SEDIMENT TRANSPORT

PROCESSES AND THEIR

MODELLING
APPLICATIONS

Edited by Andrew J. Manning

Sediment Transport Processes and Their Modelling Applications
[Link]
Edited by Andrew J. Manning

Contributors
Alberto Sanchez, Ma. Concepción Ortiz-Hernández, Levent Yilmaz, Xiao Hua Wang, Fernando Pinheiro Andutta,
Mohammed Achab, Yu-Hai Wang, Yun-Chih Chiang, Sung-Shan Hsiao, Wilson Mahera Charles, Narsis Anton Mtega,
Prashanth Reddy Hanmaiahgari, Ram Balachandar, Katerina Kombiadou, Yannis Krestenitis, Arnaud Hequette, Adrien
Cartier, Philippe Larroudé, Rabin Bhattarai, Vasileios Kitsikoudis, Epaminondas Sidiropoulos, Vlassios Hrissanthou,
Xiaobo Chao, Andrew Manning, Jeremy Spearman, Richard Whitehouse, Emma Pidduck, John Baugh, Kate Spencer

Published by InTech
Janeza Trdine 9, 51000 Rijeka, Croatia

Copyright © 2013 InTech

All chapters are Open Access distributed under the Creative Commons Attribution 3.0 license, which allows users to
download, copy and build upon published articles even for commercial purposes, as long as the author and publisher
are properly credited, which ensures maximum dissemination and a wider impact of our publications. After this work
has been published by InTech, authors have the right to republish it, in whole or part, in any publication of which they
are the author, and to make other personal use of the work. Any republication, referencing or personal use of the
work must explicitly identify the original source.

Notice
Statements and opinions expressed in the chapters are these of the individual contributors and not necessarily those
of the editors or publisher. No responsibility is accepted for the accuracy of information contained in the published
chapters. The publisher assumes no responsibility for any damage or injury to persons or property arising out of the
use of any materials, instructions, methods or ideas contained in the book.

Publishing Process Manager Iva Simcic

Technical Editor InTech DTP team
Cover InTech Design team

First published March, 2013

Printed in Croatia

A free online edition of this book is available at [Link]

Additional hard copies can be obtained from orders@[Link]

Sediment Transport Processes and Their Modelling Applications, Edited by Andrew J. Manning
p. cm.
ISBN 978-953-51-1039-2
Contents

Preface VII

Section 1 Sediment Transport Processes 1

Chapter 1 Sediment Transport Dynamics in Ports, Estuaries and Other

Coastal Environments 3
X. H. Wang and F. P. Andutta

Chapter 2 Longshore Sediment Transport Measurements on Sandy

Macrotidal Beaches Compared with Sediment Transport
Formulae 37
Adrien Cartier, Philippe Larroudé and Arnaud Héquette

Chapter 3 Sediment Transport Patterns Inferred from Grain-Size Trends:

Comparison of Two Contrasting Bays in Mexico 59
Alberto Sanchez and Concepción Ortiz Hernández

Chapter 4 Sediment Transport Modeling Using GIS in Bagmati

Basin, Nepal 77
Rabin Bhattarai

Section 2 Sediment Dynamic Processes 93

Chapter 5 Composition and Transport Dynamics of Suspended Particulate

Matter in the Bay of Cadiz and the Adjacent Continental Shelf
(SW - Spain) 95
Mohammed Achab

Chapter 6 Flocculation Dynamics of Mud: Sand Mixed Suspensions 119

Andrew J. Manning, Jeremy R. Spearman, Richard J.S. Whitehouse,
Emma L. Pidduck, John V. Baugh and Kate L. Spencer
VI Contents

Chapter 7 The Gravel-Bed River Reach Properties Estimation in Bank Slope

Modelling 165
Levent Yilmaz

Chapter 8 Scour Caused by Wall Jets 177

Ram Balachandar and H. Prashanth Reddy

Section 3 Numerical Modelling of Sediment Transport 211

Chapter 9 Modelling Cohesive Sediment Dynamics in the Marine

Environment 213
Katerina Kombiadou and Yannis N. Krestenitis

Chapter 10 Quasi-3D Modeling of Sediment Transport for Coastal

Morphodynamics 247
Yun-Chih Chiang and Sung-Shan Hsiao

Chapter 11 Derivation of Sediment Transport Models for Sand Bed Rivers

from Data-Driven Techniques 277
Vasileios Kitsikoudis, Epaminondas Sidiropoulos and Vlassios
Hrissanthou

Chapter 12 Modelling of Sediment Transport in Shallow Waters by

Stochastic and Partial Differential Equations 309
Charles Wilson Mahera and Anton Mtega Narsis

Chapter 13 Numerical Modeling Tidal Circulation and Morphodynamics in

a Dumbbell-Shaped Coastal Embayment 329
Yu-Hai Wang

Chapter 14 Numerical Modeling of Flow and Sediment Transport in Lake

Pontchartrain due to Flood Release from Bonnet Carré
Spillway 357
Xiaobo Chao, Yafei Jia and A. K. M. Azad Hossain
Preface

Sediment Transport Processes and their Modelling Applications is a book which covers a
wide range of topics. The effective management of many aquatic environments, requires a
detailed understanding of sediment dynamics. This has both environmental and economic
implications, especially where there is any anthropogenic involvement. Numerical models
are often the tool used for predicting the transport and fate of sediment movement in these
situations, as they can estimate the various spatial and temporal fluxes. However, the physi‐
cal sedimentary processes can vary quite considerably depending upon whether the local
sediments are fully cohesive, non-cohesive, or a mixture of both types. For this reason for
more than half a century, scientists, engineers, hydrologists and mathematicians have all
been continuing to conduct research into the many aspects which influence sediment trans‐
port. These issues range from processes such as scour, erosion and deposition, to how sedi‐
ment process observations can be applied in sediment transport modelling frameworks.
This book reports the findings from recent research in applied sediment transport which has
been conducted in a wide range of aquatic environments. The research was carried out by
researchers who specialise in the transport of sediments and related issues.
It is a great pleasure to write the preface to this book published by InTech. It comprises 14
chapters written by a truly international group of research scientists, who specialise in areas
such as sediment dynamics, hydrology, morphology and numerical sediment transport mod‐
elling. The majority of the chapters are concerned with sediment transport related issues in
estuarial, coastal or freshwater environments. For example: sediment dynamics in ports and
estuaries; sediment transport modelling of Bagmati Basin in Nepal using geographical infor‐
mation systems (GIS); numerical modelling of flow and sediment transport in Lake Pontchar‐
train; longshore sediment transport on sandy macrotidal beaches; and shallow water sediment
transport stochastic modelling. Other contributions in this book include: scour caused by wall
jets; mixed sediment flocculation dynamics; and fractal dimension of meandering rivers. Au‐
thors are responsible for their views and subsequent concluding statements.
In summary, this book provides an excellent source of information on recent research on
sediment transport, particularly from an interdisciplinary perspective. I would like to thank
all of the authors for their contributions and I highly recommend this textbook to both scien‐
tists and engineers who deal with the related issues.

Dr Andrew J. Manning
HR Wallingford Ltd, Coasts & Estuaries Group, Wallingford, UK
University of Plymouth, School of Marine Science & Engineering, Plymouth, UK
 Section 1

Sediment Transport Processes

 Chapter 1

Sediment Transport Dynamics in Ports, Estuaries and

Other Coastal Environments

X. H. Wang and F. P. Andutta

Additional information is available at the end of the chapter

[Link]

1. Introduction

Given ever expanding global trade, the international economy is linked to the well-being of
major coastal infrastructures such as waterways and ports. Coastal areas comprise about
69% of the major cities of the world; therefore the understanding of how coastal aquatic en‐
vironments are evolving due to sediment transport is important. This manuscript discusses
topics from both modelling and observation of sediment transport, erosion and siltation in
estuarine environments, coastal zones, ports, and harbour areas. It emphasises particular
cases of water and sediment dynamics in the high energy system of the Po River Estuary
(Italy), the Adriatic Sea, the Mokpo Coastal Zone (South Korea), the Yangtze Estuary and
the Shanghai Port, the Yellow Sea (near China), and Darwin Harbour (Northern Australia).
These systems are under the influence of strong sediment resuspension/deposition and
transport that are driven by different mechanisms such as surface waves, tides, winds, and
density driven currents.

The development of cities around ports is often associated with the expansion of port activi‐
ties such as oil, coal, and gas exportation. Such development results in multiple environ‐
mental pressures, such as dredging to facilitate the navigation of larger ships, land
reclamation, and changes in the sediment and nutrient run-off to catchment areas caused by
human activities [1]. The increase in mud concentrations in coastal waters is a worldwide
ecological issue. In addition, marine sediment may carry nutrients and pollutants from land
sources. An understanding of sediment transport leads to a better comprehension of pollu‐
tion control, and thus helps to preserve the marine ecosystem and further establish an inte‐
grated coastal management system [e.g., 2-3]. [4] observed that many historical sandy coasts
have been replaced by muddy coasts, and is considered permanent degradation. Addition‐
ally, [5] reported that recreational and maritime activities may be adversely impacted by
4 Sediment Transport Processes and Their Modelling Applications

processes of sediment resuspension and deposition. It was shown by [6] that increased sedi‐
ment concentration in the Adriatic Sea has affected the growth of phytoplankton at the sub‐
surface, because sunlight penetration is considerably reduced.

Before proceeding with the key issues about the transport of sediment in the previously
mentioned systems, a brief and summarized overview of the main characteristics and dy‐
namics of sediment transport is provided to contribute to the understanding of this chap‐
ter. In general, sediment particles considered in transport of sediment cycle, consist of
non-cohesive and cohesive sediment types (Fig. 1a). (a) Sediments of particle size d50< 4
µm, mud or clay, are classified as a cohesive sediments. In contrast, (b) particle size d50>
64 µm may be weakly cohesive; however, these particles are included in the non-cohesive
group, and range from mud through to sand [7]. The dynamics of sediment transport rely
upon water circulation, salinity concentration, biological interaction, and sediment type.
Cohesive sediments, such as clay and small-particle mud, are often transported in the wa‐
ter column, as these sediments are easily suspended by water currents. Alternatively, non-
cohesive sediments, such as sand, are usually transported along the bottom by the
processes of saltation, rolling, and sliding. Many numerical models include these process‐
es and are based on empirical experiments, often performed in laboratories. These experi‐
ments provide estimates of the bed load transport according to particle size, bottom
stress, and a threshold stress for initial bed movement [7].

The interaction between sediments is also an important feature pertinent to the transport of
sediment. The interaction among cohesive sediments (mud) is different from that of non-co‐
hesive sediments (sand). Cohesive sediments may aggregate, forming flocs of typical sizes
of 100-200 µm. This aggregation process is called flocculation, and is caused by chemical or
biological interaction. Flocculation is important for increasing the settling velocity; flocculat‐
ed sediment particles settle faster on the bottom. “Chemical flocculation” is started by salinity
ions that attach to the small mud particles, causing electronic forces between these particles,
which start aggregating and thus forming a larger mud floc. In contrast, “Biological floccula‐
tion” is caused by bacteria and plankton, which produce exopolymer (i.e. a transparent mu‐
cus) that acts as glue between mud particles. This mucus results in the formation of
extremely large flocs (~1000 µm in size), known as snow flocs [7].

The concentration of sediment near the surface may affect the formation of snow flocs, because
sunlight penetration in the water column is decreased due to increased suspended-sediment
concentration (SSC) and thereupon the reduced light penetration inhibits the production of
plankton. In high turbidity waters, i.e. SSC > 0.5 g l-1, marine snow is scarce; however, in less
turbid water, i.e. SSC < 0.1 g l-1, marine snow is common. Also, algae mats formation may influ‐
ence the degree of erosion, because they decrease the propensity of sediment resuspension. In
contrast, the influence of animal burrows may facilitate erosion [7]. Because of the different
types of sediments and the flocculation process, the profile of the vertical distribution of SSC
varies considerably. This vertical profile may indicate a well-mixed distribution, a smooth in‐
crease in sediment concentration with depth, or a depth-increase concentration with a step
shape, called lutocline (Fig. 1b). The lutocline inhibits vertical mixing and thus conserves a
nepheloid layer (i.e. bottom layer of high sediment concentration).
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 5
[Link]

Figure 1. (a) Cycle of deposition and resuspension of cohesive sediment involved in particle aggregation and breakup.
(b) Three typical vertical profiles of suspended sediment concentration in estuaries, where a, b and c denote a nearly
well-mixed, partially-mixed and step shape profiles, respectively [source: 7-8].

This chapter gives an overview of four important suspended sediment transport processes that
occur in ports, estuaries and other coastal environments. The following topics are investigated,
based upon research on sediment dynamics at the University of New South Wales, Australia:

• The importance of including wave-currents when modelling sediment transport, showing

the effect of waves generated during Bora events on SSC and net sediment flux in the
Northern Adriatic Sea;

• The effect of increased SSC, combined with increased irradiance factor (Fc) of photosyn‐
thetically active radiation (PAR), on phytoplankton blooms (PB), with analysis of the PB
event that occurred between January and April 2001 in the Mokpo Coastal Zone (Korea);

• The effect of coastal constructions on sediment transport, with analysis of the effect of
dikes on the Yangtze River Delta, and problems with silting in the navigation channel of
Shanghai Port (China);
6 Sediment Transport Processes and Their Modelling Applications

• Tidal circulation modelling, specifically the role of mangrove and tidal flat areas in caus‐
ing tidal asymmetry, and the effect on the transport of suspended sediment in Darwin
Harbour (Australia).

2. Description of study sites

2.1. The Po River and the Adriatic Sea

The Po River (~12.5o E and ~45o N) is 680 km in length, and is located in the northern area of the
Adriatic Sea. It provides up to 50% of the total fresh water discharge into the Adriatic Sea (Fig.
2). The annual mean river flow is ~ 46 km3/year, with the maximum river discharge events typ‐
ically occurring during the spring and a few times in autumn. The climate is temperate, with
average temperatures of over 10° C in summer, and over 0° C in winter, and a runoff of 250-750
mm/year. Strong northeasterly winds prevail in winter, known as Bora events (typical wind
speed ~30 m s-1). These winds are usually ~10 o C cooler than the water in the Adriatic Sea. In
contrast, the southeasterly winds, which are often less intense and occur during summer and
autumn, are known as Scirocco. The Bora and Scirocco wind conditions result in downwelling
and upwelling events, respectively, in the western Adriatic coast [9-11].
The Adriatic Sea is a semi-enclosed sea, being one of the arms of the Mediterranean Sea. The
Adriatic is connected to the eastern part of the Ionian Sea through the Ontranto Strait (Fig.
2). This sea is approximately 800 km long and 200 km wide. Depths vary from less than 200
m in the northern area, up to 1320 m in the southern area – with such depths covering an
entire ~120 km wide expanse (i.e. South Adriatic Pit, [12]), and reduce to less than 800 m at
the 70 km wide Ontranto Strait [10]. The eastern coast of the Adriatic Sea comprises numer‐
ous islands varying in main diameter from a few tens of meters up to tens of kilometres, and
this coastline has many zones of high steepness. In contrast, the western coast has isobaths
running parallel to the coastline and a smoother slope compared to the eastern coast (Fig. 2).
The Adriatic Sea receives the runoff of 28 rivers, mostly located along the coast of Italy. The
main river inflow to the Adriatic is from the Po River; however, the rivers Tagliamento,
Piave, Brenta, and Adige together contribute a runoff of ~15.2 km3/year, which is nearly one
third of the Po’s total runoff. The remaining 23 rivers in the Adriatic provide an average
runoff of ~8 km3/year [11].
The general circulation in the Adriatic Sea has been studied using field data and numerical
simulations by [1, 10, 13-16], and is observed to be a cyclonic (anti-clockwise) circulation that
is highly variable with the seasons [10, 13-17]. The annual water temperature excursion ex‐
ceeds 15o C. [1] observed the intense boundary current on the western side, the Western
Adriatic Coastal Current (WACC), which is both thermohaline and wind driven. The
WACC reaches maximum velocities during winter, under the influence of strong northeast‐
erly wind stress, i.e. Bora events [18]. The thermohaline component of the WACC is mainly
forced by river discharge from the Po River, and thus reaches maximum intensity during
spring and autumn [9, 19]. The position of the WACC is deflected from the inshore areas in
winter, towards the shelf slope during the summer by an opposite wind-driven current due
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 7
[Link]

to Scirocco events. Similar processes showing boundary currents being pushed offshore by
opposing wind-driven currents have also been observed at other shelves, such as: the Great
Barrier Reef, the continental shelf north of the Monterey Bay, and New Jersey shelf [20-22].

Figure 2. The Adriatic Sea location and a synthesized description of the main circulation.

Thermal balance in the Adriatic Sea is complex and influenced by river discharge (e.g. the
Po River), surface heat flux by the wind (Bora and Scirocco events), and heat flux through
the Otranto Strait. Water masses of the Northern Adriatic are renewed each year when the
colder and denser water mass sinks and moves along the seabed to the deep basin of the
Adriatic [1]. This northern Adriatic water mass forms a “denser cascade water”, which, for the
Adriatic Sea, is caused by temperature gradients, while for many aquatic systems, located in
Tropical and Sub-tropical areas, this is often caused by hypersaline waters [e.g. 23-30]. Dur‐
ing the spring and summer, however, the water mass in the northern Adriatic is warmed up
and forms a well-defined thermocline. Furthermore, the water discharge from the Po River
is an important controlling factor to the baroclinic currents in the basin of the Adriatic Sea
[9]. The thermal balance within the Adriatic Sea is also maintained by the net heat inflow
through the Otranto Strait from the Ionian Sea [1].

The annual load of sediment from the Po River is 10-15 x 106 tons/year. The sediment in the
Northern Adriatic Sea is mainly formed by sand with grain size varying from 50 to 2000 µm,
and silt with grain size between 2 and 50 µm. The smaller sediment particles, i.e. clay, are
also observed, however, they do not provide the major contribution of fine sediment in the
8 Sediment Transport Processes and Their Modelling Applications

northern area [1]. This chapter concerns the sediment transport in the Adriactic Sea of two
classes: sediment particles larger and smaller than 50 µm grain size. [31] suggested that fine
sediments such as silt and clay are mostly supplied from the Northern Adriatic Sea Rivers
(e.g. Po River). Sediment is supplied into the sea and later dispersed through local circula‐
tion. Because the general circulation of the Northern Adriatic Sea is cyclonic, with the pres‐
ence of the WACC on the western coast, there is a possibility that the sediment input from
the rivers in the Northern Adriatic is transported southward by the coastal current. There‐
fore, the bottom sediment distribution would be predominantly sorted by the grain size ac‐
cording to their respective settling velocities.

2.2. The Youngsan River and the Mokpo Coastal Zone (Korea)
The Youngsan River Estuary (YRE) is located in the Mokpo coastal zone (MKZ), in the
southwestern area of South-Korea (Fig. 3). The annual mean river flow is ~1.5 km3/year, and
the sediment load to the Yellow Sea is 0.7 x 106 tons/year. The climate is temperate, with
average temperatures typically between 1.7o C and 4.4o C during winter and between 21.4o C
and 26.1o C in summer. Maximum rainfall generally occurs during summer, accounting for
50 to 70% of the annual precipitation. Annual runoff is 250-750 mm. [11, 32-33]. The Mokpo
area is located at the southeastern boundary of the Yellow Sea, and the YRE is connected to
the Yellow Sea through four narrow inlets (i.e. ~1-3 km wide).

Figure 3. The domain of the hydrodynamic-sediment transport model at the southwest coast of Korea. The inset
shows the location of the 1-D ecosystem model (●) in the Youngsan River Estuarine Bay (YREB) [source: 33].

Tidal features in the YRE are mixed, but predominantly semidiurnal according to the criteria of
A. Courtier of 1938 [34], with the tidal form number [Nf=(K1+O1)/(M2+S2)=0.28]. Although there
is the presence of many island and tidal flat areas, the tidal currents of the YRE are ebb domi‐
nant. Ebb/flood dominance is characterized by a shortened ebbing/flooding period, resulting
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 9
[Link]

in stronger ebb/flood currents, respectively. In addition, the flooding periods are nearly twice
as large as the ebbing periods [35-37]. The ebb dominance is likely to be caused by important
features such as the many scattered islands, combined with the extensive tidal flats [35]. More‐
over, [38] observed that ebb dominance is likely to appear in regions of abundant tidal flats.

To add complexity to such tidal asymmetry problems (e.g. flood and ebb dominance), the
MCZ has three important sea structures: the dike built in 1981, the Youngam seawall built in
1991, and the Geumho seawall built in 1994. Since the construction of these structures, changes
in the tidal characteristics such as the increased amplitudes have been observed [37, 39-40].

This chapter section aims to show that in order to properly predict the variability in phyto‐
plankton mass production in the turbid waters of the MCZ, it is important to use a 3D sedi‐
ment transport model, coupled with the ecosystem model. This solves the variable vertical
dynamics of sediment resuspension and mixing [33].

2.3. Yangtze River and the Shanghai Port in the East China Sea

The Yangtze River or Changjiang River (Fig. 4) is the third longest river in the world (6300
km), and the fourth in terms of both water flow (~900 km3/year) and sediment discharge
(470-490 x 106 tons/year), with the transport of a dissolved load of 180 x 106 tons/year [11,
41-44]. The climate of this area is temperate, with temperatures of over 10o C in summer and
0o C winter, and an average runoff of 250-750 mm/year, the maximum river discharge occur‐
ring in summer [11]. The Yangtze is a mesotidal estuary according to the criteria of A. Cour‐
tier of 1938 [34], with a mean tidal range of 2.7 m [43].

Figure 4. The Yangtze River or Changjiang River in China, and the indication of the navigation channel used for the
trades of Shanghai Port [source: 44].

The sediment of the Yangtze River Estuary (YRE) mainly consists of small sediment particles of
less than 63 μm (over 95%). The system is dominated by small sediment particles that lead to a
highly turbid environment, and therefore the near bottom SSC can reach or exceed 4 g/l [44-47].
10 Sediment Transport Processes and Their Modelling Applications

The Yangtze connects to the coastal zone through four inlets, namely North Branch, North
Channel, North Passage, and South Passage. The main physical mechanics driving the trans‐
port of suspended sediment (TSS) varies between the four inlets: (a) in the South Passage TSS is
mainly driven by tidal distortion, (b) in the North Passage TSS is dominated by gravitational
circulation and tidal distortion, (c) in the North Channel TSS is dominated by gravitational cir‐
culation, and (d) for the North Branch the main mechanisms are not well described [44, 48-49].
For the North Passage other mechanisms are also suggested to contribute to the TSS and for‐
mation of the estuarine turbidity maximum zone (ETM), which include advective transport
and turbulence suppression by salinity or suspended sediment induced stratification [50]. [51]
performed a large analysis of the temporal and spatial variation of fluid mud, and flocculated
settling. However, the joint contribution of the different TSS driving mechanisms with geome‐
try is quite complex and requires further investigation.
The TSS in the Yangtze River Estuary (YRE) has been studied for many years [e.g. 41, 48-49,
52-59]. However, since the completion of the Deep Navigation Channel in 2011, important
changes to the local hydrodynamics, and thus to the transport of sediment, are expected. In
addition, there is the effect of the fluvial sediment trap by the Three Gorges Dam, which
caused a significant decrease to fluvial sediment load [59-60]. Although the reduction of flu‐
vial sediment has been reported, the silting problem attracted attention because the estimate
deposition of sediment in the navigation channel was over 100% of the original yearly aver‐
age predicted value, i.e. 30 million m3 [62]. Recently, [44, 63] have reported that the greater
siltation within the delta of the YRE is mostly influenced by the redistribution of local sedi‐
ment through processes such as erosion and deposition within the delta area.
On Yangtze Estuary is the Shanghai Port, the world’s busiest container port, which is extreme‐
ly important to the economy of China. During 2010 and 2011, this port handled nearly 30 mil‐
lion container units per year. To facilitate local navigation, the Deepwater Navigation Channel
(DNC) was built, 92 km in length and 12.5m deep. Although the channel comprises two dikes
of nearly 50 km each, as well as 19 groins built to increase speed along the DNC, silting is still
an issue, and dredging maintenance is greater than originally predicted [44, 55, 64-67].
A 1-DV model was applied to study the fine suspended sediment distribution at the South
Channel-North Passage of the YRE [68]. Then, a 2D vertical integrated model was used to sim‐
ulate, and subsequently to investigate the characteristics of tidal flow and suspended sediment
concentration at this channel [69]. From these studies it was observed that new features had
formed after the finalization of the shipping channel; however, the model used did not include
the baroclinic component, which is an important factor in the transport of sediment.

2.4. Darwin Harbour (Australia)

Darwin Harbour (DH) is a shallow estuary, with a typical depth of less than 20 m and a
maximum depth of up to ~40m. The harbour is situated in the Northern Territory (NT) of
Australia, and connects to the Timor Sea. The land surrounding DH is occupied by the cities
of Palmerston and Darwin (the latter is the capital of NT). DH is defined as the water body
south of a line from Charles Point (west point) to Gunn Point (east point), and comprises the
Port Darwin, Shoal Bay and the catchments of the West Arm, Middle Arm and East Arm
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 11
[Link]

[Darwin Harbour Advisory Committee 2003]. DH forms two adjacent embayments. The
western embayment receives the freshwater inputs predominantly from the Elizabeth River
(flowing into the East Arm), the Darwin River, Blackmore River and Berry Creek (flowing
into the Middle Arm), while the eastern embayment receives freshwater input from the Ho‐
ward River [70]. DH area comprises numerous tidal flats and mangroves, with nearly 5% of
the whole mangrove area in the Northern Territory, i.e. ~274 km2. [71-72]
Darwin Harbour is forced by semi-diurnal tides, and is classified as a macro-tidal estuary
(tidal form number Nf = 0.32). The maximum observed tidal range is 7.8 m, with mean
spring and neap tidal ranges of 5.5 m and 1.9 m, respectively [11, 73-76].
Evaporation usually exceeds rainfall throughout the year, except during the wet season.
From February to October, the evaporation rate ranges from 170 mm to 270 mm, respective‐
ly, with an average annual evaporation rate of ~2650 mm. The fresh-water input into DH is
negligible in the dry season, and evaporation exceeds river discharge. Therefore, in the dry
season salinity concentrations in the harbour may become at least 0.8 psu higher than the
adjacent coastal waters [77].

Figure 5. (a) Model domain of Darwin Harbour with indication of the harbour areas and data available to calibrate
and validate the model (yellow dots). (b) Unstructured numerical mesh used for the simulations, where colour corre‐
sponds to depth in meters, and numbered points indicate location of sampling stations used to analyse tidal asymme‐
try along the harbour [source: 76].

The climate of this region is tropical savannah, with average monthly temperatures of
over 20o C throughout the year. DH is located in a subarid/humid area with a typical
rainfall of 1500-1600 mm/year (rainfall of 2500 mm in exceptionally wet years). Runoff
typically varies between 100 and 750 mm/year, with maximum runoff usually occurring
between October and April [11]. Although DH is of great economic importance to the NT,
most of the current knowledge about the main driving forces for the local hydrodynamics
12 Sediment Transport Processes and Their Modelling Applications

is due to efforts by [75-76, 78-80]. To add complexity to the understanding of the hydro-
dynamical and morphological changes in DH, the combined effect from the headlands,
rivers, and embayments create a complicated bathymetry that leads to the formation of
many tidal jets within narrow channels, eddies etc.
[76] Conducted some research at the western embayment of DH (Fig. 5a), and provided a
calibrated and validated model to study the hydrodynamics in the harbour (Fig. 5b). From
this study, the role that the mangrove and tidal flat areas play on the tidal asymmetry could
be verified. It was thus confirmed that a decrease in area of the tidal flat and mangroves
would lead to increased tidal asymmetry of flood dominance, and, because of this, result in
the net sediment transport to the inner harbour area.

3. Methods

This chapter addresses the different study regions, followed by independent research and
numerical modelling. As such, we have provided the methodology in separate sub-sections.
Each of the following sub-sections summarizes the field work conducted, the calibration,
and validation of the model for the four study sites.

3.1. Setting up of the numerical model for the Northern Adriatic Sea
For the Adriatic Sea, a sediment transport model similar to that of [81] was used, with im‐
provements made by incorporating the effect of wave current [1, 82-84]. The Adriatic Inter‐
mediate Model was based on the Princeton Ocean Model (POM) [85]; with the horizontal
resolution of 5 km applied to a structured mesh. The model had 21 vertical layers that used
the sigma coordinate, with a high vertical resolution was used near the surface and bottom.
The simulations had the time steps of 7 and 700 seconds for the external and internal modes.
The 2.5 turbulent closure method of Mellor-Yamada was used, and the diffusivity coefficient
for SSC was assumed to be equal to that of heat and salt, and viscosity according to [86].
The flocculation of fine suspended sediment is mostly observed near the Po River mouth,
and in areas before reaching the ocean [87]. Because of that, flocculation or aggregation
processes were neglected, and thus all sediment behaves as a non-cohesive type and moving
as a Newtonian fluid. For the fine sediment, i.e. silt and fine sand (20 < d < 60 µm), resuspen‐
sion was caused by turbulence. Inertia of sediment particles was also neglected, and their
vertical velocity parameterized by a small settling velocity (w s). For more information about
the settling velocity, sediment source in the Adriatic Sea, and all the physical and numerical
parameters used in the model, please refer to [1, 82-84].
The tides are known to be relatively week in the Northern Adriatic Sea; however, [83] in‐
cluded the tides to observe the tidal current effect on sediment transport. For the bottom
stress two expressions were applied, an expression that considered the wave orbital velocity
on the bottom, and the other expression that neglected this effect. The third version of the
SWAN model was used to simulate the waves. The model was used in the stationary mode
to compute the wave fields under the forcing of 6 hour interval.
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 13
[Link]

The suspended sediment concentration was assumed not to affect water density. It is impor‐
tant to note that this last assumption is only valid for low concentrations of SSC, such as
those lower than 1 g/l [e.g., 88-90]. The conservation of SSM in the water column was ap‐
plied and the fluid considered incompressible. [2] showed that the Adriatic Sea is supplied
by a riverine sediment input that is ~ 1.67 Mt/month, with the Po River contributing nearly
70% of that. The other rivers along the Adriatic coast had the equal contribution, which rep‐
resented the remaining 30 % of the sediment input.

The sediment dynamics in the Northern Adriatic Sea is induced by riverine sources or resus‐
pended sediment from the seabed. Simulations were conducted to quantify the different
mechanisms responsible for the transport of suspended sediment [1], and the simulations
examining in details the wave-current interaction [84].

The numerical simulations by [1] were: (a1) simulation forced only by the Po River plume, (b1)
simulation forced by the Po River plume and wind stress. (c1, d1 and f1) simulation forced by
the Po River plume, wind stress, and additional wave forcing. For the simulation assuming
wind effect, the assumed wind conditions were the Bora and Scirocco, which are typical wind
conditions of the region [e.g., 9, 14-15, 19, 91] and summarized in [Table 2, in 1]. A homogene‐
ous field with initial temperature of 12o C, and salinity 38 psu were assumed in the model.
These are representative of ambient winter conditions without stratification. Simulations for a
30 day period were made, assuming continuous discharge from the Po River.

In contrast, the numerical simulations by [84] were: (a2) simulation forced without waves, tides,
and SSC effect on water density; (b2) simulation forced by waves, but without tides and SSC ef‐
fect on water density; (c2) similar to b2, except with waves assumed to be aligned with bottom
currents; (d2) simulation forced with waves and tides, but without SSC effect on water density;
(e2) simulation forced with waves and SSC effect on water density, but without tides. River run‐
off was assumed to be continuous from 1 January 1999 to 31 January 2001. The initial conditions
were obtained from climatological simulation of the Adriatic Sea circulation in [17], and the
sediment model was coupled with the hydrodynamical model from 1 December 2000.

3.2. Setting up of the numerical model for the Mokpo Coastal Zone

The simulation for MCZ consists of a 3D hydrodynamical model coupled with the sediment
transport model, and a 1-D biogeochemical model [33]. The Princeton Ocean Model (POM)
was chosen [85]. This model used the 2.5 turbulence closure scheme [92], and included the ef‐
fect of sediment concentration on water density, and the stability function on the drag bottom
coefficient [93]. The biological 1-D Modular Ecosystem Model (MEM) is based on the Europe‐
an Regional Sea Ecosystem Model [94]. This model constrains the physical and geophysical en‐
vironmental conditions such as photosynthetically active radiation, temperature, and salinity.
It also includes the trophic interactions between biological functional groups [95-96].

The simulations were run from January to April 2001, and the vertical salinity and tempera‐
ture data used to calibrate/validate the hydrodynamical model were obtained by Mokpo
National University at seven stations in the Youngsan River Estuary. The period of simula‐
14 Sediment Transport Processes and Their Modelling Applications

tion partially covers the winter to spring seasons, and the distribution of the 7 sampling sta‐
tions covers areas near the river mouth and upstream regions.

Hourly data from the hydrodynamic and sediment model were provided to the biogeo‐
chemical model. Specifically, temperature was used to compute the metabolic response to
the biota, salinity was used for oxygen saturation concentration, the vertical diffusion coeffi‐
cient was used for the biogeochemical-state variables, and the combination of sea surface el‐
evation with suspended sediment concentration was used to estimate light penetration in
the water column. The river discharge from the Youngsan Reservoir was also included. The
water depth at the Youngsan River Estuarine Bay was assumed to be 21 m. The horizontal
grid resolution of the model was 1km, with 18 vertical sigma layers. The open boundary
was forced with the four main tidal components, i.e. M2, S2, K1 and O1. Nodal corrections
and astronomical arguments were included to predict tides during the period of simulation.

The initial concentration values of pelagic biogeochemical are listed in (Table 1). The phyto‐
plankton population, biomass content of carbon, nitrogen, phosphorus, and silicon for each
phytoplankton group were obtained from [97]. Model sensitivity tests were performed in 8 dif‐
ferent simulations, by assuming different parameters for light attenuation and vertical mixing
rates, see Table 4 in [33]. A complete description of the whole setting of the model, the model‐
ling experiments, and additional numerical and physical parameters is provided in [33].

Nitrate Phosphate Ammonia Silicate DO

Surface 2.25 0.26 1.15 6.02 333

Bottom 1.36 1.20 1.20 7.36 340

Table 1. Initial condition assumed for concentrations of pelagic biogeochemicals (unit in mmol m−3) [source: 33].

3.3. Setting up of the numerical model for the Yangtze Estuary

To study the hydrodynamics and transport of sediment, the 3D Princeton Ocean Model
(POM) was used. This model uses a structured mesh and resolves the equations for momen‐
tum, temperature, and salinity using the finite differences method. The vertical coordinate is
sigma [85, 98-99], and the turbulent closure method is described in [92, 100], while to com‐
pute the vertical mixing processes [101] was used. To compute the horizontal diffusion of
momentum, the Smagorinsky diffusion scheme [86] was used. The complete description of
the model is shown by [99]. The wetting and drying scheme for the domain is implemented
in the model, with a minimum water depth established to avoid negative values [102-103].

To calibrate and validate the model in order to study the transport of sediment in the Deep
Navigation Channel DNC of the YRE, field data measured in 2009 were used. The data were
collected after the construction of the two dikes and 19 groins; however, the water depth
was about 10.5 m at that time [43-44]. The equation used in the model, the initial conditions
for the hydrodynamics, and initial sediment distribution are all described in [44]. The physi‐
cal and numerical parameters are summarized in table 2.
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 15
[Link]

Parameter Description Value

ws50 free settling velocity -1.715×10-5(ms-1)

m1 empirical settling coefficient -0.014

n1 empirical settling coefficient 2.20

m2 empirical settling coefficient 2.89

n2 empirical settling coefficient 2.80

C0 Flocculate empirical coefficient 0.20 (kgm-3)

α empirical coefficient 10.0

β empirical coefficient 0.5

E0 empirical erosion coefficient 2.0×10-5(kgm-2s-1)

τc critical shear stress for erosion or deposition 0.05(kgm-1s-2)

Table 2. Parameters used in the sediment transport model [source: 44].

Tidal harmonic components from 8 sites were used to verify the model, and the root
mean square error (RMSE). The tidal components used in the model were observed to
represent nearly 95% of the tidal oscillation (i.e. M2, S2, K1 and O1). Tidal currents were
used to verify the water speed, and a good agreement was achieved. Salinity measure‐
ments were used to verify the proper simulation of mixing in the YRE, and aside from
the periods of highly vertical stratification during ebb currents, the model properly simu‐
lated the temporal variation in salinity at the sampling sites. The final validation of the
model was to verify the proper simulation of the transport of SSC, and in general the
model could reproduce the physical mechanism driving the transport of sediment well. In
summary, aside from drawbacks such as over-mixing of salinity during the neap tide due
to the 2.5 Mellor-Yamada turbulence closure scheme, the model was calibrated and veri‐
fied, and thus was still a valuable tool to study and understand the influence the naviga‐
tion channel has on the transport of sediment within YRE.

3.4. Setting up of the numerical model for the Darwin Harbour

To simulate the hydrodynamics and transport of sediment for Darwin Harbour, the unstruc‐
tured numerical model FVCOM was applied [104]. The mesh was formed by 9,666 horizon‐
tal grid cells, and 20 vertical layers using sigma coordinate. The horizontal resolution varied
between ~ 20 to ~3,300 m, with the higher resolution areas in the inner harbour and lower
resolution in the outer harbour [76].
To force the model at the external open boundary, tidal forcing was used in the coastal area
between Charles Point and Lee Point. The tidal components were obtained from TPXO7.2
global model. The semi-diurnal components used to force the model were (M2, S2, N2 and
K2), while the diurnal components were (K1, O1, P1 and Q1). Three shallow-water compo‐
nents (M4, MS4, MN4) and two extra tidal components of low frequency were also used, i.e.
16 Sediment Transport Processes and Their Modelling Applications

Mf and Mm. For the internal boundary, e.g. upstream river zones, there are three sources of
fresh water in the domain (Elizabeth River, Blackmore River and Berry Creek); however, the
simulation was for the dry season and thus river discharge was negligible [76]. In the dry
season, the small presence of density-driven currents is often confined upstream of Darwin
Harbour, and they are often less than 3% of the maximum tidal current intensity. At the sur‐
face, the wind is an important mechanism to cause sediment resuspension, by wind-driven
currents and waves [105]. The macro-tides in Darwin Harbour (typical tidal oscillation be‐
tween 3.7 and 7.8 m), however, dominate the transport of sediment with tidal currents of up
to ~ 3m s-1 [106]. The additional effects from wind, river discharge and the heat flux at the
free-surface boundary were negligible, allowing the simulation to be forced by tides alone.

The bottom drag coefficient (C d) was set to be a function of the water depth (see Eq. 2 in 76).
The mangrove area was treated differently, because the influence of roots and trees signifi‐
cantly increase the friction and thus reduce water speed [107]. From empirical experiments
C d was observed to vary between 1 and 10, and its value relies upon tidal conditions, man‐
grove species, and patchiness of mangrove distribution. Therefore, the main value for C d
was set to 5. The remaining numerical and physical parameters, such as the viscosity and
diffusion coefficient, are all described in more detail in [76].

For the initial conditions, constant values for salinity (33 psu) and temperature (25°C)
were used. These are characteristic values during the dry season, and, with the zero river
input, result in a barotropic model. The simulation started on 20th of June 2006 ([Link]),
with a one second time step, and duration of 31 days. Six different simulations were ana‐
lysed, different sizes of tidal flats and mangrove areas were assumed, and one simulation
excluding the presence of tidal flats and mangrove areas. These simulations provide an
understanding of the independent effects of tidal flats and mangroves in the tidal asym‐
metry of Darwin Harbour. There were three main numerical experiments, namely: (Exp.
1) where tidal flats and mangrove areas were considered, (Exp. 2) where mangrove areas
were removed from the domain, and (Exp. 3) where both mangrove and tidal flat areas
were removed from the domain.

4. Results and discussion

4.1. Sediment transport in the Adriatic Sea

The key results from [84] are summarized as follows:

The Bora wind generated barotropic southward longshore currents that connected to the
partially buoyancy driven WACC. This resulted in surface water currents of up to 1.3 m s-1
near the Po River mouth, and maximum bottom currents of 0.3 m s-1 near Ortona. These
general features were all in concordance with [108]. The smooth wind conditions resulted in
a reduced interior vorticity, which is caused by the orographic incisions around the Dinaric
Alps [109]; however, the good representation of the WACC in the Nothern Adriatic Shelf
combined with the wave-currents provided a realistic physical representation of sediment
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 17
[Link]

transport during the Bora event. The Bora wind caused higher wave heights on the western
coast than on the eastern coast. The wave direction was mainly aligned with the wind direc‐
tion in the Adriatic Sea; however, the direction was mainly perpendicular when approach‐
ing the western coast because of wave refraction.

