting Maximum Scour Depth

The proposed method for estimating the maximum scour depth at sub-
marine pipelines requires an iterative procedure for given flow and geometric
parameters. The first step is to assume a value of the equilibrium scour depth,
(ds)esl, with which the bed shear stress in the hole can be evaluated. For a
given value of y0/D, the value of the gap flow rate can be determined by
using Fig. 6. The mean velocity beneath the pipe, Vbot, can be calculated
from
ffbot
Vb* = (6)
(ds).
and the bed shear stress in the scour hole, Tb0„ can be evaluated from
P/VL
Tbot — ' (7)
8
where p = fluid density; and / = friction factor. The friction factor can be
estimated using the Moody diagram where the relative roughness = d50/ds,
and the Reynolds number = Vhol X djv (v = kinematic viscosity). Once
Tbot is calculated it can be compared with the critical shear stress. In this
analysis, the critical shear stress was computed using the Shields entrainment
function. The iteration continues until
Tbot — ( T c ) (8)
462

J. Hydraul. Eng. 1991.117:452-466.

Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

Measured scour depth, mm

FIG. 7. Comparison of Measure and Computed Scour Depth

An example is shown to demonstrate how this method is used to evaluate

the maximum scour depth. Find the maximum scour depth at a 100-mm-
diameter pipeline subjected to the following conditions: d50 = 0.6 mm, y„
= 400 mm, V„ = 0.35 m/s and v = 0.85 X 10"6 m 2 /s.
y0 400 obot
- = = 4 -> —• = 0.29 (Fig. 6) (9a)
D 100 q0
q„ = 0.4 x 0.35 = 0.14 m2/s .-. qbM = 0.0406 m2/s (9b)
0.0406
Try ds = 120 mm: Vbot = = 0.338 m/s. (9c)
0.12
0.6
Relative roughness = = 0.005 (M)
120
0.12
Reynolds number = 0.338 x x 10~6 = 4.8 x 104 (9e)
0.85
From Moody diagram, / = 0.032 (9/)
(f
'• Tbot = Pl~l X 0.338' = 0.457 Pa (9g)

0.324 Pa for dso = 0.6 mm (from Shield's Function) (9ft)

Thus, the estimated value of 120 mm is too low, and a revised (ds)<.a is
necessary. Try (dX* = 141 mm. By repeating the procedure as outlined,
ybot = 0.288 m/s a n d / = 0.0312; .-. Tbot = 0.323 Pa - TC. The computed
scour depth means that for the given flow and geometric conditions, the
maximum scour depth is 141 mm.
463

J. Hydraul. Eng. 1991.117:452-466.

 Fig. 7 contains the comparison between the computed maximum scour
depth and the experimental data. The curve shows that most of the computed
data points lie within a scatter of ± 2 0 % .
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

CONCLUSIONS

The study proposes an iterative method for estimating maximum scour

depth at submarine pipelines. A maximum scour depth occurs when the ap-
plied shear stress is equal to the critical shear stresses for sediment entrain-
ment. The pipeline should be subjected to unidirectional current only. An
empirical function relating the flow depth ratio, ya/D, with the gap-flow rate
ratio can be used to determine the amount of gap flow through the scour
hole at equilibrium conditions. The predicted maximum scour depth com-
pares well with the experimental data obtained from the study.

ACKNOWLEDGMENTS

The writer wishes to express his sincere appreciation to Y. C. Lee for

collecting most of the experimental data. The project was financially sup-
ported by the Applied Research Grant of the Nanyang Technological Insti-
tute.

APPENDIX I. REFERENCES

Bijker, E. W. (1986). "Scour around structures." Proc. 20th Coast. Engrg. Conf.,
Taipei, Taiwan, 1754-1768.
Bijker, E. W., and Leeuwestein, W. (1984). "Interaction between pipelines and the
seabed under the influence of waves and currents." Seabed Mechanics, Proc. Symp.
IUTAM/IUGG. International Union of Theoretical Applied Mechanics/Interna-
tional Union of Geology arid Geophysics, 235-242.
Carstens, T. (1983). "Scour around an offshore structure." Oc. Sci. and Engrg.,
8(2), 157-172.
Chao, J. L., and Hennessy, P. V. (1972). "Local scour under ocean outfall pipe-
lines." J. Water Pollution Control Federation, 44(7), 1443-1447.
Chiew, Y. M., and Melville, B. W. (1987). "Local scour around bridge piers." J.
Hydr. Res., 25(1), 15-26.
Chiew, Y. M. (1990). "Mechanics of local scour around submarine pipelines." J.
Hydr. Engrg., ASCE, 116(4), 515-529.
Chiew, Y. M. (1991). "Flow around horizontal circular cylinder in shallow flows."
J. Wtrway., Port, Coast., and Oc. Engrg., ASCE, 117(2), 120-135.
Dahlberg, R. (1983). "Observations of scour around offshore structures." Can. Geo-
tech. J., 20(4), 617-628.
Freds0e, J., and Hansen, E. A. (1987). "Lift forces on pipelines in steady flow."
J. Wtrway., Port, Coast., and Oc. Engrg., ASCE, 113(2), 139-155.
Hansen, E. A., Freds0e, J., and Mao, Y. (1986). "Two-dimensional scour below
pipelines." Proc. 5th Int. Symp. on Offshore Mech. and Arctic Engrg., Tokyo,
Japan, 670-678.
Herbich, J. B. (1981). "Scour around pipelines and other objects." Offshore pipeline
design elements, Marcell Dekker, Inc., New York, N.Y., 43-96.
Herbich, J. B. (1985). "Hydromechanics of submarine pipelines: Design problems."
Can. J. Civ. Engrg., 12(4), 863-887.
Herbich, J. B., Schiller, R. E., Jr., Watanabe, R. K., and Dunlap, W. A. (1984).
"Scour around pipelines." Sea floor scour—Design guidelines for ocean founded
structures (Ocean engineering 4), Marcell Dekker, Inc., New York, N.Y., 203-
210.

