Cross-shore distribution of longshore sediment transport:

comparison between predictive formulas and field measurements

Atilla Bayram
a
, Magnus Larson
a,
*
, Herman C. Miller
b
, Nicholas C. Kraus
c
a
Department of Water Resources Engineering, Lund University, Box 118, S-22100, Lund, Sweden
b
Field Research Facility, U.S. Army Engineer Research and Development Center, 1261 Duck Road, Duck, NC, 27949-4471, USA
c
Coastal and Hydraulics Laboratory, U.S. Army Engineer Research and Development Center, 3909 Halls Ferry Road,
Vicksburg, MS, 39180-6199, USA
Received 29 September 2000; received in revised form 12 July 2001; accepted 18 July 2001
Abstract
The skill of six well-known formulas developed for calculating the longshore sediment transport rate was evaluated in the
present study. Formulas proposed by Bijker [Bijker, E.W., 1967. Some considerations about scales for coastal models with
movable bed. Delft Hydraulics Laboratory, Publication 50, Delft, The Netherlands; Journal of the Waterways, Harbors and
Coastal Engineering Division, 97 (4) (1971) 687.], EngelundHansen [Engelund, F., Hansen, E., 1967. A Monograph On
Sediment Transport in Alluvial Streams. Teknisk Forlag, Copenhagen, Denmark], AckersWhite [Journal of Hydraulics
Division, 99 (1) (1973) 2041], BailardInman [Journal of Geophysical Research, 86 (C3) (1981) 2035], Van Rijn [Journal of
Hydraulic Division, 110 (10) (1984) 1431; 110(11) (1984) 1613; 110(12) (1984) 1733], and Watanabe [Watanabe, A., 1992. Total
rate and distribution of longshore sand transport. Proceedings of the 23rd Coastal Engineering Conference, ASCE, 25282541]
were investigated because they are commonly employed in engineering studies to calculate the time-averaged net sediment
transport rate in the surf zone. The predictive capability of these six formulas was examined by comparison to detailed, high-
quality data on hydrodynamics and sediment transport from Duck, NC, collected during the DUCK85, SUPERDUCK, and
SANDYDUCK field data collection projects. Measured hydrodynamics were employed as much as possible to reduce
uncertainties in the calculations, and all formulas were applied with standard coefficient values without calibration to the data
sets. Overall, the Van Rijn formula was found to yield the most reliable predictions over the range of swell and storm conditions
covered by the field data set. The EngelundHansen formula worked reasonably well, although with large scatter for the storm
cases, whereas the BailardInman formula systematically overestimated the swell cases and underestimated the storm cases. The
formulas by Watanabe and Ackers White produced satisfactory results for most cases, although the former overestimated the
transport rates for swell cases and the latter yielded considerable scatter for storm cases. Finally, the Bijker formula systematically
overestimated the transport rates for all cases. It should be pointed out that the coefficient values in most of the employed
formulas were based primarily on data from the laboratory or from the river environment. Thus, re-calibration of the coefficient
values by reference to field data from the surf zone is expected to improve their predictive capability, although the limited amount
of high-quality field data available at present makes it difficult to obtain values that would be applicable to a wide range of wave
and beach conditions. D 2001 Elsevier Science B.V. All rights reserved.
Keywords: Longshore sediment transport; Predictive formulas; Field measurements
0378-3839/01/$ - see front matter D 2001 Elsevier Science B.V. All rights reserved.
PII: S0378- 3839( 01) 00023- 0
*
Corresponding author. Fax: +46-46-222-44-35.
E-mail address: [email protected] (M. Larson).
www.elsevier.com/locate/coastaleng
Coastal Engineering 44 (2001) 7999
Report Documentation Page
Form Approved
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it
does not display a currently valid OMB control number.
1. REPORT DATE
JUL 2001
2. REPORT TYPE
3. DATES COVERED
00-00-2001 to 00-00-2001
4. TITLE AND SUBTITLE
Cross-shore distribution of longshore sediment transport: comparison
between predictive formulas and field measurements
5a. CONTRACT NUMBER
5b. GRANT NUMBER
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)
U.S. Army Engineer Research and Development Center,Coastal and
Hydraulics Laboratory,3909 Halls Ferry Road,Vicksburg,MS,39180-6199
8. PERFORMING ORGANIZATION
REPORT NUMBER
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITORS ACRONYM(S)
11. SPONSOR/MONITORS REPORT
NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES
14. ABSTRACT
see report
15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF
ABSTRACT
Same as
Report (SAR)
18. NUMBER
OF PAGES
21
19a. NAME OF
RESPONSIBLE PERSON
a. REPORT
unclassified
b. ABSTRACT
unclassified
c. THIS PAGE
unclassified
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
1. Introduction
During the past three decades, numerous formulas
and models for computing the sediment transport by
waves and currents have been proposed, ranging from
quasi-steady formulas based on the traction approach
of Bijker, and the energetics approach of Bagnold, to
complex numerical models involving higher-order
turbulence closure schemes that attempt to resolve
the flow field at small scale. There are relatively few
high-quality field data sets on the cross-shore distri-
bution of the longshore sediment transport rate avail-
able to evaluate existing predictive formulas. Kraus et
al. (1989) and Rosati et al. (1990) measured the
longshore transport rate across the surf zone using
streamer traps (i.e., DUCK85 and SUPERDUCK field
experiments). Miller (1999) measured the cross-shore
distribution with optical backscatter sensors (OBS)
combined with current measurements (i.e., SANDY-
DUCK field experiment). The measurements reported
by Miller (1999) covered a number of storms, thus
complementing the measurements by Kraus et al.
(1989) and Rosati et al. (1990) that were made in
milder swell waves.
The objective of the present study is to evaluate the
predictive capability of six well-known sediment
transport formulas, adapted to calculate the cross-
shore distribution of the longshore sediment transport
rate, based upon the above-mentioned three field data
sets. We selected formulas that have gained world-
wide acceptance in confidently predicting longshore
sediment transport rates. This, however, should not be
interpreted as a sign of disagreement or a lessening of
the importance of formulas not discussed here. Only
sand transport was investigated in this study, and
focus is on computing the time-averaged net long-
shore transport rate.
Background to the investigated sediment transport
formulas are given in Section 2 and their main
characteristics are summarized (the equations used
to calculate the transport rate are given in Appendix
A). Next, in Section 3, the longshore sediment trans-
port data sets are described. In Section 4 are shown
the results of the comparisons between the formulas
and the field data including a wide range of wave and
current conditions. An overall discussion of the results
from the comparisons is provided in Section 5, where
the strength and weaknesses of the investigated for-
mulas are assessed, as well as their limitations, using
various statistical measures. Finally, the conclusions
of the study are presented in Section 6.
2. Longshore sediment transport formulas
Longshore sand transport is typically greatest in
the surf zone, where wave breaking and wave-induced
currents prevail, although a pronounced peak can be
found in the swash zone as well (Kraus et al., 1982).
Typically the total (or gross) longshore sediment
transport rate is computed with the CERC formula
(SPM, 1984) in engineering applications. However, as
ability to predict the surf zone hydrodynamics has
improved, the need for reliable formulas that spatially
better resolves the sediment transport rate has
increased, both concerning the cross-shore distribu-
tion of the transport rate and the concentration dis-
tribution through the water column.
In this investigation, the skill of six published
formulas proposed for calculating the cross-shore
distribution of the longshore sediment transport rate
was investigated. Transport rates were calculated for
the utilized cases using standard coefficient values (as
given in the literature) without calibration. In the
present comparison the formulas proposed by Bijker
(1967, 1971), Engelund and Hansen (1967), Ackers
and White (1973), Bailard and Inman (1981), Van
Rijn (1984), and Watanabe (1992) (as they chrono-
logically appeared in the literature) were employed,
representing the most common approaches for calcu-
lating the time-averaged net sediment transport rate.
The Bijker and Van Rijn formulas also calculate the
suspended sediment concentration distribution
through the water column, which allowed for addi-
tional comparisons for these two formulas with some
of the field cases for which concentration measure-
ments were made. This will be discussed in a forth-
coming paper (Larson et al., in preparation).
The six formulas are summarized in Table 1, where
the formulas, coefficient values, and wave and beach
conditions of the data originally used for verification
of the formulas are listed. Further details regarding the
equations are given in Appendix A. This is necessary
because some variants of the formulas have appeared
in the literature. In the following, a short background
to the formulas is presented together with their main
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 80
Table 1
Longshore sediment transport formulas (for notation see Appendix A)
Formula Longshore sediment Coefficients Verification data
D (mm) tanb Exp. condition
Bijker q
b;B
Ad
50
V
C