For low and moderate wind conditions, the modelled waves showed good agreement with
observed waves. The measurements used to verify the model were obtained at the buoys at
Ancona and Ortona. During strong wind conditions, such as Bora events, the model showed
good results compared to observations of the waves at Ortona, while for Ancona the wave
response was underestimated by 50%. This was caused by the low horizontal scale resolu‐
tion of 40 km ECMWF wind fields. Due to the complex orography, the model is incapable of
resolving the fine wind variability [91, 110].

Figure 6. The bottom stress on 15 January 2001, predicted by Experiment 1 (a), Experiment 2 (b), and their differ‐
ence (c) [source: 84].

Waves and currents have been shown to affect sediment resuspension in the Bottom Boun‐
dary Layer (BBL). Recent field studies conducted near the Po River delta were used to ana‐
lyse the effect of wave-current interaction [e.g., 111-113]. The simulation without the wave
effect (experiment 1) showed the bottom current reaching ~ 0.34 m s-1 during the Bora event.
The Bora event caused the bottom stress to increase from 0.01 N m-2 to 0.66 N m-2 (Fig. 6a).
18 Sediment Transport Processes and Their Modelling Applications

This increased bottom stress results in considerable erosion, and a subsequent increase in
concentration of both fine and course sediment near the bottom. During periods without
Bora winds, the resuspension was weak, and the fine sediment from the Po River dis‐
charge dominated the SSC in the water column. For experiment 2, in which the effect of
wave-current interaction was considered (Fig. 6b), the bottom stress reached a maximum
value of 2.2 N m-2, and the concentration of fine and coarse sediments increased by 80%.
(Fig. 7b). The wave-current interaction increased the bottom drag coefficient from 0.0048
to values of up to 0.015 (not shown).

Figure 7. The surface (layer 1) fine sediment concentration on 15 January 2001, predicted by experiment 1 (a), Experi‐
ment 2 (b), and their difference (c) [source: 84].

In experiment 1, a high bottom stress was predicted in the north-east shelf along the Italian
coast. Therefore, high SSC was obtained (i.e. 28 g m-3) near the Po River delta, with similar con‐
centration of coarse and fine sediments, which were very well vertically mixed. In experiment
2 the distribution patterns were very similar to those of experiment 1. The bottom stress and
sediment concentration increased in magnitudes to 1.3 N m-2 and 50 g m-3, respectively. This in‐
crease was verified along the coast (~ 200 km long), southwards from the Po River nearshore
zone. It was observed that the prediction of the fine suspended sediment concentration
showed good agreement with [112]. [113] has also obtained SSC in the range of 70-100 g m-3.
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 19
[Link]

The sediment flux was analysed in experiment 1, using results from a cross-sectional area.
During the Bora event on 15th of January, an upwelling net sediment flux was observed, de‐
spite the downwelling currents along the Italian coast. The horizontal fields near the surface
and bottom were also analysed, and a net sediment flux with maximum values of 20 g m-2
s-1, was observed southwards near the Italian coast. The predictions were in agreement with
estimates by [113] in a different Bora event. In experiment 1, the southward flux of 10.5 t s-1
(fine sediment) and 9.3 t s-1 (coarse sediment) was calculated at a cross-section area (N) near
the Po River delta in the Adriatic Sea. In contrast, for experiment 2, the flux of fine and
coarse sediment increased to 25.6 t s-1 and 24.1 t s-1, respectively.

Experiment 3 was conducted to verify the effect from the wave and current aligned to the
sediment transport. The net sediment flux of fine and coarse sediment was over predicted
by 8% and 9%, respectively, from experiment 2. These small differences are evidence that
the wave propagation direction had little effect on sediment flux. This was shown for strong
wave conditions by [114].

To observe the effect of the tides, experiment 4 was conducted. It considered the same con‐
ditions as experiment 2, with the additional influence of tidal currents. The four main semi-
diurnal tidal components were used (i.e. M2, K2, N2, S2), and the three main diurnal
components (i.e. K1, O1, P1). The sediment flux at the cross-sectional area (N) was observed
to be reduced by 1%, compared with results from experiment 2.

The final experiment was similar to experiment 2. It included, however, the additional in‐
fluence of sediment concentration on water density, using a simple bulk relation from
[83]. Because of the small SSC, the effect on sediment flux was negligible, with slight
changes of less than 1%.

4.2. Factors driving the phytoplankton bloom in the Mokpo coastal zone

The results showed little change in phytoplankton biomass throughout the entire water col‐
umn during January. In contrast, in February the concentration of phytoplankton biomass
increased at the surface. In the first two weeks of March this biomass increased significantly,
and reached maximum values at the surface after two weeks (i.e. mid March). The model
simulated the timing of the phytoplankton blooms well; however, the maximum biomass
obtained from the theoretical results was 2-3 times lower than observed values. Some possi‐
ble explanations for this underestimation are the short period of simulation (4 months), and
the use of a 1-D biogeochemical model instead of a 3D model. Moreover, when the sluice
gates located upstream are opened, the system temporally becomes a salt-wedge estuary.
This increased fresh water input may cause the sinking riverine phytoplankton detritus that
flows along the bottom. Additionally, there is a likely effect from the resuspended phyto‐
plankton and/or detritus that were not incorporated in the model [33].

A good correlation was found between the variation in phytoplankton biomass and diatom
concentrations. Despite the observed increase in radiance Q (W m-2), the diatom bloom fin‐
ished in April. Diatom concentration decreases mostly because of lack of dissolved phos‐
phate (P) and silicate (Si) in the euphotic zone. This depletion is caused by phytoplankton
20 Sediment Transport Processes and Their Modelling Applications

uptake combined with stratification-induced limitation of nutrient supply near the bottom.
It has been shown that phytoplankton blooms develop due to a decrease in vertical mixing
rates. Therefore, the increased vertical mixing that occurs from early January to mid-Febru‐
ary inhibits the permanence of phytoplankton cells in the euphotic area [33]. Vertical mixing
was increased due to the colder wind decreasing surface water temperatures, which then
caused convective overturn in the water column.

The features of phytoplankton biomass (PB) variation were examined for light attenuation
coefficients (Fc) of 0.43, 0.46 and 0.49 (Fig. 8). The PB over the euphotic zone increased slow‐
ly. All the blooms started at the same time, reaching nearly the same maximum of PB, i.e. 35
mg m-2. However, the rates of increase were different, with maximum growth rate verified
to occur for Fc = 0.49. Moreover, variation between neap and spring tides has also been
shown to affect the PB. Water turbidity is increased due to larger sediment resuspension
caused by the strong spring tidal current, and thus the PAR attenuation may affect phyto‐
plankton production. The model results showed that the absence of the effect from SSC re‐
duces PAR attenuation, and therefore changes phytoplankton production (not shown).
Vertical mixing also affects the PB, because phytoplankton cells are taken to deeper layers
where light attenuation is higher [33]. [115] proposed that blooms happen when phyto‐
plankton growth exceeds the rate of vertical movement. Therefore, a decrease in vertical
mixing results in phytoplankton blooms in coastal and oceanic waters [e.g. 116-118].

Figure 8. Temporal variation of the vertical integrated chlorophyll-a concentration in the euphotic zone. The numeri‐
cal experiment for different values of Fc and the simulation start time at January 2001 [source: 33].

4.3. The effect of dikes on sediment transport in the Yangtze River Estuary

Tidal harmonic components were calculated using measurements from 8 water level stations,
and later compared with results of the model. The root mean square error obtained for the 4
main components was under 10%, and the shift of tidal phases less than 10 degrees. In general,
the model showed reasonable agreement compared with measurements of tidal oscillation,
water current, and temporal variation of salinity and SSC. Details of the model calibration and
verification are shown in [44]. The results of the SSC calculated from the model are shown in
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 21
[Link]

(Fig. 9a). Differences between simulations and observations of the speed increase of the surface
flood currents, during the rising sea level, were verified. Observations showed more steepness
during flood tides, and, for a short period of time, measured flood current was approximately
20% higher than the model result. In contrast, water currents near the bottom consistently
showed good agreement between modelled and observed values at any given time. Even with
the minor differences found, the model was properly calibrated, and thus shown to be a valua‐
ble tool for studying sediment transport mechanisms in the YRE [44].

Figure 9. (a) Measured (red line) and model simulated (blue dotted line) suspended sediment concentration at 62
cm above the seabed. (b) The tidally averaged residual suspended sediment flux and its three components, i.e. re‐
sidual flow (Fr), tidal pumping (Ft), and shear dispersion (Fs), where the positive value indicates ebb transport of
SSC. (c) and (d) show the tidally averaged suspended sediment concentration after and before the DNC construc‐
tion, respectively. The location of the dikes is indicated in both figures (c and d), and the location of the sampling
station in the DNC (black dot).

Figure 9b shows field data results of the sediment transport near the bottom. During the
spring tides between 30th of March and 2nd of April, there was more suspended sediment
present at the onset of the ebb tide than at the onset of the flood tide. This resulted in a resid‐
ual and tidal pumping transport of sediment downstream, where residual transport domi‐
nated the total transport of suspended sediment (TSS). Commonly, high residual TSS causes
a downstream shear dispersion TSS, whereas low residual transport causes an upstream
shear dispersion transport. In contrast, during the neap tide observed from the 2nd of April
onwards, the bottom horizontal water velocities were much smaller than those observed
during spring tides. Because of this, sediment resuspension decreased considerably, and
22 Sediment Transport Processes and Their Modelling Applications

therefore the residual transport of SSC was reduced. In addition, tidal pumping was ob‐
served to transport SSC mainly landward. The residual transport and shear dispersion were
observed to alternate SSC transport between upstream and downstream directions.
The results showed that the maximum SSC is generated in the sand bar area (Figs. 9c,d). The
simulation in which the effect of the two dikes was applied (Fig. 9c) showed a discontinuity
in the high SSC caused by the DNC, and the maximum SSC was observed adjacent to the
north dike (e.g. SSC ~ 6 kg m-3). In contrast, the simulation without the presence of dikes
showed high SSC within the entire sand bar area near the estuarine mouth. The SSC de‐
creasing upstream indicated a low local resuspension of sediment and a low sediment input
from river discharge. In addition, the Three Gorges Dam caused a significant decrease in
sediment load to the Yangtze River [60-61]. This local sediment resuspension is therefore
evidence that the silting occurring within the DNC is caused by redistribution of sediment
between the shoals on each side of the dikes.
Comparison between the simulations with and without the dikes, found enhanced ebb dom‐
inance to occur with the inclusion of the dikes (not shown). These changes have previously
been observed and reported [63], by comparing field measurements obtained before and af‐
ter construction of the dikes. After construction, the traditional sediment transport path
across the estuarine mouth was blocked; however, some sediment may still be transported
around the dike edges, and a small amount may enter the channel. This sediment intruding
into the DNC is likely to cause siltation near the seaward side of the DNC.

4.4. Tidal asymmetry in Darwin Harbour

The amplitude of the M2 tidal component was observed to gradually increase from the open
boundary (i.e. outer harbour) to nearly 1.7 m in the arms, and decrease from the arms to
nearly 1.0 and 0 meters in the tidal flat and mangrove areas, respectively. The decrease in
amplitude for the tidal flat and mangrove areas was caused by large energy dissipation due
to bottom friction. The phase shifted in areas of the outer harbour and harbour arms, and
this shift was caused by reduced wetting-drying areas. The maximum horizontal velocity is
about 3.0 m s-1 in the Middle Arm (see Fig. 5). The variation of M2 tidal current ellipses were
observed at many vertical layers and locations of the domain. In the outer harbour the M2
currents were up to 0.3 m s-1, and up to 0.6 m s-1 in the channel. The horizontal currents were
observed to decrease by almost 30 % from the surface to the bottom [76].
Comparison of the three numerical experiments from simulations that neglected the tidal
flat or mangrove areas, showed an increase in the tidal amplitude for the inner and outer
harbour. The maximum increase was obtained when both tidal flats and mangroves areas
were not considered in the domain (0.02 m), while there was an increase of 0.01 m for the
simulation neglecting only the mangrove areas. Although the variation of tidal amplitude
was observed to be quite small, for the M4 tidal component, there was an increase in amplifi‐
cation of almost 50% when the mangrove areas were removed, and almost 75% when both
the tidal flats and mangrove areas were removed. The phase of the M4 tidal component was
observed to advance almost 20 degrees when the mangrove areas and tidal flats were not
included in the domain [76].
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 23
[Link]

To verify the effect of mangrove and tidal flat areas on tidal asymmetry, the tidal asymme‐
try skewness parameter γ was calculated. This allows identification of the major factors con‐
trolling tidal asymmetry. For γ, it was verified that the main tidal components controlling
tidal asymmetry were M2 and M4, and so the expression to calculate γ was:

3 2 2
2 M 2 M 4 ( M 2 M 4)
a a sin 2ϕ − ϕ
γM 2/M 4 = (1)
1 2 3

2 ( aM 2 + 4aM2 4) 2

where a and ϕ are respectively the amplitudes, and phases of the astronomical tides M2 and M4.

Figure 10. Predicted changes of the coefficient ( γ ), with results obtained from the difference between experiment 2
and 1 (A), and 3 and 1 (B). (C) Parameter γ calculated for different percentages of mangrove area removed in the East
Arm near Station Blay [source: 76].

Figure 10 shows that the removal of mangroves and tidal flat areas resulted in an increased
tidal asymmetry. In turn, Darwin Harbour would have more flood dominance in tidal cur‐
rents (Figs. 10A,B). The maximum increase in asymmetry was observed in experiment 3. The
experiments in which different percentages of the mangrove areas were removed are shown
in figure 10C. The relation between removed mangrove area and increased tidal asymmetry
skewness factor was observed to be almost linear, and increased by 0.1 if 100% of the man‐
24 Sediment Transport Processes and Their Modelling Applications

grove areas were removed. These results demonstrate that the tidal flats and mangrove
zones function as sponge zones for dampening tidal asymmetry [76].

5. Conclusions

This chapter provided information about four important suspended sediment transport
processes. It showed how wave current interaction increases SSC and therefore affects the
net transport of sediment, and demonstrated the importance of applying a coupled hydro‐
dynamical and biogeochemical numerical model to better simulate phytoplankton blooms.
In addition, the effect of coastal construction on sediment transport was described, along
with the role of tidal flats and mangrove areas in causing tidal asymmetry. Specific conclu‐
sions for each case study are as follows:

5.1. Effect of wave current interaction on sediment transport in coastal zones

The general features of the model during the Bora event in January 2001 predicted the
WACC, and the large wave fields generated by the strong winds agreed with observations
on the western Adriatic shelf. The bottom boundary layer was resolved, and the wave-cur‐
rent interaction was able to be implemented in the model. Moreover, the tidal effect on sedi‐
ment transport was also examined in the Northern Adriatic Sea.

The model showed that the Bora event occurring from 13–17 January 2001 caused waves
with height of 2 m, and period of 5 s. The wave-current near the bottom layer resulted in
strong sediment resuspension. The wave effect, combined with the longshore coastal cur‐
rents and the turbulent driven vertical flux, caused a large southward sediment flux. The
flux was estimated using data from a cross-section area of the Adriatic Sea, near the Po Riv‐
er. The experiment including the wave-current interaction showed larger southward sedi‐
ment flux than the experiment with waves neglected. The sediment flux was clearly
maintained by the strong vertical mixing and the bottom sediment resuspension. The results
also showed that the southward flux was more confined to the northern area of the Adriatic
Sea, and sediment plumes were confined to the western Adriatic shelf north of Ancona.

Simulation results have shown that in areas between the Po River and Ancona, sedimenta‐
tion and erosion rates doubled due to the combined motion of wave–current interaction in
the BBL during the Bora event. Bora events typically occur 9 times during winter seasons
[119], and thus the influence of waves is important in the long-term sediment transport at
this coastal area. The annual sedimentation rate near the Po River mouths was found to be
2-6 cm yr-1 [120]. The sedimentation rate predicated during the Bora event of 13-17 January
2001 represented nearly 10% of the annual rate. This prediction agrees with observations if
we assume the average of ~ 10 Bora events per year. In addition, from experiment 2, it was
verified that the high concentration of suspended sediment along the western coast, north of
Ancona, was locally driven by waves. Moreover, the wave-current interaction with the BBL
enhanced the bottom stress and caused increased sediment resuspension. From the third ex‐
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 25
[Link]

periment it was observed at the cross-sectional area (N), that wave direction and tides had
little effect on sediment resuspension in the northeastern coast of the Adriatic Sea.

5.2. Effect of SSC on phytoplankton bloom

Simulation of phytoplankton biomass dynamics was obtained during the winter and spring
seasons. The model was run using irradiance forcings and physical oceanographic forcings
such as SSC, temperature, salinity, tides, and vertical diffusivity. The results from winter re‐
vealed that cooling of the surface waters induced vertical mixing, and thus inhibited the
growth of phytoplankton biomass. In contrast, periods of neap tides combined with in‐
creased freshwater discharge decreased vertical mixing, and in turn triggered PB spring
bloom. In the tidally turbid coastal waters, such as the MCZ, the small neap tidal currents
were not capable of increasing SSC. As a result, more light was observed though the water
column, favouring increases in phytoplankton biomass [33].

Results revealed that prediction of phytoplankton blooms are very sensitive to the light at‐
tenuation factor (Fc), vertical diffusion (Kv), and suspended sediment concentration (SSC).
Moreover, simulations using depth-averaged diffusivity and monthly averaged vertical dif‐
fusivity, may hinder the primary production processes. Finally, the simulations revealed the
importance of coupling the 3D hydro-sediment model with the ecosystem model, which
leads to more realistic estimates of phytoplankton biomass oscillation [33].

5.3. Influence of coastal construction on sediment transport

The simulations were well calibrated and validated using field data measured in the DNC
in Shanghai Port on March 2009. Combining the field measurements with the extrapolat‐
ed results for the whole domain of the YRE, we have verified that the system is highly
stratified during neap tides; however, close to well mixed during spring tides. Calculation
of the Richardson number revealed the dominance of fresh water inflow over vertical
mixing during neap tides; while, in spring tides, vertical mixing overcomes the buoyancy
of the fresh water river inflow [44].

The simulations revealed that the transport of sediment into the DNC comes from open
areas, rather than from upstream areas as is commonly expected. Therefore, the input from
the Yangtze River is not the major source of sediment, causing siltation, in this navigation
channel. The dikes were observed to inhibit the majority of the alongshore sediment trans‐
port near the delta zone; however some of the sediment transported towards the dikes veers
south, and as it approaches the edges of the dikes a small amount may enter the channel. It
was verified that, in the North Channel, the estuarine maximum turbidity zone was generat‐
ed predominantly by gravitational circulation, whereas, in the South Passage, it was gener‐
ated mainly by tidal distortion effects. For the DNC there was no maximum turbidity
generated locally. Some turbid waters in adjacent areas were observed to be transported up‐
stream in this channel. In summary, the simulation results revealed the magnitude of the
maximum turbidity zone in the North Passage, and showed that most of the sediment de‐
positing into the DNC is caused by sediment redistribution from adjacent zones [44].
26 Sediment Transport Processes and Their Modelling Applications

5.4. Tidal asymmetry modulated by tidal flats and mangrove areas

The results obtained from the simulations, combined with field observations, showed that
the circulation in DH is mainly driven by tides during the dry season, with negligible wind
and river influence. DH is a semidiurnal system, dominated primarily by the M2 tidal com‐
ponent, and secondarily by the S2 component. The maximum velocities observed were in the
Middle Arm, with peaks of over 3.0 m s-1 at the surface layer. The Middle Arm also had the
maximum and minimum vertical shear of the horizontal velocities, which would therefore
result in some of the maximum vertical mixing zones of DH.
The sensitivity test evidenced that tidal flats and mangrove areas play an important role in
modulating the amplitude and phase of some tidal components. In particular, the amplitude
of the M 4 component was observed to increase by almost 75%, and the phase was verified to
advance/delay a few degrees in the outer/inner harbour respectively. In addition, the ampli‐
tude and phase of the M2 component were observed to change in the inner harbour, with an
amplitude increase of 1% and phase advance of 4 degrees.
The parameter gamma was calculated, showing variation in the tidal elevation skewness.
Gamma was observed to increase by 100% in the absence of mangrove areas, and by 120% in
the absence of both tidal flats and mangrove areas. Furthermore, the increase in the eleva‐
tion skewness correlated almost linearly with the decrease in mangrove area. It was there‐
fore suggested that a similar effect would be found with a decrease in tidal flat areas. The
findings of this study show how important flooding estuarine areas, e.g. tidal flats and man‐
groves, are in modulating tidal asymmetry. These findings could be further used to verify
similar effect in other estuaries, bays, and harbour areas. As tidal asymmetry strongly affects
sediment transport in the estuaries, care must be taken in terms of the reclamation of the
mangrove areas and tidal flats around the harbour watershed.

Acknowledgements

X. H. Wang and F. P. Andutta were supported by a 2011 Australian Research Council/ Link‐
age Project – LP110100652. This work was also supported by the National Computational In‐
frastructure National Facility at the Australian National University. This is a publication of
the Sino-Australian Research Centre for Coastal Management, paper number 10.

Author details

X. H. Wang* and F. P. Andutta

*Address all correspondence to: [Link]@[Link]

School of Physical, Environmental and Mathematical Sciences, University of New South

Wales at Australian Defence Force Academy UNSW-ADFA, Australia
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 27
[Link]

References

[1] Wang, X. H., & Pinardi, N. (2002). Modeling the dynamics of sediment transport and
resuspension in the northern Adriatic Sea. Journal of Geophysical Research, 107(C12),
3225, doi:10.1029/2001JC001303.

[2] Frascari, F., Frignani, M., Guerzoni, S., & Ravaioli, M. (1988). Sediments and pollu‐
tion in the Northern Adriatic Sea. Annals of the New York Academy Sciences, 534,
1000-1020.

[3] Giordani, P., Hammond, D. E., Berelson, W. M., Montanari, G., Poletti, R., Milandri,
A., Frignani, M., Langone, L., Ravailoi, M., Rovatti, G., & Rabbi, E. (1992). Benthic
fluxes and nutrient budgets for sediments in the Northern Adriatic Sea: Burial and
recycling efficiencies. Science of the Total Environment, Supplement, 251-275.

[4] Wolanski, E. (2007). Estuarine Ecohydrology, Elsevier, Amsterdam, 157.

[5] Warren, R., & Johnsen, J. (1993, June 14-16, 1993). Cohesive sediment modelling for
coastal lagoons. Kuala Lumper, Malaysia. paper presented at International Colloquium
and Exposition on Computer Applications in Coastal and Offshore Engineering (ICE-CA
COE’93).

[6] Vichi, M., Pinardi, N., Zavatarelli, M., Matteucci, G., Marcaccio, M., Bergamini, M. C.,
& Frascari, F. (1998). One-dimensional ecosystem model tests in the Po Prodelta area
(Northern Adriatic Sea). Environmental Modelling & Software, 13, 471-481.

[7] Wolanski, E., Andutta, F. P., & Delhez, E. (2012). Estuarine hydrology. In: Encyclope‐
dia of Lakes and Reservoirs. Lars Bengtsson, Reginald Herschy, Rhodes Fairbridge (eds.),
Springer, In press.

[8] Maggi, F. (2005). Flocculation dynamics of cohesive sediment. PhD-Thesis, Delft Uni‐
versity of Technology, Delft, 136.

[9] Kourafalou, V. H. (1999). Process studies on the Po River plume, North Adriatic Sea.
Journal of Geophysical Research, 104, 29963-29985.

[10] Wang, X. H. (2005). A numerical study of sediment transport in a coastal embayment

during a winter storm. Ocean Modelling, 10, 253-271.

[11] Milliman, J. D., & Farnsworth, K. L. (2011). River Discharge to the Coastal Ocean: A
Global Synthesis, Cambridge University Press, 392.

[12] Vilibić, I. (2003). An analysis of dense water production on the North Adriatic shelf.
Estuarine, Coastal and Shelf Science, 56, 697-707.

[13] Malanotte-Rizzoli, P., & Bergamasco, A. (1983). The dynamics of the coastal regions
of the Northern Adriatic Sea. Journal of Physical Oceanography, 13, 1105-1130.
28 Sediment Transport Processes and Their Modelling Applications

[14] Artegiani, A., Bregant, D., Paschini, E., Pinardi, N., Raicich, F., & Russo, A. (1997a).
The Adriatic Sea general circulation. Part I: air-sea interactions and water mass struc‐
ture. Journal of Physical Oceanography, 27, 1492-1514.
[15] Artegiani, A., Bregant, D., Paschini, E., Pinardi, N., Raicich, F., & Russo, A. (1997b).
The Adriatic Sea general circulation. Part II: Baroclinic circulation structure. Journal of
Physical Oceanography, 27, 1515-1532.
[16] Zavatarelli, Z., Pinardi, N., Kourafalou, V. H., & Maggiore, A. (2002). Diagnostic and
prognostic model studies of the Adriatic Sea general circulation: Seasonal variability.
Journal of Geophysical Research, 107, 3004, doi:10.1029/2000JC000210.
[17] Zavatarelli, M., & Pinardi, N. (2003). The Adriatic Sea modelling system: a nested ap‐
proach. Annales Geophysicae, 21, 345-364.
[18] Poulain, P. M. (2001). Adriatic Sea surface circulation as derived from drifter data be‐
tween 1990 and 1999. Journal of Marine System, 29, 3-32.
[19] Raicich, F. (1996). On the fresh water balance of the Adriatic Sea. Journal of Marine
System, 9, 305-319.
[20] Haley, J. P. J., & Lermusiaux, P. F. J. (2010). Multiscale two-way embedding schemes
for free-surface primitive equations in the “Multidisciplinary simulation, estimation
and assimilation system”. Ocean Dynamics, 60, 1497-1537.
[21] Ramp, S. R., Lermusiaux, P. F. J., Shulman, I., Chao, Y., Wolf, R. E., & Bahr, F. L.
(2011). Oceanographic and atmospheric conditions on the continental shelf north of
the Monterey Bay during August 2006. Dynamics of Atmospheres and Oceans, 52,
192-223.
[22] Andutta, F. P., Ridd, P. V., & Wolanski, E. (submitted). Age and the flushing time in the
Great Barrier Reef coastal waters.
[23] Nunes, Vaz. R. A., Lennon, G. W., & Bowers, D. G. (1990). Physical behaviour of a
large, negative or inverse estuary. Continental Shelf Research, 10, 277-304.
[24] Diop, E. S., Soumare, A., Diallo, N., & Guisse, A. (1997). Recent change of mangroves
of the Saloum river estuary, Senegal. Mangroves and Salt Marshes, 1, 163-172.
[25] Largier, J. L., Hollibaugh, J. T., & Smith, S. V. (1997). Seasonally hypersaline estuaries
in Mediterranean-climate regions. Estuarine, Coastal and Shelf Science, 45, 789-797.
[26] Lavin, M. F., Godinez, V. M., & Alvarez, L. G. (1998). Inverse-estuarine features of
the upper gulf of California. Estuarine, Coastal and Shelf Science, 47, 769-795.
[27] Webster, I. T. (2010). The hydrodynamics and salinity regime of a coastal lagoon- The
Coorong Australia- Seasonal to multi-decadal timescales. Estuarine Coastal and Shelf
Science, 90, 264-274.
[28] Andutta, F. P., Ridd, P. V., & Wolanski, E. (2011). Dynamics of hypersaline coastal
waters in the Great Barrier Reef. Estuarine, Coastal and Shelf Sciences, 94, 299-305.
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 29
[Link]

[29] Winant, C. D., & de Velasco, G. G. (2012). Dynamics of hypersaline estuaries: Laguna
San Ignacio, Mexico. Chapter 12 in, 2, In: Uncles R J, Monismith S G (Ed.), Water and
Fine Sediment Circulation. In: Wolanski E, McLusky D (Series Ed.), The Treatise on estuar‐
ine and coastal Science, Elsevier.

[30] Nunes, Vaz. R. A. (2012). The salinity response of an inverse estuary to climate
change & desalination. Estuarine, Coastal and Shelf Sciences, 98, 49-59.

[31] Brambati, A., Bregant, D., Lenardon, G., & Stolfa, D. (1973). Transport and sedimen‐
tation in the Adriatic Sea. Museo Friulano di Storia Nat., Udine, Italy, Publ. 20, 60.

[32] Byun, D. S., Wang, X. H., & Holloway, P. E. (2004). Tidal characteristic adjustment
due to dyke and seawall construction in the Mokpo Coastal Zone, Korea. Estuarine,
Coastal and Shelf Science, 59, 185-196.

[33] Byun, D. S., Wang, X. H., Zavantarelli, M., & Cho, Y. K. (2007). Effects of resuspend‐
ed sediments and vertical mixing on phytoplankton spring bloom dynamics in a ti‐
dal estuarine embayment. Journal of Marine Systems, 67, 102-118.

[34] Defant, A. (1960). Physical Oceanography, Pergamon Press, New York, 2, 598.

[35] Kang, J. W. (1999). Changes in tidal characteristics as a result of the construction of

sea-dike/sea-walls in Mokpo coastal zone in Korea. Estuarine, Coastal and Shelf Science,
48, 429-438.

[36] Kang, J. W., & Jun, K. S. (2003). Flood and ebb dominance in estuaries in Korea. Es‐
tuarine, Coastal and Shelf Science, 56, 187-196.

[37] Kang, J. W., Moon, S. R., & Lee, K. H. (2009). Analyzing sea level rise and tide charac‐
teristics change driven by coastal construction at Mokpo Coastal Zone in Korea.
Ocean Engineering, 36, 415.

[38] Speer, P. E., & Aubrey, D. G. (1985). A study of nonlinear tidal propagation in shal‐
low inlet/estuarine systems. Part II: theory. Estuarine, Coastal and Shelf Science, 21,
207-224.

[39] Lee, S. W. (1994). Change in tidal height at Mokpo Harbour due to the construction
of Youngsan River Estuary Dyke. Korean harbour hydraulics, 18, 27e37, in Korean.

[40] Jeong, M. S., Jeong, D. D., Shin, S. H., & Lee, J. W. (1997). Tidal changes in the harbor
due to the development of Mokpo Coastal Zone. Journal of Korean Harbour, 11, 1e8, in
Korean.

[41] Milliman, J. D., Shen, H., Yang, Z., & Meades, R. H. (1985). Transport and deposition
of river sediment in the Changjiang estuary and adjacent continental shelf. Continen‐
tal Shelf Research, 4(1/2), 37-45.

[42] Milliman, J. D., & Syvitski, J. P. M. (1992). Geomorphic/tectonic control of sediment

discharge to the ocean: the importance of small mountainous rivers. Journal of Geolo‐
gy, 100(5), 525-544.
30 Sediment Transport Processes and Their Modelling Applications

[43] Cao, Z., Wang, X. H., Guan, W., Hamilton, L. J., Chen, Q., & Zhu, D. (submitted). Ob‐
servations of nepheloid layers in the Yangtze estuary, China. Marine Technology Soci‐
ety Journal.

[44] Song, D., & Wang, X. H. (inpreparation). Observation and Modelling Study of the Tur‐
bidity Maximum in the Deepwater Navigation Channel in the Yangtze River Estuary.

[45] He, Q., Li, J., Li, Y., Jin, X., & , Y. (2001). Field measurements of bottom boundary lay‐
er processes and sediment resuspension in the Changjiang estuary. Science in China
(Series B), 44(Supp.), 80-86.

[46] Shi, J. Z., Zhang, S. Y., & Hamilton, L. J. (2006). Bottom fine sediment boundary layer
and transport processes at the mouth of the Changjiang Estuary, China. Journal of Hy‐
drology, 327(1/2), 276-288.

[47] Liu, G., Zhu, J., Wang, Y., Wu, H., & Wu, J. (2011). Tripod measured residual currents
and sediment flux Impacts on the silting of the Deepwater Navigation Channel in the
Changjiang Estuary. Estuarine, Coastal and Shelf Science, 93(3), 192-201.

[48] Li, J., & Zhang, C. (1998). Sediment resuspension and implications for turbidity maxi‐
mum in the Changjiang Estuary. Marine Geology, 148(3-4), 117-124.

[49] Shi, Z., & Kirby, R. (2003). Observations of fine suspended sediment processes in the
turbidity maximum at the North Passage of the Chanjiang estuary, China. Journal of
Coastal Research, 19(3), 529-540.

[50] Shi, Z. (2004). Behaviour of fine suspended sediment at the North passage of the
Changjiang Estuary, China. Journal of Hydrology, 293(1-4), 180-190.

[51] Shen, H., & Pan, D. (2001). Turbidity maximum in the Changjiang Estuary, China Ocean
Press, Beijing, in Chinese with English introduction, 194.

[52] Yang, Z., Milliman, J. D., & Fitzgerald, M. G. (1982). Transfer of Water and Sediment
from the Yangtze River to the East China Sea, June 1980. Canadian Journal of Fisheries
and Aquatic Sciences, 40(S1), 72-82.

[53] Milliman, J. D., Hsueh, Y., Hu, D., Pashinski, D. J., Shen, H., Yang, Z., & Hacker, P.
(1984). Tidal phase control of sediment discharge from the Yangtze River. Estuarine,
Coastal and Shelf Science, 19(1), 119-128.

[54] Beardsley, R. C., Limeburner, R., Yu, H., & Cannon, G. A. (1985). Discharge of the
Changjiang (Yangtze River) into the East China Sea. Continental Shelf Research, 4(1-2),
57-76.

[55] Su, J., & Wang, K. (1986). The suspended sediment balance in Changjiang Estuary.
Estuarine Coastal and Shelf Science , 23(1), 81-98.

[56] Shen, H., Li, J., Zhu, H., Han, M., & Zhou, F. (1993). Transport of the suspended sedi‐
ment in the Changjiang Estuary. International Journal of Sediment Research, 7, 45-63.
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 31
[Link]

[57] Shen, J., Shen, H., Pan, D., & Xiao, C. (1995). Analysis of transport mechanism of wa‐
ter and suspended sediment in the turbidity maximum of the Changjiang Estuary.
Acta Geographica Sinica, 50(5), 411-420, in Chinese with English abstract.
[58] Hamilton, L. J., Shi, Z., & Zhang, S. Y. (1998). Acoustic backscatter measurements of
estuarine suspended cohesive sediment concentration profiles. Journal of Coastal Re‐
search, 14(4), 1213-1224.
[59] Gao, J., Yang, Y., Wang, Y., Pan, S., & Zhang, R. (2008). Sediment dynamics of turbid‐
ity maximum in Changjiang River mouth in dry season. Frontiers of Earth Science in
China, 2(3), 249-261.
[60] Chen, X., & Zong, Y. (1998). Coastal erosion along the Changjiang deltaic shoreline,
China: History and prospective. Estuarine Coastal and Shelf Science, 46(5), 733-742.
[61] Yang, Z., Wang, H., Saito, Y., Milliman, J. D., Xu, K., Qiao, S., & Shi, G. (2006). Dam
impacts on the Changjiang (Yangtze) River sediment discharge to the sea: The past
55 years and after the Three Gorges Dam. Water Resource Research, 42, W04407, doi:
10.1029/2005WR003970.
[62] Wu, H., & Guo, W. (2004). The physical model study on suspend sediment back silt‐
ing in navigation channel of Yangtze River Estuary. Shanghai Estuarine and Coastal
Science Research Centre, in Chinese.
[63] Tan, Z., Fan, Q., Zheng, W., & Zhu, J. (2011). Analysis of reasons for the siltation in
North Passage of Yangtze Estuary. Port & Waterway Engineering [1], 29-40, in Chinese
with English abstract.
[64] He, S., & Sun, J. (1996). Characteristics of suspended sediment transport in the tur‐
bidity maximum of the Changjiang River Estuary. Oceanologia et Limnologia Sinica,
27(1), 60-66, in Chinese with English abstract.
[65] Chen, J., Li, D., Chen, B., Hu, F., Zhu, H., & Liu, C. (1999). The processes of dynamic
sedimentation in the Changjiang Estuary. Journal of Sea Research, 41(1-2), 129-140.
[66] Wu, J., Wang, Y., & Cheng, H. (2009). Bedforms and bed material transport pathways
in the Changjiang (Yangtze) Estuary. Geomorphology, 175-184.
[67] Ge, J., Ding, P., & Chen, C. (2010). Impacts of Deep Waterway Project on local circula‐
tions and salinity in the Changjiang Estuary, China. Shanghai, China. paper presented
at the 32nd International Conference on Coastal Engineering.
[68] Shi, J. Z. (2010). Tidal resuspension and transport processes of fine sediment within
the river plume in the partially-mixed Changjiang River estuary, China. A personal
perspective Geomorphology, 121(3-4), 133-151.
[69] Shi, J. Z., Zhou, H. Q., Liu, H., & Zhang, Y. G. (2010). Two-dimensional horizontal
modeling of fine-sediment transport at the South Channel-North Passage of the par‐
tially mixed Changjiang River estuary. China Environment Earth Sciences, 61(8),
1691-1702.
32 Sediment Transport Processes and Their Modelling Applications

[70] Wilson, D., Padovan, A., & Townsend, S. (2004). The Water Quality of Spring and
Neap Tidal Cycles in the Middle Arm of Darwin Harbour during the Dry Season.
Water Monitoring Branch, Natural Resource Management Division, Department of Infra‐
structure, Planning and Environment, Darwin.

[71] Brocklehurst, P., & Edmeades, B. (1996). The Mangrove Communities of Darwin Har‐
bour. Tech Memo. [96].

[72] Tien, A. T. (2006). Influence of deep aquifer springs on dry season stream water qual‐
ity in Darwin Rural Area. Water Monitoring Branch Natural Resource Management Divi‐
sion Report, 6/2006D, 54.

[73] Michie, M. G. (1987). Distribution of foraminifera in a macrotidal tropical estuary:

Port Darwin, Northern Territory of Australia. Australian Journal of Marine and Fresh‐
water Research, 38, 249-259.

[74] Woodroffe, C. D., Bardsley, K. N., Ward, P. J., & Hanley, J. R. (1988). Production of
mangrove litter in a macrotidal embayment, Darwin Harbour, N.T., Australia. Es‐
tuarine, Coastal and Shelf Science, 26(6), 581-598.

[75] Li, L., Wang, X. H., Sidhu, H., & Williams, D. (2011). Modelling of three-dimensional
tidal dynamics in Darwin Harbour, Australia. in Proceedings of the 15th Biennial Com‐
putational Techniques and Applications Conference (CTAC-2010), 52, C103-C123.

[76] Li, L., Wang, X. H., Williams, D., Sidhu, H., & Song, D. (2012). Numerical study of
the effects of mangrove areas and tidal flats on tides: A case of study of Darwin Har‐
bour, Australia. Journal of Geophysical Research, In press.

[77] Padovan, A. V. (1997, December). The water quality of Darwin Harbour: October
1990 - November 1991. Water Quality Branch, Water Resources Division, Department of
Lands, Planning and Environment, NT Government, Report [34/1997D].

[78] Williams, D., Wolanski, E., & Spagnol, S. (2006). Hydrodynamics of Darwin Harbour.
In: Wolanski E, editor. The Environment in Asia Pacific Harbours, Springer, Printed in
The Netherlands, 461-476.

[79] Williams, D. (2009). Part 1: hydrodynamics and sediment transport. In Dredging of

sand from Darwin Harbour, hydrographic and marine life, Australian Institute of Marine
Science, Arafura Timor Research Facility, Brinkin, Northern Territory, 1-33.

[80] Ribbe, J., & Holloway, P. E. (2001). A model of suspended sediment transport by in‐
ternal tides. Continental Shelf Research, 21, 395-422.

[81] Wang, X. H. (2001). A numerical study of sediment transport in a coastal embayment

during a winter storm. Journal of Coastal Research, 34, 414-427.