464

J. Hydraul. Eng. 1991.117:452-466.

 Hulsbergen, C. H. (1984). "Stimulated self-burial of submarine pipelines." Proc.
16th Offshore Tech. Conf., Houston, Tex., 171-175.
Ibrahim, A. (1986). "Scour around pipeline under marine environments," thesis sub-
mitted to the University of Newcastle upon Tyne, at Newcastle upon Tyne, En-
gland, in partial fulfillment of the requirements for the degree of Doctor of Phi-
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

losophy.
Ibrahim, A., and Nalluri, C. (1986). "Scour prediction around marine pipelines."
Proc. 5th Int. Symp. on Offshore Mech. and Arctic Engrg., Tokyo, Japan, 679-
684.
Jensen, B. L., Sumer, B. M., Jensen, H. R., and Freds0e, J, (1988). "Flow around
and forces on a pipeline near a scoured bed." Proc. 7th Int. Conf. on Offshore
Mech. and Arctic Engrg., Houston, Tex., 39-48.
Kjeldsen, S. P., Gj0rsvik, O., Bringaker, K. G., and Jacobsen, J. (1973). "Local
scour near offshore pipelines." Proc. 2nd Int. Conf. on Port and Oc. Engrg. under
Arctic Conditions, University of Iceland, 308-331.
Leeuwestein, W., Bijker, E. W., Peerbolte, E. B., and Wind, H. G. (1985). "The
natural self-burial of submarine pipelines." Proc. 5th Int. Conf. on Behaviour of
Offshore Struct., 717-728.
Littlejohns, P. S. G. (1977). "A study of scour around submarine pipelines." Report
No. INT 113, Hydr. Res. Station, Wallingford, England.
Mao, Y. (1986). "The interaction between a pipeline and an erodible bed." Series
Paper No. 39, Inst, of Hydrodynamics and Hydr. Engrg., Tech. Univ. of Den-
mark, Lyngby, Denmark.
Mao, Y. (1987). "The flow induced pipe vibration during its sagging process." J.
Hydr. Res., 25(5), 565-582.
Mao, Y. (1988). "Seabed scour under pipelines." Proc. 7th Int. Conf. on Offshore
and Arctic Engrg., Houston, Tex., 33-38.
Meyer-Peter, E., and Muller, R. (1948). "Formulas for bed-load transport." Report
on 2nd Meeting, International Association for Hydraulic Research, 3, 39-64.
Raudkivi, A. J., and Sutherland, A. J. (1981). "Scour at bridge crossing." Road
Research Unit Bulletin No. 54, Nat. Roads Board, Wellington, New Zealand.

APPENDIX II. NOTATION

The following symbols are used in this paper:

D = pipe diameter;
d50 = mean particle size;
ds = scour depth;
fr)«t = estimated scour depth;
/ = friction factor;
G = gap size;
9 = gravitational acceleration;
H = scour depth measured from center of pipe;
=
q' ratio of unit flow rate through scour hole (or gap) over undis-
turbed unit flow rate;
<7bot = unit flow rate through scour hole (or gap);
1o = undisturbed approach unit flow rate;
9top = unit flow rate over top of cylinder;
R = pipe radius;
u* = shear velocity;
not = mean velocity in scour hole beneath pipe;
vc = critical mean velocity;
v0 = undisturbed mean velocity;
465

J. Hydraul. Eng. 1991.117:452-466.

 yn = undisturbed approach flow depth;
v = kinematic viscosity;
p = fluid density;
T =
bot critical shear stress in scour hole;
Downloaded from ascelibrary.org by IMPERIAL COLLEGE LONDON on 05/07/13. Copyright ASCE. For personal use only; all rights reserved.

tc = critical shear stresses in scour hole; and

T„ - undisturbed bed shear stress.

466

J. Hydraul. Eng. 1991.117:452-466.