g
p
exp
0:27s 1d
50
qg
ls
b;wc
_ _
q
s;B
1:83q
b;B
I
1
ln
33h
r
_ _
I
2
_ _
A= 15
(non-breakingbreaking)
0.23 0.07 H
0
= 1.6 m;
T=12.0 s;
a = 13
EngelundHansen q
t;EH
V
0:05Cs
2
b;wc
s 1
2
d
50
q
2
g
5=2
0.190.93
Watanabe q
t;W
A
s
b;wc
s
b;cr
V
qg
_ _
A= 0.52 0.22.0 0.20.01 H
0
= 0.022.4 m;
(regular irregular) T=1.018.0 s;
a = 1545
Ackers White
(not modified)
q
t;AW
V
1
1 p
d
35
V
V

_ _
n
C
d;gr
A
m
F
C
A
m
A, m, n, C
d,gr
, F
C
(see Appendix A)
0.20.61 h = 0.187.17 m
Van Rijn
q
b;VR
0:25cq
s
d
50
D
0:3

s
0
b;wc
q

s
0
b;wc
s
b;cr
s
b;cr
_ _
1:5
q
s;VR
c
a
Vh
1
h
_
h
a
v
V
c
c
a
dz c
a
VhF
0.10.2 H
0
= 0.070.2 m;
T=1.02.0 s;
a = 90
BailardInman q
t;BI
0:5qf
w
u
3
0
e
b
q
s
qgtanc
d
v
2
d
3
v
_ _
0:5qf
w
u
4
0
e
s
q
s
qgw
s
d
v
u

e
b
=0.1; e
s
= 0.02 0.1750.6 0.0340.138 H
0
= 0.051.44 m;
T=1.011.0 s;
a = 2.818.9
transport formula
A
.
B
a
y
r
a
m
e
t
a
l
.
/
C
o
a
s
t
a
l
E
n
g
i
n
e
e
r
i
n
g
4
4
(
2
0
0
1
)
7
9