[82] Wang, X. H. (2002). Tide-induced sediment resuspension and the bottom boundary
layer in an idealized estuary with a muddy bed. Journal of Physical Oceanography, 32,
3113-3131.
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 33
[Link]

[83] Wang, X. H., Pinardi, N., & Malacic, V. (2007). Sediment transport and resuspension
due to combined motion of wave and current in the northern Adriatic Sea during a
Bora event in January 2001: A numerical modelling study. Continental Shelf Research,
27, 613-633.
[84] Blumberg, A. F., & Mellor, G. L. (1987). A description of a three-dimensional coastal
ocean circulation model. in Three dimensional Coastal Ocean Models, edited by N. S.
Heaps, American Geophysical Union, Washington D. C., 1-16.
[85] Smagorinsky, J. (1963). General circulation experiments with the primitive equations,
I: the basic experiment. Monthly Weather Review, 91, 99-164.
[86] Nittrouer, C., Miserocchi, S., & Trincardi, F. (2004). The PASTA project: investigation
of Po and Apenine sediment transport and accumulation. Oceanography, 17(4), 46-57.
[87] Adams, C. E., Jr , , & Weatherly, G. L. (1981). Some effects of suspended sediment
stratification on an oceanic bottom boundary layer. Journal of Geophysical Research, 86,
4161-4172.
[88] Trowbridge, J. H., & Kineke, G. C. (1994). Structure and dynamics of fluid muds over
the Amazon continental shelf. Journal of Geophysical Research, 99, 865-874.
[89] Fohrmann, H., Backhaus, J. O., Laume, F., & Rumohr, J. (1998). Sediments in bottom-
arrested gravity plumes: numerical case studies. Journal of Physical Oceanography, 28,
2250-2274.
[90] Cavaleri, L., & Bertotti, L. (1997). In search of the correct wind and wave fields in a
minor basin. Monthly Weather Review, 125, 1964-1975.
[91] Mellor, G. L., & Yamada, T. (1982). Development of a turbulence closure model for
geophysical fluid problems. Reviews of Geophysics and Space Physics, 20(4), 851-875.
[92] Byun, D. S., & Wang, X. H. (2005). Numerical studies on the dynamics of tide and
sediment transport in the western tip of the southwest coast, Korea. Journal of Geo‐
physical Research, 110, C03011.
[93] Baretta, J. G., Ebenhöh, W., & Ruardij, P. (1995). The European regional seas ecosys‐
tem model, a complex marine ecosystem model. Netherlands Journal of Sea Research,
33, 233-246.
[94] Legendre, L., & Rassoulzadegan, F. (1995). Plankton and nutrient dynamics in ma‐
rine waters. Ophelia, 41, 153-172.
[95] Vichi, M., Oddo, P., Zavatarelli, M., Coluccelli, A., Coppini, G., Celio, M., Fonda,
Umani. S., & Pinardi, N. (2003). Calibration and validation of a one-dimensional
complex marine biogeochemical fluxes model in different areas of the northern Adri‐
atic Sea. Annales Geophysicae, 21(2), 413-437.
[96] Yoon, Y. H. (2000). On the distributional characteristics of phytoplankton community
in Mokpo Coastal Waters, Southwestern Korea during low temperature season. Jour‐
nal of Institute for Basic Sciences, Yosu Natinia University, 2, 71-82, in Korean.
34 Sediment Transport Processes and Their Modelling Applications

[97] Blumberg, A. F., & Mellor, G. L. (1983). Diagnostic and prognostic numerical circula‐
tion studies of the South Atlantic Bight. Journal of Geophysical Research, 88(C8),
4579-4592.
[98] Mellor, G. L. (2004). Users guide for a three-dimensional, primitive equation, numeri‐
cal ocean model. Program in Atmospheric and Oceanic Sciences, Princeton University,
Princeton, NJ, 08544-0710.
[99] Mellor, G. L., & Yamada, T. (1974). A hierarchy of turbulence closure models for
planetary boundary layers. Journal of the Atmospheric Sciences, 31(7), 1791-1806.
[100] Mellor, G. L. (2001). One-dimensional, ocean surface layer modeling: a problem and
a solution. Journal of Physical Oceanography, 31(3), 790-809.
[101] Oey, L. Y. (2005). A wetting and drying scheme for POM. Ocean Modelling, 9(2),
133-150.
[102] Oey, L. Y. (2006). An OGCM with movable land-sea boundaries. Ocean Modelling,
13(2), 176-195.
[103] Chen, C., Liu, H., & Beardsley, R. C. (2003). An Unstructured Grid, Finite-Volume,
Three-Dimensional, Primitive Equations Ocean Model: Application to Coastal Ocean
and Estuaries. J. Atmos. Oceanic Technol., 20, 159-186.
[104] Wright, L. D., Boon, J. D., Xu, J. P., & Kim, S. C. (1992). The bottom boundary layer of
the bay stem plains environment of lower Chesapeake Bay. Estuarine, Coastal and
Shelf Science, 35(1), 17-36, doi: 10.1016/S0272-7714(05)80054-X.
[105] Mehta, A. J. (1988). Laboratory studies on cohesive sediment deposition and erosion.
in Physical processes in estuaries, edited by J. Dronkers and W.V. Leussen, Springer Verlag,
Berlin, 427-445.
[106] Mazda, Y., Magi, M., Kogo, M., & Hong, P. N. (1997). Mangroves as a coastal protec‐
tion from waves in the Tong King delta, Vietnam. Mangroves and salt marshes, 1(2),
127-135, doi: 10.1023/a:1009928003700.
[107] Lee, C. M., Askari, F., Brook, J., Carniel, S., Cushman-Rosin, B., Dorman, C., Doyle, J.,
Flament, P., Harris, C. K., Jones, B. H., Kuzmic, M., Martin, P., Ogston, A., Orlic, M.,
Perkins, H., Poulain, P., Pullen, J., Russo, A., Sherwood, C., Signell, R. P., & Thaler,
Detweiler. D. (2005). Northern Adriatic response to a wintertime Bora wind event.
EOS Transactions, 86(16), 19.
[108] Pullen, J., Doyle, J. D., Hodur, R., Ogston, A., Book, J. W., Perkins, H., & Signell, R.
(2003). Coupled ocean-atmosphere nested modeling of the Adriatic Sea during win‐
ter and spring 2001. Journal of Geophysical Research, 108, C103320.
[109] Signell, R. P., Carniel, S., Cavaleri, L., Chiggiato, J., Doyle, J., Pullen, J., & Sclavo, M.
(2005). Assessment of operational wind models for the Adriatic Sea using wind and
wave observations: a first analysis focused on the Venice lagoon region. Journal of
Marine System, 53, 217-233.
 Sediment Transport Dynamics in Ports, Estuaries and Other Coastal Environments 35
[Link]

[110] Matteucci, G., & Frascari, F. (1997). Fluxes of suspended materials in the north Adri‐
atic Sea (Po Prodelta area). Water, Air and Soil Pollution, 99, 557-572.

[111] Boldrin, A., Langone, L., Miserocchi, S., Turchetto, M., & Acri, F. (2005). Po River
plume on the Adriatic continental shelf: dispersion and sedimentation of dissolved
and suspended matter during different river discharge rates. Marine Geology, 222-223,
135-158.

[112] Traykovski, P., Wiberg, P. L., & Geyer, W. R. (2006). Observations and modeling of
wave-supported sediment gravity flows on the Po prodelta and comparison to prior
observations from the Eel shelf. Continental Shelf Research, 111, C03S17, doi:
10.1029/2005JC003110.

[113] Grant, W. D., & Madsen, O. S. (1979). Combined wave and current interaction with a
rough bottom. Journal of Geophysical Research, 84, 1797-1808.

[114] Huisman, J., Von Oostveen, P., & Weissing, F. J. (1999). Critical depth and critical tur‐
bulence: two different mechanism for the development of phytoplankton blooms.
Limnology and Oceanography, 44, 1781-1787.

[115] Gran, H. H., & Braarud, T. (1935). A quantitative study of the phytoplankton in the
Bay of Fundy and the Gulf of Marine. Journal of the Biological Board of Canada, 1,
279-467.

[116] Sverdrup, H. U. (1953). On conditions for vernal blooming of phytoplankton. Journal

du Conseil- Conseil International Pour L’exploration de La Mer, 18, 287-295.

[117] Cloern, J. E. (1991). Tidal stirring and phytoplankton bloom dynamics in an estuary.
Journal of Marine Research, 49, 203-221.

[118] Wang, X. H., Oddo, P., & Pinardi, N. (2006). On the bottom density plume on coastal
zone off Gargano (Italy) in the southern Adriatic Sea and its inter-annual variability.
Journal of Geophysical Research, 111, C03S17, doi:10.1029/2005JC003110.

[119] Palinkas, C. M., Nittrouer, C. A., Wheatcroft, R. A., & Langone, L. (2005). The use 7Be
to identify event and seasonal sedimentation near the Po River delta, Adriatic Sea.
Marine Geology, 222-223, 95-112.
 Chapter 2

Longshore Sediment Transport Measurements on

Sandy Macrotidal Beaches Compared with Sediment
Transport Formulae

Adrien Cartier, Philippe Larroudé and

Arnaud Héquette

Additional information is available at the end of the chapter

[Link]

1. Introduction

In a context of global climate change, local sea level rise could affect the different coastal
processes as erosion, transport and deposition which are responsible in maintaining the
coastline. The study of sediment transport processes is one of the key for a better under‐
standing of the coastal evolution which is needed for effective design of coastal engineering
or to protect anthropogenic activities and population from marine submersion. One of the
main processes that control coastal evolution is sediment transport. A number of studies
have been focused on this topic, but they were mostly restricted to micro- to mesotidal
beaches [1-3] and field investigations on sandy macrotidal beaches appear to be more limit‐
ed, notably because these environments are less common along the worldwide coastline [4].

Only a few studies have been conducted for quantifying sediment flux on macrotidal beach‐
es [5, 6] where sediment transport results from the complex interactions of tidal currents
with longshore currents generated by obliquely incident breaking waves, this complexity
being further increased by the large variations in water level that induce significant horizon‐
tal translations of the surf zone. Although a number of studies were recently conducted on
the morphodynamics of the barred macrotidal beaches of Northern France [7-12], relatively
little effort has been dedicated to measuring longshore sediment transport on these beaches,
even though it is largely recognized that they are affected by significant longshore transport
that plays a major role in the morphodynamics of the intertidal zone [10, 13]. Apart from
some attempts to make estimates of longshore sediment transport from fluorescent tracers
[10, 14-16] and to infer transport directions using grain-size trend analysis techniques [17,
38 Sediment Transport Processes and Their Modelling Applications

18] no studies were conducted up to now for trying to quantify accurately longshore sand
transport on these sandy macrotidal beaches.

Very recent field experiments conducted on macrotidal beaches of Northern France showed
that, at a very short time scale (minutes), cross shore sediment flux is generally higher than
longshore flux, suggesting that shore-perpendicular sediment transport associated with
wave oscillatory currents probably represents a major factor controlling the cross-shore mi‐
gration of intertidal bars [19]. Further analysis highlighted the strong dependence of long‐
shore sediment transport (LST) on instantaneous hydrodynamic conditions that are
extremely variable from one hydrodynamic zone to the other, notably between the non-
breaking zone of wave shoaling and the surf zone [20]. Although such field experiments can
provide very useful results that contribute to a better understanding of beach morphody‐
namic and sediment transport dynamics, in situ experiments are hard to undertake due to a
series of technical and environmental factors. During the last decades, numerical modeling
of coastal sediment transport and morphodynamics has grown substantially and is now
largely used by the coastal scientific community. As a first step, models have to be calibrated
in order to correspond as close as possible to natural phenomenae. Thus, a major focus of
nearshore research is to relate (measured and predicted) sediment transport rates to mor‐
phological change, with the aim of improving our understanding and modeling capabilities
of beach morphodynamics.

This study is based on previous field investigations conducted on sandy barred macrotidal
beaches of northern France by Cartier and Héquette [19-22] during which longshore sedi‐
ment fluxes were estimated using streamer traps, following the method proposed by Kraus
[23]. Longshore sediment transport rates were compared with several sediment transport
formulae integrated in a numerical model. This numerical model is characterized by a cou‐
pling of three codes consisting in enchained Artemis for swells, Telemac2d for the currents
and Sisyphe for the morphodynamic evolution [24]. The aim of this contribution is to
present the results of the field measurements carried out on these multi-barred sandy beach‐
es and to discuss the abilities of numerical models to predict longshore sediment transport
on these macrotidal environments.

2. Study area

This study has been conducted on three sandy barred beaches of Northern France from No‐
vember 2008 to March 2010. The first field experiment site (Zuydcoote, ZY) is located near
the Belgian border, facing the North Sea; the second site (Wissant Bay, WI) is on the shore of
the Dover Strait, while the third study site (Hardelot, HA) is located on the eastern English
Channel coast (Figure 1).
 Longshore Sediment Transport Measurements on Sandy Macrotidal Beaches Compared with Sediment Transport 39
Formulae
[Link]

Figure 1. Location of the three study sites along the coast of Northern France: A) Zuydcoote; B) Wissant Bay; C) Harde‐
lot Beach.

The study sites consist of 300 to 800 m wide dissipative beaches characterized by extensive
intertidal bar-trough systems (Figure 2). The coasts of Northern France are exposed to rela‐
tively low-energy waves that are refracted by numerous offshore sand banks. Dominant
wave directions are from southwest to west, originating from the English Channel followed
by waves from the northeast to north, generated in the North Sea. Offshore modal signifi‐
cant wave heights are similar for all the study sites and are less than 1.5 m, but may exceed 4
m during storms [25, 26]. The presence of several sand banks on the shoreface and the inner
shelf and the gentle beach slopes that characterize the coasts of Northern France are respon‐
sible for strong wave energy dissipation, resulting in modal significant wave heights lower
than 0.6 m in the intertidal zone [11, 25]. Wave heights can nevertheless reach 2 m on the
foreshore during extreme events [27]. During such high wave energy conditions, substantial
volumes of sediment can be transported on these beaches as revealed by the formation and
migration of large megaripples across the intertidal zone [10].
40 Sediment Transport Processes and Their Modelling Applications

Figure 2. Shore perpendicular beach profiles, panoramic and aerial vertical photographs (©Orthophoto, 2005) of
each study site (See Figure 1 for location).

The study sites are affected by semi-diurnal tides with mean spring tide ranging from about
6 m at Zuydcoote to almost 10 m at Hardelot (Figure 3). This high tidal range is responsi‐
ble for relatively strong tidal currents that flow almost parallel to the shoreline in the coast‐
al zone, at Wissant and Zuydcoote, the ebb is directed westward and the flood is flowing
eastward, while at Hardelot Beach, the ebb and flood are directed southward and north‐
ward respectively. The reversing of tidal currents does not occur at high or low tide, but
typically after a delay of two to three hours. Current measurements conducted in previous
studies revealed that the speeds of flood currents exceed those of the ebb, resulting in a
flood-dominated asymmetry responsible for a net regional sediment transport to the east-
northeast [9, 28].
 Longshore Sediment Transport Measurements on Sandy Macrotidal Beaches Compared with Sediment Transport 41
Formulae
[Link]

Figure 3. General characteristics of each field site. A) Water level variations for mean spring tide. B) Summary of hy‐
drodynamic and morphodynamic characteristics at each field sites (see Figure 1 for location)

3. Field measurements

3.1. Field methodology

In order to determine the ability of numerical models to predict sediment transport and
morphodynamics over sandy beds, several field measurements of sediment transport have
been carried out on three different sandy macrotidal beaches. Although sediment fluxes
can be estimated using acoustic or optical backscatter instruments, this study was based
on direct sediment transport measurements using sediment traps rather than these techni‐
ques. Previous studies highlighted that acoustic or optical backscatter sensors can often
be problematic in the coastal zone due to bubbling in the breaker and surf zones [29] and/
or to the presence of organic matter in the water column and to grain size variability [30].
Thus, streamer traps appeared to be the most adapted and the most accurate method to
measure longshore sediment transport in the shoaling, breaking and surf zone on these
macrotidal beaches.
42 Sediment Transport Processes and Their Modelling Applications

Figure 4. Field methodology. A) In situ measurements of longshore sediment transport during low to moderate wave
energy conditions. B) Schematic representation of streamer trap used during the experiments. C) Locations of sedi‐
ment trap deployment along a shore-perpendicular transect. D) Hydrographic instruments used during the field ex‐
periments. E) Shore-perpendicular profile showing the location of the hydrographic instruments. Codes refer to beach
morphology where UB and LB are upper and lower beach respectively; B and T correspond to bars and troughs. Eleva‐
tions are relative to the French topographic datum (IGN69).

Longshore sediment fluxes were estimated using streamer traps, following the method pro‐
posed by Kraus [23]. The sediment traps consisted of a vertical array of five individual
streamer traps with 63 µm mesh size sieve cloth that collects sand-size particles at different
elevations above the bed. First streamer trap (F5) is located 0.05 m above the bed and the last
one (F1) is at approximately 0.90 m (Figure 4, B). Measurements of LST with the sediment
traps were undertaken during 10 minutes.

Calculations of the sediment flux from sand traps were carried out according to the proce‐
dure of Rosati and Kraus [31]. The sediment flux Q(f), in kg.s-1.m-2, at a streamer trap (f) is
equivalent to:
S( f )
Q( f ) =
w *h *t

Where S(f) is the dry weight of sediment collected in the streamer (f), h is the height of the
streamer opening (0.07 m), w is the streamer width (0.14 m), and t is the sampling period (≈
 Longshore Sediment Transport Measurements on Sandy Macrotidal Beaches Compared with Sediment Transport 43
Formulae
[Link]

10 minutes). The sediment flux between neighboring streamers QE(f) corresponds to the lin‐
ear interpolation between two adjacent traps:
QE ( f ) = 0, 5 * (F ( f ) + F ( f + 1))
The depth integrated flux (Q) in kg.s-1.m-1 is:
N N
Q = h * ∑ Q( f ) + ∑ a( f ) * QE ( f )
i=1 i=1

Where h is the height of the streamer opening in meters, Q(f) is the sediment flux at a
streamer f, a(f) is the distance between neighboring streamers, QE(f) is the sediment flux be‐
tween neighboring streamers and N is the total number of streamers (N = 5).
Measurements of LST were carried out at several locations across the intertidal zone during
rising and falling tides in order to obtain estimates of longshore sediment flux from the low‐
er to the upper beach during flood and ebb (Figure 4, C).Although the sediment traps were
usually deployed in similar water depths, ranging from 0.8 to 1.4 m, sediment transport
measurements took place in various hydrodynamic zones including shoaling, breaker and
surf zones, depending on wave activity during sampling. For safety reasons, sand transport
measurements were conducted only under low to moderate wave energy conditions (Hs
max ≈ 0.7 m).
Two field experiments were conducted on each study site: Zuydcoote in November 2008
(ZY08) and December 2009 (ZY09), Wissant in March 2009 (WI09) and March-April 2010,
(WI10), and Hardelot in June 2009 (HA09) and January-February 2010 (HA10), resulting in
the collection of 172 depth-integrated sediment flux measurements.
Coastal hydrodynamics were measured at different locations across the bar-trough mor‐
phology along cross shore transects using various hydrographic instrument (Figure 4, D).
Waves and currents were measured using three different hydrographic instruments: an
Acoustic Doppler Current Profiler (ADCP), and two electromagnetic wave and current me‐
ters (Midas Valeport©, and InterOcean ADW S4). All instruments operated during 9 mi‐
nutes intervals every 15 minutes at a frequency of 2 Hz, providing almost continuous
records of significant wave height (Hs), wave period and direction, longshore current veloci‐
ty (Vl), and mean current velocity (Vm) and direction. Current velocity was measured at dif‐
ferent elevations above the bed depending on each instrument. The ADW S4 and Valeport
current meters recorded current velocity at 0.4 m and 0.2 m above seabed respectively, while
the ADCP measured current velocity at intervals of 0.2 m through the water column from
0.4 m above the bed to the water surface. Current velocity at 0.2 m above the bed was esti‐
mated using the ADCP data by applying a logarithmic regression curve to the measured ve‐
locities obtained at different elevations in the water column.
Beach morphology plays a major role in the variation of sediment transport rates, especially
on bar-trough topography [9, 10, 13, 32, 33]. Thus, during each field experiment, beach mor‐
phology was surveyed using a very high resolution Differential Global Positioning System
(DGPS) with horizontal and vertical error margins of ± 2 cm and ± 4 cm respectively. A 300
m wide zone of the beach was systematically surveyed on each study site whereas the cross-
shore extent of the surveyed area was variable depending on tidal range.
44 Sediment Transport Processes and Their Modelling Applications

3.2. Sediment transport: role of the main physical forcings

3.2.1. Hydrodynamic conditions during the field experiments

The field experiments have been conducted under different conditions of wave energy and
tidal range at each study site. Because sand transport measurements were restricted to mod‐
erate to low wave energy conditions, the range of wave heights recorded during sediment
trapping is relatively constrained. A classification of wave energy conditions during the
field experiments was adopted in which Hs< 0,2 m represents low wave energy conditions,
0,2 ≤ Hs< 0,4 m refers to moderate wave energy conditions and Hs ≥ 0,4 m represents higher
energetic conditions (Figure 5).

Figure 5. Examples of hydrodynamic conditions during some field experiments. A) Time series of significant wave
height (m) and wave direction (°) during WI10 and HA09 experiments. B) Photographs of low and high wave energy
conditions during the experiments.

The lowest wave energy conditions took place during the HA09 experiment. 99% of the
significant wave heights were under 0.4 m and 46% were under 0.2 m for a mean longshore
current velocity of 0.2 m.s-1. High wave energy conditions occurred during several field
experiments with a maximum wave height of 2.4 m reached at Wissant in 2010 (WI10) while
longshore current velocity reached 2 m.s-1. Such high energy conditions lasted over only two
tidal cycles however during this field experiment (Figure 5). In comparison, the maximum
significant wave height during the HA10 experiment was 2.1 m, but the duration of high energy
conditions was considerably longer as 80% of the recorded wave heights was higher than 0.4
 Longshore Sediment Transport Measurements on Sandy Macrotidal Beaches Compared with Sediment Transport 45
Formulae
[Link]

m for an average Hs of 0.70 m, which is remarkably high for the relatively low-energy coasts
of Northern France where wave heights are generally lower in the intertidal zone [11, 13].

3.2.2. Longshore sediment transport rates

During the six field experiments, more than 700 sediment samples were collected, which
were used to compute 172 depth-integrated sediment fluxes. Among these, 79 depth-inte‐
grated sediment fluxes were obtained close to a hydrographic instrument, allowing a com‐
parison between LST and hydrodynamic parameters. Throughout the 6 field experiments,
most of the longshore sediment transport rates (> 50%) ranged from 1 x 10-4 kg.s-1.m-1 to 1 x
10-3 kg.s-1.m-1 (Figure 6A). Sediment transport rates show a high variability depending on the
study site and during each field experiment due to variations in hydrodynamic conditions
(Figure 6B). Longshore sediment flux reached values up to 2.1 x 10-1 kg.s-1.m-1 during the
ZY09 experiment, which was the most energetic event during which sediment sampling
took place. Lower rates of sediment transport were measured in the vicinity of current me‐
ters where sediment flux nevertheless reached approximately 1.6 x 10-1 kg.s-1.m-1 for a mean
flow velocity of 0.5 m.s-1. Significantly higher transport rates were observed during the most
energetic conditions, however, notably during the HA10, ZY09 and WI09 field experiments
(Figure 6B).

Figure 6. Range of LST for (A) all field experiments and (B) each field experiment.

Longshore sediment transport rates measured during this study are in the same order of
magnitude as other studies conducted on microtidal beaches [23, 31, 34-37] as well as on
macrotidal beaches [5] (Figure 7). The fact that ranges of values are similar whatever the ti‐
46 Sediment Transport Processes and Their Modelling Applications

dal conditions are, suggests that tidally-induced currents are not the main forcing and do
not act significantly on longshore sediment transport magnitude.

Figure 7. Maximum and minimum values of LST measured using streamer traps on microtidal beaches[35, 38] and
macrotidal beaches [5]. Numbers of samples are located just above the bar charts.

3.2.3. Relationship between longshore sediment transport and hydrodynamics

Comparisons of transport rates with hydrodynamic data showed that longshore sediment
transport increases with both significant wave height and mean current velocity (Figure 8).
Low sediment transport (< 1.0 x 10-4 kg.s-1.m-1) is mainly associated with small wave heights
(< 0.3 m), but only a small increase in wave height appears to induce significantly larger
sediment transport. However, high sediment transport values can also be associated with
low wave conditions, such as during the WI09 experiment when a sediment transport rate of
2.4 x 10-2 kg.s-1.m-1 was measured with a significant wave height of about 0.2 m for example.
The variability in sediment flux values obtained during conditions of equivalent wave
heights, and the fact that similar transport rates may be associated with different wave
heights, even on the same beach and during the same field experiment (e.g., WI09), suggest
that waves do not represent the only factor controlling sediment transport.

Least-square regression analyses show that longshore sediment flux is better correlated with
current velocity than with wave height as revealed by higher determination coefficients
(Figure 8). Similarly to what was observed with significant wave heights, high sediment
transport rates can also be observed with lower current velocities. Further analysis detailed
in Cartier and Héquette [20] highlighted that variations in LST are better explained by these
two forcing parameters in the surf zone, where currents generated by obliquely incident
breaking waves are acting, than in the shoaling zone.

Longshore sediment transport in the nearshore zone is commonly related to the longshore
wave energy flux (Pl) evaluated at the breaker zone, sand transport being expressed as an
immersed-weight transport rate (Il) and related to Pl [39]. Conversely to what was observed
in other studies [40], however, previous analyses of our data showed no relationship be‐
tween LST and wave breaking angle, which is directly involved in the computation of Pl [19,
 Longshore Sediment Transport Measurements on Sandy Macrotidal Beaches Compared with Sediment Transport 47
Formulae
[Link]

20, 22]. These results may be explained by the influence of tidal currents that interact with
wave-induced longshore currents on these macrotidal beaches

Figure 8. Relationship between LST and significant wave height (Hs), and with mean current velocity (Vm) at 0.2 m (tri‐
angle) and 0.4 m (circles) above the bed using all the samples collected near a current meter during all field experiments.

4. Sediment transport modelling

4.1. Methodology

The methodology used during field experiments did not allow us to compare sand transport
rates with well known formulae such as the CERC formula [41] or the Kamphuis’s formula
[42] for example. These formulations, which are essentially based on a wave energy flux ap‐
proach, provide estimates of total sediment transport rates across the entire surf zone. A
large amount of studies have compared that kind of numerical model with in situ measure‐
ments [29, 43-47], or with laboratory measurements [48-50], and even between several nu‐
merical models in order to understand their behaviour with virtual data [51-53]. However,
our measurements can not be compared with these numerical models because sand trap‐
ping took place at only one location in the surf or shoaling zone while these formulae esti‐
mate the total longshore sediment transport across the surf zone.

There is, nevertheless, a number of sediment transport formulae that can be compared with
localized measurements of sand flux. In the present study, calculations of sediment trans‐
port rates have been realized using a coupling of three codes [54]. The sedimentary evolu‐
tion is modeled under the action of the oblique incident waves and is coupled with different
numerical tools dedicated to the other process involved in the nearshore zone. We can men‐
tion the following modules:

The wave module takes into account the surge energy dissipation (hyperbolic equation of
extended Berkhoff). The Artemis code (Agitation and Refraction with Telemac2d on a Mild
Slope) solves the Berkhoff equation taken from Navier-Stokes equations with some other hy‐
potheses (small wave steepness of the surface wave, small slope...). The main results are, for
48 Sediment Transport Processes and Their Modelling Applications

every node of the mesh, the height, the phase and the incidence of the waves. Artemis can
take into account the reflection and the refraction of waves on an obstacle, the bottom fric‐
tion and the breakers. One of the difficulties with Artemis is that a fine mesh must be used
to have good results whereas Telemac2d does not need such a fine mesh.

The hydrodynamic module calculates currents induced by means of the surge of the waves,
from the concept of radiation constraints obtained according to the module of waves. Tele‐
mac2d is designed to simulate the free surface flow of water in coastal areas or in rivers.
This code solves Barré Saint-Venant equations taken from Navier-Stokes equations vertical‐
ly averaged. Then, the main results are, for every node of the mesh, the water depth and the
velocity averaged over the water column. Telemac2d is able to represent the following phys‐
ical phenomena: propagation of long periodic waves, including non-linear effects, wetting
and drying of intertidal zone, bed friction, turbulence…

The sedimentary module integrates the combined actions of the waves and the wave cur‐
rents (2D or 3D) on the transport of sediment [24].

The Sisyphe code solves the bottom evolution equation which expresses the mass conserva‐
tion by directly using a current field result file given by Telemac2d (Figure 9). Several of the
most currently used empirical or semi-empirical formulas are already integrated in Sisyphe.

Figure 9. Diagram of the model ATS (Artemis-Telemac-Sisyphe) used in our simulations, showing the principle of ex‐
ternal coupling to make a loop over one hydro-meteorological event time step (between t1 and t2).

A hydrodynamic simplified model (called Multi1DH) uses the following assumptions: a

random wave approach and a 1DH (cross-shore) direction. An offshore wave model (shoal‐
ing + bottom friction + wave asymmetry) is used with the break point estimation. The waves
in the surf zone are modeled with the classic model of Svendsen (1984) with an undertow
model (roller effect) [55, 56]. The longshore current model is the Longuet-Higgins’s model
[57]. The model is included in the Sysiphe code to calculate the sea bed evolution with sever‐
al sediment transport formulas.
 Longshore Sediment Transport Measurements on Sandy Macrotidal Beaches Compared with Sediment Transport 49
Formulae
[Link]

It is generally accepted that the estimate is acceptable when the flux is between 0.5 and 2
times the in situ measurement [58]. Thus, in the following figures, three lines symbolize the
extent of data that are significant at 0.5 Qsm<Qsc<2 Qsm. The standard error (Srms) was also cal‐
culated using the following equation to characterize the dispersion of data, where the higher
the values, the higher the data are scattered.
∑iN=1 Log (QSc ) − Log (QSm) ²
Srms = N −2

where Qsc is the flux calculated, Qsm the measured flux and N the number value.

4.2. Results

4.2.1. Potential effects of water depth

Sediment flux measurements were performed in a water depth between approximately 1 m

and 1.5 m, but because of the excursion of the tide, these measurements were made at sever‐
al locations on the foreshore (Figure 4). Calculation of sediment load at the same location on
the field requires a lot of computer manipulations. So initially, the sediment loads are mod‐
elled in the middle of the digital domain. The water level in the middle of the simulated
field rarely matches the exact water level measured during sampling. A test was therefore
carried out to assess the potential effects of water depth on computed sediment flux using
the data measured during the two field experiments at Hardelot beach (2009 and 2010). For
the numerical simulations, bathymetry has been considered as stable during and between
the two field campaigns.

The following graphs show the control of the water level on the accuracy of modeled flux
(Figure 10). The example shows calculations based on the expression of Bijker [59], similar
observations having been observed for the other formulations.

The results show that sediment fluxes tend to be better estimated with decreasing water
depth, which is consistent with the depths at which measurements were made. The correla‐
tions are even better for water depths just below the water level of the in situ measurement
(<1 m). In a water depth between 2.70 m and 1.74 m, the computed values seem to line up
around 1x10-3kg.s-1.m-1. There is very little change in calculated flux in comparison with
those measured in situ. As soon as the water level is similar to that of measurement at the
time of trapping (1.51 m), the distribution of points tends to align the right Qsc = Qsm. How‐
ever, when the water column is much lower than reality, many errors appear in the calcula‐
tion and the estimation of sediment flux becomes completely erroneous. In fact, the RMS
errors are quite high when the water level is far from the actual water depth (Figure 10). It
should be emphasized, however with all the initial approximations, the mere fact of posi‐
tioning the water level at the same level as during the sampling resulted in relatively better
results. In particular, when h = 0.91 m, which resulted in a percentage of acceptable values of
32%. Overall calculated sediment fluxes are nevertheless generally overestimated.
50 Sediment Transport Processes and Their Modelling Applications

Figure 10. Comparison between measured (Qsm) and computed (Qsc) transport rates using Bijker's formula [59] for dif‐
ferent water depths (h).

The water level acts directly on the current profile and associated sediment transport mech‐
anisms. Thus, when the water column is higher than that at the time of measurement, the
transport in suspension is favoured in the modelling, leading to an overestimation of the in‐
tegrated flux in the water column. In contrast, a shallower water column could lead to high‐
er sediment fluxes due to increased bed shear. However, it appears that the formulas are
struggling to express sediment transport in very shallow water. Despite very simplified ini‐
tial conditions, the results obtained in this study show that the coupling of these three codes
is not so far from reality when we take into account the water level measured in situ. More
precise calibrations will likely improve the accuracy of the results in the future.

4.2.2. Calculated sediment flux against in situ data

Using the water levels measured during the field experiments, the calculations of sediment
transport were calibrated for each case to be as close as possible to the in situ measurement.
Sediment transport modeling was carried out on the beaches of Wissant and Hardelot, and
sediment loads calculated by several formulas have been extracted and compared to field
data. The calculations concerning the Zuydcoote field site are not presented because of prob‐
lems related to the computational domain. Because the bathymetry input was not broad
enough, the waves spread by the code Artemis underwent many artifacts and measurement
errors that did not allow reliable sediment transport calculations.
 Longshore Sediment Transport Measurements on Sandy Macrotidal Beaches Compared with Sediment Transport 51
Formulae
[Link]

Sediment fluxes were compared with the expressions of Bijker [59], noted BI68, and a cou‐
pling of Van Rijn formula [60] with an expression modelling the transport in suspension
[61], this coupling being called the Soulsby-Van Rijn formula noted SVR97 (Figure 11). The
reader is referred to [58, 62]for details on the different formulae.

Figure 11. Comparisons between in situ (Qsm) sand transport rates measured during HA09, HA10, WI09 and WI10,
with calculated sediment fluxes (Qsc) following Bijker and Soulsby – Van Rijn formulae.

The results were analyzed according to the study site and the mathematical expression used.
The results are better on the site of Hardelot for both formulas, the error Srms being less than
1.0 with a minimum of 0.53 with SVR97. Moreover, the percentage of significant values
reaches 32% and 46% for SVR97 and BI68, respectively. On the site of Wissant, it does not
exceed 30% whatever the formula. RMS errors are associated with a greater dispersion of
52 Sediment Transport Processes and Their Modelling Applications

data readily observable in the graphs (Figure 11). When considering all the data, the formula
for SVR97 is the expression that is most satisfactory with a Srms of only 1.04 and 36% of ac‐
ceptable values. The proportion of sediment flux of low intensities (<1 x 10-3 kg.s-1.m-1) is
higher during the campaigns carried out in Hardelot than those held in Wissant. Converse‐
ly, the flux measurements acquired during the field investigations in the Bay of Wissant
proved to be higher. Morphological changes were also more significant in Wissant than dur‐
ing the campaigns conducted in Hardelot.

5. Discussion

In the present study, measurements of sediment flux were carried out at successive posi‐
tions between the lower and the upper beach during rising or falling tide in order to meas‐
ure sediment transport in comparable water depths. Our results showed that sand transport
was mainly dependent on the mean flow, especially above a velocity threshold of approxi‐
mately 0.4 m.s-1 (Figure 8). Further results have shown that sediment transport was also con‐
trolled by wave action, but correlation analyses between LST and significant wave height
showed a better relationship in the surf zone than in the non-breaking zone [20]. These ob‐
servations highlight the mere fact that sediment transport processes are strongly different
from a hydrodynamic zone to another and it underlines the need to use appropriate formu‐
lae in order to model sediment transport properly. Although a large amount of studies com‐
pared sediment transport calculation against in situ data, few of them have been undertaken
using sediment transport measurements from macrotidal beaches.

Results obtained in a previous study by Camenen and Larroudé [62] showed that the Bijker
formula generally tends to underestimate the sediment transport when there is interaction
of waves and current, which is generally the case in the coastal zone. In this formula, the
swell is considered to be the only mechanism responsible for suspending sediment. There‐
fore, when wave height is low, the simulated sediment transport remains insignificant even
if the average current velocity is high. Although the strength of currents in the intertidal
zone is usually related to the conditions of agitation, it may sometimes be forced by wind or
induced by the combination of tidal currents and those generated by the incident swell [10].
The suspended particles are then provided by currents and waves. Our results show, how‐
ever, that the higher sediment fluxes (> 1 x 10-3 kg.s-1.m-1) are underestimated, which may be
due to low wave height and mean current that are powerful enough to induce substantial
sediment transport, which cannot be modeled by this expression. Conversely, the lowest
sediment transport rates occur when the swell and the mean current are of low intensity.
The direction of tidal currents is also directly involved in the magnitude of the sediment flux
since it can easily reduce or conversely increase wave-induced currents, depending on the
phase of the tidal cycle and the direction of the longshore current. Such types of case are not
considered in the model, however, which may explain a part of the variability observed be‐
tween measured and computed sediment fluxes.
 Longshore Sediment Transport Measurements on Sandy Macrotidal Beaches Compared with Sediment Transport 53
Formulae
[Link]

The formula of Soulsby-Van Rijn comes from the coupling of the Van Rijn formula [60] with
an expression modeling the transport in suspension. It takes into account many physical pa‐
rameters for estimating the bed load and suspended load. Even though the calculations are
more complicated and time consuming, the estimate is generally better, but significant er‐
rors may occur when the wave direction is opposite to that of the current [62].
In this study, the results are particularly satisfactory for the data obtained during the field
experiments at Hardelot (Figure 11) where beach morphology changes were more limited
than at Wissant. When taking into account all values, it clearly appears that the most signifi‐
cant fluxes can nevertheless be largely underestimated since some values may be up to four
times lower than the measured transport rates. During high wave energy conditions, beach
morphology and bed roughness change rapidly due to an increase in bed load sediment
transport. The impact of these bottom changes on the distribution of sediment in the water
column is crucial and plays an important role in the mechanisms of suspended sediment
transport. Because bed roughness is variable, largely depending on the local morphology of
the beach, it is necessary to incorporate changes in beach morphology in the process of cal‐
culation which has not been done yet in this first attempt to model longshore transport on
macrotidal beaches.

6. Conclusion

Despite several simplifications in the modeling procedure, comparisons of longshore sedi‐

ment transport fluxes measured on sandy macrotidal beaches with computed sand fluxes
gave encouraging results. It was shown that water depth is one of the major parameter af‐
fecting modeled sediment transport rates, as calculated sand fluxes were more comparable
with in situ measurements when simulated water depth was similar to the actual water
depth measured in the field. The height of the water column therefore represents a key term
to consider in modeling sediment transport on these beaches. The best modeling results
were obtained with the data collected during low energy conditions at Hardelot beach
where the beach morphology was the most stable. A limitation of the modeling approach
used in this study is related to the fact that beach morphology changes are not taken into
account in the calculations, which should be considered in future modeling studies. Because
sediment traps mainly collect sediments transported in suspension, future investigations us‐
ing this data set will be aimed at de-coupling suspended and bed load transport calculation
in the different sediment transport formulae in order to evaluate only the suspended sedi‐
ment flux, which should result in more accurate comparisons between modeled and meas‐
ured sediment transport rates.

Acknowledgements

This work was supported by French Research National Agency (ANR) through the Vulnera‐
bility Milieu and Climate program (project VULSACO, n° ANR-06-VMC-009) and by the
54 Sediment Transport Processes and Their Modelling Applications

French Centre National de la Recherche Scientifique (CNRS) through the PLAMAR and MI‐
CROLIT projects of the Programme “Relief de la Terre”. Additional funding was provided
by the “Syndicat Mixte de la Côte d’Opale” through a doctoral scholarship to Adrien Carti‐
er. The authors would like to thank Aurélie Maspataud, Vincent Sipka and Antoine Tresca
as well as all the students and permanent staff for their help during the field experiments.

Author details

Adrien Cartier1*, Philippe Larroudé2 and Arnaud Héquette3

*Address all correspondence to: [Link]@[Link]

1 LOG - UMR CNRS 8187, University of Littoral Côte d’Opale,, France

2 LEGI-UMR 5519 UJF, University of Grenoble,, France

3 LOG - UMR CNRS 8187, University of Littoral Côte d’Opale,, France

References

[1] Wang, P., Kraus, N. C., & Davis, R. A. (1998). Total longshore sediment transport rate
in the surf zone-field measurement and empirical predictions. J Coast. Res., 14,
269-282.
[2] Bayram, A., Larson, M., Miller, H. C., & Kraus, N. C. (2001). Cross-shore distribution
of longshore sediment transport: comparison between predictive formulas and field
measurements. Coast. Eng., 44(2), 79-99.
[3] Kumar, V. S., Anand, N. M., Chandramohan, P., & Naik, G. N. (2003). Longshore
sediment transport rate-measurement and estimation, central west coast of India.
Coast. Eng., 48, 95-109.
[4] Davies, A. G. (1980). Geographical variations in coastal development. London, Long‐
man, 212.
[5] Levoy, F., Montfort, O., & Rousset, H. (1994). Quantification of longshore transport in
the surf zone on macrotidal beaches. Fields experiments along the western coast of
Cotentin (Normandy, France). Paper presented at 24 th International Conference on
Coastal Engineering., Kobé, Japon. 2282-2296.
[6] Corbau, C., Ciavola, P., Gonzalez, R., & Ferreira, O. (2002). Measurements of Cross-
Shore Sand fluxes on a Macrotidal Pocket Beach (Saint-Georges Beach, Atlantic
Coast, SW France). J. Coast. Res. [36].
[7] Reichmüth, B., & Anthony, E. J. (2008). Dynamics of intertidal drainage channels on a
multi barred macrotidal beach. Earth Surf. Process. Landf., 33(1), 142 -151 .
 Longshore Sediment Transport Measurements on Sandy Macrotidal Beaches Compared with Sediment Transport 55
Formulae
[Link]

[8] Reichmüth, B., & Anthony, E. J. (2002). The variability of ridge and runnel beach
morphology: examples from Northern France. J. Coast. Res. [36].
[9] Anthony, E. J., Levoy, F., & Montfort, O. (2004). Morphodynamics of intertidal bars
on a megatidal beach, Merlimont, Northern France. Mar. Geol., 208, 73-100.
[10] Sedrati, M., & Anthony, E. J. (2007). Storm-generated morphological change and
longshore sand transport in the intertidal zone of a multi-barred macrotidal beach.
Mar. Geol., 244, 209-229.
[11] Héquette, A., Ruz, M. H., Maspataud, A., & Sipka, V. (2009). Effects of nearshore
sand bank and associated channel on beach hydrodynamics: implications for beach
and shoreline evolution. J. Coast. Res. [56].
[12] Maspataud, A., Ruz, M. H., & Héquette, A. (2009). Spatial variability in post-storm
beach recovery along a macrotidal barred beach, southern North Sea. J. Coast. Res.
[56].
[13] Reichmüt, B., & Anthony, E. J. (2007). Tidal influence on the intertidal bar morpholo‐
gy of two contrasting macrotidal beaches. Geomorphology, 90(1- 2), 101-114.
[14] Levoy, F., Montfort, O., & Larsonneur, C. (1997). Quantification des débits solides sur
les plages macrotidales à l’aide de traceurs fluorescents, application à la côte ouest
du Cotentin. Oceanol. Acta, 20(6), 811-822.
[15] Voulgaris, G., Simmonds, D., Michel, D., Howa, H., Collins, M. B., & Huntley, D.
(1998). Measuring and modelling sediment transport on a macrotidal ridge and run‐
nel beach: an intercomparison. J. Coast. Res., 14(1), 315 .
[16] Stépanian, A., Vlaswinkel, B., Levoy, F., & Larsonneur, C. (2001). Fluorescent tracer
experiment on a macrotidal ridge and runnel beach : a case study at Omaha beach,
North France. Coastal dynamics’01 Lund, Suède, 1017-1027.
[17] Reichmüth, B. (2003). Contribution à la connaissance de la morphodynamique des
plages à barres intertidales: Approche expérimentale, Côte d’Opale, Nord de la
France. Thèse de doctorat Université du Littoral Côte d’Opale, 247.
[18] Héquette, A., Hemdane, Y., & Anthony, E. J. (2008). Determination of sediment trans‐
port paths in macrotidal shoreface environments: a comparison of grain-size trend
analysis with near-bed current measurements. J. Coast. Res., 24, 695-707.
[19] Cartier, A., & Héquette, A. (2011). Estimation of longshore and cross shore sediment
transport on sandy macrotidal beaches of Northern France. Paper presented at Pro‐
ceedings Coastal sediments’11, Miami, Florida, USA. 2130-2143.
[20] Cartier, A., & Héquette, A. (2011). Longshore and cross shore variation in sediment trans‐
port on barred macrotidal beaches, Northern France.
[21] Cartier, A. (2011). Evaluation des flux sédimentaires sur le littoral du Nord Pas-de-
Calais: Vers une meilleure compréhension de la morphodynamique des plages mac‐
rotidales. Thèse de Doctorat, Université du Littoral Côte d’Opale.
[22] Cartier, A., & Héquette, A. (2011). Variation in longshore sediment transport under
low to moderate conditions on barred macrotidal beaches. J. Coast. Res. [64], 45 -49 .
56 Sediment Transport Processes and Their Modelling Applications

[23] Kraus, N. C. (1987). Application of portable traps for obtaining point measurements
of sediment transport rates in the surf zone. J. Coast. Res., 3, 139-152.
[24] Hervouet, J. M. (2007). Hydrodynamics of Free Surface Flows: Modelling With the Fi‐
nite Element Method. John Wiley & Sons, 360, 10.1002/9780470319628.
[25] Clique, P. M., & Lepetit, J. P. (1986). Catalogue sédimentologique des côtes français‐
es, côtes de la Mer du Nord et de la Manche. 133.
[26] Ruz, M. H., Héquette, A., & Maspataud, A. (2009). Identifying forcing conditions re‐
sponsible for foredune erosion on the northern coast of France. J. Coast. Res. [56].
[27] Maspataud, A., Idier, D., Larroudé, Ph., Sabatier, F., Ruz, M. H., Charles, E., Leca‐
cheux, S., & Héquette, A. (2010). L’apport de modèles numériques pour l’étude mor‐
phodynamique d’un système dune-plage macrotidal sous l’effet des tempêtes : plage
de la dune Dewulf, Est de Dunkerque, France. XIèmes. Journées Nationales Génie Civil-
Génie Côtier Les Sables d’Olonne, 353-360.
[28] Héquette, A., Hemdane, Y., & Anthony, E. J. (2008). Sediment transport under wave
and current combined flows on a tide-dominated shoreface, northern coast of France.
Mar. Geol, 249, 226-242.
[29] Esteves, L. S., Lisniowski, M. A., & Wiliams, J. J. (2009). Measuring and modelling
longshore sediment transport. Est. Coast. Shelf Sci., 83, 47-59.
[30] Battisto, G. M., Friedrichs, C. T., Miller, H. C., & Resio, D. T. (1999). Response of OBS
to mixed grain-size suspensions during Sandyduck’97. Coastal sediment 99 Long Is‐
land, New York, 1, 297-312.
[31] Rosati, J. D., & Kraus, N. C. (1989). Development of a portable sand trap for use in
the nearshore Department of the army, U.S. Corps of Engineers. Technical report
CERC [89-91], 181.
[32] Sipka, V., & Anthony, E. J. (1999). Morphology and Hydrodynamics of a macrotidal
ridge and runnel beach under modal low conditions. J. Res. Oceanogr., 24, 25-31.
[33] Cartier, A., Larroudé, Ph, & Héquette, A. (2012). Comparison of sediment transport
models with in-situ sand flux measurements and beach morphodynamic evolution.
ICCE, Santander 2012. Submitted
[34] Kraus, N. C., & Dean, J. L. (1987). Longshore sediment transport rate distributions
measured by trap. Coastal sediment’87 ACE, 881-898.
[35] Kraus, N. C., Gingerish, K. J., & Rosati, J. D. (1989). Duck 85 surf zone sand transport
experiment. US Army Corps of Engineers, Waterways Experiment Station, Coastal Engi‐
neering Research Center, Vicksburg, Mississippi., 48.
[36] Sabatier, F. (2001). Fonctionnement et dynamiques morpho-sédimentaires du littoral
du delta du Rhone. Thèse de Doctorat Université Aix-Marseille III.
[37] Sabatier, F., Stive, M. J. F., & Pons, F. (2004). Longshore variation of depth of closure
on a micro-tidal wave-dominated coast. Paper presented at International Conference
of Coastal Engineering American Society of Civil Engineering, Lisboa. 2329-2339.
[38] Sabatier, F., Samat, O., Chaibi, M., Lambert, A., & Pons, F. (2004, 7-9 septembre 2004).
Transport sédimentaire de la dune à la zone du déferlement sur une plage sableuse
 Longshore Sediment Transport Measurements on Sandy Macrotidal Beaches Compared with Sediment Transport 57
Formulae
[Link]

soumise à des vents de terre. Paper presented at VIIIèmes Journées Nationales Génie
Civil- Génie Côtier Compiègne,. 223-229.
[39] Komar, P. D. (1998). Beach processes and sedimentation. Pearson Education, 544.
[40] Komar, P. D., & Inman, D. L. (1970). Longshore sand transport on beaches. Journal of
Geophysical Research, 75(30), 5514-5527.
[41] Manual, S. P. (1984). Coastal engineering Research Center, in: U.S Army Corps of En‐
gineers, t.E.W., DC 20314 (Ed.).
[42] Kamphuis, J. W., Davies, M. H., Nairn, R. B., & Sayao, O. J. (1986). Calculation of lit‐
toral sand transport rate. Coast. Eng., 10, 1-21.
[43] Davies, A. G., Ribberink, J., Temperville, A., & Zyserma, J. A. (1997). Comparisons
between sediment transport models and observations made in wave and current
flows above plane beds. Coast. Eng., 31, 163-198.
[44] Miller, H. C. (1999). Field measurements of longshore sediment transport during
storms. Coast. Eng., 36, 310-321.
[45] Eversole, D., & Fletcher, C. H. (2002). Longshore sediment transport rates on a reef-
fronted beach: field data and empirical models Kaanapali Beach, Hawaii. J. Coast.
Res. In press, 19.
[46] Zengh, J., & Hu, J. (2003, November 9-11). Calculation of longshore sediment trans‐
port in Shijiu Bay. Paper presented at International Conference on Estuaries and
Coasts,, Hangzhou, China.
[47] Rogers, A. L., & Ravens, T. M. (2008). Measurement of Longshore Sediment Trans‐
port Rates in the Surf Zone on Galveston Island, Texas. J. Coast. Res. [2], 62-73.
[48] Smith, E. R., & Wang, P. (2001). Longshore sediment transport as a function of ener‐
gy dissipation. Paper presented at ASCE Conf. Proc. Ocean Wave Measurement and
Analysis.
[49] Wang, P., Ebersole, B. A., Smith, E. R., & Johnson, B. D. (2002). Temporal and spatial
variations of surf-zone currents and suspended sediment concentration. Coast. Eng.,
46, 175-211.
[50] Singh, A. K., Deo, M. C., & Kumar, V. S. (2007). Prediction of littoral drift with ar‐
tificial neural networks. Hydro. Earth Syst. Sciences Disc., 4, 2497-2519.
[51] Schoonees, J. S., & Theron, A. K. (1995). Evaluation of10 cross-shore sediment trans‐
port morphological models. Coast. Eng., 25, 1-41.
[52] Davies, A. G., van Rijn, L. C., Damgaard, J. S., van de Graaff, J., & Ribberink, J. S.
(2002). Intercomparison of research and practical sand transport models. Coast. Eng,
46(1), 1-23.
[53] Van Maanen, B., Ruiter, d. P. J., & Ruessink, B. G. (2009). An evaluation of two along‐
shore transport equations with field measurements. Coast. Eng., 56, 313-319.
[54] Larroudé, Ph. (2008). Methodology of seasonal morphological modelisation for nour‐
ishment strategies on a Mediterranean beach. Mar. Pollut. Bull., 67, 45-52.
58 Sediment Transport Processes and Their Modelling Applications

[55] Dally, W. R., Dean, J. L., & Dalrymple, R. A. (1984). A model for breaker decay on
beaches. Paper presented at In 19th Coastal Engineering Conference Proceedings.
82-88.

[56] Svendsen, I. A. (1984). Mass flux and undertow in the surf zone. Coast. Eng., 8,
347-365.

[57] Longuet-Higgins, M. S. (1970). Longshore currents generated by obliquely Incident

waves 1. Journal of Geophysical Research, 75, 6778-6789.

[58] Camenen, B. (2003). Comparison of sediment transport formulae for the coastal envi‐
ronment. Coast. Eng., 48, 111-132.

[59] Bijker, E. (1968). Littoral drift as function of waves and current. Paper presented at
Coastal Engineering Conference Proceedings, London, UK. 415-435.

[60] Van Rijn, L. C. (1993). Principles of sediment transport in rivers, estuaries and coastal seas.,
Publ. Aqua Publications, Zwolle, the Netherlands.

[61] Soulsby, R. (1997). Dynamics of marine sands, a manual for practical applications., Tho‐
mas Telford,, Wallingford, England.

[62] Camenen, B. (2000). Numerical comparison of sediment transport formulae. Sand‐

wave Dynamics Workshop Lille, France. , 37-42.
 Chapter 3

Sediment Transport Patterns Inferred from Grain-Size

Trends: Comparison of Two Contrasting Bays in Mexico

Alberto Sanchez and Concepción Ortiz Hernández

Additional information is available at the end of the chapter

[Link]

1. Introduction

The analysis and description of geological processes in sedimentary environments have been
widely defined by the frequency distribution of grain size (Friedman, 1961) whereas the change
in the sediment textural characteristics has been used to evaluate the net sediment transport (Mc
Laren and Bowles, 1985). The evaluation as a grain-size trend analysis (GSTA) is defined in Gao
and Collins (1991, 1992) and LeRoux (1994a,b). The use of textural characteristics (grain size,
sorting, and skewness) to infer the sediment transport was originally shown by a decrease in
particle size in the direction of flow. In Sunamura and Horikawa (1971) a combination of grain-
size and sorting identified four possible examples where it is possible to infer the direction of the
sediment transport. In the early 1980s, the use of grain size, sorting, and skewness were pro‐
posed to infer sediment transport on the basis of the statistical analysis of sediment-transport paths
along the transect (Friedman, 1961). Later bidimensional models of sediment transport pro‐
posed by Gao and Collins (1992), LeRoux (1994a) and Poizot et al. (2008) are supported by analytic
geometry, vector analysis, and statistics to obtain more robust results of the magnitude and
direction of the transport vectors. The GSTA is an excellent approach for establishing sediment
transport in a variety of environments such as rivers, beaches, harbors, estuaries, continental shelf,
and submarine canyons (Carriquiry and Sánchez, 1999; Carriquiry et al., 2001; Sanchez et al., 2008,
2009, 2010; Sanchez and Carriquiry, 2012).

1.1. Bahia Chetumal

Bahia Chetumal is located in the Mexican Caribbean at the mouth of the Rio Hondo, which defines
the border between Mexico and Belize and is one of the few surface runoffs from the Yucatan
Peninsula. Organic and inorganic wastes of the extensive cane crops adjacent to the river are
discharged directly into the Rio Hondo (Ortiz Hernández and Sáenz Morales, 1999) and subse‐
60 Sediment Transport Processes and Their Modelling Applications

quently are transported and deposited into the bay. Although studies have been made in Bahia
Chetumal concerning the sedimentology (De Jesús Navarrete et al., 2000; Sanchez et al., 2008), the
distribution of metals (García Ríos and Gold Bouchot, 2003; Díaz López et al., 2006; Buenfil Rojas
and Flores Cuevas, 2007), and aromatic hydrocarbons (Álvarez Legorreta and Sáenz Morales,
2005) only in last study there is a good relationship between fine sediments and the concentra‐
tion of aromatic hydrocarbons in the deepest parts of the bay reported.

1.2. Bahia Magdalena

The lagoon complex Bahia Magdalena-Almejas is an ecosystem with high biodiversity,
fisheries, and tourism on the peninsula of Baja California. Recently, a synthesis and integration
of studies on biology, ecology, physical oceanography, and social sciences in this lagoon
complex was made for the purpose of providing accessible information for making decisions
on the use and sustainable exploitation of the natural resources and to identify possible areas
for ecological protection (Funes Rodríguez et al., 2007). However, the sedimentary processes
in this bay were not integrated into this synthesis, but they are of importance in the assessment
of the ecological risk and the recovery and rehabilitation of marine environments (Carriquiry
and Sánchez, 1999; Sanchez et al., 2008). In contrast to Bahia Chetumal, in the lagoon complex
there is no evidence of impacts derived from human activities, at least for metal contamination
(Shumilin et al., 2005).
In our work, the transport and dispersion of the surface sediments are compared in these two
contrasting bays. The interpretation of spatial trends of textural characteristics and their
comparison with hydrographic records may form the basis of a framework for implementing
environmental monitoring schemes in both bays, in addition to expanding our knowledge of
the biogeochemical processes and their relationship with the sedimentary dynamics that
determine the functioning of the ecosystem.

2. Setting

2.1. Bahia Chetumal

Bahia Chetumal is semielongated (~ 110-km long and ~ 20-km wide) with a maximum of 49
km in its central area and a minimum of 5 km at the head (Fig. 1). The bathymetry of the bay
is relatively shallow (4-m on average) with a center channel from 6- to 8-m depth with a SW
direction. There are some narrow, deep depressions known locally as pozas. The Rio Hondo
and some smaller streams that flow into the bay cause estuarine conditions with a salinity of
10 to 18, decreasing from the mouth to the head of the bay. In the summer, rainfall is highest
and accentuates the estuarine conditions in the bay (Carrillo et al., 2009). In this Caribbean
region, the tides are mixed semidiurnal with a microtidal range between 10 and 20 cm (Kjerfve,
1981). Local winds are dominated by easterly and southeasterly trade winds and by large-scale
perturbations, such as tropical storms, hurricanes, and cold fronts (Gallegos et al., 1997; Mooers
and Maul, 1998). The winds are easterlies-south easterlies with a mean speed of 3.1 m s-1 and
the maximum air temperature (> 34.6 °C) is during August (Carrillo et al., 2009). The dry season
 Sediment Transport Patterns Inferred from Grain-Size Trends: Comparison of Two Contrasting Bays in Mexico 61
[Link]

lasts from late March to early June and the winds are mainly from the southeast. The mean
annual river discharge into the bay is about 1,500 million m3. Recently, from observations
during 2005–2006, it was estimated that the Rio Hondo discharge varies from about 9 to 24 to
over 78 m3 s-1 during the dry and the wet seasons. It can be as high as 220 m3 s-1 during the wet
season (SARH-CNA, 1987).

Figure 1. Study area and sampling stations in Bahia Chetumal. The dotted lines denote the bathymetry of the bay.

2.2. Bahia Magdalena

Bahia Magdalena is on the southwestern coast of the peninsula of Baja California (Fig. 2). The bay
is characterized by a relatively shallow area with marshes, lagoons, and channels with a depth <
10 m. In the central part of the bay, the depth is greater than 20 m, with a channel that connects the
bay with the ocean (Álvarez Borrego et al., 1975). In the ocean there is an area of seasonal upwell‐
ing (Zaitsev et al., 2003) that transports nutrients into the bay during spring tides. Inside the bay,
the speed of the tidal current is ~23 cm s-1 during the flood tide and 20 cm s-1 during the ebb tide
(Acosta Ruíz and Lara Lara, 1978), with a maximum tidal-current speed of 1.09 m s-1 in the mouth
(Obeso Nieblas et al., 1999). In Morales Zárate et al. (2006) is described the circulation and passive
transport of particles in Bahia Magdalena. Seeded particles tend to concentrate in the shallow and
internal bay, along Isla Margarita and specific areas in the northern part of the bay. The maxi‐
62 Sediment Transport Processes and Their Modelling Applications

mum particle concentration occurred off the northern part of Isla Margarita, and was associated
with transport generated by wind and the residual tidal flow.

Figure 2. Study area and sampling stations in Bahia Magdalena. The dotted lines denote the bathymetry of the bay.

3. Methods

3.1. Sampling and analysis of surface sediments

In September 1998, the first 2 cm of the surface sediment was collected by scuba diving and
Van Veen grab samples at the 43 sampling stations in Bahia Chetumal and the 58 in Bahia
Magdalena (Fig. 1, 2). The sieve analysis was made by the sieving method (Ingram, 1971). The
textural characteristics (grain size, sorting, and skewness) were calculated using granulometric
data (Folk, 1974).

3.2. Multivariate statistical analysis

The principal component analysis (PCA) is a classical statistical method. This linear transform
has been widely used in data analysis (Carriquiry et al., 2001). If X is a n X m data-matrix (n
samples of m variables, here we choose grain size, sorting coefficient, and skewness, and X is
demeaned and the covariance matrix is R, then a set of orthogonal eigenvectors U – [u1; u2;…;
un] exists:

RU − UΛ
 Sediment Transport Patterns Inferred from Grain-Size Trends: Comparison of Two Contrasting Bays in Mexico 63
[Link]

Define:

Y - XU

Then Y – [y1; y2;…; yn] also forms an orthogonal set. The {yi} are the principal components (PC),
the {ui} are the eigenvectors of the covariance matrix, and the proportion of the total variance
that each eigenvector ‘‘accounts for’’ is given by the magnitude of the eigenvalue. Here, we do
not provide the principal components, but show the principal component weights (eigenvec‐
tor) of the first PC that indicates the dominate relation among the three grain-size character‐
istics (Davis, 1986).

3.3. Sediment-transport model

Sediment-transport models have allowed coastal oceanographers to infer the residual-

sediment transport based on spatial trends of sediment (Mc Laren and Bowles, 1985; Gao and
Collins, 1992; LeRoux, 1994a; Poizot et al., 2008). In our study, the model proposed by LeRoux,
(1994a,b), based on the principles of analytic geometry and vector analysis of textural data,
was used. With this method, the magnitude and direction of the vector of transport were
obtained by comparison of the textural characteristics of five neighboring sampling stations
(one central and four satellites). The general considerations of the model are (1) textural trends
resulting from the hydrodynamic conditions of the environment, (2) applicable in the coastal
zone and shelf where sediment transport is unidirectional, (3) the gradient between textural
parameters is constant in the area where we compared the five sampling stations, (4) textural
parameters in the model have the same weight and importance, and (5) the distance between
the five stations (interseasonal) is not critical, especially if there is a clear textural gradient
between stations.

4. Results

4.1. Grain size trends

4.1.1. Bahia Chetumal

The average grain size was 1.6 ϕ with maximum grain size of -0.43 ϕ and minimum of 2.4 ϕ.
In general, the spatial trend of grain size is to decrease toward the central part and head of the
bay and in the area around the mouth of the Rio Hondo and the city of Chetumal, whereas the
central area of the coast has a finer grain size than the eastern coastal area (Fig. 3A). The
sediments with better sorting are associated with stations where the grain size is fine and vice
versa (Fig. 3B). The coefficient of determination between these variables (grain size vs. sorting)
is R2 = 0.52 (F1, 41, α = 0.05 = 43.7, P = 0.0000). The skewness (Fig. 3C) of the surface sediments is
negative (toward coarse sediments) throughout the bay, except in the coastal margin, south of
Isla Tamalcab (Fig. 3C), where skewness is moderately positive (toward fine sediments). The
surface sediments are dominated 87% on average by sand (minimum 76% and maximum 96%)
and a minor proportion of mud 13% on average (minimum 4% and maximum 24%).
64 Sediment Transport Processes and Their Modelling Applications

Figure 3. Spatial distribution of textural characteristics (A) Grain size (phi), (B) Sorting (phi), and (C) Skewness (phi) of
sediments in Bahia Chetumal.

4.1.2. Bahia Magdalena

The grain-size distribution is relatively homogeneous in the bay with an interval in grain size
from 2.5 to 3.5 ϕ with the exception of a textural gradient (increase in grain size) measured
toward the mouth of the bay and Isla Margarita (Fig. 4A). In general, the sediments are well-
sorted. Sorting is poor toward where the grain size increases (mouth of the bay and Isla
Margarita (Fig. 4B). The skewness characteristics were systematically negative throughout the
bay with certain trends toward positive values in the west, east, and toward the mouth of the
bay and Isla Margarita (Fig. 4C).
 Sediment Transport Patterns Inferred from Grain-Size Trends: Comparison of Two Contrasting Bays in Mexico 65
[Link]

Figure 4. Spatial distribution of textural characteristics (A) Grain size (phi), (B) Sorting (phi), and (C) Skewness (phi) of
sediments

4.2. Principal component analysis

4.2.1. Bahia Chetumal

The principal component analysis used for the textural data of sediment (grain size, sorting,
and skewness) is shown in Figure 5A-C. The first eigenvalue, the grain size, and sorting explain
67% of the variance (Fig. 5A, B) and the second eigenvalue, skewness explains 28% of the
variance (Fig. 5C). These two eigenvalues explain 95% of the variability of the textural
characteristics. Only 5% of the variability is caused by other factors. The grain size and sorting
showed a negative correlation. Thus, fine sediments are related to well-sorted clastic material
and coarse-grained sediments are poorly sorted, whereas the skewness coefficient has a poor
correlation with those textural characteristics.

4.2.2. Bahia Magdalena

The multivariate statistical analysis of principal components used for the Bahia Magdalena
textural data indicated that two factors explain 98% of the total variability (Fig. 5D-F). Factor 1 is
constituted by the grain size and sorting, and explains 74% of the variability (Fig. 5D, E), and factor
2, with 24% of the variability, corresponds to the skewness of the sediment (Fig. 5F). The grain size
is inversely correlated with sorting, i.e. fine-grained sediments are better sorted with an skew‐
ness towards negative values. Only 2% of the variability is caused by other factors.
66 Sediment Transport Processes and Their Modelling Applications

Figure 5. Graphic representation of the main factors and their respective loads obtained from the principal compo‐
nent analysis of sediment textural-data. GS: Grain-size, S: Sorting, and Sk: Skewness.

4.3. Net sediment transport

4.3.1. Bahia Chetumal

The vectors for the residual sediment transport are shown in Figure 6. In the north of the bay,
the transport vectors show a SW direction. In the central region of the bay, transport vectors
have a S-SE direction, except in the central-eastern margin, where the vector has a SW transport
(Fig. 6). In the southern zone, the residual transport vectors showed a preferential direction to
the S-SE. Stations 28, 29A, and 34 near the mouth of the Rio Hondo have a net sediment
transport to the E-NE. In general, the surface sediment transport in Bahia Chetumal during
sampling was from the head to the mouth of the bay (SW to SE direction). The smaller
magnitudes of the transport vectors are located in the central part of the bay (Fig. 6) and this
coincides with those areas of the bay in which the grain size tends to finer sizes.

4.3.2. Bahia Magdalena

The vectors of the residual-sediment transport described a cyclonic gyre in the central part and
deep bay, whereas in the southeastern they describe an anticyclonic gyre (Fig. 7). In the
northwestern margin of the bay, the residual-transport vectors showed a southeasterly
direction (Fig. 7). On the coastal margin of Isla Margarita, the residual transport vectors
denoted a pathway of sediment particles in a southwesterly direction (Fig. 7). The sediment
transport to the mouth of the bay was not defined because only one sample was collected.
 Sediment Transport Patterns Inferred from Grain-Size Trends: Comparison of Two Contrasting Bays in Mexico 67
[Link]

5. Discussion

5.1. Frequency distribution of magnitude vector

The vector analysis of the textural characteristics is a qualitative property that indicates the
relative magnitude and the predominant direction of each vector, where two neighboring
stations are exchanging material, without needing to provide quantitative data on the amount
of material exchanged. The frequency distribution of the magnitude of the vectors was in the
range of 0.4 to 0.7 (90% of the vector quantities) obtained in Bahia Chetumal. For Bahia
Magdalena, the frequency distribution of the magnitude of the vector was heterogeneous, with
only 34% of the vector magnitude in the range of 0.4 to 0.5 and the remaining 66% varied
between magnitudes 0.1 to 0.4 and 0.6 to 1.0. Other studies have shown that 80% of vectors
were characterized by vector magnitudes between 0.4 and 0.5 in the North Sea (Gao and
Collins, 1994) and 0.9-1.0 for the northern Gulf of California (Carriquiry et al., 2001). This may
suggest that hydrodynamic conditions in Bahia Magdalena are less homogeneous than those
of Bahia Chetumal, the North Sea, and the northern Gulf of California, and therefore reflect
the sedimentological environmental gradients. This suggests a higher variability in the
exchange of materials between nearby stations (Carriquiry and Sanchez, 1999).

Figure 6. Dispersion of sediments in Bahia Chetumal inferred by the transport vectors. The vectors describe the main
trajectory of the sedimentary material.
68 Sediment Transport Processes and Their Modelling Applications

5.2. Transport and dispersion of sediments

The grain-size trend analysis has been widely used in marine and coastal environments to
establish the net transport of sediments. In these studies, the net transport and dispersion of
sediment were validated by comparing the residual vector transport with ocean currents
(Carriquiry and Sanchez, 1999; Van Wesenbeck and Lanckneus, 2000; Carriquiry et al., 2001;
Liu et al., 2002; Jia et al., 2002; Duman et al., 2004, 2006; Friend et al., 2006; Lucio et al., 2006;
Duc et al., 2007). Thus, the GSTA is a useful tool to infer the movement of sediment particles
in places where environmental hydrodynamics are poorly understood.

Figure 7. Dispersion of sediments in Bahia Magdalena inferred by the transport vectors. The circular arrows describe
the main trajectory of the sedimentary material and residual tidal circulation.

The residual transport vectors allowed us to define that the net sediment transport is S-SW
and S-SE inside Bahia Chetumal with a convergence towards the central part of bay consistent
with the contours of the grain size and sorting. Hydrographic information is limited for Bahia
Chetumal. The only earlier data come from Morales Vela et al. (1996) who measured the surface
and bottom currents over a year inside the bay. The average surface current was 12 cm s-1 with
a direction 183° ± 88°. On the bottom, the currents were slightly lower at 9 cm s-1 and a dominant
direction of 182° ± 81°. Although the directions of surface currents varied between 137° and
240° and the bottom was 136° and 236°, both streams were in the S-SW and S-SE direction.
Bahia Chetumal has a microtidal regime and tidal currents are expected to be weak. Therefore,
the wind remains as a candidate to mix the water column. The prevalent wind direction was
E–SE in the 70% of the total observed data. The role of prevailing winds and wind events is
related to the orientation of the estuary. In Bahia Chetumal, the role of wind forcing in the
circulation needs to be studied in detail (Carrillo et al., 2009). The persistence of relatively
 Sediment Transport Patterns Inferred from Grain-Size Trends: Comparison of Two Contrasting Bays in Mexico 69
[Link]

mixed conditions caused by strong wind events and the response of the plume to the direction
of the wind deserve further study. However, the dispersion of sediment (S-SW and S-SE) is
consistent with the direction of surface and bottom currents and winds reported for Bahia
Chetumal (Morales Vela et al., 1996; Carrillo et al., 2009). The spatial distribution of fine to
coarse sand suggests that the hydrodynamic conditions are sufficiently intensive to limit the
deposition of silt and clay in the central and deep bay (Morales Vela et al., 1996; Sanchez et al.,
2008). In Carrillo et al. (2009) it has been suggested that the magnitude of the surface and bottom
currents are relatively low, consistent with Lankford (1977) who placed the Bahia Chetumal
as a low energy environment. However, the currents (Morales Vela et al., 1996) inside the bay
are able to resuspend (Shepard and Keller, 1978) and transport (Komar, 1977) sediment of 3
ϕ to 5 ϕ in the S-SW and S-SE directions (Sanchez et al., 2008).

The net sediment transport described a cyclonic gyre in the central part and deep portion and
an anticyclonic gyre in the southeastern part of Bahia Magdalena. Studies of the circulation
inside the bay are limited but useful to validate the sedimentary transport inferred from
textural trends. In Sánchez Montante et al. (2007) the residual currents that result from the
rectification of the forced circulation with the M2 tidal component were calculated. The results
of residual tidal currents pointed to the presence of a cyclonic circulation in the central region
of the bay and an anticyclonic circulation in the southeastern region. An experiment using the
release of 58 particles inside the bay demonstrated the presence of a cyclonic gyre (Sánchez
Montante et al., 2007), which corresponds to the residual tidal circulation and sediment
transport and dispersion in the interior of the bay (Sanchez et al., 2010). An experiment with
particles seeded in the bay shows they tend to remain inside and concentrate in areas that
corresponded with the actual distribution of fish stocks. The areas of particle concentration
were located along the coastal margin of the bay and the coasts of Islas Margarita and
Magdalena. The sediment transport vectors defined convergence zones, which coincide with
those areas in which the particles tend to be concentrated (Morales Zárate et al., 2006).

5.3. Variability of textural characteristics

The textural trends of the sediment have been extensively used to infer the possible paths of clastic
material in different coastal environments. The combination of textural characteristics defines the
existence of several examples for the inference of sediment transport and in all these, the sorting
was better in the direction of the current (Sunamura and Horikawa, 1971; Mc Laren and Bowles,
1985; Gao and Collins, 1991, 1992; LeRoux, 1994a,b; Carriquiry and Sánchez, 1999; Carriquiry et
al., 2001; Poizot et al., 2008; Sanchez et al., 2008, 2009, 2010; Sanchez and Carriquiry, 2012). The
principal component analysis used for the textural characteristics of this study indicated that the
spatial trends of grain size and sorting explain 67% and 74% of the variability, whereas the
skewness explains 28% and 24% of the variability. The correlation analysis of grain size vs. sorting
was significant (F1,101,α=0.05 = 430, P <0.0000) with a coefficient of determination (R2 = 0.88). For grain
size vs. skewness, the coefficient of determination (R2 = 0.28) is consistent with the explained
variability by the principal component analysis (22%). The results of Bahia Chetumal and
Magdalena contrast with the variability in spatial trends of grain size and skewness that explain
95% of the variability in the Yellow Sea (Cheng et al., 2004).
70 Sediment Transport Processes and Their Modelling Applications

In areas influenced by discharges from rivers, such as the Yellow Sea, the flocculation of the
particles is one of the important factors affecting sorting and, to a lesser degree, grain size of the
particles (Kranck and Milligan, 1991). This allows us to define the conditions of deposition on the
basis of the analysis of the grain size (Fox et al., 2004). The difficulty of establishing a trend in the
sediment texture is because flocculation can lead to a preference of the deposition of fine parti‐
cles in the settling of selective sites of sediment that are related to the hydrodynamic conditions.
The difference between the principal component analyses can be caused by processes derived
from the flocculation of sediment, which allowed the establishment of poorly sorted material in
the Yellow Sea. The discharge of ephemeral streams into Bahia Magdalena and the low dis‐
charge of the Rio Hondo into Bahia Chetumal promote well-sorted sediment deposition.

5.4. Implication of grain-size trend analysis

In general, the spatial trends of textural characteristics corroborate that the well-sorted fine-
grained material is distributed over the central part of the bay and agreed with the preferential
accumulation of organic matter, metals, and hydrocarbons in Bahia Chetumal (Ortiz Hernán‐
dez and Sáenz Morales, 1999; García Ríos and Gold Bouchot, 2003; Álvarez Legorreta et al.,
2005). This contrasts with Bahia Magdalena, which has no pollution problems, at least from
potentially toxic elements (Shumilin et al., 2005) and organic matter (Sanchez, 2010, 2011).
The results obtained in the present and previous studies are of great significance for the region
generally and particularly for the use and sustainable exploitation of natural resources, and
for the definition of possible areas of environmental protection. By using the hypothesis that
the contaminants preferentially associated with the fine particles of sediment, i.e. silts and
clays, would follow the path of the sediment transport, the textural-trend analysis helps to
identify the relationship between the discharge of pollutants and their sources, and helps to
predict the transport and fate of contaminated sediments in marine environments. The
development of Bahia Chetumal and Magdalena, caused by an increase of anthropogenic
activities, undoubtedly will contribute to deterioration of these environments. The integration
of research studies, including biological, chemical, and sedimentological, can be the basis for
proposing monitoring programs, especially in areas where there are concerns about possible
sources of pollution or are subject to a possible environmental impact, e.g. in the spawning
and larval rearing areas of marine species of economic importance and the habitat and feeding
sites of marine mammals such as the manatee and whales.

6. Conclusions

The grain size of surface sediments shows that particles are finer and better sorted with a
skewness towards coarse particles in the central region and head of both bays. The principal
component analysis indicated that the correlation between grain size and sorting was signifi‐
cant and explained 67% and 74% of the variability of the textural trend of grain size. The
remaining 28% and 24% is explained by the skewness. In Bahia Chetumal, the net transport of
sediment suggests that clastic material and particles (inorganic and organic) of anthropogenic
or natural origin have a transport in the S-SW and S-SE directions, except near the mouth of
 Sediment Transport Patterns Inferred from Grain-Size Trends: Comparison of Two Contrasting Bays in Mexico 71
[Link]

the Rio Hondo where the net transport of sediment is preferentially E-NE. This indicates that
there can be a greater influence on the dispersion of sediments and particulate pollutants that
are discharged directly into the river, through the drainings of agricultural fields and urban
waste. The convergence of the transport vectors in the central part of the bay is consistent with
previous studies that described the spatial distribution of metals and hydrocarbons.

The net sediment transport described a cyclonic gyre in the central part and deep portion and
an anticyclonic gyre in the southeastern part of Bahia Magdalena. These results agree with the
residual tidal currents that showed the presence of a cyclonic circulation in the central region
of the bay and an anticyclonic circulation in the southeastern region. The release of 58 particles
inside the bay demonstrated the presence of a cyclonic gyre that corresponds to the residual
tidal circulation and sediment transport and dispersion in the interior of the bay. The textural-
trend analysis is a technique that can be useful for predicting the transport and fate of polluting
sediments. The results of our research and the preliminary studies of geochemistry could be
the basis to suggest new research and the monitoring of water quality and sediment, with
special attention of areas where pollution problems exist or may be subjected to an eventual
environmental impact.

Acknowledgements

We thank an anonymous reviewer for their valuable suggestions to this work. Financial
support SIP 20110143 and 20120689, Instituto Politécnico Nacional, México. Thanks to Dr. Ellis
Glazier for editing this English-language text.

Author details

Alberto Sanchez1 and Concepción Ortiz Hernández2

*Address all correspondence to: alsanchezg@[Link]

1 Centro Interdisciplinario de Ciencias Marinas, Instituto Politécnico Nacional, La Paz, Baja

California Sur, México

2 El Colegio de la Frontera Sur, Unidad Chetumal, Chetumal, Quintana Roo, México

References

[1] Acosta Ruíz J, Lara Lara J. Resultados fisicoquímicos en un estudio de variación diur‐
na en el área central de Bahía Magdalena, B.C.S. Ciencias Marinas 1978; 5(1) 37-45.
72 Sediment Transport Processes and Their Modelling Applications

[2] Álvarez Borrego S, Galindo Bect L, Chee Barragan A. Características hidroquímicas

de Bahía Magdalena, B.C.S. Ciencias Marinas 1975; 2(1) 94-110.

[3] Álvarez Legorreta T, Sáenz Morales R. Hidrocarburos aromáticos policíclicos en sedi‐

mentos de la Bahía de Chetumal. In: Botello AV, Rendón von Osten J, Gold Bouchot
G, Agraz Hernández C. (eds.) Golfo de México, Contaminación e Impacto Ambiental:
Diagnóstico y Tendencias. Mexico: Universidad Autónoma de Campeche, Universi‐
dad Nacional Autónoma de México, Instituto Nacional de Ecología; 2005. p299-310.

[4] Buenfil Rojas M, Flores Cuevas N. Determinación de metales pesados (As, Cd, Hg y
Pb) presentes en el Río Hondo, Quintana Roo. In: VI Congreso Internacional y XII
Nacional de Ciencias Ambientales: Memorias en extenso, 6-8 June 2007, Chihuahua,
Chihuahua. Mexico: Universidad Autónoma de Chihuahua, Academia Nacional de
Ciencias Ambientales; 2007.

[5] Carriquiry JD, Sánchez A. Sedimentation in the Colorado River delta and Upper Gulf
of California after nearly a century of discharge loss. Marine Geology 1999; 158(1-4)
125-145.

[6] Carriquiry JD, Sánchez A, Camacho-Ibar VF. Sedimentation in the northern Gulf of
California after the elimination of Colorado River discharge. Sedimentary Geology
2001; 144(1-2) 37-62.

[7] Carrillo L, Palacios Hernández E, Yescas M, Ramírez Manguilar AM. Spatial and sea‐
sonal patterns of salinity in a large and shallow tropical estuary of the western carib‐
bean. Estuaries and Coasts 2009; 32(5) 906-916.

[8] Cheng P, Gao S, Bokuniewicz H. Net sediment transport patterns over the Bohai
Strait based on grain size trend analysis. Estuarine, Coastal and Shelf Science 2004;
60(2) 203-212.

[9] Davis J., editor. Statistics and data analysis in geology. New York: Wiley; 1986.

[10] De Jesús Navarrete A, Oliva Rivera JJ, Valencia Beltrán V, Quintero López N. Distrib‐
ución de sedimentos en la Bahía de Chetumal, Quintana Roo, México. Hidrobiológica
2000; 10(1) 61-67.