9
9
8
1
characteristics as well as references to the original
publications. Also, Van De Graaff and Van Oveerem
(1979) can be consulted for a comprehensive sum-
mary of some formulas. They compared three formu-
las for the net longshore sediment transport, namely
the formulas by Bijker, EngelundHansen, and
AckersWhite, although they focused on the gross
rate and made comparisons for a number of selected
hypothetical cases.
2.1. The Bijker formula
Bijkers (1967, 1971) sediment transport formula is
one of the earliest formulas developed for waves and
current in combination. It is based on a transport
formula for rivers proposed by KalinskeFrijlink
(Frijlink, 1952). Bijker distinguishes between bed
load and suspended load, where the bed load transport
depends on the total bottom shear stress by waves and
currents. The suspended load is obtained by integrat-
ing the product of the concentration and velocity
profiles along the vertical, where the reference con-
centration for the suspended sediment is expressed as
a function of the bed load transport. In its original
form, the bed-load formula does not take into account
a critical shear stress for incipient motion, implying
that any bed shear stress and current will lead to a net
sediment transport. The Bijker transport formula
(hereafter, called the B formula) is, in principle,
applicable for both breaking and non-breaking waves.
However, different empirical coefficient values are
needed in the formula.
2.2. The Engelund and Hansen formula
Engelund and Hansen (1967) originally derived a
formula to calculate the bedload transport over dunes
in a unidirectional current by considering an energy
balance for the transport. Later, this formula (here-
after, called EH formula) was applied to calculate the
total sediment transport under waves and currents, and
modifications were introduced to account for wave
stirring (Van De Graaff and Van Overeem, 1979).
However, their theory has limitations when applied to
graded sediments containing large amount of fine
fractions, causing predicted transport rates to be
smaller than the actual transport rates. Similar to the
Bijker formula, no threshold conditions for the initia-
tion of motion was included in the original formula-
tion. The same coefficient values are used for
monochromatic and random waves.
2.3. The Ackers and White formula
Ackers and White (1973) developed a total load
sediment transport formula for coarse and fine sedi-
ment exposed to a unidirectional current. Coarse
sediment is assumed to be transported as bed load
with a rate taken to be proportional to the shear stress,
whereas fine sediment is considered to travel in
suspension supported by the turbulence. The turbu-
lence intensity depends on the energy dissipation
generated by bottom friction, which makes the sus-
pended transport rate related to the bed shear stress.
The empirical coefficients in the AckersWhite for-
mula (hereafter, called AW formula) were calibrated
against a large data set covering laboratory and field
cases (HR Wallingford 1990; reported in Soulsby,
1997). Van De Graaff and Van Overeem (1979)
modified the AW formula to account for shear exerted
by waves.
2.4. The Bailard and Inman formula
Bailard and Inman (1981) derived a formula for
both the suspended and bed load transport based on
the energetics approach by Bagnold (1966). Bagnold
assumed that the work done in transporting the sedi-
ment is a fixed portion of the total energy dissipated
by the flow. The BailardInman formula (hereafter,
called BI formula) has frequently been used by
engineers because it is computationally efficient, takes
into account bed load and suspended load, and the
flow associated with waves (including wave asymme-
try) and currents can be incorporated in a straightfor-
ward manner. A reference level for the velocity
employed in the formula (normally taken to be 5.0
cm above the bed) must be specified.
2.5. The Van Rijn formula
Van Rijn (1984) proposed a comprehensive theory
for the sediment transport rate in rivers by considering
both fundamental physics and empirical observations
and results. The formulations were extended to estua-
ries as summarized by Van Rijn (1993) (hereafter,
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 82
called VR formula). Bed load and suspended load are
calculated separately, and the approach of Bagnold
(1966) is adopted for computing the bed load. Sus-
pended load is determined by integrating the product
of the vertical concentration and velocity profiles,
where the concentration profile is calculated in three
layers. Different exponential or power functions are
employed in these layers with empirical expressions
that depend on the mixing characteristics in each
layer.
2.6. The Watanabe formula
Watanabe (1992) proposed a formula for the total
load (bed load and suspended load) based on the
power model concept. The volume of sediments set
in motion per unit area is proportional to the com-
bined shear stress of waves and currents, if the
critical value for incipient motion is exceeded, and
this volume is transported with the mean flow
velocity. This formula has been widely used in Japan
for the prediction of, for example, beach evolution
around coastal structures and sand deposition in
harbors and navigation channels. The Watanabe for-
mula (hereafter, called W formula) and its coefficient
values have been calibrated and verified for a variety
of laboratory and field data sets during the last
decade (e.g. Watanabe, 1987; Watanabe et al.,
1991). However, it has not yet been established
whether the value of the non-dimensional coefficient
in the formula (A) is a constant or it depends on the
wave and sediment conditions. Different values are
employed for laboratory and field conditions,
whereas the same value is typically used for mono-
chromatic and random waves.
3. Longshore sediment transport data sets
3.1. DUCK85 surf zone sand transport experiment
The DUCK85 surf zone sand transport experi-
ment was performed at the U.S. Army Corps of
Engineers Field Research Facility at Duck, NC in
September, 1985. Kraus et al. (1989) measured the
cross-shore distribution of the longshore sediment
transport rate using streamer traps. Eight runs were
made where the amount of sediment transported at a
specific location in the surf zone during a certain
time was collected using streamer traps oriented so
that the traps opposed the direction of the longshore
current. The traps, each consisting of a vertical
array of polyester sieve cloth streamers suspended
on a rack, were deployed across the surf zone. The
polyester cloth allowed water to pass through but
retained grains with diameter greater than the 0.105
mm mesh, which encompasses sand in the fine
grain size region and greater. From knowledge of
the trap mouth area, the trap efficiency, and the
measurement duration, the local transport rate was
derived. The trapping efficiency has been exten-
sively investigated through laboratory experiments
(Rosati and Kraus, 1988) allowing for confident
estimates of the local longshore sediment transport
rate.
Wave height and period were measured using the
photopole method described by Ebersole and Hughes
(1987). This method involved filming the water
surface elevation at the poles placed at approxi-
mately 6-m intervals across the surf zone utilizing
as many as eight 16-mm synchronized cameras. The
bottom profile along the photopole line was surveyed
each day. Fig. 1 shows surveyed bottom topography
along the measurement transect on September 6,
1985. The root-mean-square (rms) wave height
(H
rms
) at the most offshore pole was in the range