[11] Díaz López C., Carrión Jiménez J.M., González Bucio J.L. Estudio de la contamina‐
ción por Hg, Pb, Cd y Zn en la Bahía de Chetumal, Quintana Roo, México. Revista
Sociedad Química del Perú 2006; 72(1) 19-31.

[12] Duc DM, Nhuan MT, Ngoi CV, Nghi T, Tien DM, van Weering TjCE, van den Bergh
GD. Sediment distribution and transport at the nearshore zone of the Red River del‐
ta, northern Vietnam. Journal Asian Earth Science 2007; 29(1) 558-565.

[13] Duman M, Avci M, Duman S, Demirkurt E, Duzbastilar MK. Surficial sediment dis‐
tribution and net sediment transport pattern in Izmir Bay, western Turkey. Continen‐
tal Shelf Research 2004; 24(9) 965-981.
 Sediment Transport Patterns Inferred from Grain-Size Trends: Comparison of Two Contrasting Bays in Mexico 73
[Link]

[14] Duman M, Duman S, Lyons TW, Avci M, Izdar E, Demirkurt E. Geochemistry and
sedimentology of shelf and upper slope sediments of the south-central Black Sea.
Marine Geology 2006; 227(1-2) 51-65.

[15] Folk RL, editor. Petrology of Sedimentary Rock. Austin, TX: Hemphill Publishing
Company; 1974.

[16] Fox JM, Hill PS, Milligan TG, Boldrin A. Flocculation and sedimentation on the Pô
River Delta. Marine Geology 2004; 203(1-2) 95-107.

[17] Friedman GM. Distinction between dune, beach, and river sands from their textural
characteristics. Journal of Sedimentary Petrology 1961; 31(4) 514-529.

[18] Friend PL, Velegrakis AF, Weatherston PD, Collins MB. Sediment transport path‐
ways in a dredged ria system, southwest England. Estuarine, Coastal and Shelf Sci‐
ence 2006; 67(3) 491-502.

[19] Funes Rodríguez R, Gómez Gutiérrez J, Palomares García R. editors. Estudios ecoló‐
gicos en Bahía Magdalena. Mexico: Centro Interdisciplinario de Ciencias Marinas. In‐
stituto Politécnico Nacional, México; 2007.

[20] Gallegos A, Czitrom S. Aspectos de la oceanografía física regional del Mar Caribe. In:
Lavin MF. (ed) Contribuciones a la oceanografía física en México. Mexico: Unión Ge‐
ofísica Mexicana; 1997. p225-242.

[21] Gao S, Collins MB. A critique of the Mc Laren method for defining sediment trans‐
port paths: discussion. Journal of Sedimentary Petrology 1991; 61(1) 143-146.

[22] Gao S, Collins MB. Net sediments transport patterns from grain size trends, based
upon defi nition of ‘transport vectors’. Sedimentary Geology 1992; 81(1-2) 47-60.

[23] Gao S, Collins MB. Analysis of grain size trends, for defining sediment transport
pathways in marine environments. Journal of Coastal Research 1994; 10(1) 70-78.

[24] García Ríos V.A., Gold Bouchot G. Trace metals in sediments from Bahía de Chetu‐
mal, Mexico. Bulletin of Environmental Contamination and Toxicology 2003; 70(6)
1228-1234.

[25] Ingram RL. Sieve analysis. In: Carver RE. (ed.) Procedures in Sedimentary Petrology.
New York: Wiley Interscience; 1971. p49-68.

[26] Jia JJ, Gao S, Xue YC. Sediment dynamic processes of the Yuehu inlet system, Shan‐
dong Peninsula, China. Estuarine, Coastal and Shelf Science 2003; 57(5-6) 783-801.

[27] Kjerfve B. Tides of the Caribbean Sea. Journal of Geophysical Research 1981; 86(C5)
4243-4247.

[28] Komar DP. Selective longshore transport rates of different grain size fractions within
a beach. Journal of Sedimentary Petrology 1977; 47(4) 1444-1453.
74 Sediment Transport Processes and Their Modelling Applications

[29] Kranck K, Milligan TG. Grain size in oceanography. In: Syvitski JPM. (ed.) Principles,
methods and application of particle size analysis. New York: Cambridge University
Press; 1991. p332-345.

[30] Lankford RR. Coastal lagoons of Mexico: Their origin and classification. In: Wiley M.
(ed.) Estuarine Processes. New York: Academic; 1977. p182-215.

[31] LeRoux JP. An alternative approach to the identification of the end sediment trans‐
port paths based on grain size trends. Sedimentary Geology 1994a; 94(1-2) 97-107.

[32] LeRoux JP. A spreadsheet template for determining sediment transport vectors from
grain size parameters: Computer and Geoscience 1994b; 20(3) 433-440.

[33] Liu JT, Liu K, Huang JC. The effect of a submarine canyon on the river sediment dis‐
persal and inner shelf sediment movement in southern Taiwan. Marine Geology
2002; 181(4) 357-386.

[34] Lucio PS, Bodevan EC, Dupont HS, Ribeiro LV. Directional kriging: a proposal to de‐
termine sediment transport. Journal of Coastal Research 2006; 22(6) 1340-1348.

[35] Mc Laren P., Bowles D. The effects of sediment transport on grain size distributions.
Journal of Sedimentary Petrology 1985; 55() 457-470.

[36] Mooers CK, Maul GA. Intra-Americas sea circulation. In: Robinson A.R., Brink K.H.
(ed) The Sea. New York: Wiley; 1998. p183–208.

[37] Morales Vela B, Olivera Gómez D, Ramírez García P. editors. Conservación de los
manatíes en la región del Caribe de México y Belice. Mexico: El Colegio de la Fron‐
tera Sur (ECOSUR)-Consejo Nacional de Ciencia y Tecnología (CONACyT); 1996.

[38] Morales Zárate MV, Aretxabaleta AL, Werner FE, Lluch Cota SE. Modelación de la
circulación invernal y la retención de partículas en el sistema lagunar Bahía Magdale‐
na-Almejas (Baja California Sur, México). Ciencias Marinas 2006; 32(4) 631-647.

[39] Obeso Nieblas M., Gaviño Rodríguez J.H., Jiménez Illescas A Modelación de la mar‐
ea en el sistema lagunar Bahía Magdalena-Almejas, B.C.S., México. Oceánides 1999;
4(1) 79-98.

[40] Ortiz Hernández MC, Sáenz Morales R. Effects of organic material and distribution
of fecal coliforms in Chetumal Bay, Quintana Roo, Mexico. Environmental Monitor‐
ing and Assessment 1999; 55(3) 423-434.

[41] Poizot E, Méar Y, Biscara L. Sediment Trend Analysis through the variation of granu‐
lometric parameters. A review of theories and applications. Earth-Science Reviews
2008, 86(1-4) 15-41.

[42] Sanchez A. editor. La Productividad Marina y su relación con los componentes bio‐
génicos en el complejo lagunar de Bahía Magdalena-Almejas, Baja California Sur:
Fase II. Mexico: Centro Interdisciplinario de Ciencias Marinas. Instituto Politécnico
Nacional; 2010. p1-8.
 Sediment Transport Patterns Inferred from Grain-Size Trends: Comparison of Two Contrasting Bays in Mexico 75
[Link]

[43] Sanchez A. editor. Distribución y composición de la materia orgánica en sedimentos

del complejo lagunar de Bahía Magdalena-Almejas, Baja California Sur, México: Fase
I. Mexico: Centro Interdisciplinario de Ciencias Marinas. Instituto Politécnico Nacio‐
nal; 2011. p1-7.

[44] Sánchez A, Carriquiry J. Sediment transport patterns in Todos Santos Bay, Baja Cali‐
fornia, Mexico, inferred from grain-size trends. In: Mannig A. (ed) Sediment trans‐
port in Aquatic environments. Croatia: Intech; 2012. p1-18.

[45] Sánchez A, Alvarez Legorreta T, Sáenz Morales R, Ortiz Hernández MC, López Ortiz
BE, Aguiñiga S. Distribución de parámetros texturales de los sedimentos superfi‐
ciales en la Bahía de Chetumal: Implicaciones en la inferencia de transporte. Revista
de la Sociedad Geológica Mexicana 2008; 25(3) 523-532.

[46] Sánchez A, Carriquiry J, Barrera J, López Ortiz BE. Comparación de modelos de

transporte de sedimento en la Bahía Todos Santos, Baja California, México. Boletin de
la Sociedad Geologica Mexicana 2009; 61(1) 13-24.

[47] Sánchez A, Shumilin E, López Ortiz BE, Aguíñiga S, Sánchez Vargas L, Romero Gua‐
darrama A, Rodriguez Meza D. Sediment transport in Bahía Magdalena, inferred of
grain-size trend analysis. Journal Latinoamerican of Aquatic Science 2010; 38(2)
167-177.

[48] SARH-CNA. Sinopsis geohidrológica del Estado de Quintana Roo. México: Secretaría
de Recursos Hidraúlicos. 1987.

[49] Sánchez Montante O, Zaitsev O, Saldívar Reyes M. Condiciones hidrofísicas en el sis‐

tema lagunar Bahía Magdalena-Almejas. In: Funes-Rodríguez R, Gómez Gutiérrez J,
Palomares García R. (eds.) Estudios ecológicos en Bahía Magdalena. Mexico: Centro
Interdisciplinario de Ciencias Marinas. Instituto Politécnico Nacional; 2007. p1-28.

[50] Shepard FP, Keller GH. Currents and sedimentary processes in submarine canyons
off the northeast United States. In: Stanley DJ, Kelling GK. (eds.) Sedimentation in
Submarine Canyons, Fans and Trenches. Stroudsburg, Pennsylvania: Dowden,
Hutchinson & Ross Inc.; 1978. p15-32.

[51] Shumilin E, Rodríguez Meza GD, Sapozhnikov D, Lutsarev S, Murrillo de Nava J.

Arsenic concentrations in the surface sediments of the Magdalena-Almejas Lagoon
complex, Baja California Peninsula, Mexico. Bulletin Environmental Contamination
and Toxicology 2005; 74(3) 493-500.

[52] Sunamura T, Horikawa K. Predominant direction of littoral transport along Kujyu‐

kuri Beach, Japan. Coastal Engineering in Japan 1971; 14 107-117.

[53] Van Wesenbeck V, Lanckneus J. Residual sediment transport paths on a tidal sand
bank: a comparison between the modified Mc Laren model and bedform analysis.
Journal of Sedimentary Research 2000; 70 (3) 470-477.
76 Sediment Transport Processes and Their Modelling Applications

[54] Zaitsev O, Cervantes Duarte R, Sánchez Montante O, Gallegos García A. Coastal up‐
welling activity on the Pacific shelf of the Baja California Peninsula. Journal Ocean‐
ography 2003; 59(4) 489-502.
 Chapter 4

Sediment Transport Modeling Using GIS in Bagmati

Basin, Nepal

Rabin Bhattarai

Additional information is available at the end of the chapter

[Link]

1. Introduction

Soils play an important role in the maintenance of global food supplies, an ever increasingly
important role as total population expands. The first ‘Global Assessment of Human-Induced
Soil degradation’ (GLASOD) was published in 1990 and estimated that 1.97 billion hectares,
equivalent to an area of 15% of total land cover, suffered degradation from the mid 1940’s
up to 1990. The more recent GLASOD (Global Assessment of Soil Degradation) survey has
indicated more than 109 ha of the land surface of the world are currently experiencing seri‐
ous soil degradation as a result of water erosion. For total suspended sediment yield from
the land to the oceans, values closer to 15-20 x 109 tons year −1 have been most frequently
cited, whereas average global specific total sediment load is approximately 140-188 tons
km-2 year-1 [23]. On a global scale, the loss of 75 billion tons of soil costs the world about US
$400 billion/year (at US$3/ton of soil for nutrients and US$2/ton of soil for water), or approx‐
imately US$70/person/year [13].

Erosion prediction is the most widely used and most effective tool for soil conservation
planning and design. Because it is impossible to monitor the influence of every farm and
ranch management practice in all ecosystems under all weather conditions, erosion predic‐
tions are used to rank alternative practices with regard to their likely impact on erosion.
These erosion predictions are thus an essential part of soil conservation programs. Assess‐
ment of soil erosion as to how fast soil is being eroded is helpful in planning conservation
work. Estimates of the rate of soil loss may then be compared with what is considered ac‐
ceptable and the effects of different conservation strategies can be determined. Modeling can
be an effective method of predicting soil loss under a wide range of conditions as it can pro‐
vide a quantitative and consistent approach to estimating soil erosion and sediment trans‐
78 Sediment Transport Processes and Their Modelling Applications

port. Using remote sensing and GIS to parameterize such models allows them to be applied
over local, regional and global scales.
Two main types of model: empirically based and process based are available for predicting soil
erosion and sediment transport. Empirically based technology means regression or lumped
mathematical models, which were developed using the experimental data of plot studies on
erosion by water. Zingg [29] and Musgrave [18] equations are examples of initial steps to‐
wards the empirical soil erosion models. Universal Soil Loss Equation, USLE [26], later re‐
vised as Revised USLE or RUSLE [20] is one such model developed in the USA with more than
10,000 plot years of research data and experience of soil scientists. It is the most widely used
model for soil erosion estimation because of the simplicity. It is based on the set of mathemati‐
cal equations that estimate average annual soil loss from inter-rill and rill erosion. In addition,
the equation combines interrelated physical and management parameters such as soil type,
rainfall pattern, and topography that influence the rate of erosion. Erosion Productivity Im‐
pact calculator (EPIC) model [25], which was developed to assess the effect of soil erosion on
soil productivity, also uses USLE and Modified USLE (MUSLE) model [24] to simulate erosion
process. Chemical, Runoff and Erosion from Agricultural Management Systems (CREAMS)
model [11], Agriculture Non-point Source Pollution model (AGNPS) model [28], and Soil and
Water Assessment Tool (SWAT) model [1] are the examples of hybrid models which are based
on USLE/MUSLE/RUSLE for the erosion estimation but use the sediment transport approach
on the basis of continuity equation for sediment yield estimation.
Physically or process based models are intended to represent the essential mechanisms con‐
trolling erosion and sediment transport process. These models are the synthesis of individu‐
al component that affect the erosion and transport process. Aerial Non-point Source
Watershed Environmental Response Simulation (ANSWERS) model [2], Kinematic Runoff
and Erosion model (KINEROS) model [27], European Soil Erosion model (EUROSEM) mod‐
el [17], and Water Erosion Prediction Program (WEPP) [19] are examples of process based
models. Although physically based models try to emulate the physical processes involved in
soil erosion and sediment transport, the weakness of these models is numerous parameters
they need for calibration and also suffer from the problem of equifinality [3].
The overall aim of the study is the modeling of soil erosion and transport processes in dis‐
tributed manner so that erosion, deposition and sediment yield can be computed and veri‐
fied with the observations in data limited conditions. To achieve this objective, an empirical
model was framed within Geographic Information System (GIS) to predict soil erosion in
distributed manner. Then, the sediment delivery approach is used to predict sediment yield
in this study. For the empirical approach, the revised form of the USLE model, RUSLE, is
used to predict erosion potential on a cell-by-cell basis in conjunction with SEDD model to
determine the catchment sediment yield by using the concept of sediment delivery ratio [7].

2. Methodology

A very popular empirical model, known as USLE is used to estimate soil erosion in this
study. Then, sediment delivery approach is used to estimate the sediment yield which a part
 Sediment Transport Modeling Using GIS in Bagmati Basin, Nepal 79
[Link]

of eroded sediment that appears at watershed outlet. Empirical methods such as the USLE
have been found to produce realistic estimates of surface erosion (and also sediment yield)
over areas of small size [26, 10 4]. Sediment delivery distributed (SEDD) model couples
USLE with a spatial disaggretion criterion of sediment delivery processes. The revised form
of USLE, commonly known as RUSLE, is expressed as:

A= R * K * L * S * C * P (1)

Where, A = average annual soil loss predicted (ton ha-1), R = rainfall runoff erosivity factor
(MJ mm ha-1 hr -1), K = soil erodibility factor (ton ha hr MJ-1 ha-1 mm-1), L = slope length fac‐
tor, S = slope steepness factor, C = cover management factor and P = support practice factor.

The value of RUSLE factors are computed using the following methods as described in the
Agricultural Handbook 703 [20].

R=
1 n
∑
n i=1 (∑ E (I ) )
m

j=1
j 30 j (2)

Where, n = total number of years, m = total number of rainfall storms in i th year, I 30 = maxi‐
mum 30 minutes intensity (mm hr -1), E j = total kinetic energy (MJ ha-1) of j th storm of i th
year and it is given as:

p
Ej = ∑ ek * dk (3)
i=1

Where, p = total number of divisions of j th storm of i th year, d k = rainfall depth of k th division

of the storm (mm), e k = kinetic energy (MJ ha-1 mm-1) of k th division of the storm and is given
as [20]:

(−0.05ik )
ek = 0.29(1 − 0.72e ) (4)

Where, i k = intensity of rainfall of k th division of the storm (mm hr-1)

If λ is the horizontal projection of the slope length (in meter), then L factor is given as:

L = ( 22.1
λ
) m
(5)

Where, λ = contributing slope length (in meter), m = variable slope length exponent.

The slope-length exponent ‘m’ is related to the ratio β of rill erosion (caused by flow) to in‐
terrill erosion (principally caused by raindrop impact) by the following equation:
80 Sediment Transport Processes and Their Modelling Applications

m = β / (1 + β) (6)

For moderately susceptible soil in both rill and inter-rill erosion, McCool et al. (1989) sug‐
gested the equation:

(Sinθ / 0.0896)
β= (7)
3.0(Sinθ)0.8 + 0.56

Where, θ = slope angle (degrees).

The slope steepness factor S is evaluated from the following equations (McCool et al., 1987):

S = 10.8 sin θ + 0.03 for s 9%

(8)
S = 16.8 sin θ − 0.50 for s ≥ 9%

Where, s = slope in percentage.

C and P factors are assigned to different grid according to land cover while K factor is esti‐
mated using the soil data.
In a catchment, not all eroded soil reaches the catchment outlet but a part of the soil eroded in
an overland region gets deposited within the catchment. The values of ratio of sediment yield
to total surface erosion, which is termed as sediment delivery ratio (D R), for an area are found
to be affected by catchment physiography, sediment sources, transport system, texture of
eroded material, land cover etc. [23]. However, variables such as catchment area, land slope
and land cover have been mainly used as parameters in empirical equations for D R [9, 12].
Ferro & Minacapilli [5] and Ferro [1997] hypothesized that D R in grid cells is a
strong function of the travel time of overland flow within the cell. The travel time
is strongly dependent on the topographic and land cover characteristics of an area
and therefore its relationship with D R is justified. Based on their studies on proba‐
bility distribution of travel time, the following relationship was assumed herein for
a grid cell lying in an overland region of a catchment:

DR = exp( − γti ) (9)

Where, t i = travel time (hr) of overland flow from the i th overland grid to the near‐
est channel grid down the drainage path and γ = coefficient considered as constant for
a given catchment.
The travel time for grids located in a flow path to the nearest channel can be estimated if the
lengths and velocities for the flow paths are known. The direction of flow from one cell to a
neighboring cell is often ascertained by using an eight direction pour point algorithm in
grid-based GIS analysis. Once the pour point algorithm identifies the flow direction in each
cell, a cell-to-cell flow path is determined to the nearest stream channel and thus to the
catchment outlet. If the flow path from cell i to the nearest channel cell traverses m cells and
 Sediment Transport Modeling Using GIS in Bagmati Basin, Nepal 81
[Link]

the flow length of the i th cell is l i (which can be equal to the length of a square side or to a
diagonal depending on the direction of flow in the i th cell) and the velocity of flow in cell i is
v i, the travel time t i from cell i to the nearest channel can be estimated by summing up the
time through each of the m cells located in that flow path:

m li
ti = ∑ (10)
i=1
vi

In this study, the method of determination of the overland flow velocity proposed by the US
Soil Conservation Service was chosen due to its simplicity and the availability of the infor‐
mation required (SCS, 1975). The flow velocity (v i) is considered to be a function of the land
surface slope and the land cover characteristics:

vi = ai * Si b (11)

Where, b = a numerical constant equal to 0.5 [22, 5], S i = slope of the i th cell and a i = a coeffi‐
cient related to land use [8]. Introducing equations (10) and (11) into equation (9) gives

m li
DR = exp( − γ ∑ ) (12)
i=1
ai Si 0.5

It should be noted that l i / S i 0.5 is the definition of travel time used by Ferro & Minacapilli
[5]. Values of the coefficient a i for different land uses were adopted from [8] and are pre‐
sented in Table 4.
If S E is the amount of soil erosion produced within the i th cell of the catchment estimated
using equation (1), then the sediment yield for the catchment, S y, was obtained as follows:

n
Sy = ∑ DR * SE (13)
i=1

Where, n = the total number of cells over the catchment and the term D R = the frac‐
tion of S E that ultimately reaches the nearest channel. Since the D R of a cell is hypothe‐
sized as a function of travel time to the nearest channel, it implies that the gross erosion
in that cell multiplied by the D R value of the cell becomes the sediment yield contri‐
bution of that cell to the nearest stream channel. The D R values for the cells marked
as channel cells are assumed to be unity.

3. Study Area

The study area selected for this study is Bagmati Basin, Nepal. The basin is chosen because
of its bio-climatic diversity due to elevation differences from valley floors to mountain sum‐
82 Sediment Transport Processes and Their Modelling Applications

mits, and related land use changes having influence on soil erosion, which is considered
typical for the Middle Mountains of Nepal. Bagmati is the draining river from the Kathman‐
du city which is the capital of Nepal. The Bagmati basin covers an area of 3,500 km2 in total
and drains out of Nepal across the Indian State Bihar to reach the Ganges. The watershed
with the elevation ranging from 57 m to 2,913 m is situated at latitude of 26° 30’ to 28°N and
longitude 85° to 86°E. The watershed can be divided into three main areas: the upper, mid‐
dle and the lower Bagmati watershed areas (BWA). The Upper Bagmati Watershed Area
covers the whole of the Kathmandu valley including its source at Shivapuri. From the Cho‐
var gorge, the river flows into the Middle Bagmati watershed Area across the Mahabharat
and Siwalik ranges. The catchment area of upper and middle Bagmati basin is about 2,800
km2. The terrain of the upper and middle BWA is rugged and comprised of several steep
mountains except Kathmandu valley. The area of upper and middle Bagmati basin draining
to Karmaiya is considered in the study on the basis of data availability.

Figure 1. Bagmati basin, Nepal

The climate of the Bagmati watershed can be subdivided into three altitude/climate zones.
These are: (a) Subtropical sub humid zone below 1,000 m: the southern most parts of the
Bagmati watershed area including the Siwaliks region lie in this zone, (b) Warm temperate
humid zone between 1,000-2,000 m: a large part (more than 60%) of the BWA lies in warm
temperate humid zone between 1,000 – 2,000m altitudes and (c) Cool temperate humid zone
between 2,000-3,000 m: only a small portion (about 5%) of the Bagmati watershed falls above
 Sediment Transport Modeling Using GIS in Bagmati Basin, Nepal 83
[Link]

2,000 m. The annual average rainfall in the watershed is about 1,800 mm and it produces
1,400 mm of runoff per year on average, which accounts for about 75% of annual average
rainfall. In the basin, steep slope in mountainous area and land use change are the major fac‐
tors of soil erosion, which is considered typical for the Middle Mountains of Nepal. Total
population in the catchment is about 1.5 millions. Figure 1 shows the map of the catchment
along with streams and tributaries.

SN Data type Stations Location Duration Remarks

Lat. (N) Long.(E)

1 Rainfall Daman 27° 36’ 85° 05’ 1987-97 Daily

Hetauda 27° 25’ 85° 03’ 1987-97 Daily

Godavari 27° 35’ 85° 05’ 1987-97 Daily

Airport 27° 42’ 85° 22’ 1990-97 Daily

1993-97 Hourly

Nagarkot 27° 42’ 85° 31’ 1990-97 Daily

Sindhuligadhi 27° 17’ 85° 58’ 1990-97 Daily

Karmaiya 27° 07’ 85° 28’ 1990-97 Daily

2 Sediment Karmaiya 27° 07’ 85° 28’ 1990-91,93,95-97 Daily

Table 1. Description of hydrologic data set

SN Description Scale or Source Remarks

grid resolution

1 DEM 90 m (SRTM DEM) USGS Raster

2 Landuse 1:25,000 BIWMP Vector

3 Soil 1:25000 BIWMP Vector

4 Watershed boundary 1:25,000 BIWMP Vector

5 River network 1:25000 BIWMP Vector

Table 2. Description of spatial data set

Hydrologic data (rainfall, evaporation, suspended sediment concentration) for the basin are
obtained from Department of Hydrology and meteorology (DHM). Digital Elevation Model
(DEM) data, in 90 m resolution, was obtained from obtained from United States Geological
Survey (USGS) (available at: [Link] STRM DEM provides comprehensive
and consistent global coverage of topographically derived data sets, including streams,
drainage basins and ancillary layers. Other spatial data set such as: soil, land use, basin
boundary, river network are obtained from Bagmati Integrated Watershed Management
84 Sediment Transport Processes and Their Modelling Applications

Programme (BIWMP). The details of hydrologic data are provided in Table 1 while Table 2
contains the details about spatial data set.

As observed in the DEM of the watershed (Figure 2), the elevation varies significant‐
ly from as low as 137 m to as high as 2913 m from mean sea level. Lower part of
the watershed is relatively flat compared to the upper and middle part. Kathmandu,
the capital of Nepal lies in the upper part of the watershed. One third of the water‐
shed is relatively flat as 34% of the watershed area has slope in the range of 0 - 10%.
About 50% of the area has mild slope ranging from 10 - 30%. Remaining 15% water‐
shed contains high slope with slope value more than 30%.

Figure 2. Digital Elevation Model (DEM) of Bagmati basin (SRTM DEM, 90 m resolution)

The land use in the watershed is observed to be mixed type. Cultivated land is ma‐
jor land use pattern in the upper part of the watershed while in middle and lower
part of the watershed, forest area is seen to be dominant land use type. Majority of
built-up area falls on the upper part of watershed, which represents Kathmandu. The
land use pattern in the watershed is presented in Figure 4. 5. More than half of the wa‐
tershed area (58%) is covered by forest. Cultivated land accounts for 38% of the area
of the watershed while nearly 4% of the land in the watershed is barren. The land use
distribution in the watershed is presented in Figure 3. The most extensive soils in the
area are Dystrochrepts, Hapludalfs and Haplumbrepts, which occupy most of the hilly
and mountaineous land. The texture of these soils is sandy/loamy in nature that var‐
 Sediment Transport Modeling Using GIS in Bagmati Basin, Nepal 85
[Link]

ies from sandy clay to loam. The Dystrochrepts are also the most important soils in
the inner Terai valleys. Soil type Rhododtalfs is commonly found in the gently undulat‐
ing slopes and restricted to scattered, quasi-subtropical areas in the lower Hiamlayas.
These soils are prone to severe soil erosion. The soil in the south face on the low alti‐
tude Mahabharat range is Dystrochrepts and Hapludalfs. These soils are mostly cultivat‐
ed. The Haplaquepts are the dominant soils in the Terai plain as well as on paddy
fields in hilly areas and elsewhere. Major soil types in the mountainous lands are Ha‐
plumbrepts and Dystrochrepts. Loamy soil texture is dominant in the watershed as dem‐
onstrated in Figure 4.

Figure 3. Land use distribution in Bagmati basin

Figure 4. Soil map of Bagmati basin

86 Sediment Transport Processes and Their Modelling Applications

Data Preparation And Simulation

Revised Universal Soil Loss Equation is one of the simplified models, which predicts soil
erosion from hillslopes. The factors such as rainfall runoff erosivity factor (R) associated
with the model represent the effects of climatic parameters in soil erosion while soil erodibil‐
ity factor (K) represents the nature of the soil, its characteristics and influence in soil erosion.
Topography and land use practices are other major factors incorporated in the model to ac‐
count their effects in soil erosion.

Out of seven rainfall stations in Bagmati basin, one station measures hourly rainfall while
remaining six other stations measures daily rainfall. So, rainfall data from these seven sta‐
tions are analyzed to find the correlation in the rainfall pattern. The analysis of daily,
monthly and annual rainfall trends of these stations showed that the trend was similar for
all these stations. This helped to in disaggregating daily rainfall data into hourly data for the
remaining six stations. For the basin, rainfall erosivity index “R” value was computed for
monthly basis R value for was computed using equations (2), (3) and (4) since sediment
yield information was available on monthly basis. Soil erodibility (K) factor values were as‐
signed on grid by grid basis on the basis of soil texture [21] of the basin and assigned K
values are presented in Table 3. Topographical parameters (L, S) were extracted from 90 m
resolution SRTM digital elevation model (DEM) obtained from USGS ([Link]
Equation (5) was used for L factor calculation while S factor was computed using equation
(8) for each cell. Similarly, C value, which depends on land use, was obtained from different
literature [21], [16]. The C values used assigned for different land use in the basin are tabu‐
lated in Table 4. In case of P factor, the value is taken 0.5 for agricultural land and for rest of
the land use; P value is assigned to be 1.

Textural Class Organic matter content (%)

0.5 2 4

Fine sand 0.0211 0.0184 0.0132

Very fine sand 0.0553 0.0474 0.0369

Loamy sand 0.0158 0.0132 0.0105

Loamy very fine sand 0.0580 0.0501 0.0395

Sandy loam 0.0356 0.0316 0.0250

Very fine sandy loam 0.0619 0.0540 0.0435

Silt loam 0.0632 0.0553 0.0435

Clay loam 0.0369 0.0329 0.0277

Silty clay loam 0.0487 0.0422 0.0343

Silty clay 0.0329 0.0303 0.0250

Table 3. Soil Erodibility factor by soil texture in SI unit (ton ha hr MJ-1 ha-1 mm-1)
 Sediment Transport Modeling Using GIS in Bagmati Basin, Nepal 87
[Link]

SN Land Use C value basis C Value a value

1 Cultivation Crops, disturbed land 0.4000 1.55

2 Water body Depositional sinks 0.0001 3.08

3 Forest Forest 0.0020 0.76

14 Barren land Fallow 1.0000 3.08

15 Built-up area Paved, occasional construction 0.0005 6.19

Table 4. Cover management factor (C) on the basis of land use

Results And Discussion

"Once RUSLE parameters for Bagmati basin was computed following the procedure out‐
lined alobe, sediment delivery ratio (SDR) map for Bagmati basin computed using Equation
(12)". The SDR map for the basin is presented in Figure 5 below. It is observed that flat areas
around the south and north parts of the watershed has low sediment delivery ratios while
the hilly areas within the watershed had higher values for sediment delivery ratio. This find‐
ing is consistent with the fact that steep areas are supposed to have higher sediment deliv‐
ery ratio compared to flat areas. In terms of watershed management perspective, the areas
with higher values of SDR should be given higher priority compared to areas with lower
SDR values for implementation of erosion control measures in this watershed.

Figure 5. Sediment Delivery Ratio (SDR) map for Bagmati Basin

The sediment yield data are available for only few months of the year for Bagmati basin. So, it
was not possible to analyze the long term sediment yield value and thus, monthly computa‐
88 Sediment Transport Processes and Their Modelling Applications

tion is carried out. Soil erosion map and SDR map was used to compute the sediment yield val‐
ue at the watershed outlet. The Observed monthly sediment yield was compared with the
computed as seen in Figure 6 below. The simulated result using this approach is fairly consis‐
tent with the observed data although this methodology slightly overpredicted sediment yield
for the most of the observed months. There can be several reasons which can lead to overesti‐
mation of sediment yield values. For example, only one rainfall station had hourly measure‐
ment while remaining stations recorded daily values. It was assumed that the rainfall pattern
over the watershed was similar. If rainfall data with finer temporal resolution were available
for all the stations, the computed of R value would have been more reliable.

Figure 6. Comparison of simulated and observed sediment yield

The comparison of observed and computed sediment yield also indicate that great care is
required in the selection of input values for the rainfall (R) and soil erodibility (K) factors.
The USLE model was developed from the data suing the experiments that were carried out
on a standard plot of 22.1 m length of uniform 9% slope. So, USLE-based performance can
expected to be better for finer (for example 30 m) DEM resolution. Earlier studies have dem‐
onstrated that DEM resolutions can affect the outcome of RUSLE based simulations and bet‐
ter agreement can be obtained using fine DEM resolution [4]. Similarly, RUSLE results may
be improved if more detailed soil, land use/cover data are available.

The model prediction may have been improved if γ coefficient was calibrated using the
measured sediment yield values at mean annual scale for SDR computation. During SDR
calculation, the sensitivity analysis of the parameter γ showed that the computed Sy was not
very sensitive to γ in equation (12). The variation of γ value by 15 times (from 0.1 to 1.5)
changed the Sy value only 10%. Since large variation in γ affected Sy insignificantly during
sensitivity analysis, γ value was taken as 1 in the computation for simplicity. The sensitivity
analysis has supported the findings of Jain & Kothyari [10] where they had reported that Sy
was not very sensitive to γ in their study.

Conclusion
Soil erosion is a natural process. Modeling a natural process using mathematical simulation
involves use of complex relationships. The number of factors associated with such complex
process imposes their effect in various degrees. It is, thus, essential to consider only those
factors, which are likely to have dominant effects in the process while carrying out mathe‐
 Sediment Transport Modeling Using GIS in Bagmati Basin, Nepal 89
[Link]

matical simulation. This simplifies the process and is acceptable in most cases. Universal Soil
Loss Equation (USLE) (and its revised form, RUSLE) is one of such simulation model, which
predicts soil erosion from hillslopes. The factors such as rainfall runoff erosivity factor (R)
associated with the model represent the effects of climatic parameters in soil erosion while
soil erodibility factor (K) represents the nature of the soil, its characteristics and influence in
soil erosion. Topography and land use practices are other major factors incorporated in the
model to account their effects in soil erosion.

This study is an attempt to estimate soil erosion and sediment yield at Bagmati River basin us‐
ing existing conceptual methods and GIS. This methodology can be used for the identification
of sediment source areas and prediction of sediment yield at a catchment scale with available
optimum data sets. ArcGIS was used for discretizing the catchment into grid cells of different
resolutions. Grid cell slope, drainage direction and catchment boundary were generated from
DEM using pour point method. The DEM was further analyzed to classify the grid cells into
overland flow and channel region by using channel initiation threshold area approach. After
preparing different USLE parameter layers, the gross surface erosion map was computed. The
sediment delivery ratio of overland flow cell was assumed to be a function of the travel time of
overland flow from given cell to the nearest downstream channel cell. For channel cells, the
sediment delivery ratio was assumed to be unity. The computed and observed values were ob‐
served to have some discrepancy for monthly sediment yield. The variation is resulted by the
few assumptions made during the analysis. In the study, computation of soil erodibility value
(K) was based on soil texture only. Similarly, constant cover management factor (C) values
were used instead of time varying because of the lack of series of land-use map for different
years. Use of finer resolution DEM can also improve the estimation of slope length (L) and
Slope steepness (S) factor. Improved results can be expected if these enhancements are incor‐
porated. The proposed modeling framework is simple and can be a useful tool in conservation
planning with reasonable reliability at data scarce areas.

Acknowledgements

The author would like to thank Department of Hydrology and Meteorology, Nepal and Bag‐
mati Integrated Watershed Management Programme for providing the data used in this
study. The present work benefited from the input of Dr. Dushmanta Dutta (CSIRO, Austral‐
ia), who provided valuable comments and assistance to the undertaking of the research
summarised here.

Author details

Rabin Bhattarai

Department of Agricultural and Biological Engineering, University of Illinois at Urbana-

Champaign, Urbana, IL 61801, USA
90 Sediment Transport Processes and Their Modelling Applications

References

[1] Arnold, J. G., Srinivasan, R., Muttiah, R. S., & Williams, J. R. (1998). Large area hydro‐
logic modeling and assessment part I: model development. Journal of American Water
Resources Association, 34(1), 73-89.
[2] Beasley, D. B., Huggins, L. F., & Monke, E. J. (1980). ANSWERS: A model for water‐
shed planning. Trans. ASAE, 23(4), 938-944.
[3] Beven, K., & Freer, J. (2001). Equifinality, data assimilation, and uncertainty estima‐
tion in mechanistic modelling of complex environmental systems using the GLUE
methodology. Journal of Hydrology, 249(1-4), 11-29.
[4] Bhattarai, R., & Dutta, D. (2007). Estimation of soil erosion and sediment yield using
GIS at catchment scale. Water Resources Management, 21(10), 1635-1647.
[5] Ferro, V., & Minacapilli, M. (1995). Sediment delivery processes at basin scale. Hy‐
drol. Sci. J., 40(6), 703-717.
[6] Ferro, V. (1997). Further remarks on a distributed approach to sediment delivery. Hy‐
drol. Sci. J., 42(5), 633-647.
[7] Ferro, V., & Porto, P. (2000). Sediment delivery distributed (SEDD) model. J. Hydrol.
Engng ASCE, 5(4), 411-422.
[8] Haan, C. T., Barfield, B. J., & Hayes, J. C. (1994). Design Hydrology and Sedimentolo‐
gy for Small Catchments. Academic Press, New York.
[9] Hadley, R. F., Lal, R., Onstad, C. A., Walling, D. E., & Yair, A. (1985). Recent Devel‐
opments in Erosion and Sediment Yield Study. UNESCO (IHP) Publication, Paris,
France.
[10] Jain, M. K., & Kothyari, U. C. (2000). Estimation of soil erosion and sediment yield
using GIS. Hydrol. Sci. J., 45(5), 771-786.
[11] Kinsel, W. G. (1980). CREAMS: A Field Scale Model for Chemicals, Runoff, and Ero‐
sion From Agricultural ManagementSystems. U.S. Department of Agriculture, Conser‐
vation Report, 26, 640.
[12] Kothyari, U. C., & Jain, S. K. (1997). Sediment yield estimation using GIS. Hydrol. Sci.
J., 42(6), 833-843.
[13] Lal, R. (1990). Soil Erosion in the tropics, principles & management. New York.
McGraw-Hill Inc.
[14] McCool, D. K., Brown, L. C., Foster, G. R., Mutchler, C. K., & Meyer, D. (1987). Re‐
vised slope steepness factor for the universal soil loss equation. Trans. ASAE, 30,
1387-1396.
[15] McCool, D. K., Foster, G. R., Mutchler, C. K., & Meyer, L. D. (1989). Revised slope
length factor for the Universal Soil Loss Equation. Trans. ASAE, 32, 1571-1576.
[16] Morgan, R. P. C. (1995). Soil erosion & conservation. Longman, UK.
[17] Morgan, R. P. C., Quinton, J. N., Smith, R. E., Govers, G., Poesen, J. W. A., Auers‐
wald, K., Chisci, G., Torri, D., & Styczen, M. E. (1998). The European Soil Erosion
 Sediment Transport Modeling Using GIS in Bagmati Basin, Nepal 91
[Link]

Model (EUROSEM): A dynamic approach for predicting sediment transport from

fields and small catchments. Earth Surface Processes and Landforms, 23, 527-544.
[18] Musgrave, G. W. (1947). The quantitative evaluation of factors in water erosion- a
first approximation. Journal of Soil and water conservation, 2(3), 133-138.
[19] Nearing, M. A., Foster, G. R., Lane, L. J., & Flinkener, S. C. (1989). A process based
soil erosion model for USDA water erosion prediction project technology. Trans.
ASCE , 32(5), 1587-1593.
[20] Renard, K. G., Foster, G. R., Weesies, G. A., Mc Cool, D. K., & Yoder, D. C. (1996).
Predicting Soil Erosion by Water: A guide to conservation planning with the Revised
Universal Soil Loss Equation. Agricultural Handbook 703. Agricultural Research Services.
US Department of Agriculture.
[21] Schwab, Glenn. O., Frevert, Richard. K., Edminster, Talcott. W., & Barnes, Kenneth.
K. (1981). Soil and water conservation engineering. John Willey & Sons, New York,
USA.
[22] Soil Conservation Service. (1975). Urban hydrology for small watersheds. USDA
Technical release [55].
[23] Walling, D. E. (1988). Erosion and sediment yield research-some recent perspectives.
J. Hydrol., 100, 113-141.
[24] Williams, J. R., & Berndt, H. D. (1977). Sediment yield prediction based on watershed
hydrology. Trans. ASAE, 20(6), 1100-1104.
[25] Williams, J. R., Jones, C. A., & Dyke, P. T. (1984). The EPIC model and its application,
Proceeding of ICRISAT-IBSNAT-SYSS Symposium on Minimum data sets for Agro‐
technology Transfer. March 1983, Hyderabad. India.
[26] Wischmeier, W. H., & Smith, D. D. (1978). Predicting rainfall erosion losses-a guide
to conservation planning. US Department of Agriculture, Agricultural Handbook 537.
[27] Woolhiser, D. A., Smith, R. E., & Goodrich, D. C. (1990). KINEROS, A kinematic run‐
off and erosion model: documentation and user manual. ARS-77, USDA-Agricultural
Research Service.
[28] Young, R., Onstad, C. A., Bosch, D. D., & Anderson, W. P. (1987). AGNPS: Agricul‐
tural Non-Point Source Pollution Model: awatershed analysis tool. USDA-Agricul‐
tural Research Service. ConservationResearch Report 35, U.S. Department of Agriculture,
Washington, D.C.
[29] Zingg, R. W. (1940). Degree and length of land slope as it affects soil loss in runoff.
Agriculture Engineering, , 21(2), 59-64.
 Section 2

Sediment Dynamic Processes

 Chapter 5

Composition and Transport Dynamics of Suspended

Particulate Matter in the Bay of Cadiz and the Adjacent
Continental Shelf (SW - Spain)

Mohammed Achab

Additional information is available at the end of the chapter

[Link]

1. Introduction and objective

The quantification of suspended particulate matter (SPM) and the investigation of its dy‐
namics are of major importance to understand sediment transport dynamics and many
land-shelf-ocean interaction processes [1, 2, 3, 4, 5, 6]. The analysis of SPM transport process‐
es in the marine environment requires a simultaneous study of water masses dynamic and
movement, the direction and intensity of the currents as well as the characteristics of the
suspended matter and the characteristics of the bottom sediments [7, 8, 9]. The application
of remote sensing techniques to the study of suspended matter dynamics allows model for
marine and coastal water circulation, based on the use of "turbidity patterns" as natural trac‐
ers; relating parameters of water quality to satellite images [10, 11, 12]. The utility of satellite
images lies in the high frequency with which data can be taken on a point of the earth. This
allows a large volume of information for various meteorological situations, that would be
very difficult to obtain using conventional sampling methods. However, images must be
calibrated and evaluated using "in situ" data [13].