of 0.40.5 m, and the peak spectral period (T
p
) was
in the range of 912 s (see Table 2). Long-crested
waves of cnoidal form arriving from the southern
quadrant prevailed during the experiment, producing
a longshore current moving to the north with a
magnitude of 0.10.3 m/s.
3.2. SUPERDUCK surf zone sand transport experi-
ment
Patterned after the DUCK85 experiment, the
SUPERDUCK surf zone sand transport experiment
was conducted during September and October 1986
(Rosati et al., 1990). However, at SUPERDUCK a
temporal sampling method to determine transport
rates was emphasized in which traps were inter-
changed from 3 to 15 times at the same locations.
Fewer runs are available where the cross-shore dis-
tribution was measured (two runs were employed
here; see Table 2).
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 83
Waves and currents were measured in the same
manner as for DUCK85. The wave conditions during
the experiment were characterized as long-crested
swell (T
p
in the range 613 s) with a majority of
the waves breaking as plunging breakers. The wave
height (H
rms
) at the most seaward pole varied between
0.3 and 1.6 m during the experiment. A steady off-
shore wind (67 m/s) was typically present during the
measurements. The mean longshore current speeds
measured were in the range of 0.10.7 m/s. During
the experiment, the seabed elevation at each of the
photopoles was surveyed once a day and Fig. 2
shows, as an example, the surveyed bottom profile
on September 19, 1986. In contrast to the DUCK85
experiment, barred profiles occurred several times
during the SUPERDUCK experiment, although the
two runs presented here involved shelf-type profiles.
3.3. SANDYDUCK surf zone sand transport experi-
ment
The SANDYDUCK experiment was conducted in
the same location as the DUCK85 and SUPERDUCK
experiments (Miller, 1998). SANDYDUCK included
several major storms, complementing the measure-
ments made during DUCK85 and SUPERDUCK.
Fig. 1. Bottom profile surveyed along the measurement transect on September 6, 1985 (DUCK85 experiment).
Table 2
Beach and wave characteristics for runs selected from the DUCK85, SUPERDUCK, and SANDYDUCK experiments for comparison with
sediment transport formulas
Date Profile type H
rms
(m) T
p
(s) D
ref
(m) V
mean
(m/s)
Sept. 5, 1985, 09.57 Shelf 0.50 11.4 2.14 0.11
Sept. 5, 1985, 10.57 Shelf 0.46 11.2 1.80 0.17
Sept. 5, 1985, 13.52 Shelf 0.54 10.9 2.19 0.17
Sept. 5, 1985, 15.28 Shelf 0.46 11.1 1.94 0.22
Sept. 6, 1985, 09.16 Shelf 0.48 12.8 1.40 0.30
Sept. 6, 1985, 10.18 Shelf 0.36 13.1 2.14 0.29
Sept. 6, 1985, 13.03 Shelf 0.42 10.1 2.43 0.22
Sept. 6, 1985, 14.00 Shelf 0.36 11.2 2.34 0.18
Sept. 16, 1986, 11.16 Shelf 0.60 10.1 2.20 0.20
Sept. 19, 1986, 10.16 Shelf 0.59 10.1 2.66 0.17
March 31, 1997 Bar 1.36 8.0 6.70 0.49
April 1, 1997 Bar 2.92 8.0 6.82 1.10
October 20, 1997 Bar 2.27 12.8 6.44 0.53
February 04, 1998 Bar 3.18 12.8 8.59 0.60
February 05, 1998 Bar 2.94 12.8 6.83 0.45
H
rms
= root-mean-square wave height at the most offshore measurement point; T
p
= peak spectral period; D
ref
=water depth at the most offshore
measurement point; V
mean
= mean longshore current velocity in the surf zone.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 84
Sediment transport measurements were made using
the Sensor Insertion System (SIS), which is a diverless
instrument deployment and retrieval system that can
operate in seas with individual wave heights up to 5.6
m (Miller, 1999). An advantage of the SIS is that it
allows direct measurement of wave, sediment concen-
tration, and velocity together with the bottom profile
during a storm. The SIS is a track-mounted crane with
the instrumentation placed on the boom. A standard
SIS consists of OBS to measure sediment concentra-
tion, an electromagnetic current meter for longshore
and cross-shore velocities, and pressure gauge for
waves and water levels. The wave conditions and
mean longshore current velocities for five storm cases
employed in this investigation are described in Table
2. During the SANDYDUCK experiment the concen-
tration was measured at several points through the
vertical as well as at a number of cross-shore loca-
tions. Simultaneously, the velocity was recorded and
the local transport rate was derived from the product
of the concentration and velocity.
Longshore bars were typically present during
SANDYDUCK, making it possible to assess the
effects of these formations on the transport rate dis-
tribution. As an example, the bottom profile measured
during the storm on October 20, 1997, is shown in
Fig. 3. The beach at Duck is composed of sand with a
median grain size (d
50
) of about 0.4 mm at the
shoreline dropping off at a high rate to 0.18 mm in
the offshore. In the region where d
50
= 0.18 mm (most
of the typical surf zone width), sediment sampling has
yielded d
35
= 0.15 mm (diameter corresponding 35%
Fig. 3. Bottom profile surveyed along the measurement transect on October 20, 1997 (SANDYDUCK experiment).
Fig. 2. Bottom profile surveyed along the measurement transect on September 19, 1986 (SUPERDUCK experiment).
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 85
being finer) and d
90
= 0.24 mm (diameter correspond-
ing 90% being finer) as typical values (Birkemeier et
al., 1985).
4. Evaluation of the formulas
Measured hydrodynamics were employed as much
as possible in the formulas to reduce the uncertainties
in the transport calculations. The rms wave height and
peak spectral wave period was used as the character-
istic input parameters to quantify the random wave
field. Values at intermediate locations where no meas-
urements were made were obtained by linear interpo-
lation (note that this is the cause for the discontinuities
in the derivative of the calculated transport rate dis-
tributions). It was assumed that the incident wave
angle was small, implying that the angle between the
waves and the longshore current was approximately
90. In the VR formula the undertow velocity is
needed if the resultant shear stress is calculated (i.e.,
the shear stress resulting from the cross-shore and
longshore currents combined). No undertow measure-
ments were available and the model of Dally and
Brown (1995) was employed to calculate this velocity.
The influence of the shear stress from the undertow
was typically small compared to that of the longshore
current and waves. In a few cases extrapolation of the
current from the most shoreward or seaward measure-
ment point was needed. On the shoreward side the
current was assumed to decrease linearly to become
zero at the shoreline, whereas at the seaward end the
current was taken to be proportional to the ratio of
breaking waves (i.e., assuming that most of the
current was wave-generated in this region).
The roughness height (r), which determines the
friction factors for waves and current, is a decisive
parameter that may markedly influence the sediment
transport rate, especially the bed load transport. Here,
the calculation of the roughness height was divided
into three different cases depending on the bottom
conditions, namely flat bed, rippled bed, and sheet
flow. The division between these cases was made
based on the Shields parameter (h), where h < 0.05
implied flat bed, 0.05 < h < 1.0 rippled bed, and h > 1.0
sheet flow (Van Rijn, 1993). An iterative approach
was needed because the bottom conditions are not
known a priori when the roughness calculation is
performed. A Shields curve was employed to deter-
mine the criterion for the initiation of motion based on
h, which was included in the formulas that have this
feature.
The roughness height was estimated in the follow-
ing manner:

Flat bed: r is set equal to 2.5d

50
, where d
50
is
the median grain size (Nielsen, 1992)

Rippled bed: r is calculated from the ripple

height and length and the Shields parameter
according to Nielsen (1992)

Sheet flow: r is calculated from the Shields

parameter and d
90
according to Van Rijn
(1984), where d
90
is the grain size that 10%
of the sediment exceeds by weight.
The wave friction factor ( f
w
) was computed based
on r using the formula proposed by Swart (1976),
which is based on an implicit relationship given by
Jonsson (1966), assuming rough turbulent flow,
lnf
w
5:98 5:2
r
a
b
_ _
0:194
for :
r
a
b
< 0:63
f
w
0:3 for :
r
a
b
0:63
1
where a
b
is the amplitude of the horizontal near-bed
water particle excursion. For the purpose of compar-
ing the predictive capabilities of the formulas, coef-
ficient values proposed by the developers and
coworkers were employed without any particular
tuning of the coefficients. Predicted transport rates
with the formulas were converted to mass flux per unit
width before comparison with the measurements.
4.1. Comparisons with DUCK85 data
Although simulations were carried out for all runs
listed in Table 2, only four of the runs were selected
for detailed discussion here. However, the overall
conclusions given are based on the results from all
simulations. Fig. 4 shows predicted cross-shore dis-
tributions of the longshore sediment transport rate
with the different formulas for the experimental run
at 0957 on September 5, 1985 (denoted 859050957
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 86
from year/month/day/time, in accordance with Kraus
et al., 1989). It is noted that the AW, EH, and VR
formulas yield good predictions within the same order
of magnitude as the measured data, at least in the
inner part of the surf zone. Contrarily, the B, BI, and
W formulas give significantly higher transport rates
than the measurements, especially B and W. Some
overestimation is expected for the B and BI formulas
because they do not take into account the threshold of
sediment motion in their original formulations,
although this simplification would not account for
the large deviations found for the B formula. Bijker
(1971) pointed out that his formula tended to over-
estimate the transport rate using the recommended
value on the main coefficient.
Figs. 5 and 6 show calculated and measured trans-
port rates for the experimental runs 859051528 and
859060916, respectively. All formulas except B and
W predicted longshore transport rates at the correct
order of magnitude, although the EH formula some-
what underpredicted the transport rate. Bearing in
mind that the sediment has a d
50
of approximately
0.18 mm at the site, the discrepancy between the EH
formula and the measured rates might be due to
limitations in the derivation of the formula. For fine-
graded sediments large suspension modifies the veloc-
ity distribution so that the assumptions underlying the
formula are not satisfied (Engelund and Hansen,
1967). Another reason for the discrepancy might be
the relatively strong influence of the predicted rough-
ness on the transport rate that the EH formula dis-
plays. Run 859051528 (see Fig. 5) indicated a
bimodal distribution with a large peak in the outer
surf zone (at around 145 m) and a small peak in the
inner surf zone (at around 125 m). In general, the
formulas fail to correctly predict this shape, especially
the shoreward peak. This peak is probably a function
of additional breaking and current generation close to
the shoreline. Because no current measurements were
available here, extrapolation was employed, implying
that some portion of the discrepancy is probably
caused by the uncertainty in the input data rather than
the formulas themselves.
During Run 859060916 the most seaward trap was
located outside the surf zone and a relatively small
amount of sand was collected in it (Fig. 6). Thus, there
is clear evidence that the longshore transport rate
dropped off steeply seaward of the break point.
Predictions by the AW, BI, EH, and VR formulas
were in satisfactory agreement with the measured
Fig. 4. Comparison between calculated and measured cross-shore distribution of longshore sediment transport rate for Run 859050957 from the
DUCK85 experiment.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 87
rates both inside and outside the surf zone. In Fig. 7,
predictions of the transport rates are compared to
measurements from Run 859061018, and all formulas
overpredict the rates outside the break point (approx-
imately located at 140 m). In contrast, the AW, BI,
EH, and VR formulas produced satisfactory results in
the surf zone, whereas B and W produce much too
large transport rates in this zone as well.
Fig. 5. Comparison between calculated and measured cross-shore distribution of longshore sediment transport rate for Run 859051528 from the
DUCK85 experiment.
Fig. 6. Comparison between calculated and measured cross-shore distribution of longshore sediment transport rate for Run 859060916 from the
DUCK85 experiment.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 88
In general, based on the calculation results from the
DUCK85 runs summarized in Table 2, the B, BI, and
W formulas overestimated the transport, whereas the
AW, EH, and VR formulas yielded overall good pre-
dictions. Most of the formulas produced cross-shore
distributions that were more or less in agreement with
the measured distributions, although some calibration
factors might be needed to achieve quantitative agree-
ment. The observed discrepancy between the meas-
urements and predictions using standard coefficient
values is attributed to several factors: all formulas rely
on a considerable number of parameters and coeffi-
cients, where the values were typically determined
from situations not completely representative for the
field (e.g., laboratory, river environment). Also, the
transport is sensitive to the estimated bottom rough-
ness, which is a difficult quantity to determine, there-
by introducing significant uncertainty into the cal-
culations.
4.2. Comparisons with SUPERDUCK data
Four runs were conducted during the SUPER-
DUCK experiment where the cross-shore distribution
of the longshore sand transport rate was measured.
However, only two runs had sufficient information on
the local hydrodynamics to allow comparison with the
predictive formulas. Figs. 8 and 9 show the measured
and computed longshore transport rates for Runs
8609161116 and 8609191016, respectively. For Run
8609161116, similar conclusions can be drawn as for
the earlier discussed DUCK85 runs concerning all the
formulas. The B and W formulas overestimate the
transport rates, whereas the AW, BI, EH, and VR
formulas produce distributions that are in good agree-
ment with the measurements, at least shoreward of the
main break point. A relatively small amount of sand
was collected outside the surf zone (break point
located at approximately 170 m; see Fig. 8), again
displaying the sharp drop in the sand transport rate
occurring seaward of the break point. For Run
8609191016 (Fig. 9) the formulas showed an agree-
ment with the data in accordance with the previous
runs, although the over-predictions were relatively
more marked outside the surf zone for this run.
4.3. Comparisons with SANDYDUCK data
Five experimental runs from the SANDYDUCK
experiment (Miller, 1998) were available for compar-
Fig. 7. Comparison between calculated and measured cross-shore distribution of longshore sediment transport rate for Run 859061018 from the
DUCK85 experiment.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 89
ison with the predictive formulas. As opposed to the
DUCK85 and SUPERDUCK experiments, the trans-
port rates were found to be in the sheet flow regime
for all SANDYDUCK measurements. The cross-shore
distribution of the measured and calculated transport
rates on March 31, 1997 (denoted 97/03/31) is given
Fig. 8. Comparison between calculated and measured cross-shore distribution of longshore sediment transport rate for Run 8609161116 from the
SUPERDUCK experiment.
Fig. 9. Comparison between calculated and measured cross-shore distribution of longshore sediment transport rate for Run 8609191016 from
the SUPERDUCK experiment.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 90
in Fig. 10. For this run, the predicted peak transport
rates are markedly shifted shoreward relative to the
measured peak. Thus, all formulas produce unsatis-
factory results, although the BI, EH, VR, and W
formulas yield values that are more in agreement with
the data than the AW and B formulas. In Fig. 11
comparisons are made between predictions and meas-
urements for the run on April 1, 1997. As for Run 97/
03/31, the BI, EH and W formulas capture the main
features of the measured transport rate distribution.
Consequently, these formulas give better predictions
under sheet flow conditions than if low-energy swell
waves prevail, which was the case for DUCK85 and
SUPERDUCK. The AW and B formulas have a
tendency to overpredict under conditions giving large
measured transport rates, especially the AW formula.
This is partly because under high waves in the surf
zone the transport is dominated by suspended load. As
shown by Larson et al. (in preparation), the formulas
typically overestimate the time-averaged sediment
concentration, implying that the total transport rate
becomes too large.
Fig. 12 shows predicted and measured longshore
transport rates for a storm on October 20, 1997. The
measured transport rate distribution across shore is
bimodal with one peak just shoreward of the break
point and the other peak near the shoreline. However,
there is a large variability in the measured rate, both at
the scale of the surf zone and in between points, which
is difficult to explain in terms of the measured forcing.
Calculations with the AW, BI, EH, VR, and W for-
mulas yield acceptable agreement with the measured
distribution in most of the surf zone, but near the
shoreline the measured peak is not predicted. The
measured wave heights and currents at the points
closest to the shoreline do not indicate a potential
for large transport here. The AW and B formulas
overpredict the transport rate in the outer part of the
surf zone, but are doing better in the inner part, as well
as outside the break point.
Fig. 11. Comparison between calculated and measured cross-shore
distribution of longshore sediment transport rate for Run 97/04/01
from the SANDYDUCK experiment.
Fig. 12. Comparison between calculated and measured cross-shore
distribution of longshore sediment transport rate for Run 97/10/20
from the SANDYDUCK experiment.
Fig. 10. Comparison between calculated and measured cross-shore
distribution of longshore sediment transport rate for Run 97/03/31
from the SANDYDUCK experiment.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 91
Fig. 13 shows calculated and measured transport
rate distributions for the run on February 4, 1998,
which is representative for large transport on a barred
profile during a storm. The peak in the transport rate
was observed some distance shoreward of the bar
crest, whereas the formulas predicted the peak to
occur more seaward (i.e., close to the bar crest). In
this respect, the BI, EH, and W formulas yield
locations of the peak transport that are more seaward
than predicted by AW, B, and VR. In contrast to Run
98/02/04, Fig. 14 shows a comparison for Run 98/02/
05 representative of the transport during a moderate
storm. Most of the formulas underpredict the transport
rate, however, the AW and B formulas are yielding
satisfactory agreement. The general features of the
cross-shore distribution are well reproduced by all
formulas, indicating that tuning of the coefficients in
the formulas would considerably improve the predic-
tions.
Comparisons between the SANDYDUCK meas-
urements and the formulas allowed for an evaluation
of their predictive capability during storm conditions.
Overall, the AW and B formulas predicted higher
transport rates than the other formulas as well as the
measurements. Also, the VR and W formulas yielded
slightly better predictions than the BI and EH for-
mulas. The formulas often failed to accurately predict
the location of the peak in the transport rate in the surf
zone. However, seaward of this peak, all formulas
except B displayed satisfactory agreement with the
measured rates.
5. Discussion of results
Comparisons between field measurements and cal-
culations indicated that several of the formulas yield
predictions that might be considered acceptable in
many coastal engineering applications. However, to
objectively quantify the predictive capability of the
formulas, an overall comparison based on individual
point measurements was carried out for each formula.
Fig. 15 summarizes this comparison for the six
selected formulas and all measurements from the three
data sets. In the figure, the circles denote data points
from the DUCK85 and SUPERDUCK experiments,
representative of transport under low-energy condi-
tions, and the squares denote points from the SAN-
DYDUCK experiment, indicative of the transport
during storm conditions. Viewing Fig. 15, the BI
and VR formulas show best agreement regarding the
DUCK85 and SUPERDUCK data, displaying least
scatter around the line of perfect agreement (i.e., the
ratio between predicted and measured transports, q
p
/
q
m
, respectively, is one). Most of the computed values
are within a factor 5 of the measured values (see
dashed lines in Fig. 15) for these two formulas. The B
and W formulas have a tendency to overpredict the
Fig. 14. Comparison between calculated and measured cross-shore
distribution of longshore sediment transport rate for Run 98/02/05
from the SANDYDUCK experiment.
Fig. 13. Comparison between calculated and measured cross-shore
distribution of longshore sediment transport rate for Run 98/02/04
from the SANDYDUCK experiment.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 92
measured transport rates, whereas the AW and EH
formulas show considerably more scatter around the
line q
p
/q
m
= 1.0. Concerning the SANDYDUCK data,
computed values with the AW and B formulas are
typically too large and with the BI and EH formulas
too small. The VR and W formulas yield the least
scatter around the line of perfect agreement.
Quantitative and qualitative comparisons were
made between measurements and predictions regard-
ing the scatter, trend, and clustering of the calculated
Fig. 15. Comparison between calculated and measured cross-shore distribution of longshore sediment transport for all three data sets
employed.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 93
points around q
p
/q
m
= 1.0. As a measure of the scatter,
the rms error was calculated according to,
r
rms