Studies on the behaviour of suspended particulate matter have been made by many ocean‐
ographers and sedimentologists [14, 15, 16]. Several authors have studied the mineralogy of
suspended matter, in order to investigate the influence of tides in estuarine systems and the
relationship between the minerals in suspension and those deposited [17, 18]. Others studies
have been focused on the possibility of using clay minerals in suspension as tracers, allow‐
ing to follow the progression of the river flows path in marine environment [19, 20, 21]. In
the Gulf of Cadiz, Several works have been realized with the objective of determining the
suspended matter content and their influence on the recent marine sedimentation [8, 22, 23].
96 Sediment Transport Processes and Their Modelling Applications

Studies about the dynamics of fine sediments and clay minerals in the bay of Cadiz and the
adjacent marine deeper zones have been approached [24, 25, 26]. The paths by which fine
sediment are transported from different source areas to the marine environment have been
deduced, using clay minerals as dynamic tracers [27, 28]. Others studies have been focused
on the sedimentary exchange dynamics between inner area of Cadiz bay and the continental
shelf [29, 30, 31]. The main objective was the determination of the nature and origin of the
suspended particulate matter in the bay of Cadiz, as well as the hydrodynamic behaviour of
turbidity plumes and the dispersal of SPM and its effects on the inner continental shelf. The
proposed chapter is based on data of suspended particulate matter concentration, minera‐
logical compositions, degrees of turbidity, Landsat images analysis and complemented by
grain sizes and hydrodynamic data. The combined study of dispersal of suspended sedi‐
ment and degrees of turbidity by analyzing Landsat images, allows to recognize the trans‐
port paths followed by the fine sediments in the surface marine waters from the inner areas
of Cadiz bay to the external zones and the adjacent continental shelf.

2. Area description

The study area is located at the Southwest of the Iberian Peninsula, between the mouth of
the Guadalquivir River and the Trafalgar Cape (Fig.1).

Figure 1. Geographical setting and location of surface sediment and water samples in the study area.

The Bay of Cadiz is about 28.5 km long and 13.5 km wide. Three sedimentary environments
are distinguished: The outer bay (surface of about 118km2), located to the north, is divided
into two zones, a western and an eastern one, with presence of rocky outcrops in the North
(Rota) and south (Cádiz), resulting slopes of 2°. The outer bay, is well connected to the conti‐
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 97
[Link]

nental shelf, and is very affected by the waves, currents and storms. The inner bay (Surface ≈
40 km2) or Lagoon system located to the South and protected from waves and storms of the
West and Southwest. Characterized by shallow water (<5 m depth and slope of 0.15 °), ex‐
cept in the navigation channel connecting the inner and outer bay (12 m depth and slope of
2.2 °). The salt marshes and tidal flat (Surface ≈ 227km2), occupy the most internal and shel‐
tered areas, drained by a complex system of tidal creeks and channels of great importance in
the hydrodynamic. This wide marshes zone occupied by halophyte vegetation is isolated
from the open sea by sandy beach ridges and littoral spits.

2.1. Oceanographic setting

The zone is affected by a mesotidal regime, where the mean tidal range is 2.39 m. In spring
tides the highest range reaches 3.71 m and in neap tides the lowest range is 0.65 m [32]. In
the Bay of Cadiz, the bottom morphology influences the behaviour of the tidal wave, caus‐
ing a time delay of 12 minutes between the outer and the inner bay. The speed of the tidal
current is highly variable, with the highest values reached inside the Bay of Cadiz, where
the highest speeds have been determined along the Strait of Puntales [33], reaching values
over 1.5 m/s during spring tide ebbs. The input and output flows, are controlled by the tidal
regime. The tidal currents oriented SW-NE are directed into the bay, while the ebbs (NW-
SE) outwards [34, 35]. They are responsible of fine sediment transport from inner zones to‐
ward the external bay and continental shelf [36]. Wind and waves action are also essential
factor in the sedimentary dynamics. Waves present seasonal character and the storm aver‐
age frequency is of 20 days/year. The strongest storms occur in the fall-winter season and
the calm periods during the summer. The prevailing swell is from the west. The data from
the point WANA 1054046 in the WANA network, show that the waves from the W and SW
represent 70% of the time for the period 2006- 2012 (Fig. 2). Storm waves are related to
southwest Atlantic storms [37]. Mean significant wave height (H1/3) is 0.85 m and represent
45%. The most common periods range from 3 to 4 seconds. The maximum values of H1/3
reached during storms are 4 m. with periods of 6.24 s [38].

The main littoral-drifts spreads towards the southeast, generating transport in the same di‐
rection because of the coastal orientation, facing westerly and SW winter storms. Easterly
winds also generate littoral-drifts towards the North and NW. In the continental shelf, the
hydrodynamic regime is controlled by the North Atlantic Surface Water (NASW) moving
towards the southeast, and is responsible for the dispersal of fine sediments from the Gua‐
dalquivir and Guadiana Rivers (annual flux of 9200km3 and 5500 km3 respectively) [39,40,
27, 28]. The Mediterranean Outflow Water (MOW) moving west to deeper water [41, 42] and
do not have an influence on the present day sedimentation in our study area.
98 Sediment Transport Processes and Their Modelling Applications

Figure 2. Significant height and peak period rose in the study area (Wana point 1054046).

2.2. Physiographic and Geologic framework

The coastline physiography is oriented NNW-SSE, with some sectors facing E-W that give
the coast a stepped outline, strongly controlled by tectonic fractures [43]. The influence of
tectonic structures and coastal morphology constitute an essential factor in the hydrody‐
namic control of the sedimentation and distribution of facies in the bay of Cadiz and the
continental shelf [44, 45]. An Early Quaternary compressive tectonic episode has been de‐
duced from reverse faults observed in marine sediments of the continental margin [46]. The
continental shelf has a gentle slope and a slight inclination toward the west, with an average
width of 40 km and is oriented from NNW-SSE with NNE-SSW sectors. The physiography
of the sea-bottom shows a close concordance with the shoreline, the isobaths generally run‐
ning parallel to the coast. The slope break occurs at 150-200 m water depth and shows signif‐
icant variation north to south in cross section [26, 47]. The most important geological
formations present in the surrounding areas of Cadiz bay are mainly pre-orogenic units
from the Betic Mountain Range. Other units outcropping are Post-orogenic formation from
the Neogene Guadalquivir basin. Upon all those materials, appear Quaternary deposits con‐
stituted by muddy marshes, beach sands and continental deposits [48, 49, 50, 51, 52, 53, 54].
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 99
[Link]

3. Materials and methods

The study of concentration of suspended particulate matter (SPM) was based on the analysis
of 36 water samples obtained in different zones of the study area (tidal creeks and channels,
river mouths, lagoon, etc). The study of surface sediments was carried out on 250 samples
collected from different sectors of the bay of Cadiz and the adjacent continental shelf (Fig.1).
The sample position was determined by Differential Global Position System (DGPS) and er‐
rors made in the horizontal DGPS positioning were verified as being less than 2m. The ex‐
traction of surface sediments was performed by means of Van Veen dredge and gravity
cores. The water samples were taken 3hours after high tide, wind from the north and north‐
east, average speed of 55km/hr and mean wave height of 0.6 and maximum of 1.5m. The ex‐
traction of samples was executed to specific depths with oceanographic bottles,
simultaneously with the passage of the Landsat satellite over the study zone. The purpose of
this operation is to obtain a synoptic picture of the turbid flumes, by comparing the data ob‐
tained from direct methods (water samples) and those of indirect methods (satellite images).
The separation of SPM has been achieved following the method of [55], which consists to the
filtration of a volume of five litters of water through pre-weighed filters by MILLIPORE
(0.45 microns). Filters were washed with distilled water, dried at about 60° and weighed.
The <2µm fraction was separated by a standard sedimentation method [56, 57]. The use of
satellite images in this study is based on the utilization of inorganic SPM as a natural tracer.
Satellite images of the Bay of Cadiz have been recorded by the satellite TM Landsat, using
bands 2 and 5, and a spatial resolution of 30x30 m. Landsat images has been analysed to ob‐
tain extent and direction of turbidity plumes in several hydrodynamic situation in Cadiz
bay and inner shelf waters. The process of the images has been carried out according to the
methodology described by [58].

Systematic granulometric and mineralogical analyses were carried out to establish facies
distribution and mineralogical composition. The collected samples (approximately 250 g
of surface sediments) were placed in plastic bags sealed and identified. The Grain size of
coarse fraction was determined by dry sieving sediments, using sieve column ranging
from 4 Phi (0.063 mm) to -2 Phi (4mm). The fine fraction analysis was made by use of la‐
ser diffraction analyser (AMD). Grain size data were processed by GRADIST software
(version 4.0). The characteristic statistic indexes and parameters were calculated using
standard method [59, 60]. The mineralogical analysis of suspended particulate matter
(SPM) and surface sediment was made through X-ray Diffraction techniques (XRD) using
a Philips PW-1710 diffractometer, equipped by Cu-Kα radiation, automatic silt and graph‐
ite monochromator. Quantification of different mineralogical phases was calculated by the
classic method of area measurement of peaks, considering the different reflection capaci‐
ties of the minerals [61, 62]. Factor Analysis (Principal Components Analysis, PCA) was
used to determine the mineral assemblages and to establish possible sediment transport
paths from the bay toward the continental shelf.
100 Sediment Transport Processes and Their Modelling Applications

4. Results

4.1. Surface Sediment Characteristics

Grain size analysis of modern sediments of Cadiz bay bottoms show that samples are main‐
ly composed of sand and mud, and subordinate amounts of gravel. The grain size distribu‐
tion shows the predominance of the coarser fractions in the outer bay, more exposed to
wave action and currents. The finer fractions appear in more sheltered zones of the bay.

Taking into account the textural characteristics of the marine deposits and the sedimentary
environments, we can differentiate various types of sediment (Fig. 3).

Figure 3. Surface sediments distribution on the Bay of Cadiz and transport paths of fine particules.
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 101
[Link]

The sand is the dominant fraction of the outer bay sediments, with an average of 75% espe‐
cially in the littoral zone. In some areas, this fraction changes laterally to muddy sand facies;
due to the recent action of transport processes taking place in this area of the Cadiz bay. This
facies extend into the 20-30 m deep inner shelf, being configured in two bands, one by the
north margin of the bay and another one more to the South, bordering the city of Cadiz. The
silt appears in low proportions (10%), giving the highest values in some sectors of the outer
bay and the adjacent continental shelf. It distribution is of great importance to understand
the modern sediment dynamics in the Bay of Cadiz, especially the sediment exchange be‐
tween the inner and outer bay. The Clays are an important sedimentary fraction, especially
in low-energy sedimentary environments where concentrations reach 90% such as inner bay
and the tidal channels that drain the salt marshes. In the outer bay, the contents are very low
(<5%). The gravels (less than 5%), are mainly composed of bioclasts and rock fragments, de‐
rived from erosion of rocky shoals and coastal cliffs.

Deposit environment Bay of Cadiz and inner shelf Salt marshes and tidal creeks

Clay minerals Factor 1 (99% of the variance) Factor 1 (100% of the variance)

Illite 2.15 2.16

Smectite 0.31 0.25

Interstratified I-S 0.28 0.19

Chlorite 0.29 0.38

Kaolinite 0.29 0.28

Mineral assemblage I"/"/Sm"/Cl"/K"/"/I-Sm I"/"/Cl"/K"/Sm"/I-Sm

Table 1. Factor scores of the clay minerals in the bay of Cadiz-inner shelf and the salt marshes- tidal creeks, Q-mode
factor analysis.

Figure 4. Clay mineralogy of surface sediments in the bay of Cadiz and the salt marsh.
102 Sediment Transport Processes and Their Modelling Applications

In what concerns the mineralogical composition, the main minerals in terrigenous sediments
of the bay of Cadiz are quartz (55%), calcite (20%) and plagioclases (5-10%). The mineralogi‐
cal analysis of clay fraction indicate the predominance of Illite (60%), followed by Smectite
(13%), interstratified Illite-smectite (10%), kaolinite (8%) and Chlorite (7%) (Fig. 4). In the salt
marshes, the most abundant clay mineral is Illite (66%). Other clay minerals found in lesser
quantities are Kaolinite (9%) and chlorite (10%) especially in the high tidal zone, and smec‐
tite (7%) in tidal creeks and tidal channels bottoms. Factor analysis was used to establish the
relationships between different clay minerals and their associations. The Q-mode factor
analysis results are summarized in Table 1.

4.2. Suspended sediment concentration

In the study area, significant concentrations of suspended matter are frequently observed,
due to the existence of different sources of fine sediments (river mouths,e.g Guadalquivir
and Guadalete rivers, tidal channels, etc.), and the action of the hydrodynamic regime
(Winds, waves and tidal currents). The concentration of suspended particulate matter under
the hydrodynamic conditions at the sampling time (winds of N and NE, average speed of
55km/hr) shows an average dry weight content of 6.5 mg/l (Fig.5). The highest values were
found in the outer bay, especially in the oriental sector, near the Guadalete and San Pedro
river mouths (16 and 25 mg/l respectively) and to the south in Sanctipetri tidal creek
(13mg/l). The concentrations of SPM are also relatively high in the central part of the bay,
particularly north of the city of Cadiz (14mg/l) and in the navigation channel (12.87 mg / l)
connecting the inner and the outer bay of Cadiz. The higher concentrations of SPM are con‐
sistent with the pattern of tidal currents and the distribution of fine facies on the sea bottom.
The lowest values (1.5-5 mg/l) appear in the adjacent continental shelf waters characterized
by low sediment input, as well as in parts of the inner bay less affected by tidal currents.

Figure 5. Distribution of suspended sediment concentration (in mg/l of dry weight) in the surface waters of the study
area (see Fig. 6 for location of water samples).
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 103
[Link]

4.3. Clay minerals composition of SPM

In the present study, clay minerals have been used as tracers, due to their small size making
them the only particles susceptible of being transported away from the bay, out towards the
continental shelf; and can be useful in the determination of the pathways of fine sediments and
the suspended matter. Analysis of the mineralogical composition of suspended matter shows
that the most abundant clay mineral consists mainly of illite (41%) followed by smectite (38%),
chlorite (11%), interstratified llite-smectite (7%) and kaolinite (6%) (Fig.6 & table.2).

Clay minerals Outer bay Inner bay Tidal creeks Continental shelf

Illite 37 43 43 38

Smectite 36 40 35 40

Interstratified I-S 6 9 5 10

Chlorite 12 15 14 9

Kaolinite 7 5 6 7

Table 2. Means values in percentage of clay minerals in SPM of different deposit environments.

Figure 6. Distribution of principal clay minerals of suspended particulate matter in the study area.
104 Sediment Transport Processes and Their Modelling Applications

The relations between the clay minerals were established by Q-mode factor analysis. The
main clay minerals assemblage obtained in suspended matter is I>S>Cl>I-S>K (Fig. 7). The
clay minerals distribution indicates high concentration of illite in the inner and protected
part of Cadiz bay, while in the adjacent marine zones predominates the smectite.

Figure 7. Factor scores (factor1: 99% of the variance) of the clay minerals in SPM obtained by Q- mode factor analysis.

The R-mode factor analysis results provide three factors explaining the 100% of the total var‐
iance. Factor 1 (52%) groups exclusively smectite (negative loading: -0. 95), is well represent‐
ed on all bottoms of the bay of Cadiz. Factor 2 (28%) associates interstratified IS (positive
loading:0.89) to kaolinite (0.4). This factor scores better in the western sector of the inner bay
and central part of the external as well as in front of the river mouths of the Guadalete and
San Pedro. Factor 3 (20%) with only the mineral chlorite (0.95), dominates next to the
mouths of the Sancti Petri tidal creek and the San Pedro and Guadalete Rivers.

4.4. Landsat images analysis

The use of satellite images in studies of transport dynamics of sediment in suspension is of

great interest. The usefulness of this technique is based on the monitoring of turbidity
plumes, consisting mainly of suspended inorganic particles. Geologists, geographers and
oceanographers, also applied this technique, for coastal processes study [10].

Landsat TM images used in this study are property of the Spain Ministry of Environment
(Junta de Andalucía). The analyses of these images for different hydrodynamic and mete‐
orological situations illustrate the existence of water masses of different degrees of turbidity,
oriented from the inner zone towards the outer bay extending to the continental shelf (Fig.
8). The highest turbidity is observed in the coastal areas of the bay, as well as in front of the
mouths of the Sancti Petri tidal creek and the San Pedro and Guadalete Rivers.
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 105
[Link]

Figure 8. Turbidity plume trayectories in several hydrodynamic situations from the analysis of Landsat TM images of
Cadiz bay. 8a: date: 22-05-1996, Wind: Western-WWS, Speed: 39km/h, Hs: 0.6m, Tidal coef: 0.55 ; 8b: date:
13-08-1991, Wind: East, Speed: 37km/h, Hs: 0.54m, Tidal coef: 0.78; 8c: date: 07-06-1996, Wind: North-East, Speed:
55km/h, Hs: 0.6m, Tidal coef: 0.64; 8d: recent satellite image during ebb tide.

In situation of strong Easterly winds (Fig. 8a), suspended matter derived from resuspension
of the inner bay bottoms are subsequently transported by ebb tidal currents, with an aver‐
age speed of 1.02 m /s, which is reduced to 0.77 m / s in the outer bay. These turbid flumes
(SPM flows) appear as branch oriented northwards along the eastern edge of the bay and
then turn west, influenced by coastline and bottom morphology. Part of the suspended mat‐
ter tends to leave the bay being deposited in the inner continental shelf. In the continental
106 Sediment Transport Processes and Their Modelling Applications

shelf, the North Atlantic Surface water current transport a large volume of fine sediments
from the Guadalquivir rivers toward the SE. Part of this current can transport fine particles
in suspension toward the bay of Cadiz. Satellite image obtained in situation of Northeast
winds, shows the existence of several turbidity flows that appear to move from the inner‐
most zones of the bay to the outer one (Fig 8b). These flows are configured in three bands:
one oriented towards the NNW following the north margin of the bay. Other band oriented
towards the West and the third one goes towards the SSW bordering the Cadiz city. To the
south of the bay, turbidity plumes are also observed and coming from the Sanctipetri tidal
creek mouth that head towards the inner continental shelf.

On the other hand, during Western (SSW) Wind Conditions, turbidity in open coastal areas
adheres to breaking waves zone, being highly concentrated within 20 meters near the beach
and at depths less than 2 meters. outcrops rock located to the NW of the city cause, the exis‐
tence of turbidity plumes in this area (Fig. 8c).

The observation of recent satellite image during ebb tide shows that turbid flumes extend
from the inner bay out to the continental shelf, following the coastline and sea bottom
morphology (Fig. 8d).

5. Discussion

5.1. Sedimentary processes

The distribution and transport of sediments on the marine environment is function of the
wave-current and grain size interactions [63, 64]. They can also reflect the direction of water
mass movement [65, 66]. In Cadiz bay bottoms, the grain size distribution, shows the pre‐
dominance of fine fractions, (silt and clay), toward the more sheltered and internal zones of
the bay. While the coarse fractions (sand and gravel), appear in external zones, more opened
to the sea and exposed to waves and currents action. Grain size parameters show the preva‐
lence of unimodale distribution (fine and very fine sand), however bimodal and polymodal
distributions are present. The mode values increase in coastal areas and near rocky shoals,
decreasing towards the central zone of the outer bay and in the inner bay. The general trend
and the variability of different grain-size parameters, reflect the control that physiographic
and hydrodynamic factors exert on different types of sediment [31].

The abundance and the progressive deposition of fine materials in the inner bay and the
salt marshes is related to the existence of sedimentary environment of very low hydrody‐
namic energy. Their hydrodynamic regime is dominated by tidal currents, especially ebb
tides and wind of the East sector [30, 36]. These fine sediments are also found in the out‐
er bay, occupying the central and eastern sectors, covering sandy bottoms. They derive
from resuspension of fine-grained materials in the marshlands of the bay and from fine
materials supplied by the Guadalete River during periods of rainfall and floods. The pres‐
ence of mud and sandy mud facies covering sandy bottoms, possibly will indicates actual
processes of deposit and transport of fine sediments from the inner bay toward the exter‐
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 107
[Link]

nal zone reaching the inner continental shelf (Fig.3). Muddy facies are also present in the
continental shelf as a prodeltaic muddy zone situated to the north and deposited in low
energy environment. These fine-grained sediments are related to supplies coming from
the Guadalquivir River [26, 28, 67].

The grain size distribution allowed the differentiation of particles transport modes, ac‐
cording to its size and the hydrodynamic conditions. The suspended transport predomi‐
nates in the inner part of the Bay (including tidal channels and marshland areas) and to
the north of the continental shelf characterized by mud-clay bottoms deposited in low en‐
ergy conditions. In general, this type of transport predominates on many marine environ‐
ments including continental shelves [68]. In the outer bay, especially the western sector
that coincide with sandy-mud and mud facies bottoms, the saltation-suspension transport
mode predominates and may possibly indicate the transport paths of fine sediments and
suspended matter in the study area. The distributions of these different grain size classes
reflect the energy level of each depositional environment and the processes of sediment
transport; as well as the action of hydrodynamic agents, which control recent sedimentary
dynamics between different sedimentary environments of the bay of Cadiz and the conti‐
nental shelf [33, 44,69].

5.2. Suspended matter dynamics

On coastal marine environments, rivers are the major sources of supplies of suspended par‐
ticulate matter. Most of great rivers discharge important quantities of SPM, carried by long‐
shore currents parallel to the coast, forming permanent turbidity plume. The spatial
distribution of suspended sediments is a consequence of hydrodynamic forcing acting on
the unconsolidated sediments of the shelf and the coastline [70, 71]. In general, waves are
more important for resuspension in shallow water (< 10 m) and currents become more im‐
portant in deeper water (> 10 m) [72, 73, 74].

The measures of concentration of suspended matter obtained in the Bay of Cadiz and the
continental shelf waters, show values varying from one area to another. The highest values
of concentration of suspended matter are given in the outer bay, due to the influence of
Guadalete and San Pedro River mouths. These higher concentrations are consistent with the
pattern of tidal currents and the distribution of fine facies on the sea bottom. In the inner
bay, the concentrations of suspended matter have lower values, due to increased settling of
particles in sheltered environments from waves. In the continental shelf, are given the low‐
est values found in the study area, corresponding to the lower turbidity of these waters. Nu‐
merous studies about the mineralogy of the clay fraction present in surface sediment and
suspended matter, show the existence of a strict relationship between clay minerals trans‐
ported by rivers and those in the sediments of the drainage basin [75, 76]. In our case, com‐
paring the clay mineral composition of SPM and in surface sediment, a significanct change
is observed in the particulate matter with an important increase of smectite content from
13% in surface sediment to 38% in SPM and decrease of illite content (from 60% to 41%).
This effect can be explained by the tendency of illite to focus on the sea bottoms, while smec‐
tites tend to stay longer in suspension. The remaining minerals like chlorite, Kaolinite and I-
108 Sediment Transport Processes and Their Modelling Applications

S has low contents, between 5 and 10%. The main clay mineral association established in
SPM is I>S>>Cl>I-S>K, which basically coincides with that obtained for the bay and the con‐
tinental shelf and differs significantly from that found only in the bay of Cadiz [28]. There is
a greater importance of illite in samples of internal areas of the bay and tidal creeks. This
mineral, decrease in open water and the continental shelf, favoring the increase of smectite
contents. Those differences can be explained considering processes such as settling and se‐
lective transport of clay minerals in marine environments [77], in relation with hydrody‐
namics regime, and the location of the main sources of sediment supply that control the
mineralogical composition and concentration of suspended matter.

Figure 9. Transport dynamics model of fine sediments and suspended matter between the inner and the outer bay of
Cadiz and the adjoining continental shelf.

Clay minerals and assemblages have been used as dynamic tracers to deduce transport path
and the process of sedimentation of fine sediment and SPM [27, 28, 77, 78, 79, 80,]. Others
studies indicate that clay minerals can be transported large distance by rivers, wind and cur‐
rents, indicating the dominant trajectories of fine sediments and the suspended matter [22,
25, 42]. Data from the distribution of SPM concentration in the bay of Cadiz and clay miner‐
als contents and assemblages have been used to establish the transport paths model of SPM
in different area of the study zone, through the trace that different minerals have left in ma‐
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 109
[Link]

rine bottom surface. Two flows paths have been differentiated (Fig.8): i) The inflows coming
from external marine areas located to the north, in particular the Guadalquivir river mouth
and other sources. These flows can transport suspended matter and fine sediments to the
Cadiz bay bottoms by the action of marine currents, specially the littoral and the Atlantic
Surface water currents of SE direction. ii) The outflows coming from Cadiz bay and littoral
zones; can reach the continental shelf by mean of ebb tide currents. Two possible transport
paths can be deduced; the first one runs preferably by the northern margin of the bay of Ca‐
diz and reaches the Rota city. Another flow oriented towards the west, bordering the city of
Cadiz, eventually extending to the continental shelf. These bands might correspond to sea
floor marks generated by flows between the bay and the continental shelf; agreement with
the tidal flow pattern established by [33] in the Cadiz bay.

5.3. Turbidity flow patterns

The analysis of satellite images can provide information about size and direction of the
turbidity plumes and the effect of winds and tidal currents in their distribution; and esti‐
mate the concentration of particles in water column, [81, 82, 83]. The turbidity caused by
suspended particles is detectable by the reflective bands of Landsat satellite [84, 85].
Based on Landsat images, [34] and [86] show that In the Bay of Cadiz, depending to hy‐
drodynamic conditions, Turbidity plumes follow different directions and can cross the
bay area. They reach the inner shelf by the action of tidal ebb, depositing part of its SPM.
Once in the open sea, SPM are moved by currents and interact with the general hydrody‐
namic system affecting coastal areas and the Gulf of Cadiz [27, 28]. According to informa‐
tion obtained from the analysis of Landsat-TM images and the concentration of
suspended matter, four levels of turbidity has been differentiated in the sea area between
the town of Rota (northwestern boundary of the Bay), and the mouth of Sanctipetri tidal
creek (southeastern boundary of the bay) (Fig. 8):

i. High and very high turbidity Waters appears in the eastern part of the outer bay,
near the Guadalete and San Pedro river mouths, whose turbidity plumes are ori‐
ented toward the West and NW. Very high turbidity can be observed in the inner
bay and the central part of the outer bay, when the spring tide and the southeast
wind coincide. South of the bay, the Sanctipetri tidal creek shows also very high
turbidity that is oriented toward the SSW.

ii. Medium turbidity waters are found along the coast and occupy a variable band be‐
tween 0.5 and 2 km. They are specially represented in the eastern part of the outer
bay, configured in a large band following the morphology of the coast and the sea
bottom. In situation of Easterly winds, these waters can exceed the environment of
the bay of Cadiz and reach the inner shelf by the action of ebb tidal currents.

iii. Low turbidity are observed in the western sector of the inner bay (except at mo‐
ments of spring tide and east winds). The degree of turbidity was found to be relat‐
ed to a confined environment, less affected by the action of tidal currents leading to
precipitation of suspended matter. The turbidity level in this area may be due to
110 Sediment Transport Processes and Their Modelling Applications

their muddy bottoms occupied by algae, whose roots and leaves act to retain the
sediment and preventing their resuspension. Low turbidity can be observed in the
outer bay, under certain conditions, in this case, suspended particulate matter com‐
ing from the inner bay, do not reach the outer bay, characterized by profound bot‐
toms making difficult the process of resuspension of fine particles.
iv. Very low turbidity waters, correspond to Atlantic waters, characterized by little in‐
fluence of coastline and located at several kilometers from the coast. Under certain
conditions, these water, can be mixed with other more turbid found near shore, be‐
cause of the presence of suspended matter flows generated by the tides.

Taking into account the above data, three geographical sectors have been identified:

a. Proximal sector. Is a shallow area of high and very high turbidity. Is affected by the
physiography of the coast and the sea bottoms. The variation of degree of turbidity in
this area are related to continental supply and sedimentary nature of the bottoms (mud‐
dy or sandy); as well as the direct action of waves and tidal currents.
b. Middle Sector. Corresponds to the maritime area limited between the proximal and
distal sectors. This is an area of high variability and is affected in its inner part by the
action of tidal flows, which alter their limits and water quality up to 4 times daily.
The outer boundary of this sector consists of less turbid water, flowing through this
area of the Gulf of Cadiz.
c. External sector. is the maritime zone located about 5 km away from the virtual line be‐
tween the Rota and Cadiz cities, and characterized by low and very low turbidity. The
internal limit is still somewhat affected by coastal dynamics, but less than in the middle
and proximal sectors. The degree of turbidity present little change, except during peri‐
ods of flood (espacially in Autumn and Winter) in which the rivers pouring into this
part of the Gulf of Cadiz, transport lots of suspended matter, giving way to large tur‐
bidity plumes. water turbidity may be episodically affected by the action of tidal flows
during spring tides, favored by Southeast winds.

6. Conclusion

i. The high concentrations of suspended matter present in Cadiz bay waters are
linked to major sources of contributions, like rivers and tidal creeks. Fine sediments
coming from the remobilization of the inner bay bottoms and the marshlands by
the effect of waves and winds, can supply large quantity of suspended material de‐
pending on the season. This material is then transported by the effect of tidal ebb
currents to the external areas of the bay, and even continental shelf. The use of clay
minerals as natural tracers has allowed the determination of suspended matter
flows paths taking place in different sectors of the study area.
ii. The general transport pattern of suspended particulate matter (SPM) is affected
by local processes, which take place in littoral zones, in particular in Cadiz bay
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 111
[Link]

and the Guadalquivir estuary. Part of the Atlantic waters rich in SPM coming
from the Guadalquivir Rivers reaches the bay of Cadiz and can be deposited in
lagoons and salt marshes. The resuspension of fine-grained material in the inner
zone of the bay during southeast wind and ebb tidal current generate suspended
matter outflows towards the outer bay. Considerable quantity of this SPM is in‐
jected in the Atlantic waters.

iii. Landsat images show that the turbidity pattern coincides generally with the area of
the muddiest facies present on the outer bay bottoms and with the geographical lo‐
cations of the sampling stations providing the highest contents of suspended sol‐
ids. The combined study of suspended matter concentration and Lansat images
analyses, allows recognize the transport paths followed by the sediments in the
surface marine waters. these images show that turbidity plumes extend from the
inner bay out to the continental shelf by action of the tidal ebb currents, following
the coastline and sea bed physiography.

iv. The transport dynamics of fine sediment and SPM from the inner bay toward the
external zone and the inner continental shelf was found to be related to the loca‐
tion and diversity of supply sources, to the coastal and sea bottoms morphology;
as well as to the influence of the hydrodynamic system, fundamentally ebb tide
currents. Other transport agents are the currents generated by waves. In situa‐
tion of easterly storm, currents are generated towards the West, which combined
with the tidal ebb, give rise to large flumes of turbidity, which are directed off‐
shore. In contrast, the wind and westerly waves are opposed to ebb tide currents
favouring the inflow.

Acknowledgements

This paper was supported by projects 2002-01142/MAR of the Interministerial Commis‐

sion of Science and Technology (CICYT) of the Spanish Government; as well as by the
project SVT12-2010 of the Ministry of Higher Education of the Moroccan Government.
The author thanks Dr. J.M Gutierrez-Mas of the University of Cadiz for his encourage‐
ment and availability.

Author details

Mohammed Achab*

Address all correspondence to: achab@[Link]

Department of Earth Sciences, Scientific Institute, University Mohammed V-Agdal, Rabat,

Morocco
112 Sediment Transport Processes and Their Modelling Applications

References

[1] Allen, G. P. (1971). Etude des processus sédimentaires dans l´estuaire de la Gironde.
PhD thesis, Univ. Bordeaux, I.(353), 314.
[2] Latouche, C. M. (1972). La sédimentation argileuse marine au voisinage de l´embo‐
chure de la Gironde. Interprétation et conséquences. C. R. Acad. Sci. Paris, t 274,
2929-2932.
[3] Castaing, P. (1981). Le transfer á l´ocean des suspension estuariennes. Cas de la Gi‐
ronde. PhD thesis, Univ. Bordeaux, I.(701), 530.
[4] Jouanneau, J. M. (1982). Matiéres en suspension et oligo-élément métalique dans le
systeme estuarien girodin: Comportement et flux. PhD thesis, Univ. Bordeaux, I.(32),
150.
[5] Webwro, O., Jouanneau, J. M., Ruch, Py., & Mirmand, M. (1991). Grain-size relation‐
ship between Suspended matter originating in the Gironde Estuary and shelf Mud-
Patch deposits. Mar. Geol, 96, 195-165.
[6] Wegner, C. (2003). Sediment Transport on Arctic Shelves- Seasonal Variations in Sus‐
pended Particulate Matter Dynamics on the Laptev Sea Shelf (Siberian Arctic). Ber.
Polarforsch. Meeresforsch., 1618- 3193, 455.
[7] Maldonado, A., & Stanley, D. (1981). Clay mineral distribution patterns as influenced
by deposicional processes in the southeastern Levantine Sea. Sedimentology, 28, 21-32.
[8] Palanques, A., Plana, F., & Maldonado, A. Estudio de la materia en suspensión en el
Golfo de Cádiz. Acta. Geol. Hisp, t.21-22, 491-497.
[9] Chamley, H. (1993). Marine sedimentation of clay minerals. in : Sedimentologie et geo‐
chimie de la surface, Strasbourg (France), Academie des sciences, Paris (France), 27-28.
[10] Balopouls, E. T., Collins, M. B., & James, A. E. (1986). Satelite images and their use in
the numerical modeling of coastal processes. Inter. Jour. of Remote Sensing, 7, 905-519.
[11] Fernández-Palcios, A., Moreira Madueño, J. M., Sánchez, Rodriguez E., & Ojeda Zu‐
jar, J. (1994). Evaluation of different methodological approaches for monitoring water
quality parameters in the coastal waters of Andalusia (Spain) using [Link]‐
ta. Earsel workshop on Remot Sensing and GIS for Coastal zone Managment, Rijks Water‐
Staat, Survy Department. Delft, The Netherlands, 114-123.
[12] Shimwell, S. (1998). Remote sensing of suspended sediment off the Holderness
Coast, UK. ARGOSS Report A5., 23, unpublished.
[13] Baban, S. M. J. (1993). Detecting water quality parameters in the Norfolk Broads,
U.K., using Landsat imagery. Int. J. Remote Sensing, 14(N17), 1247-1267.
[14] Manheim, F. T., Meade, R. H. y, & Bond, G. C. (1970). Suspended matter in surface
waters of the Atlantic continental margin from Cape Cod to the Florida Keys. Science,
167, 371-376.
[15] Milliman, J. D., Summerhayes, C., P.y, , & Barretto, H. T. (1975). Oceanography and
suspended matter of the Amazon River. Jour. Sedim. Petrol, 45, 189-206.
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 113
[Link]

[16] Feely, R. A., Baker, E. T., Schumacher, J. D., Mascoth, G., J.y, , & Landing, W. M.
(1979). Processes affecting the distribution and transport of suspended matter in the
Notheast Gulf of Olaska. Deep-Sea Res., 26(4A), 445-464.
[17] Rudert, M. y, & Müller, G. (1981). Mineralogy and provenance of suspended solids in
estuarines and nearshores areas of the southern North Sea. Senckenberg. Mar., 13,
57-64.
[18] Lixiangyun, Chen Hongxun., Lichuanrong, Wang Xiangzhen y Qiu Chuanzhu.
(1989). Distribution characteristic of major minerals of suspended matters and sedi‐
ments in Daya Bay. Tropical Oceanology, 8(1), 34-42.
[19] Monaco, A. (1970). Sur quelques phénoménes d´échange ionique dans les suspension
argileuse an contact de léau de mer. C. R. Acad. Sci., Paris, 270, 1743-1746.
[20] Arnoux, A. y, & Chamley, H. (1974). Mineraux des argiles et détergents des eaux in‐
terstitielles dans les sediments superficiels du Golfe de Lion. C. R. Acad. Sci., Paris,
278, 999-1002.
[21] Aloisi, J. C., Monaco, A., Millot, C. y, & Pauc, H. (1979). Dynamique des suspensions
et micanismes sédimentogénetique sur le plateau continental du Golfe du Lion. C. R.
Acad. Sci., Paris, ser. D, 189, 879-882.
[22] Gutiérrez Mas, J. M., Sánchez Bellón, A., Achab, M., Fernández Palacios, A., & Sán‐
chez Rodríguez, E. (1998). Influence of the suspended matter exchange between the
Cadiz Bay and the continental shelf on the recent marine sedimentation. Alicant,
Spain. 15 th Inter. Sediment. Congr., Abstract, 404-405.
[23] Achab, M., Gutiérrez Mas, J. M., & Luna del Barco, A. (2000a). Concentration and
mineralogic composition of suspended matter in the Bay of Cadiz and adjacent conti‐
nental shelf. Geotemas, 1(14), 81-86.
[24] Gutiérrez Mas, J. M., Achab, M., Sánchez Bellón, A., Moral Cardona, J. P., & López
Aguayo, F. (1996b). Clay minerals in recent sediments of the bay and their relation‐
ships with the adjacent emerged lands and the continental shelf. Advances in Clay
minerals, 121-123.
[25] Achab, M., Gutiérrez Más, J. M., & Sánchez Bellón, A. (1998). Transport of fine sedi‐
ments and clay minerals from the tidal flats and salt marshes in Cadiz bay towards
outer marine zones. European Land-Ocean Interaction Studies, Huelva-Spain, 93-94.
[26] Lopez, Galindo A., Rodero, J., & Maldonado, A. (1999). Surface facies and sediment
dispersal patterns: southeastern Gulf of Cadiz, Spanish continental margin. Marine
Geology, 83-98.
[27] Gutiérrez Más, J. M., López Aguayo, F., & Achab, M. (2006). Clay minerals as dynam‐
ic tracers of suspended matter dispersal in the Gulf of Cadiz (SW Spain). Clay miner‐
als, 727-738.
[28] Achab, M., Gutiérrez Más, J. M., & López Aguayo, F. (2008). Utility of clay minerals
in the determination of sedimentary transport patterns in the bay of Cadiz and the
adjoining continental shelf (SW-Spain). Geodinamica Acta, 5-6.
[29] Gutiérrez Mas, J. M., Luna del Barco, A., Achab, M., Muñoz Pérez, J. J., González
Caballero, J. L., Jódar Tenor, J. M., & Parrado Román, J. M. (2000). Controlling factors
114 Sediment Transport Processes and Their Modelling Applications

of sedimentary dynamics in the littoral domain of the Bay of Cadiz. Geotemas, 1(14),
153-158.
[30] Achab, M., El Moumni, B., El Arrim, A., & Gutierrez Mas, J. M. (2005b). Répartition
des faciès sédimentaires récents en milieu marin côtier : exemple des baies de Tanger
(NW-Maroc) et de Cadix (SW-Espagne). Bull. Inst. Sci., sect. Sci. Terre [27], 55-63.
[31] Achab, M. (2011). Dynamics of sediments exchange and transport in the Bay of Cadiz
and the adjacent continental shelf (SW-Spain). In : Manning, A. (Ed) Sediment Trans‐
port in aquatic environments, Intech, Vienna, 19-44.
[32] Benavente, J. (2000). Morfodinámica litoral de la Bahía Externa de Cádiz. Tesis Doc‐
toral, Univ de Cadiz, 533.
[33] Alvarez, O., Izquierdo, A., Tejedor, B., Mañanes, R., Tejedor, L., & Kagan, B. A.
(1999). The influence os sediments load on tidal dynamics, a case study: Cadiz Bay.
Estuarine Coastal and shelf Science, 48, 439-450.
[34] Guillemot, E. (1987). Teledetection des milieux litoraux de la Baie de Cadix. PhD the‐
sis, Univ. Paris, 146.
[35] Muñoz Pérez, J. L. y, & Sánchez de Lamadrid, A. (1994). El medio físico y biológico
de la bahía de Cádiz: Saco interior. Informaciones técnicas, 28/94. Consejeria de Agri‐
cultura y Pesca (Junta de Andalucía), 161.
[36] Achab, M., Gutiérrez Mas, J. M., Sanchez Bellon, A., & Lopez Aguayo, F. (2000b).
Fine sediments and clay minerals embodiment and transport dynamics between the
inner and outer sectors of Cadiz Bay. Geogaceta, 27, 3-6.
[37] Jodar, J. M., Voulgaris, G., Luna del Barco, A., & Gutiérrez-Mas, J. M. (2002). Wave
and Current Conditions and Implications for the Distribution of Sediment in the Bay
of Cadiz (Andalucia, SW Spain) En: Littoral. The Changing Coast, 3, 287-291.
[38] Rodríguez-Ramírez, A., Ruiz, F., Cáceres, L. M., Rodríguez Vidal, J., Pino, R., & Mu‐
ñoz, J. M. (2003). Analysis of the recent storm record in the southwestern Spanish
coast: Implications for littoral management. The Science of the Total Environment,
189-201.
[39] Meliéres, F. (1974). Recherche sur la dynamique sedimentaire du Golfe de Cadix. Es‐
pagne PhD thesis Univ. Paris.; , 235.
[40] Maldonado, A., & Nelson, C. H. (1988). Dos ejemplos de márgenes continentales de
la península Ibérica: el margen del Ebro y el Golfo de Cádiz. Rev. de la Soc. Geológica
Esp, 1(3-4), 318-325.
[41] Baringer, M. O., & Price, J. F. (1999). A review of the physical oceanography of the
Mediterranean out-flow. Marine Geology, 55-63.
[42] Maldonado, A., & Stanley, D. (1981). Clay mineral distribution patterns as influenced
by depositional processes in the southeastern Levantine Sea. Sedimentology, 21-32.
[43] Gracia, F. J., Rodríguez Vidal, J., Benavente, J., Cáceres, L., & López Aguayo, F.
(1999). Tectónica cuaternaria en la bahía de Cádiz. Avances en el estudio del cuaternario
español, Girona, 67-74.
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 115
[Link]