N
1
_
logq
p
logq
m

2
N 1
_

_
_

_
1 2 =
2
where N is the number of data points. The computed
r
rms
values for all formulas are listed in Table 3,
where a smaller r
rms
value implies a smaller scatter.
From the table it can be seen that the BI formula
shows the smallest scatter for the DUCK85 and
SUPERDUCK data, followed by the VR and EH
formulas. The W formula shows the smallest scatter
for the SANDYDUCK data, followed by the BI and
VR formula. Taking an average for all data, the VR
and BI formulas display the least scatter.
Based on visual observations (Fig. 15), the for-
mulas were subjectively ranked from 1 (i.e., weak) to
5 (i.e., strong) concerning trends and clustering (see
Table 3). Also, a relative rating of the predictions was
assigned to the formulas utilizing a mean discrepancy
ratio, given by the percentage of the measurement
points lying between 1/5 to 5 of the predictions by the
formulas (this value was subtracted from 100% to
yield a small number for good agreement). The BI
formula produce the smallest discrepancy ratio (16%)
for DUCK85 and SUPERDUCK experiments, fol-
lowed by the VR and AW formulas (19% and 20%,
respectively). For the SANDYDUCK cases, the W
formula has a discrepancy ratio of only 4% with the
BI and VR formulas yielding ratios of 8% and 16%,
respectively. Taking an average for all experimental
cases, the VR formula produces the lowest discrep-
ancy ratio, whereas the other formulas yield compa-
rable ratios.
6. Conclusions
The VR formula gave the most reliable predic-
tions over the entire range of wave conditions
(swell and storm) studied, based on criteria involv-
ing the scatter, trend, and clustering of the predic-
tions around the measurements. The AW formula
gave satisfactory results for all conditions, but
scatter was marked both for swell and storm.
Regarding the scatter, the BI formula yielded
improved predictions compared to AW, although
the transport was systematically overestimated dur-
ing swell and underestimated during storm. The EH
formula displayed similar tendency as the AW
formula, producing reasonable results over the
entire range of wave conditions investigated, but
displaying significant scatter. The W formula
yielded the best predictions for the storm condi-
tions, but markedly overestimated the transport rates
for swell waves. Finally, the B formula systemati-
cally overestimated the transport rates for all con-
ditions.
The coefficient values in the sediment transport
formulas employed were the original ones as recom-
mended by the authors. In most cases, these values
were derived based upon laboratory data or data from
a river environment, involving no or limited field
Table 3
Summary of accuracy of all the formulas
Formula Scatter Trend Clustering Data with discrepancy ratio distribution
between 1/5 and 5
(DUCK85 +
SUPERDUCK)
(SANDYDUCK)
(%)
Bijker 0.868 0.608 2 1 32 8
EngelundHansen 0.705 0.519 4 3 29 18
Ackers White 0.812 0.724 4 3 20 22
BailardInman 0.659 0.485 2 4 16 24
Van Rijn 0.662 0.518 3 4 19 16
Watanabe 0.864 0.349 2 1 38 4
(DUCK85 +
SUPERDUCK)
r
rms
(SANDYDUCK)
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 94
measurements pertaining to longshore sediment trans-
port. Thus, additional calibration of the formulas
against the available data sets would increase their
predictive capability, but the modifications would be
weighted by the particular data sets. For example the
field data considered here only encompassed one
median grain size (0.18 mm; i.e., fine sand).
At the present time, there is no well-established
transport formula that takes into account all the differ-
ent factors that control longshore sediment transport
in the surf zone, although the VR evidently accounts
for many of those factors. A complete formula should
quantify bed load and suspended load, describe ran-
dom waves as well as the effects of wave breaking,
and include transport in the swash zone.
Acknowledgements
The research presented in this paper was carried
out under the Coastal Inlets Research Program of the
U.S. Army Corps of Engineers. Permission was
granted to N.C.K. and H.C.M. by the Chief of
Engineers to publish this information. Additional
support from the Swedish Natural Science Research
Council is also acknowledged (M.L. and A.B.).
Appendix A. Longshore sediment transport for-
mulas
A.1. Bijker formula (1967, 1971)
Bijker (1967) modified the KalinskeFrijlink for-
mula (Frijlink, 1952) for bed load together with Ein-
steins method for evaluating the suspended load
transport to be applied in a coastal environment. Thus,
Bijkers formula, popular among European engineers,
takes into account both waves and currents. The bed
load transport rate ( q
b,B
; in m
3
/s/m, including pores)
is calculated from,
q
b;B
Ad
50
V
C

g
p
exp
0:27s 1d
50
qg
ls
b;wc
_ _
where A is an empirical coefficient (1.0 for non-
breaking waves and 5.0 for breaking waves), d
50
the
median particle diameter, V the mean longshore cur-
rent velocity, C the Chezy coefficient based on d
50
, g
the acceleration of gravity, s ( = q
s
/q) the relative
sediment density, q
s
the density of the bed material,
q the density of water, l a ripple factor, and s
b,wc
the
bottom shear stress due to the waves and current. The
first part of the above expression represents a trans-
port parameter, whereas the second part (the expo-
nent) is a stirring parameter. The ripple factor, which
indicates the influence of the form of the bottom
roughness on the bed load transport, is expressed as,
l
C
C
90
_ _
1:5
where C
90
is the Chezy coefficient based on d
90
,
which is the particle diameter, exceeded 10% by
weight. The combined shear stress at the bed (s
b,wc
)
induced by waves and current is (valid for a 90 angle
between the waves and current),
s
b;wc
s
b;c
1
1
2
n
u
0
V
_ _
2
_ _
in which s
b,c
is the bed shear stress due to current only
and u
o
the maximum wave orbital velocity near the
bed. The coefficient n is given by,
n C

f
w
2g

in which f
w
is the wave friction factor (Jonsson, 1966).
To calculate the suspended load, Bijker (1967)
assumed that the bedload transport occurred in a
bottom layer having a thickness equal to the bottom
roughness (r). The concentration of material in the
bed load layer (c
b
; assumed to be constant over the
thickness) is:
c
b

q
b;B
6:34

s
b;c
q
_
r
The concentration distribution is obtained from,
cz c
b
r
h r
h z
z
_ _
w

q
p
j

s
b;wc
p
where z is the elevation, h the water depth, w the
sediment fall speed, and j von Karmans constant. By
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 95
integrating along the vertical from the reference height
to the water surface, the total suspended sediment load
is determined as,
q
s;B
1:83q
b;B
I
1
ln
33h
r
_ _
I
2
_ _
where I
1
and I
2
are the Einstein integrals (e.g., Van
Rijn, 1993). The total load is computed as the sum of
bed load and suspended load ( q
t,B
= q
b,B
+ q
s,B
).
A.2. Engelund and Hansen (1967) formula
Engelund and Hansen (1967) developed a formula
to compute the bed load transport under a current.
This formula was later used to compute the total load
and also modified to take into account wave stirring.
Applied to calculate the longshore sediment transport,
the formula yields:
q
t;EH
V
0:05Cs
2
b;c
1
1
2
n
u
0
V
_ _
2
_ _
2
s 1
2
d
50
q
2
g
5=2
This formula is also composed of a stirring term and a
transporting term, much in accordance with Watanabe
(1992). The same coefficient value ( = 0.05) apply for
both monochromatic and random waves in the orig-
inal formula.
A.3. Ackers and White (1973) formula
Similarly to Engelund and Hansen (1967), the
formula proposed by Ackers and White (1973) ini-
tially predicted the total load transport under a current,
but was later enhanced by Van De Graaff and Van
Overeem (1979) to describe the effects of waves. The
original AckersWhite formula may be written,
q
t;AW
V
1
1 p
d
35
V
V

_ _
n
C
d;gr
A
m
F
C
A
m
where p is the porosity of the sediment, d
35
the
particle diameter exceeded by 65% of the weight,
V
*
the shear velocity due to the current, n, m, C
d,gr
,
and A dimensionless parameters, and F
C
a sediment
mobility number. The dimensionless parameters are
written, respectively,
n 1 0:2432lnd
gr

m
9:66
d
gr
1:34
C
d;gr
exp
_
2:86lnd
gr
0:4343lnd
gr

2
8:128
_
A
0:23

d
gr
_ 0:14
where,
d
gr
d
35
gs 1
v
2
_ _
1=3
and m is the kinematic viscosity. The sediment mobi-
lity number is defined as,
F
C

V
V

V
_ _
n
C
n
d
C
d
g
n=2

s 1d
35
_
in which:
C
d
18 log
10h
d
35
_ _
The modified equation by Van De Graaff and Van
Overeem (1979) to take into account waves is written,
q
t;AWM
V
1
1 p
d
35
V
0
wc
V
;wc
_ _
n
C
d;gr
A
m

V
0
wc
V
;wc
V
0
wc
_ _
n
C
n
d
C
d
g
n=2

s 1d
35
_ A
_

_
_

_
m
where,
V
;wc
V

1
1
2
n
u
0
V
_ _
2
_ _
1=2
and:
V
0
wc
V 1
1
2
n
0
u
0
V
_ _
2
_ _
1=2
In the above formulation, n
0
is based on d
35
and n
on the bed roughness r.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 96
A.4. Bailard and Inman (1981, 1984) formula
Bailard and Inman (1981) extended the formula
introduced by Bagnold to oscillatory flow in combi-
nation with a steady current over a plane sloping
bottom. The instantaneous bed load ( q
0
b
,
BI
) and sus-
pended load ( q
0
s,BI
) transport rate vectors are ex-
pressed as,
q
0
b;BI

0:5f
w
qe
b
q
s
qg tan c

U
0
t

2
U
0
t

tanb
tanc

U
0
t

3
ib
_ _
q
0
s;BI

0:5f
w
qe
s
q
s
qgw

U
0
t

3
U
0
t

e
s
w
tanb

U
0
t

5
ib
_ _
in which tanb is the local bottom slope, tanc a
dynamic friction factor, U
t
0
the instantaneous velocity
vector near the bed (wave and current) and i
b
is a unit
vector in the direction of the bed slope. Averaging
over a wave period, the total transport rate and
direction are obtained containing both the wave- and
current-related contributions. Assuming that a weak
longshore current prevails, neglecting effects of the
slope term on the total transport rate for near-normal
incident waves, the local time-averaged longshore
sediment transport rate is (Bailard, 1984),
q
t;BI
0:5qf
w
u
3
0
e
b
q
s
q g tanc
d
v
2
d
3
v
_ _
0:5qf
w
u
4
0
e
s
q
s
q gw
s
d
v
u

where e
b
and e
s
are efficiency factors, and:
d
v

V
u
0
u

3

hjU
0
t
j
3
i
u
0
The following coefficient values are typically used
in calculations: e
b
=0.1, e
s
= 0.02, tanc = 0.63. Thus,
the efficiency factors are assumed to be constant,
although work has indicated that e
b
and e
s
are related
to the bed shear stress and the particle diameter. It
should also be noted that the formula is derived for
plane bed conditions.
A.5. Van Rijn (1984, 1993) formula
Van Rjin (1984) presented comprehensive formu-
las for calculating the bed load and suspended load,
and only a short description of the method is given in
the following. For the bed load he adapted the
approach of Bagnold assuming that sediment particles
jumping under the influence of hydrodynamic fluid
forces and gravity forces dominate the motion of the
bed load particles. The saltation (jumps) character-
istics were determined by solving the equation of
motion for an individual sediment particle. The bed
load can be defined as the product between the
particle concentration (c
b
; a reference concentration
for the bed load different from the reference concen-
tration for suspended load c
a
), the particle velocity
(u
b
), and the layer thickness (d
b
; taken to be equal to
the reference level a) according to,
q
b;VR
c
b
u
b
d
b
where,
c
b
c
0
0:18
T
D

d
50
s 1g
v
2
_ _
1=3
T
s
0
b;wc
s
b;cr
s
b;cr
in which c
0
( = 0.65) is the maximum bed load
concentration, D
*
the dimensionless grain diameter,
T the excess bed shear stress parameter, and s
0
b,wc
is
the effective bed shear stress for waves and current
combined (calculated according to Van Rijns own
method, not discussed here). Substituting the above
expressions into the bed load transport formula
together with some other relationships not given
yields,
q
b;VR
0:25cq
s
d
50
D
0:3

s
0
b;wc
q

s
0
b;wc
s
b;cr
s
b;cr
_ _
1:5
where,
c 1

H
s
h
_
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 97
in which H
s
is the significant wave height. The depth-
integrated suspended load transport in the presence of
current and waves is defined as the integration of the
product of velocity (v) and concentration (c) from the
edge of the bed-load layer (z = a) to the water surface,
yielding:
q
s;VR