[44] Achab, M., & Gutiérrez Mas, J. M. (1999a). Characteristics and controlling factor of
the recent sedimentary infill on the bottom of the Bay of Cadiz. Geogaceta, 27, 3-6.
[45] Gutiérrez Más, J. M., Achab, M., & Gracia, F. J. (2004). Structural and physiographic
control on the Holocene marine sedimentation in the bay of Cadiz (SW-Spain). Geodi‐
namica Acta, 153-161.
[46] Maldonado, A. y, & Nelson, C. H. (1999). Interaction of tectonic and depositional
processes that control the evolution of the Iberian Gulf of Cádiz margin. Marine Geol‐
ogy, 217-242.
[47] Lobo, F. J., Hernandez-Molina, F. J., Somoza, L., Rodero, J., Maldonado, A., & Barno‐
las, A. (2000). Patterns of bottom current flow deduced from dune asymetries over
the Gulf of Cadiz Shelf (southwest Spain). Marine Geology, 164, 91-117.
[48] Mabesoone, J. M. (1966). Deposicional and provenance of the sediment in the Guada‐
lete estuary (Spain). Geologie en Minjbouw-Amesterdam, 45, 25-32.
[49] Viguier, C. (1974). Le néogéne de l´andalousie Nord-occidentale (Espagne). Histoir
geologique du bassin du bas Guadalquivir. PhD thesis, Bordeaux, 449.
[50] Zazo, C., Goy, J. L., & Dabrio, C. (1983). Medios marinos y marino-salobres en la ba‐
hía de Cádiz durante el pleistoceno. Rev. Mediterránea. Ser. Geol, 2, 29-52.
[51] Gutiérrez Mas, J. M., Martín Algarra, A., Domínguez Bella, S., & Moral Cardona, J. P.
(1990). Introducción a la Geología de la provincia de Cádiz. Servicio de Publica‐
ciones, Universidad de Cádiz , 315.
[52] Moral Cardona, J. P., Achab, M., Domínguez, S., Gutiérrez-Mas, J. M., Morata, D., &
Parrado, J. M. (1996). Estudio comparativo de los minerales de la fracción pesada en
los sedimentos de las terrazas del Río Guadalete y fondos de la Bahía de Cádiz. Geo‐
gaceta, 20(7), 14-17.
[53] Dabrio, C. J., Zazo, C., Goy, J. L., Sierro, F. J., Borja, F., Lario, J., Gonzalez, A., &
Flores, J. A. (2000). Deposicional history of estuarine infill during the last postglacial
transgression (Gulf of Cadiz, Southern Spain). Marine Geology, 162, 381-404.
[54] Achab, M., & Gutiérrez Más, J. M. (2005a). Nature and distribution of the sand frac‐
tion components in the Cadiz bay bottoms (SW Spain). Revista de la Sociedad Geologica
de España, 18(3-4), 133-143.
[55] Greenberg, A. E., Clescer, L. S., & Eaton, A. D. (1992). Standard Methodes for the Ex‐
amination of Water and Wastewater. 18th Edición, American Public Heath Association.
[56] Holtzapffel, T. (1985). Les minéraux argileux. Préparation. Analyse diffractométrique
et détermination. Bull. Soc. Géol. Nord, 12, 136.
[57] Tucker, M. (1988). Techniques in Sedimentology. Blackwell, Oxford, UK, 394.
[58] Ojeda, J., Sanchez, E., Fernandez-Palacios, A., & Moreira, J. M. (1995). Study of the
dynamics of estuarine and coastal waters using remote sensing: the Tinto-Odiel estu‐
ary, SW Spain. [Link] Conserv, 1, 109-118.
[59] Folk, R. L., & Ward, W. C. (1957). Brazos River bar: a study in the significance of
grain-size parameters. Journal of Sedimentary Petrology, 27, 3-26.
116 Sediment Transport Processes and Their Modelling Applications

[60] Martins, L. R. (2003). Recent sediments and grains-size analysis. GRAVEL, 1, 90-105.
[61] Pevear, D. R., & Mumpton, D. R. (1989). Quantitative Mineral Analysis of clays. Col‐
orado. CMS Workshop Lectures1. The Clay Mineral Society.
[62] Ortega-Huertas, M., Palomo, I., Moresi, M., & Oddone, M. (1991). A mineralogical
and geochemical approach to establishing a sedimentary model in a passive conti‐
nental margin (Subbetic zone, Betic Cordilleras, SE Spain). Clay Minerals, 389-407.
[63] Van Rijn, L. C. (1984). Sediment transport part I: bed load transport. Journal. Hydraul.
Eng., 110(10), 1431-1456.
[64] Gao, S., & Collins, M. B. (1994). Análysis of grain size trends, for defining sediment
transport pathways in marine environments. Jour. Coast. Res, 10(1), 70-78.
[65] Meade, R. H. (1972). Transport and deposition of sediments in estuaries. Geological
Society of America, 133, 91-120.
[66] Poulos, S. E., Collins, M. B., & Shaw, H. F. (1996). Deltaic sedimentation, including
clay mineral deposition patterns marine embayment of Greece (SE. Alpine Europe).
Journal of Coastal Research, 12(4), 940-952.
[67] Gutiérrez-Mas, J. M., Hernández-Molina, J., & Lopez-Aguayo, F. (1996a). Holocene
sedimentary dynamics on the Iberian continental shelf of the Gulf of Cadiz
(SWSpain). Continental Shelf Research, 16(13), 1635-1653.
[68] Cacchione, D. A., & Drake, D. E. (1990). Shelf sediment transport: An overview with
applications to the northem Califomia continental shelves. In: Mehaute, B. L. & Hanes,
D. M. (eds.) The Sea 7b. John Wiley and Sons, New York, 729-773.
[69] Achab, M., Gutiérrez Más, J. M., Moral Cardona, J. P., Parrado Román, J. M., Gonza‐
lez Caballero, J. L., & Lopez Aguayo, F. (1999b). Relict and modern facies differentia‐
tion in the Cadiz bay sea bottom recent sediments. Geogaceta, 27, 187-190.
[70] Mc Cave, I. N. (1975). Vertical flux of particles in the ocean. Deep-Sea Research,, 22,
491-502.
[71] Drake, D. E. (1976). Suspended sediment transport and mud deposition in continen‐
tal shelves. In: Marine sediment transport and environmental managment. Edited by D.J.
Stantey and D.J.P. Swift, Wiley, N.Y., 127-158.
[72] Cavaleri, L., & Malanotte-Rizzoli, P. (1981). Wind wave prediction in shallow water:
theory and application. J Geophys Res, 86, 10961-10973.
[73] Holthuijsen, L. H. (2007). Waves in Oceanic and Coastal Waters, Cambridge Univesity
Press, Cambridge.
[74] Karsten, A., Lettmann, Jörg, Olaf, Wolff, & Badewie, Thomas H. (2009). Modeling the
impact of wind and waves on suspended particulate matter fluxes in the East Frisian
Wadden Sea (southern North Sea). Ocean Dynamics, DOI 10.1007/s10236-009-0194-5.
[75] Griffin, G. M. (1962). Regional clay- mineral facies products of weathering intensity
and current distribution in northeastern Gulf of Mexico. Geol. Soc. Amer. Bull, 73.
[76] Biscaye, P. E. (1965). Mineralogy and sedimentation of recent deep-sea in the Atlantic
Ocean and adjacent seas and Oceans. Geol. Soc. Amer Bull., 76, 803-831.
 Composition and Transport Dynamics of Suspended Particulate Matter in the Bay of Cadiz 117
[Link]

[77] Chamley, H. (1989). Clay sedimentology, Springer-Verlag, Berlin, 623.

[78] Sawheney, B. L., & Frink, C. R. (1978). Clay minerals as indicators of sediment source
in tidal estuaries of long island sound. Clays and Clay minerals, 26(3), 227-230.

[79] Naidu, A. S., Han, M. W., Mowatt, T. C., & Wajda, D. (1995). Clay minerals as indica‐
tors of sources of terrigenous sediments their transportation and deposition: Bering
Basin Russian-Alaskan Arctic. Marine Geology, 127(1-4), 87-104.

[80] Bhukhari, S. S., & Nayak, G. N. (1996). Clay minerals in identification of provenance
of sediments of Mandovi Estuary, Goa, West Coast of India, Indian. Journal of Marine.
Science, 25(4), 341-345.

[81] Lo, C. P. (1986). Applied Remote Sensing, Longman, London, 393.

[82] Kleman, Y., & Hardisky, M. A. (1987). Remote sensing of estuaries: An overview.
Symp. Remote sensing of envirinment, Ann Arbour, Michigan, 183-204.

[83] Fiedler, P. C., & Laurs, R. M. (1990). Variability of the Columbia River plume ob‐
served in visible and infrared satellite imagery. Inter. Jour. Remote Sensing, 11(6), 999,
1010.

[84] Spitzer, D., & Driks, R. W. J. (1987). Bottom influence on the reflectance of the sea.
Int. Journal Remote Sensing, 8, 779-290.

[85] Baban, S. M. J. (1995). The use of LandSat imagenery to map fluvial sediment dis‐
charge into coastal waters. Marine Geology, 1230, 263-270.

[86] Bernal, Ristori. E. (1986). Aplicaciones de la teledetección espacial al medio marino

litoral. 1ª Jornadas de Ingeniería Geográfica. Madrid. Bol. Inf. Serv. Geográf. Ejerc., 62,
37-56.
 Chapter 6

Flocculation Dynamics of Mud: Sand Mixed Suspensions

Andrew J. Manning, Jeremy R. Spearman,

Richard J.S. Whitehouse, Emma L. Pidduck,
John V. Baugh and Kate L. Spencer

Additional information is available at the end of the chapter

[Link]

1. Introduction

Sediments present in muddy estuaries and tidal inlets are regarded as being predominantly
cohesive. These muds are usually composed of both clay and silt minerals combined with
organic matter (Winterwerp and van Kesteren, 2004), and with the exception of very low
particle concentrations or extremely high energy flow conditions, muddy particles occur as a
spectra of floc sizes (D) when entrained into suspension (Kranck and Milligan, 1992).

In reality, natural sediments tend to comprise a mixture of different particle sizes, non-cohesive
sediment including fine sands and, because of the interaction between these different fractions,
the mixture behaves in a different way than the constituent parts (Whitehouse et al., 2000).
Uncles et al. (1998) found that the proportion of mud and sand in subtidal and intertidal
sediments can vary both temporally and spatially (e.g. Uncles et al, 1998). Fig. 1 shows an
example of mud and sand in close proximity in the Eden Estuary (east coast of Scotland).

Very little is quantitatively known about how mixtures of cohesive and non-cohesive sedi‐
ments, of different ratios and concentrations, interact whilst in suspension in turbulent flows
and the effect this has on the resultant flocs formed and their flocculation properties, in
particular the settling velocity. This has important implications for sediment transport
modelling. Drawing on key literature and new data, this chapter will provide an overview of
mixed sediment flocculation dynamics and how they can influence sediment transport.

The first part of this chapter reviews the theoretical aspects relating to the flocculation of
mud:sand mixtures. It commences with a brief review of flocculation processes (2), followed
by an overview of segregation environments verses flocculating suspensions (3), and then the
biological influences on mixed sediment flocculation are summarised (4). The second part of
120 Sediment Transport Processes and Their Modelling Applications

Figure 1. Sand and muddy sediments in close proximity, Eden Estuary, Fife (east coast of Scotland).

the chapter (5-7) draws on the findings of recent empirical studies assessing mixed sediment
floc behaviour. The laboratory experimental protocols and findings are reported with floc data
in spectral and parameterised formats presented and discussed. The potential implications of
mud:sand flocculation on sediment transport modelling are also discussed (8-9).

2. Flocculation factors

From a sediment transport perspective, knowledge of the settling rate of sediments in

suspension is vital in determining depositional fluxes and sediment transport rates. Sand is a
non-cohesive material and therefore does not flocculate in pure sand suspensions. The settling
velocity (Ws) is generally proportional to the square of the particle size or diameter (D).
Conversely, mud is strongly cohesive and flocculates forming small, compact microflocs as
well as larger, more porous macroflocs (Eisma, 1986; Manning, 2001; Manning and Dyer,
2002a,b) – Fig. 2. Flocculation is a dynamically active process which readily reacts to changes
in hydrodynamically generated turbulent shear stresses (τ) (e.g. Krone, 1962; Parker et al.,
1972; McCave, 1984; van Leussen, 1994; Winterwerp, 1998; Manning, 2004a), suspended
particulate matter (SPM) concentration, together with salinity, mineralogy and biological
stickiness.
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 121
[Link]

A B

1.2 mm

2 mm

1.25 mm

0.85 mm

C D

1.9 mm

0.5 mm

Figure 2. A selection of floc images from a predominantly muddy origin. A) A ragged cluster-type macrofloc (top) and
Fig. 2. A of
a simple stringer composed selection of floc images
two macroflocs from a by
interlinked predominantly muddy
organic fibres origin.B)
(bottom); A)a A ragged
‘string ofcluster-type
pearls’ type macro‐
macrofloc
floc; C) a long interlinked (top) and
stringer a simple two
comprising stringer composed
clustered of two macroflocs
macroflocs; D) raggedinterlinked
macroflocs bysettling;
organic and
fibresE) a selec‐
(bottom); B) a ‘string of pearls’ type macrofloc; C) a long interlinked stringer comprising two
tion of small slow settling microflocs, some of which are probably the result of macrofloc fracturing and subsequent
clustered macroflocs; D) ragged macroflocs settling; and E) a selection of small slow settling
break-up during a microflocs,
turbulent event
some ofwhich
whichexceeded the original
are probably the resultmacrofloc structural
of macrofloc integrity
fracturing threshold.
and subsequent break-up
during a turbulent event which exceeded the original macrofloc structural integrity threshold.
Flocculation can significantly alter the sediment transport patterns throughout an estuary, and
floc properties can vary both in time and space. For example, Manning et al. (2006) showed
that during spring tidal conditions in the Tamar Estuary (UK), macroflocs can typically reach
1-2 mm in diameter. These flocs demonstrate settling velocities up to 20 mm.s-1, but their
effective densities ρe (i.e. the floc bulk density less the water density) are generally less than
50 kg.m-3, which means they are prone to break-up when settling through a region of high
turbulent shear.
122 Sediment Transport Processes and Their Modelling Applications

There are, however, many estuarial environments where mud and fine sand co-exist as a single
mixture (Mitchener et al., 1996) and this creates the potential for these two fractions to combine
and exhibit some degree of interactive flocculation (Manning et al., 2007, 2009). The erosion
and consolidation of mixtures of mud and sand has been thoroughly reviewed (Williamson,
1991; and Whitehouse et al, 2000), and there have been some studies that have examined mixed
sediment settling (e.g. Dankers et al., 2007). However, very little investigation has been devoted
to the potential flocculation that may occur when mud and sand mixtures are entrained into
suspension, as it was not considered to be an important factor. This could be a valid assumption
for a segregational environment, where the mud and sand do not combine into a single matrix.

When we refer to ‘mixed sediment flocculation’ in this chapter, we are primarily referring to
suspension mixtures of mud (typically composed of clay minerals and fine silts up to 63 µm
in diameter together with organic matter) and predominantly non-cohesive sediments
(typically up to the size of fine sands, i.e. about 100-200 µm, as larger grains are unlikely to
directly interact with mud).

Previous research has shown that a clay content of between 5 – 10% can cause natural sediment
mixtures to behave in a cohesive manner (Dyer, 1986; Raudviki, 1998). Thus, different ratios
of mud and sand can vary the level of cohesion, which will influence the resultant level of
flocculation. Biological activity, more commonly associated with cohesive sediments, has been
highlighted to play an important role in the cohesion of sediments (e.g. Paterson and Hager‐
they, 2001). However, it is extremely difficult to quantify such a complex sedimentary matrix
in a fundamental manner, primarily as a result of a lack of verification data.

Of the various processes that occur during a tidal cycle, flocculation of the sediment is regarded
as one of the primary mechanisms that can affect the deposition, erosion and consolidation
rates. An individual floc may comprise up to 106 individual particulates. As flocs grow in size
their effective densities generally decrease (Koglin, 1977; Tambo and Watanabe, 1979; Klimpel
and Hogg, 1986) and their settling speeds rise due to a Stokes’ Law relationship (Dyer and
Manning, 1998) between D and Ws. Furthermore, low density flocs also demonstrate settling
velocities that are significantly quicker than the individual cohesive particles (~ 1-5 µm in
diameter). The cohesive nature of these particulates is a combination of both the electrostatic
charging of the clay minerals as they pass through brackish to highly saline water, and various
sticky biogenic coatings, such as mucopolysaccharides (e.g. Paterson 1989).

Van Leussen (1988) theoretically assessed the comparative influence of the three main
collision mechanisms: Brownian motion, turbulent shear and differential settling, and
deduced that turbulent shear stresses, principally those generated by velocity gradients
present in an estuarine water column, were the dominant flocculation mechanism. This
mechanism was deemed most effective for turbulent shear stresses ranging between
0.03-0.8 Pa. These stresses are representative of those typically experienced in the near bed
region of many European macrotidal and mesotidal estuaries and hence estuaries are ideal
environments for flocculation.

The energy for turbulent mixing is derived from the kinetic energy dissipated by the water
flowing across the sediment bed. The frictional force exerted by the flow per unit area of the
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 123
[Link]

bed is the shear stress (turbulent shear stress during turbulent flow conditions). The efficiency
with which the particles flocculate is a reflection of the stability of the suspension (van Leussen,
1994). A suspension is classified as unstable when it becomes fully flocculated, and is stable
when all particles remain as individual entities.

As low to medium levels of turbulent shear stress can promote floc growth, high levels of
turbulence that occur during a tidal cycle, can cause disruption to the flocculation process by
instigating floc break-up, and eventually pull the constituent components of a floc apart. As
turbulent activity increases, both turbulent pressure differences and turbulent shear stresses
in the flow rise. If the floc structural integrity is less than the imposing turbulent induced forces,
the floc will fracture. Also, aggregate break-up can occur as a result of high impact particle
collisions during very turbulent events. Floc break-up by three-particle collisions tends to be
the most effective (Burban et al., 1989). Hence, turbulent shear stress can impose a maximum
floc size restriction on a floc population in tidal waters (McCave, 1984). Eisma (1986) observed
a general agreement between the maximum floc size and the smallest turbulent eddies as
categorised by Kolmogorov (1941a, b).

3. Segregation and flocculation

This section looks at how mud and sand can co-exist within an aquatic environment. Mud:sand
sediment mixtures may behave either in a segregated way, or interact through flocculation.
The phenomenon of mud:sand segregation considers the mud and sand to operate as two
independent suspensions (van Ledden, 2002) and, as such, very little bonding occurs, and
flocculation interactions between the cohesive and non-cohesive sediment fractions are non-
existent. Mixed sediment experiments have shown that mud particles and sand grains which
behave in a segregated manner, settle simultaneously but as independent fractions to form
two well sorted layers at the bed/water interface (Ockenden and Delo, 1991; Migniot, 1968;
Williamson and Ockenden, 1993).

Williamson (1991) reviewed a number of the characteristics of mud:sand mixtures in the

natural environment (some of the key findings are summarised in this paragraph). The review
investigated the distributions and characteristics of mud and sand mixtures based on a
literature search and a review of relevant fieldwork data. Some of the features common to both
mud and sand, such as: spatial distributions, vertical layering, bioturbation, depositional
characteristics and flocculation, were described. The review suggested that muddier sediments
were generally found in regions of lower dynamic activity and sandier sediments in higher
energy regions. However, the local distributions could only be explained by local hydrody‐
namic analysis and these data were often lacking, which did not allow a complete picture to
be obtained. Flocculation and the effects of salinity distributions were found to be important
in governing the mud distributions, with a muddy reach often being found in the flocculation
zone. The vertical profile of settled mud and sand was also investigated, with laminations of
mud and sand often being found. The thickness of the layers in the laminated sediment profiles
were typically sub-millimetre to a few millimetres. The process of bioturbation (i.e. the
124 Sediment Transport Processes and Their Modelling Applications

reworking of the bed sediments by living organisms) can potentially produce a mixing of bed
sediment particles prior to resuspension (e.g. Nowell et al., 1981; Paterson et al., 1990; Widdows
et al., 2004). Thus a bed which is initially deposited as a discretely segregated layering of mud
and sand may be transformed into a quasi-homogeneous mixture.

Van Ledden (2003) states that mud and sand can be deposited as mixtures or in alternating
layers in estuaries. An example of this is visible in the upper part of Fig 1. Additionally,
biological activity such as bioturbation (i.e. the reworking of the bed sediments), can mix the
sediment particles. As a result the mud content in many parts of an estuary may not be uniform,
but can become segregated both vertically and horizontally – a phenomenon known as
mud:sand segregation (van Ledden, 2003).

Mud:sand segregation can have a direct influence on the settling velocity of the sediments once
entrained. For instance, the settling velocity of individual sand grains could be reduced as they
pass through a layer of flocculating muddy sediments in close proximity to the sea bed. Van
Ledden (2003) provides three examples which illustrate the importance of why a physical
understanding of the distribution of mud and sand in estuarine systems is important:
• Large mud content variations at the bed surface indicate that both mud and sand contribute
to bed level changes in estuaries and tidal inlets. These will affect the navigable depth and
high water levels.

• Cohesive muddy sediments have the propensity to adsorb contaminants (Förstner and
Wittmann,1983). This, in turn, has a direct effect on water quality and related environmental
issues (e.g. Uncles et al., 1998). The amount of segregation present on both temporal and
spatial scales will provide an indication to the potential degree of pollution in bed sediments.

• The mud content in sediment beds is a crucial habitat parameter, which controls the
distribution of flora and fauna in estuarine systems (e.g. Reid and Wood, 1976; Kennish,
1986; Widdows et al., 2004). Dyer et al. (2000), for example, showed that the sediment type
and grain size are the best physical descriptors of floral and faunal assemblages in the upper
zone of intertidal mudflats.
Van Wijngaarden (2002a, 2002b) examined the mud:sand content distributions in the upper
300 mm of the bed in the Haringvliet – Holland Diep (The Netherlands). Mud content varied
from less than 15% at the mouths of most of the river branches feeding into the system, to
nearly two thirds mud in the channels of the Holland Diep. Fast settling sand grains accumu‐
lated at the end of river branches whereas the slower-settling muddy suspensions were
transported further downstream due to settling lag into the central part of the Holland Diep.
The segregation is, to a large extent related to varying bed levels throughout the system and
variations in the turbulent shear stresses (van Ledden, 2003), which influence erosion,
deposition and transport.

There are also many locations where mud and sand co-exist as a mixture (Mitchener et al.,
1996) and this creates the potential for these two fractions to combine within a flocculation
matrix when re-entrained into suspension (Manning et al., 2007). When sand is added to a
predominantly muddy matrix, Mitchener et al. (1996) found that this increased the binding
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 125
[Link]

potential between the clay particles, for example as found in the subtidal mud patches off
Sellafield in the Irish Sea (Feates and Mitchener, 1998). Thus the physical effect of adding
cohesive mud to a sandy environment can create increased bed stability, which can potentially
lead to mixed sediment flocs forming when the eroded bed is entrained (Kamphuis and Hall,
1983; Alvarez-Hernandez, 1990; Williamson and Ockenden, 1993; Torfs, 1994; Mitchener et al.,
1996; and Panagiotopoulus et al., 1997). Even where sand and mud are considered to be fairly
well segregated at the bed, sand and mud can co-exist in suspended sediment transport.
Spearman et al. (2011) describe an example in the outer Thames Estuary (UK), renowned for
being a sandy area, where the flux of suspended sediment of mud and sand are of the same
magnitude.

Therefore, in a segregated environment, both mud and sand are present acting in a completely
independent manner. In a flocculating environment, the mud and sand particles are interacting
to form flocs which demonstrate very different characteristics (e.g. D, Ws, ρe) from their
compositional base. The nature of the sedimentary regime is best determined by observational
measurements rather than being able to be determined a priori. This can pose additional
problems for the prediction and modelling of suspended sediment transport in mixed
sediment estuarine environments and this will be considered in Section 9.

4. Role of biology in mud: Sand mixtures

Although not directly examined in the laboratory experiments which will be discussed later
in this chapter, it is important to consider other effects of which a key one is due to biological
factors influencing the grains in suspension. These factors work in addition to the primary
chemico-physical ones to make mixed sediment flocculation possible. In predominantly
muddy/silty environments, benthic microphytobenthos contribute up to half the total auto‐
trophic production in an estuarine system (Underwood and Kromkamp, 1999; Cahoon, 1999).
Biostabilisation can increase particle cohesion, for example: epipelic diatoms (e.g. Paterson and
Hagerthey, 2001) secrete extra-cellular polymeric substances (EPS; Tolhurst et al., 2002) as they
move within the sediments. EPSs are regarded as highly effective stabilisers of muddy
sediments (e.g. de Brouwer et al. 2005; Gerbersdorf et al. 2009; Grabowski et al., 2012).

The influence of biology on sand is reported to a much lesser extent in the literature, however
sand grains that are exposed to long-term biological activity, may also develop a cohesive bio-
coating which could increase the particle collision efficiency when they are entrained. Hickman
and Round (1970) reported that sand particles can be joined by 'epipsammic' diatoms which
attach to sand grains. Epipsammic macro-algal forms either adnate to the grain surface or
attach to sand grains by their mucilage stalks. Epipsammic diatoms which are attached to sand
grains, demonstrate strong adhesive properties to the grain surface (Harper and Harper,
1967). When fine sand and biology are combined into a single matrix, they can form “microbial
mats” and the binding strength of these mats can be extremely high. Little (2000) states that
because these types of algal threads are sticky with EPS, they can efficiently trap fine sand
grains. These sticky bio-coatings can increase the collision efficiency (Edzwald and O’Melia,
126 Sediment Transport Processes and Their Modelling Applications

1975) of particles when entrained into suspension, thus allowing fine sand grains to adhere
with the clay fraction and form the cage-like structure around fine sand particles. Through
microscopic photography, Wolanski (2007) observed the formation of large muddy flocs
formed by mud creating a sticky membrane around large non-cohesive silt particles.

5. Experimental approaches

When investigating the role sand may play in the flocculation process, several important
research questions need to be considered, including:
i. How does the settling velocity of mixed sediment flocs vary in response to different
mud:sand mixtures?

ii. What effect does turbulence have on mixed sediment flocculation?

iii. Do resuspended sand particles favour interacting with microflocs or macroflocs

more, and enhance their settling dynamics?

iv. If mixed sediment flocculation occurs, are sand grains directly incorporated into both
microfloc and macrofloc fractions?

v. Does flocculation have an effect on the distribution of the particle mass and the mass
settling flux (MSF) of different suspended mud:sand mixtures?
In order to address aspects of the above questions, a series of new controlled laboratory
environment research were initiated to quantitatively examine the flocculation and interaction
between suspended sand and mud sediment mixtures. Other aspects of mud:sand behaviour
have been assessed in laboratory environment measurements (e.g. Ockenden and Delo, 1988;
Williamson and Ockenden, 1993; Torfs, 1994; Torfs et al., 1996; Dankers et al., 2007). During
the new experiments, suspensions of mud and sand, of different total concentrations, were
sheared at different rates in a mini-annular flume and the resultant floc properties observed.
The new experimental runs primarily comprised pre-determined mud:sand mixtures com‐
plemented with some additional data from naturally occurring mud:sand sediment mixtures.

5.1. Annular flume simulations

This study utilised a mini-annular flume to create a consistent and repeatable turbulent
environment (see Fig. 3A) (Manning and Whitehouse, 2009). The annular flume has an outer
diameter of 1.2 m, a channel width of 0.1 m and a maximum depth of 0.15 m, along with a
detachable motor driven rotating roof (10 mm thick) to create the flow for cohesive sediment
experiments (e.g. Manning and Dyer, 1999). Maximum flow speeds of approximately 0.7
m.s-1 can be produced in the lower half of the water column, created by 10 mm deep paddles
attached to the underside of the roof. A Nortek mini-ADV (Acoustic Doppler Velocimeter)
probe was used to calibrate the flow in terms of velocity and turbulent kinetic energy (TKE)
at a distance of 22 mm (the floc extraction height) above the flume channel base.
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 127
[Link]

(A)

(B)

Figure 3. The mini-annular flume (A) and the LabSFLOC instrument set-up (B).

5.2. Floc property measurements

Representative floc populations were measured using the LabSFLOC version 1.0 –
Laboratory Spectral Flocculation Characteristics – instrument (Manning, 2006). This utilises
128 Sediment Transport Processes and Their Modelling Applications

a high magnification Puffin (model UTC 341) monochrome all-magnetic Pasecon tube video
camera (Manning and Dyer, 2002a), to observe particles settling in a Perspex settling
column (see Fig. 3B), allowing for minimal disruption of the particles. The video camera,
positioned 75 mm above the base of the column, views all particles in the centre of the
column that pass within a 1 mm depth of field, 45 mm from the lens. The video camera
has an annulus of six high intensity red 130 mW LED’s (light emitting diodes) posi‐
tioned around the camera lens, which results in the flocs being viewed as silhouettes and
produces a clear image of their size and structure. Whilst other studies may refer to muddy
and/or mud-sand mixture particles as aggregates, for simplicity this study will refer to all
aggregated combinations of particles as flocs.

5.3. Flume experimental protocols

The flume was filled with 45 litres of saline water (salinity = 20 ±0.2), to a depth of 0.13 m. The
mixed sediments (both pre-determined and natural) were introduced into the flume as slurries
of known SPM (suspended particulate matter) concentrations. Gravimetric analysis of
extracted water samples was used to monitor the ambient concentration during the flume runs
and check they were within the required experimental tolerances. For each run, different
rotation speeds were used to shear the sediment slurries at shear stresses (τ) ranging from
0.06-0.9 Pa ±5% (equivalent Kolmogorov microscale values are: 381 - 138 µm ; equivalent G-
values, the root mean square of the gradient in the turbulent velocity fluctuations, are: 7.1 –
54.2 s -1) at the floc sampling point. Manning and Whitehouse (2009) report the calibration of
the mini-flume hydrodynamics. Each run was initiated at the fastest rotational velocity and
decreased towards the slowest speed as the run progressed. Further details of the experimental
protocols are outlined by Manning et al. (2007).

The mixed sediment slurries were sheared in the flume for 30 minutes at each stress level. This
duration of shearing, which was pre-determined in accordance with theoretical flocculation
time (TF), allowed each sediment suspension to attain floc equilibrium. Van Leussen (1994)
defines TF as the time required to decrease the number of individual unflocculated particles in
a suspension, to just 10% of the initial number as a result of flocculation.

Floc population sampling comprised careful extraction of a suspension sample from the
same height in the water column as the ADV calibration using a bespoke glass pipette. To
obtain a floc sample, the rotation was stopped for approximately 6-8 seconds, although
flow in the flume still continued through inertia, maintaining particles in suspension
throughout this period. Manning and Whitehouse (2009) showed that the flow does not
significantly slow until at least 15-20 seconds after stopping the drive motor. The floc
sample was then transferred to the LabSFLOC Perspex settling column, whereby each
individual floc was observed by the video camera as it was settling. Parameters of
individual floc size (D) and settling velocity (Ws) were recorded during settling and the
values obtained by video image post-processing. The experimental flow speeds generat‐
ed in the flume were sufficient to keep the fine sand in suspension. The aperture of the
pipette was brought into contact with the settling column water surface and held in place
(vertically) allowing the captured flocs to undergo gravitational settling through the still
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 129
[Link]

water column. Extensive testing of this sampling protocol during the EC COSINUS project
(e.g. Gratiot and Manning, 2004) revealed that this technique created minimal floc
disruption during acquisition. Once floc samples were extracted, the flume lid rotation
continued at the next selected velocity.

5.4. LabSFLOC data processing

Parameters D and Ws, for all settling flocs viewed by the LabSFLOC video camera (for
each sample), were measured simultaneously from the video recordings. Digitisation of the
calibrated images resulted in a pixel resolution of 6.3 µm to determine floc size and
position, from which settling velocity is determined by analysis of sequential images at a
sampling rate of 25 Hz. The effective density (ρe) of each floc was calculated by applying
Stokes’ Law relationship; ρe is the difference between the floc bulk density (ρf) and the
water density (ρw). To apply Stokes’ Law, it is assumed that each sampled floc that fell
through the still water enclosed within the settling column was within the viscous Reynolds
region; i.e. when the individual floc Reynolds number (Re) was less than 0.5. For instan‐
ces where Re exceeded 0.5, the Oseen modification, as advocated by ten Brinke (1994), was
applied in order to correct for the increased inertia during settling. It is assumed that the
measured particle is spherical; that is, it is as ‘deep’ as the measured D size.

The observed flocs were measured within a reference volume of water. By implementing a
sequence of algorithms, originally derived by Fennessy et al. (1997) and modified by Manning
(2004b), the dry mass of a floc population could be compared with the measured SPM
concentration. This provides an estimate of the efficiency of the sampling procedure, and
yielded corresponding rates of MSF. By definition, the data obtained from LabSFLOC are both
of qualitative and quantitative value.

The floc data is presented as individual scatterplots and also as spectral size-banded (SB)
distributions of floc mass and MSF; SB1 represents microflocs less than 40 µm in size and
SB12 are macroflocs greater than 640 µm in diameter. Sample mean values are quoted. To
provide a quantitative framework for population comparisons, the macrofloc and micro‐
floc range of properties were assessed (Eisma, 1986; Manning, 2001), as these parameters
are often used in flocculation modelling. The demarcation point for the macrofloc:micro‐
floc fractions was a floc size of 160 µm (Manning, 2001) and was chosen for two main
reasons: i) this was found to be the most statistically significant separation point for the
majority of the mixed sediment floc populations in terms of mass settling properties; ii) it
also provides computational continuity with previously derived flocculation algorithms for
pure mud suspensions, such as the Manning Floc Settling Velocity (MFSV) algorithms
which describe floc settling at different concentrations within turbulent flow (Manning and
Dyer, 2007). Strictly it should be noted that microflocs are cohesive sediment flocs resistant
to break-up by shear, however, in this study, many pure sand particles fall within the
microfloc size range. Therefore, in this chapter microflocs refer to the ‘fine particle
population’ < 160 µm in diameter. The sand used in the tests also contains a fraction with
grains greater than 160 µm (around 10% by mass). Therefore, the macrofloc fraction may
also contain a number of pure sand grains.
130 Sediment Transport Processes and Their Modelling Applications

5.5. Floc microstructure

In order to examine the floc internal microstructure (matrix) at a sub-micron level (1-2 nm;
Buffle and Leppard, 1995), use of transmission electron microscopy (TEM) was employed
in a separate series of experiments (see Spencer et al., 2010). In addition, energy disper‐
sive spectroscopy (EDS) was used to provide the elemental composition of the floc
components. Samples were prepared for TEM analysis by first stabilising the samples in
glutaraldehyde and embedding the samples in Spurr resin. The samples were polymer‐
ised at 60 ºC overnight. Ultrathin sections of the polymerised resins (50 nm) were obtained
by sectioning with a diamond knife mounted in an ultramicrotome (RMC Ultramicro‐
tome MT-7) and were then mounted on formvar copper grids for analysis. The ultra-thin
sections were then observed in transmission mode at an accelerating voltage of 80 kV using
a JEOL 1200EXIITEMSCAN scanning transmission electron microscope (STEM). The
scanning mode of the STEM was used to generate a microprobe beam for EDS of individ‐
ual floc components in sections allowing observation of minerals across the aggregates. A
Princeton Gamma Tech (PGT) Si[Li] X-ray detector and Imix multichannel analyser
provided spectra of all elements, with an atomic number greater than 10, on a “per colloid”
basis.

6. Experimental results

Sections 6.1-6.5 report findings from the laboratory studies with pre-determined (PD)
mud:sand mixtures conducted by Manning et al. (2007). Sections 6.5-6.6 report a selection of
tests on naturally occurring mud and sand mixtures (NM), and analysis of a mixed sediment
microfloc internal structure, respectively.

6.1. Sediments (PD)

The sand used in the pre-determined mixtures was named Redhill 110, which is a well-
rounded and closely graded silica sand used by HR Wallingford for model testing with
mobile sediment beds. Redhill 110 has a d50 of about 110 µm, with a d10 of 70 µm and a
d90 of approximately 170 µm (Redhill 110 size values quoted are from independent analysis
conducted at HR Wallingford). The experimental mud sample was obtained from the
surface down to a depth of about 50 mm from the Calstock region of the upper Tamar
Estuary (UK) and had an average organic content of approximately 10%. Fitzpatrick (1991)
found Tamar Estuary mud to be generally high in kaolinite clay minerals and Fennessy et
al. (1994) also report microscopic fragments of Tourmaline and Hornblende minerals
present in Calstock mud. This particular mud was used as its floc properties are widely
reported from earlier studies (e.g. Manning and Dyer, 2002b ; Mory et al., 2002 ; Bass et
al., 2006). The mud was collected only a few days before the flume experiments were
conducted, and cold stored (frozen) in a wet form to maximise organic matter preservation.
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 131
[Link]

6.2. Overview of experimental runs (PD)

These experiments comprised a series of three main flume runs, A to C, based on pre-deter‐
mined mud:sand (M:S) ratios (i.e. Run A = 75M:25S, Run B = 50M:50S and Run C = 25M:75S; units
expressed as percentages). These main runs were each divided into 12 minor runs (based on
concentration). This produced a total of 36 mixed sediment floc spectral samples. Three nominal
total SPM concentrations were used: 200 mg.l-1, 1000 mg.l-1 and 5000 mg.l-1. Four shear stresses
were used per run and these were determined by the ADV records as nominal clearwater τ values
of: 0.06, 0.35, 0.6 and 0.9 Pa. The experimental conditions are summarised in Table 1.