_
h
a
vcdz
Integrating after substituting in the longshore cur-
rent can be shown to give,
q
s;VR
c
a
Vh
1
h
_
h
a
v
V
c
c
a
dz c
a
VhF
where c is the concentration distribution, V the mean
longshore current, and,
F
V

jV
a
h a
_ _
Z
_
_
0:5
a=h
h z
z
_ _
Z
0
lnz=z
0
dz=h

_
1
0:5
e
4Z
0
z=h0:5
lnz=z
0
dz=h
_
c
a
0:015
d
50
a
T
1:5
D
0:3

Z
0
Z w
Z
w
bjV

W 2:5
w
V

_ _
0:8
c
a
c
0
_ _
0:4
b 1 2
w
V
_ _
2
in which Z is a suspension parameter reflecting the
ratio of the downward gravity forces and upward fluid
forces acting on a suspended sediment particle in a
current, w is an overall correction factor representing
damping and reduction in particle fall speed due to
turbulence, and b is a coefficient quantifying the
influence of the centrifugal forces on suspended
particles.
Van Rijn (1984) calculated the concentration dis-
tribution c in three separate layers, namely:

from the reference level a to the end of a near-

bed mixing layer (of thickness d
s
)

from the top of the d

s
-layer to half the water
depth (h/2)

from (h/2) to h
Different exponential or power functions are
employed in these regions with empirical expressions
depending on the mixing characteristics in each layer.
A.6. Watanabe (1992) formula
The formula proposed by Watanabe (1992) for the
total load was developed to calculate the longshore
sediment transport rate as combined bed and sus-
pended load according to,
q
t;W
A
s
b;wc
s
b;cr
V
qg
_ _
where A is an empirical coefficient (about 0.5 for
monochromatic waves and 2.0 for random waves) and
s
b,cr
is the critical bed shear stress for incipient motion
(determined from the Shield curve for oscillatory
flow). This formula is composed of one part repre-
senting stirring of the sediment (the shear stress term)
and another term describing the transport (the long-
shore current speed).
References
Ackers, P., White, W.R., 1973. Sediment transport: new approach
and analysis. Journals of Hydraulics Division 99 (1), 2041
2060.
Bailard, J.A., 1984. A simplified model for longshore sediment
transport. Proceedings of the 19th Coastal Engineering Confer-
ence, pp. 14541470.
Bailard, J.A., Inman, D.L., 1981. An energetics bedload model for
plane sloping beach: local transport. Journal of Geophysical
Research 86 (C3), 20352043.
Bagnold, R.A., 1966. An approach to the sediment transport prob-
lem from general physics. Geological Survey Professional Pa-
pers 422-1, Washington, USA.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 98
Bijker, E.W., 1967. Some considerations about scales for coastal
models with movable bed. Delft Hydraulics Laboratory, Publi-
cation 50, Delft, The Netherlands.
Bijker, E.W., 1971. Longshore transport computations. Journal of
the Waterways, Harbors and Coastal Engineering Division 97
(4), 687703.
Birkemeier, A.W., Miller, C.A., Wilhelm, D.S., DeWall, A.E., Gor-
bics, S.C., 1985. A users guide to the coastal engineering
research centers (CERCS) field research facility. Instruction
Report CERC-85-1, Coastal Engineering Research Center, US
Army Engineer Waterways Experiment Station, Vicksburg,
MS.
Dally, W.R., Brown, C., 1995. A modeling investigation of the
breaking wave roller with application to cross-shore currents.
Journal of Geophysical Research 100 (C12), 2487324883.
Ebersole, B.A., Hughes, S.A., 1987. DUCK85 photopole experi-
ment, Miscellaneous Paper CERC-87-18, Coastal Engineering
Research Center, US Army Engineer Waterways Experiment
Station, Vicksburg, MS.
Engelund, F., Hansen, E., 1967. A Monograph On Sediment Trans-
port in Alluvial Streams. Teknisk Forlag, Copenhagen, Den-
mark.
Frijlink, H.C., 1952. Discussion des formules de debit solide de
Kalinske, Einstein et Meyer Peter et Mueller compte tenue
des mesures recentes de transport dans les rivieres Neerlanda-
ises. 2me Journal Hydraulique Societe Hydraulique de France,
Grenoble, 98103.
Jonsson, I.G., 1966. Wave boundary layers and friction factors.
Proceedings of the 10th Coastal Engineering Conference,
ASCE, pp. 127148.
Kraus, N.C., Isobe, M., Igarashi, H., Sasaki, T., Horikawa, K., 1982.
Field experiments on longshore sand transport in the surf zone.
Proceedings of the 18th Coastal Engineering Conference,
ASCE, pp. 969988.
Kraus, N.C., Gingerich, K.J., Rosati, J.D., DUCK85 surf zone
sand transport experiment. Technical Report CERC-89-5, Coast-
al Engineering Research Center, US Army Engineer Waterways
Experiment Station, Vicksburg, MS.
Larson, M., Bayram, A., Kraus, N.C., Miller, H., 2000. Compar-
ison of time-averaged concentration profile measurements
with existing predictive models. Coastal Engineering (in prep-
aration).
Miller, H.C., 1998. Comparison of storm longshore transport rates
to predictions. Proceedings of the 25th Coastal Engineering
Conference, ASCE, pp. 29542967.
Miller, H.C., 1999. Field measurements of longshore sediment
transport during storms. Coastal Engineering 36, 301321.
Nielsen, P., 1992. Coastal Bottom Boundary Layers and Sediment
Transport. World Scientific, Singapore.
Rosati, J.D., Kraus, N.C., 1988. Hydraulic calibration of the stream-
er trap. Journal of Hydraulic Engineering 114 (12), 15271532.
Rosati, J.D., Gingerich, K.J., Kraus, N.C., 1990. SUPERDUCK surf
zone sand transport experiment. Technical Report CERC-90-10,
Coastal Engineering Research Center, US Army Engineer
Waterways Experiment Station, Vicksburg, MS.
Soulsby, R., 1997. Dynamics of Marine Sands; A Manual for Prac-
tical Applications. Thomas Telford Publications, UK.
SPM, 1984. Shore Protection Manual. Coastal Engineering Re-
search Center, US Army Engineer Waterways Experiment Sta-
tion, Vicksburg, MS.
Swart, D.H., 1976. Computation of longshore transport. Report
R968-I, Delft Hydraulics Laboratory, Delft, The Netherlands.
Van De Graaff, J., Van Overeem, J., 1979. Evaluation of sediment
transport formulae in coastal engineering practice. Coastal En-
gineering 3, 132.
Van Rijn, L.C., 1984. Sediment transport: Part I: Bed load transport;
Part II: Suspended load transport; Part III: Bed forms and allu-
vial roughness. Journal of Hydraulic Division 110 (10), 1431
1456; 110 (11) 16131641; 110 (12) 1733-1754.
Van Rijn, L.C., 1993. Principles of sediment transport in rivers,
estuaries and coastal seas. Aqua Publication, The Netherlands,
Amsterdam.
Watanabe, A., 1987. Three-dimensional numerical model for beach
evolution. Proceedings of Coastal Sediments 87, pp. 802818.
Watanabe, A., 1992. Total rate and distribution of longshore sand
transport. Proceedings of the 23rd Coastal Engineering Confer-
ence, pp. 25282541.
Watanabe, A., Shimizu, T., Kondo, K., 1991. Field application of a
numerical model for beach topography change. Proceedings of
Coastal Sediments 91, pp. 18141829.
A. Bayram et al. / Coastal Engineering 44 (2001) 7999 99