Table 1. Overview of experimental runs & samples.

132 Sediment Transport Processes and Their Modelling Applications

During a pilot study to design and refine experimental protocols on the floc population
evolution of a few pre-selected slurries, observations indicated that at a τ of 0.06 Pa the sand
in the upper part of the water column settled to the channel base. However, this preliminary
inspection indicated that there was still sufficient fine sand in suspension in the lower half of
the flume to maintain the nominal mud to sand ratio in the floc sampling region. Furthermore,
during the pilot study, checks were made on mixture homogeneity during suspension and
revealed a nominal 8% mixture deviation (in terms of the sand) for a 75% sand slurry, reducing
to less than 5% for a 75M:25S mixture. These nominal deviations are deemed acceptable for
these mixed sediment flocculation experiments, but are taken into consideration when
interpreting the study results.

During the main flume run, the total suspended concentrations were monitored by gravimetric
analysis of samples withdrawn at the floc sampling point. This analysis indicated that the 200
mg.l-1 total SPM varied the least at ±3%; the higher 5000 mg.l-1 varied by ±4.7%; and the 1000
mg.l-1 slurry nominally varying by ±4.3% by the time of floc sampling. Therefore, these
relatively small deviations demonstrate that the majority of the mixed sediment mass was
remaining in suspension for the shearing duration. Therefore the floc population characteris‐
tics were related closely to the initial total concentrations and mud:sand ratios. Further details
on the homogeneity of mud:sand mixing within the mini-annular flume is reported by
Manning et al. (2009).

6.3. Floc size and settling velocity spectra with mixtures of mud and sand (PD)

To demonstrate the floc properties for suspensions comprising 75M:25S, 50M:50S and 25M:
75S, a number of examples of the individual detailed spherical-equivalent dry mass weighted
floc sizes vs. settling velocity spectra are presented (Figs 4Ai-4Av). The plots represent the
mass-balance corrected floc distributions, thus an individual point on each graph may
represent several flocs with very similar floc characteristics. The diagonal lines on each
scatterplot represent contours of constant floc effective density, ρe, (units = kg.m-3), i.e. the bulk
density minus the water density.

For completeness the full set of D vs. Ws floc distributions for all experiments can be found in
Figs. 5, 6 and 7. By following the plots in each column, starting at the lower plot, one can track
the evolution of the floc populations formed in a constant SPM concentration as the shear stress
rises through the various increments. Similarly, by following the plots from left to right, the
effect of rising concentration on the floc dynamics can be observed. Sections 6.3 and 6.4
summarise some of the key observations from a selection of the populations.

6.3.1. Run A (75M:25S) (Fig. 5)

The flocs from the lower SPM concentration (200 mg.l-1), A1-A4, appear to produce three
separate clusters: a sub-70 µm group, a fraction greater than 160 µm; with a third group
sandwiched in between. For example, the 204 individual flocs that comprised sample A3 (Fig.
5 box A3) ranged from 42 µm to 182 µm in diameter (also Fig. 4Ai). Corresponding settling
velocities spanned 0.3 mm.s-1 to 3.4 mm.s-1 for sample A3.
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 133
[Link]

A. B. C.

i.

ii.

iii.

iv.

v.

Fig. 4. The floc size vs. settling velocity scatter plots (A, left-hand column) for five selected samples: i) A3, ii)
A11, iii) B9, iv) C6 and v) C3. Diagonal lines on figures in column A represent contours of constant Stokes
The floc size
Figure 4. equivalent vs. settling
effective velocity
density: red = scatter
1600 kgm plots
-3 (A, left-hand column)
, green = 160 kgm-3, andfor five=selected
black 16 kgm-3samples: i) A3,
. The centre (B)ii)and
A11, iii)
B9, iv) C6 right-hand
and v) C3.(C)Diagonal
columns lines on figures
represent in column A represent
the corresponding size-bandedcontours
SPM% andof constant Stokes
mass settling fluxequivalent effective
distributions.
density: red
The= size
1600 kgmare
bands -3
, green = 160
illustrated kgm
in the -3
table, and black
below the =plots.
16 kgm-3. The centre (B) and right-hand (C) columns repre‐
sent the corresponding size-banded SPM% and mass settling flux distributions (units = mg.m-2s-1). The size bands are
illustrated in the table below the plots.
134 Sediment Transport Processes and Their Modelling Applications

Figure 5. Distribution floc/aggregate size and settling velocity characteristics for the Run A (75M:25S) samples. Diago‐
nal lines represent contours of constant Stokes equivalent effective density: red = 1600 kgm-3, green = 160 kgm-3, and
black = 16 kgm-3.
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 135
[Link]

Figure 6. Distribution floc/aggregate size and settling velocity characteristics for the Run B (50M:50S) samples. Diago‐
nal lines represent contours of constant Stokes equivalent effective density: red = 1600 kgm-3, green = 160 kgm-3, and
black = 16 kgm-3.
136 Sediment Transport Processes and Their Modelling Applications

Figure 7. Distribution floc/aggregate size and settling velocity characteristics for the Run C (25M:75S) samples. Diago‐
nal lines represent contours of constant Stokes equivalent effective density: red = 1600 kgm-3, green = 160 kgm-3, and
black = 16 kgm-3.

The Run A floc growth was potentially stimulated by a greater abundance of sediment, with
DMax (maximum floc diameter) nearly reaching 700 µm at peak turbidity (5000 mg.l-1). The floc
growth signified a corresponding quickening in Ws with rising SPM, producing WsMax
(maximum settling velocities) of 7-8 mm.s-1 at 5000 mg.l-1; approximately double the speed
exhibited by the dilute sandy mud suspensions. This is demonstrated by A11 (Fig. 5 box A11
and Fig. 4Aii) where the shear stress was the same as A3 (0.35 Pa), but the particle mass in
suspension were raised by a factor of twenty five. Flocs greater than 160 µm comprised 61%
of the total population. In terms of the effects of shear stress, 0.35 Pa seems to produce the
largest, fastest settling macroflocs at 75M:25S. These inter-relationships will be further
examined in the Discussion (Section 7).
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 137
[Link]

6.3.2. Run B (50M:50S) (Fig. 6)

Increasing the sand content to equal the mud fraction (50M:50S), brought about a general
decrease in the macrofloc settling velocity across the entire shear stress range at each base
concentration increment.
In contrast to the macroflocs, the smaller 50M:50S microfloc fractions all displayed quicker fall
rates when compared to the 75% mud in settling rate for each mixed suspension run, with the
5000 mg.l-1 50M:50S mixed suspension (sample B9, Fig. 6) microfloc fraction settling velocity
peaking at a highly turbulent τ of 0.9 Pa.
The B9 size vs. settling velocity floc scatter plot (Fig.6 box B9 and Fig. 4Aiii) shows a “W” or
“double-V” pattern to the aggregates distribution. By this we mean there are small, fast settling
microflocs (nominal 20-40 µm), whose settling velocity range expands at the mid-size microfloc
fraction (nominal 40-80 µm). Then, for the microflocs nominally greater than 80 µm in size, the
spread in the microfloc Ws again reduces, thus producing a “V” shaped distribution. This “V”
pattern is repeated for the macroflocs, with their largest Ws scatter occurring between 185-230
µm for Sample B9.
The microflocs forming the first “V” spanned from 32 µm and up to 114 µm where they form
the apex with the adjacent “V” to form the “W”. At each end of the size range there are
aggregates settling at 5-7 mm.s-1, whilst the middle part of the “V” sections shows flocs falling
as slowly as 0.1 mm.s-1. In the upper left part of the D vs. Ws scatterplot, there are a number
of aggregates which appear to be between 35-50 µm in diameter, settling at 3-6 mm.s-1 and
exhibiting effective densities of 2000-5000 kg.m-3, which is up to three times the effective
density of a sand grain. It is most probable that these are individual fragments of either
Tourmaline or Hornblende; minerals native to the Tamar Estuary and its catchment. The
majority of the aggregate population between 45-90 µm appears to be dominated by sand
grains, with a minimum amount of cohesive matter (i.e. mud content) attached to the sand
grains. These would form very basic, dense, lower order floc structures, which would trap
very little interstitial water. This is indicated by high effective densities (ρe ~1200-1400 kg.m-3),
large fractal dimensions (nf of 2.8-2.9) and low porosities (~10-20%), but they are still not
characteristic of pure (i.e. unflocculated) sand grains.

6.3.3. Run C (25M:75S) (Fig. 7)

Reducing the mud content to 25%, meant the microfloc size fraction tended to dominate the
size and settling dynamics as the total concentration rose throughout Run C. At dilute
conditions, the microflocs represented less than one quarter of the individual flocs for the A1-4
samples; for example C3 (Fig 7. Box C3 and Fig. 4Av). However, with many of the sub-160 µm
C1-4 flocs settling at 4-7 mm.s-1, they were falling significantly quicker than their muddier Runs
A and B counterparts.
A five-fold rise in the total SPM concentration increased the production of smaller flocs, with
the macrofloc size fractions only accounting for 10-20% of the individual aggregates. For
example, nearly 90% of the C6 flocs (τ = 0.6 Pa, SPM = 1000 mg.l-1) were within the microfloc
range (Fig. 7 box C6 and Fig. 4Aiv). This was approximately 15-20% more microflocs when
138 Sediment Transport Processes and Their Modelling Applications

compared to the more cohesive B6 and A6 samples (see relevant boxes in Fig. 6 and Fig 5.
respectively).

The accuracy of effective density values is crucial to the determination of when mixed sediment
particles are flocculating, or if the fine sand particles remain as individual inert entities. The
reliability of the LabSFLOC effective density estimates are demonstrated by their observation
of pure sand grains (Fig. 8). The D and Ws fine sand observations produce a distribution which
closely follows the 1600 kg.m-3 density contour, generally not deviating by no more than ±100
kg.m-3 for over three hundred sand grain observations.

Figure 8. Settling settling vs. floc size for a 100% sand sample. Diagonal lines represent contours of constant Stokes
equivalent effective density: red = 1600 kgm-3, green = 160 kgm-3, and black = 16 kgm-3.

6.4. Floc composition with mixtures of mud and sand (PD)

To illustrate how the floc structure varies at different mud:sand ratios, a few examples will be
presented with the compositional properties (effective density and SPM) as size band distri‐
butions. We start with sample A3 which represents a muddier dilute concentration and the D
vs. Ws scatterplot (see Fig. 4Ai) shows that the macrofloc and microfloc fractions formed three
distinctively separate groups. From Fig. 4Ai we can determine that the microfloc effective
densities (ρe_micro ranging from 200-1580 kg.m-3) were generally an order of magnitude greater
than the macroflocs (ρe_macro from 30-100 kg.m-3). This suggests that together with some
individual sand grains, some of the sand grains may have also been included into the microfloc
structure during the flocculation process.

In terms of the mass distribution across the dilute concentration floc population, the small
microflocs for A3 represented three quarters of the mass (Fig. 4Bi). This is similar to fully
cohesive suspensions within a moderately-high shear zone (τ of 0.6-1 Pa) which suggests the
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 139
[Link]

mixture is still behaving as a cohesive suspension, even with 25% sand present in the initial
mixture. For this sample, the denser, more compact microflocs represented two thirds of the
total 254 mg.m-2s-1 mass settling flux (Fig. 4Ci).
At a concentration of 5000 mg.l-1, the 75M:25S macroflocs of sample A11 (Fig. 4Aii) the
macroflocs were observed to be delicate, low in density (ρe ranging from 20-200 kg.m-3) entities.
The A11 macroflocs now represented 84% of the mass (Fig. 4Bii), which was more than double
the A3 macrofloc mass. Higher turbidity stimulated floc growth in A11, resulting in the largest
flocs (Dmax) growing to 670 µm.
The A11 microfloc fraction consisted of higher density flocs, with the smallest flocs (40-80 µm)
demonstrating effective densities of over 1100 kg.m-3, which are indicative of sand-laden
microflocs or sand grains (where the effective density is greater than 1600 kg.m-3). With sand
accounting for one quarter of the total suspension and the microflocs representing 16% of the
A11 mass, continuity of mass dictates that a reasonable portion of the sand must have been
incorporated in many of the macrofloc structures during the flocculation process. This is very
different from some segregational theories (e.g. van Ledden, 2003) which regard suspensions
of sand and mud as completely independent entities.
Collectively, the fast settling A11 macroflocs contributed 94% of the total mass settling flux (33
g.m-2s-1; Fig. 4Cii); a result of a macrofloc settling velocity of 7.2 mm.s-1, which was nearly three
times quicker than the corresponding Wsmicro. To put this all into context, the A11 total MSF
was 13 times greater than the value computed by the use of an estimated mean settling velocity
of 0.5 mm.s-1; a typical parameterised cohesive sediment Ws value derived from the gravi‐
metric analysis of Owen tube (Owen, 1976) samples. Dearnaley, (1996) summarised the
primary drawback associated with the Owen tube and other field settling tube devices,
including the disruptive nature on flocs of the instrument sampling. Even the A11 microflocs
were settling five times quicker than a 0.5 mm.s-1 parameter value (A11 Wsmicro = 2.5 mm.s-1).
Examination of the 50M:50S sample B9 D vs. Ws scatterplot (Fig. 4Aiii), reveals the presence
of a high density sub-group of flocs (upper left-hand section). These flocs, which are only 35-50
µm in diameter, are settling at 3-6 mms-1 and exhibiting effective densities of up to 2000-3500
kg.m-3. This is up to three times the typical effective density of a sand grain. It is proposed that
these are individual fragments of either Tourmaline or Hornblende; minerals native to the
Tamar Estuary and its catchment (Fennessy et al., 1994). However, given the Tamar’s history
for shipping copper out of Calstock, and the rich mining history for everything from tin to
silver, these heavier particles could be from a number of sources. The majority of the floc
population between 45-90 µm appears to be dominated by sand grains as their effective
densities are typically greater than 1600 kg.m-3, with a minimum amount of cohesive matter
(i.e. mud content) attached to the sand grains. These would form very basic, dense, lower order
floc structures, which would trap very little interstitial water. We could ask the question; if
these high density particles were included in the mud used for all mixtures, why are they
observed only in this case? It is possibly due to uncertainty made when estimating size and
settling velocity of flocs rises as the particles become smaller (i.e. they are harder to detect as
their images are formed from less pixels). Furthermore, these very dense mineral fragments
only constitute a few percent of the total mass.
140 Sediment Transport Processes and Their Modelling Applications

To reiterate, the microflocs tended to dominate the less cohesive Run C samples (25M:75S).
The C6 (τ = 0.6 Pa and SPM = 1000 mg.l-1) macroflocs did not grow larger than 215 µm (Fig.
4Aiv). This was a 40% reduction in size when compared to the corresponding 75M:25S sample
(A6). The low density (effective densities of less than 70 kg.m-3) C6 small macroflocs fell at a
combined average Wsmacro of 1.35 mm.s-1, whilst the Wsmicro was 3.6 mm.s-1. The C6 microflocs
also represented three quarters of the SPM and 90% of the C6 MSF of 3.2 g.m-2s-1 (see Fig. 4Biv
and Fig. 4Civ, respectively). To place this MSF observation into perspective: it was approxi‐
mately double the flux produced either by pure mud or a 75% mixed mud suspension; 31%
greater than a 50:50 mixture could produce, and six times greater than the flux obtained by
using a constant 0.5 mm.s-1 Ws (a typical settling parameter used in cohesive sediment
transport modelling).

The ‘clustered’ appearance depicted by the lower concentration (SPM = 200 mg.l-1) 25M:75S
C3 sample (Fig. 4Av) is similar to Sample C6 (Fig. 4Av). The shear stress was less turbulent
(τ = 0.35 Pa) than C6, so one would assume the floc settling dynamics would improve.
However, the removal of three quarters of the cohesive matter meant that the Wsmacro was only
0.9 mms-1; half the Wsmacro for the 75M:25S run A3. As with the 1000 mg.l-1 C6 suspension, the
C3 macroflocs only represented a quarter of the SPM (Fig. 4Bv). The main difference between
the lower and the higher Run C suspension was fewer individual unflocculated sand grains
in the suspension at the lower turbidity.

6.5. Analysis of macrofloc: Microfloc trends (PD)

This section will look at the macrofloc and microfloc (Eisma, 1986) settling velocity trends (i.e.
Wsmacro and Wsmicro respectively) calculated from the pre-determined mud:sand mixture data
presented earlier in Section 6.3. A dual-modal approach is advised when assessing parame‐
terised floc settling and floc mass population data, as it tends to be more realistically repre‐
sentative than a single sample average (Dyer et al., 1996; Mietta, 2010), especially when
considering the effects of mass settling fluxes to the bed (Baugh and Manning, 2007). This
approach also permits quantitative inter-comparisons with previous pure mud flocculation
studies.

The density contours superimposed on the Ws vs. D scatterplots presented in Section 6.3
indicate that only a minimum number of sand grains remained in an unflocculated state. This
was confirmed from an assessment of both the effective density and SPM distributions.
Therefore these few grains were included in the microfloc analysis presented in this section,
as they form part of the total suspension and this provides the continuity of mass when
comparing the different samples. However, to make these assessments fully rigorous, the mud
fraction of the samples will be isolated and examined independently in the ‘modelling
implications’ section (see Section 9).

6.5.1. Run a using 75% mud: 25% sand

Fig. 9 shows the macrofloc and microfloc averaged settling velocity plots which cover both the
pre-determined mixtures experimental concentration and shear stress ranges. The solid lines
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 141
[Link]

on Figs 9.A and 9.B correspond to the 25% sand mixed suspensions; the dotted curve lines are
the contrasting 100% mud suspension outputs from the MFSV (this prediction was calibrated
principally for Tamar mud extracted from the same study location). The straight dotted lines
represent the d50 and d10 settling rates of pure sand grains determined by the SandCalc sediment
transport computational software package (HR Wallingford, 1998).

A) 75M:25S Macroflocs B) 75M:25S Microflocs

C) 50M:50S D) 50M:50S

E) 25M:75S
F) 25M:75S

Figure 9. Wsmacro (left column, y-axis, units = mm.s-1) & Wsmicro (right column, y-axis, units = mm.s-1) values for runs A
(75M:25S), B (50M:50S)
Fig. 9. Wsand
macro C (25M:75S),
(left plottedunits
column, y-axis, against shear
= mm.s -1
)&stress (x-axis,
Wsmicro (rightunits = Pa).
column, Solidunits
y-axis, lines=+ symbols indicate
-1
mixed sediment floc data points. Dashed lines indicate predictedand
mm.s ) values for runs A (75M:25S), B (50M:50S) behaviour of 100%
C (25M:75S), mud
plotted macroflocs
against shear at three concen‐
stressmud
trations, and 100% (x-axis, units = at
microflocs Pa). Solid lines
a single + symbols indicate
concentration. mixed sediment
Lines indicating SandCalc flocestimated
data points.
settling velocities of
unhindered d10Dashed
and d50lines
pureindicate predicted
sand grains behaviour
are also [Link] 100% mud macroflocs at three concentrations,
and 100% mud microflocs at a single concentration. Lines indicating SandCalc estimated
settling velocities of unhindered d10 and d50 pure sand grains are also plotted.
Substituting 25% of the pure mud suspension for sand produced a distinct change to the
macrofloc settling velocity (Fig. 9.A). Starting at the lowest concentration (200 mg.l-1), the
quiescent conditions of 0.06 Pa only produced a Wsmacro of 0.65 mm.s-1: nearly half the settling
142 Sediment Transport Processes and Their Modelling Applications

rate of pure mud. As the shear stress increased, the floc dynamics respond and the settling
velocity increased to a maximum of 1.7 mm.s-1 at 0.35 Pa, which was 0.8 mm.s-1 slower than
pure mud at the same concentration. The intermediate concentration (1000 mg.l-1) Wsmacro
closely mimicked the settling profile of pure mud macroflocs at the less turbid 200 mg.l-1. This
is primarily a result of the 75M:25S suspension lacking sufficient cohesion because it only
comprises 75% mud and the potential level of flocculation is more restricted than pure mud.
The mixed sediment macroflocs also demonstrated lower effective densities (~30-50 kg.m-3)
than their pure mud counterparts.

The smaller mixed sediment microfloc fractions all settled faster than the pure mud equiva‐
lents, at each stress increment (Fig. 9.B). Where the macrofloc mixed fraction showed settling
peaks at 0.35 Pa, similar to natural muds (Manning, 2004b), the mixed Wsmicro tended to produce
a maximum at the higher turbulent shear stress of 0.6 Pa.

At high turbidity (5000 mg.l-1), the macroflocs were nearly three time more dense than at lower
turbidity. This saw the Wsmacro peaking at 7.2 mm.s-1, which was 2.5 mm.s-1 faster than the 100%
mud equivalent, and 0.4 mm.s-1 quicker than a d50 pure sand. The corresponding Wsmicro was
2.7 mm.s-1, which was similar to a d10 sand grain and 1.7 mm.s-1 quicker than pure mud
microflocs.

6.5.2. Run B using 50% mud: 50% sand

Increasing the sand content to equal the mud fraction (50M:50S), brought about a general
decrease in the macrofloc settling velocity across the entire shear stress range at each base
concentration increment (Fig. 9.C). For the 200 mg.l-1 slurries sheared at 0.35 Pa, the equally
mixed sediment produced a Wsmacro of 1.6 mm.s-1, a reduction of 0.1 mm.s-1 from the 75% mud,
and was 0.8 mm.s-1 slower at settling than the pure mud benchmark.

At the highest suspended concentration (5000 mg.l-1), and again at a turbulent stress of 0.35
Pa, the 50M:50S slurry produced a Wsmacro of 5.4 mm.s -1. This was 0.8 mm.s-1 faster than pure
mud, but 1.8 mm.s-1 slower than the 75M:25S macroflocs. This large Wsmacro difference exhibited
between the 75M:25S and 50M:50S mixtures, decreased as the TKE dissipated to a lesser level.
However, both mixed suspension macroflocs at the low shear stress were still slower than pure
mud, which settled considerably faster.

In contrast to the macroflocs, the smaller 50M:50S microfloc fractions (Fig. 9.D) all displayed
quicker settling velocities when compared to 75M:25S. The one main exception was the 5000
mg.l-1 concentration, where Wsmicro achieved a maximum speed of 3.3 mm.s-1; which was 2.3
mm.s-1 faster than pure mud and 0.75 mm.s-1 quicker than the corresponding 75M:25S
microflocs.

6.5.3. Run C using 25%mud: 75% sand

The addition of a greater amount of sand particles in suspension significantly enhanced the
settling dynamics at their respective shearing stresses which stimulate maximum flocculation.
All 25M:75S values of Wsmicro exceeded the purely cohesive suspensions by more than a factor
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 143
[Link]

of two (Fig. 9.F), and the majority of the microfloc samples also exceeded the settling rate of a
d10 sand grain. At an SPM concentration of 200 mgl-1, the Wsmicro at 0.06 Pa was 1.8 mm.s-1 and
increased to a peak of 3.3 mm.s-1 at 0.6 Pa. By increasing the SPM concentration to 5000 mg.l-1,
the Wsmicro maximum peaked at 4.7 mm.s-1. This was approximately five times faster than the
value for 100% mud, and nearly double the equivalent 75M:25S Wsmicro (Fig. 9.F).

Conversely, all macrofloc fractions settled significantly slower within the less cohesive
suspensions. At peak turbidity, the macrofloc fraction fell at 3.5 mm.s-1; this was the sole
macrofloc fraction to exceed the settling velocity of d10 sand. In fact, this 25M:75S macrofloc
fraction was 1.2 mm.s-1 slower than the corresponding Wsmicro from the same run.

In terms of the particle mass distribution: as the percentage content of non-cohesive sediment
rose (i.e. mud content decreased), the relative contribution of the microfloc fraction to the total
SPM concentration in each population increased.

6.6. Comparative data for sediment from Portsmouth Harbour – Natural Mixture (NM)

To support the data derived from the pre-determined mud:sand slurries, a selection of
naturally occurring mixed sediment samples collected from within Portsmouth Harbour (a
tidal inlet on the southern coast of the UK) were also assessed using the same type of laboratory
flume runs (Pidduck and Manning, in prep.). The same protocols used for the pre-determined
mixture experiments, were adopted for these runs. Sediment transport in Portsmouth Harbour
has been studied by Hydraulics Research (1959), Lonsdale (1969) and Harlow (1980). Regular
dredging activities for military vessel access to the Royal Naval Base, combined with an ebb-
dominant macrotidal regime, mean that the fine mud and coarser sands that reside in the
Harbour can become mixed.

Two Portsmouth Harbour samples at a constant SPM concentration of 2000 mg.l-1 and sheared
at 0.35 Pa are described. The first suspension, 4_A (Fig. 10a), was a low cohesive sediment
composed of 38M:62S (including coarse silts). Loss-on-ignition tests indicated that sediment
4_A was approximately 6% organic. The 4_A flocs ranged in size from 29-313 µm, although
there is an absence of particles in the 33 to 69 µm range. The smallest microflocs (2% of the
population) all demonstrate effective densities of quartz and beyond, which suggests the
presence of some very dense minerals; possibly some metallic particles. The larger microflocs
were less dense (~ 700 kg.m-3).

The 4_A microflocs comprised just over half of the SPM, with their settling velocities spanning
three orders of magnitude from 0.36-34 mm.s-1. This resulted in a Wsmicro of 5.4 mm.s-1, which
was 1.3 mm.s-1 quicker than the larger macroflocs. This was due to the macroflocs demon‐
strating effective densities predominantly below 200 kg.m-3, which are more indicative of
cohesive flocs.

The second sample, 6_B (Fig. 10b), was more cohesive as it contained only 30% sand (70M:30S)
and the sediment mixture had 8.4% organic matter present within its matrix. Where the sample
4_A D vs. Ws distribution favoured the smaller size fractions, 6_B depicts a population more
characteristic of a pure mud. The microflocs were distinctly slower in settling, ranging from
144 Sediment Transport Processes and Their Modelling Applications

a) 100
Settling Velocity (mm/s)

10

0.1
10 100 1000
b) 100
Floc Size (microns)
Settling Velocity (mm/s)

10

0.1
10 100 1000
Floc Size (microns)

Figure 10. Settling settling vs. floc size for Portsmouth Harbour samples: a) 4_A (38M:62S); b) 6_B (38M:62S). Both
samples had nominal 2 g.l-1 total SPM concentrations and were sheared at a stress of 0.35 Pa. Diagonal lines represent
contours of constant Stokes equivalent effective density: red = 1600 kgm-3, green = 160 kgm-3, and black = 16 kgm-3.

2-8 mm.s-1. All flocs were also less dense than 4_A; effective densities under 740 kg.m-3, with
the largest flocs having a ρe of just 20 kg.m-3.

The macroflocs comprised nearly two thirds of 6_B population and over three quarters of the
mass. The macrofloc and microfloc settling dynamics of the Portsmouth Harbour samples, at
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 145
[Link]

the three induced shear stresses (0.06, 0.35 and 0.6 Pa; 0.9 Pa was not available for the Ports‐
mouth Harbour tests), are illustrated in Fig. 11. The data reveals some interesting settling
velocity trends and these will be discussed in Section 7.

5
Settling velocity (mm/s)

2 Ws_Macro (38M:62S) Ws_micro (38M:62S)

Ws_Macro (70M:30S) Ws_micro (70M:30S)

1

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Turbulent shear stress (Pa)

Figure 11. Wsmacro and Wsmicro values plotted against shear stress for Portsmouth Harbour samples 4_A (38M:62S) and
6_B (70M:30S). Both samples had nominal 2 g.l-1 total SPM concentrations and were sheared at a stress of 0.35 Pa.

6.7. Floc microstructure

To illustrate how both non-cohesive and cohesive sediments components can combine in
natural microflocs, electron micrographs of cross-sections through natural microflocs from the
Tamar Estuary (UK) are shown in Fig. 12. The low resolution TEM image which encompasses
the entire microfloc (Fig. 12a) shows the complex matrix of structurally interdependent
components of a typical floc section. Both organic and inorganic particles are present creating
a highly porous, high water content, three-dimensional sedimentary matrix.

7. Discussion of experimental findings

7.1. Settling velocity

This section addresses issues relating to research questions i-iii listed in Section 5. A number
of generalised trends, in terms of the settling velocity, can be deduced from the macrofloc and
microfloc data. The macrofloc settling velocities generally slowed as the sand content rose.
These macroflocs fell slightly quicker than the microflocs at low turbidity, but almost three-
times as quick at the higher suspended concentration. However, as the mud content decreased,
146 Sediment Transport Processes and Their Modelling Applications

(a)

(b)

Figure 12. Low resolution (a) and high resolution (b) TEM images of a natural microfloc composed of a mud:sand mixture.

the particle cohesion efficiency would also reduce and could potentially limit the floc growth
potential, curbing the equilibrium floc size of the macrofloc fraction.

The microfloc settling responded to a greater abundance of sand, whereby the greater the sand
content in a mixed fraction - the faster the Wsmicro. For example, for a 25M:75S suspension, the
microfloc settling velocities demonstrated a three-fold increase at low turbidity and nearly
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 147
[Link]

doubled in settling speed at high turbidity to produce Wsmicro of 3.5 mm.s-1 and 4.7 mm.s-1,
respectively. The effective density data of many of the microfloc fractions from the pre-
determined mixtures tests ranged from 800-1200 kg.m-3. This would suggest that the finer sand
grains tended to interact and bond better with the smaller floc structures, accounting for the
quicker microfloc settling velocities observed.
The flocs produced from the natural Portsmouth Harbour sediments showed similar general
settling velocity patterns to those of the pre-determined Tamar mixed suspensions. For the
less cohesive 38M:62S slurry (sample 4_A), the microfloc fraction settled quicker than the
macroflocs. By taking into account differences in SPM and M:S ratio, one can deduce that the
4_A microflocs were settling approximately 1.5 mm.s-1 quicker than their manufactured slurry
equivalents, whilst the Portsmouth macroflocs fell nearly twice as quick as their pre-deter‐
mined slurry equivalent. This could be a result of slightly larger sand grains present in the
Portsmouth 4_A sediment and also stronger bio-film coatings present in the 4_A mixture
providing extra adhesion for the sand grains permitting greater uptake within the macrofloc
fraction.
It is interesting to observe that the microflocs in 4_A produced their fastest settling velocities
at a τ of 0.6 Pa, whilst the Wsmacro peaked at a less turbulent 0.35 Pa. This can be explained by
the denser microflocs being stronger than the weaker macroflocs, hence they can survive larger
stresses. The ratio of a floc’s diameter to the corresponding dissipating eddy size, such as the
Kolmogorov microscale (1941a, b), in turbulent flow is a fundamental governing condition for
estuarine flocculation dynamics (Tomi and Bagster, 1978; Tambo and Hozumi, 1979; McCave,
1984). Furthermore, if settling velocities are large, more turbulent energy is required to keep
those flocs in suspension.

7.2. Composition and SPM distribution

Aspects relating to research question iv are now discussed. The LabSFLOC data has provided
evidence of how sand grains can be potentially included within a floc matrix. The Ws vs. D
spectra show that only a minimal amount of potentially unflocculated pure sand particles are
present in a few of the samples; this is in terms of both individual numbers and the percentage
of the total SPM (typically less than 1-2% of the total mud:sand concentration). An accurate
mass balance between the predetermined mixed suspension introduced into the flume at the
commencement of each run and the filtered SPM obtained from each sample promotes
confidence in the mixed sediment LabSFLOC floc observations.
The LabSFLOC sampling protocol of measuring D and Ws simultaneously means that data on
individual floc effective density is available. The latter provides important information about
the composition of each floc (Dyer, 1989). The data identifies that there is a wide range in
effective densities exhibited across each spectrum, particularly in the microfloc range, but most
are less than pure quartz (~1600 kg.m-3). Transmission electron microscopy (TEM) images have
also visually identified the presence of both clay minerals and quartz mineral fragments within
natural microfloc structures (Spencer et al., 2010). This leads to the suggestion that when mixed
sediments flocculate, the sand particles favour the microfloc fraction, which is logical reason‐
ing: microflocs tend to have the stronger bonding potential due to the closeness of the bonds.
148 Sediment Transport Processes and Their Modelling Applications

Uptake of individual sand particles will probably be much less in the macroflocs. This is
consistent with the order of aggregation theory (Krone, 1962; Eisma, 1986) which states that
microflocs will flocculate into macroflocs when the ambient conditions are favourable. This
provides a more efficient mechanism / pathway for the fine sand grains to move into the
macrofloc fractions.
The EDS floc structural analysis of the TEM floc images presented in Section 6.7, identified
that the microfloc inorganic constituents primarily comprised planar clay minerals (identified
by the thin dark grey objects in Fig. 12) and fine quartz fragments (all much smaller than the
mean sand grain size), evident from concoidal fracturing (the black marks in Fig. 12a and 12b).
Other minerals present included Fe and Mn oxides and opaque sub-cubic minerals (probably
pyrite), which are all typical of estuarine sediments. The organic constituents are predomi‐
nantly observed to be bacteria and their EPS (extracellular polymeric substance; see Under‐
wood and Paterson, 2003; Tolhurst et al., 2002) fibrils, which are produced by the bacteria for
attachment, assimilation of food (dissolved organic carbon) and for protection from predation
and contaminants. In the high resolution TEM image of the microfloc (Fig 12b.), the EPS can
be seen linking the biological and inorganic particles and represents a micro-structural
framework of the floc matrix (Fig. 12b). The EPS matrix is considered to be the component of
the floc that enhances floc building and provides it with its strength.
For the Tamar mixtures, with a sand d50 of 0.11 mm, it is geometrically possible that only one
sand grain may form a microfloc. The data shows that many of the microflocs exhibited
effective densities significantly less than pure quartz, but higher than most pure mud micro‐
floc. This suggests that the mixed sediment microflocs could be either combined mixtures of
very fine quartz fragments and mud, as illustrated by the TEM images, or they could be
individual larger quartz particles which are coated in organic mud. For example, Whitehouse
et al. (2000) offer a scenario where mud can create a ‘cage-work’ structure which can fully
encompass the sand grains, thus trapping the sand within a clay floc envelope. Mehta et al.
(2009) observed flocs of various sizes in Lake Apopka (Florida, USA) where the inorganic
particles are held together by embayment within a spacious exopolymeric biofilm (e.g. organic
mucus) (Fig. 13). Such flocs do not conform to the mathematical fractal description typically
attributed to predominantly inorganic flocs (e.g. Winterwerp and van Kesteren, 2004, Winter‐
werp et al, 2006), because there is no floc formation that can be described as the primary
structure. All these cases would produce microflocs which are both less dense than their
constituent minerals, but would have the potential to bond with a macrofloc due to their part-
biological matrix.

7.3. MSF distributions

By combining the settling velocity and mass distribution findings, it is possible to assess the
mass settling flux (i.e. the product of the concentration and the Ws); this enables aspects of
research question v to be discussed.
The combined effects of particle concentration and turbulent shearing have long been attrib‐
uted to the growth of mud flocs (e.g. Tsai et al., 1987; Burban, 1987; Puls et al., 1988; Kranck
and Milligan, 1992). Under optimum flocculation conditions, Mehta and Lott (1987) suggested
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 149
[Link]

Figure 13. A very porous (low density) floc, composed from a translucent organic coating eveloping a solid (opaque)
core (from Mehta et al., 2009).

that pure mud macroflocs tend to contribute most to the MSF, on account of high instability
(van Leussen, 1994) due to floc growth potential producing a greater number of larger
macroflocs with fast settling velocities. Observations in estuaries reveal these pure mud
macroflocs can typically grow to mean a diameter > 400 µm, exhibiting effective densities of
less than 40-50 kg.m-3 and becoming more than 95% porous. These macroflocs are highly
delicate entities and are easily progressively broken apart as they pass through regions of
higher turbulent shear stress (Glasgow and Lucke, 1980). However, the data presented in this
chapter indicates a trend whereby an increase in sand content, and a subsequent decrease in
mud, favours the microflocs as the dominant flux contributor.

For example, if we consider a flocculating mixture comprising 25% mud and 75% sand, at a
nominal concentration of 1000 mg.l-1 and sheared at a τ of 0.6 Pa (i.e. Sample C6 ; see Fig. 4Civ),
this results in the microflocs representing three quarters of the SPM. Therefore, the microfloc
fraction would be contributing 88% of the total MSF (3.08 g.m-2s-1). To place this MSF value
into perspective: it is approximately double the flux estimated for either a pure mud or a 75%
mixed mud suspension; nearly 30% greater than the flux for a 50M:50S mixture; and six times
greater than the MSF obtained by using a constant 0.5 mm.s-1 settling velocity.

In contrast, by maintaining the ambient SPM concentration at 1000 mg.l-1, but making the
suspension 75% cohesive (i.e. 75M:25S), when it is sheared at 0.35 Pa (Sample A7) the total
MSF (2.2 g.m-2s-1) would be weighted 73%:27% in favour of the macroflocs. This settling flux
150 Sediment Transport Processes and Their Modelling Applications

distribution is more characteristic of a fully cohesive suspension (Manning and Bass, 2006).
This suggests that with just an 8% lower MSF than pure mud, the 75M:25S mixture is behaving,
to some degree, predominantly as a cohesive suspension, even with 25% fine sand present in
the mixture.
The data shows that the greater the sand content of a mixed suspension, the higher the total
MSF. Although it is not possible to state how much, or even when, cohesive material attaches
to individual sand grains, the effective density distributions (see Figs 4Ai-v) indicate that many
of the microflocs are less dense than quartz (for a nominally mass-balanced mud:sand mixture).
The Wsmicro generally rose with rising sand content. One can see that this smaller size fraction
is extremely important in terms of the total MSF for less cohesive suspensions. By averaging
the MSF over the entire concentration and shear stress ranges for a nominally constant ratio
of mud and sand, the data reveals that for a predominantly sandy suspension (Run C - 25M:
75S), the microflocs represented the majority (~80%) of the total MSF. In contrast, the microflocs
contributed less than half (~42%) of the settling flux for the muddier 75M:25S slurry (Run A).
With the sandier 4_A Portsmouth microflocs (see Fig. 10a) representing over half of the
total 2000 mg.l-1 suspension and the macroflocs comprising three quarters of the more
cohesive sample 6_B flocs (see Fig. 10b), the Portsmouth samples displayed a similar mass
distribution to those of the Tamar pre-determined slurries. In terms of the MSF, Sample
4_A produced a resultant 9.9 g.m-2s-1, which was approximately 50% greater than the Tamar
manufactured suspension. Whilst the Sample 6_B depositional flux, 13.6 g.m-2s-1, was more
than three times the settling flux of the Tamar equivalent mixtures. The higher mass settling
fluxes were a function of the quicker settling velocities demonstrated by the Portsmouth
Harbour suspensions.
A direct comparison of the mass settling fluxes and their associated dynamics, can also provide
a practical way to illustrate the enhanced / increased flocculation with respect to turbulent
intensity. If we consider the 5000 mg.l-1 B9 floc sample from the 50M:50S suspension, the very
turbulent (τ = 0.9 Pa) environment produced a net MSF of 13.8 g.m-2s-1 (see Fig. 4Ciii), with just
half the flux attributed to the macroflocs. In comparison the more advanced flocculation of the
less turbulent (τ = 0.35 Pa) Sample B11, resulted in a MSF of 26.2 g.m-2s-1. This was nearly double
the Sample B9 flux and was primarily due to the B11 macroflocs contributing 80% of the total
flux. The fast settling (Ws of 6-14 mm.s-1) macroflocs ranging from 482 to 650 µm produced
nearly one quarter of the B11 MSF.

8. Parameterisation of mixed sediment flocculation

Since the mid-1990s, much research has been conducted in Europe on the parameterisation of
the natural flocculation process, through projects such as COSINUS - Prediction of COhesive
Sediment transport and bed dynamics in estuaries and coastal zones with Integrated NUmerical
Simulation models (see Berlamont, 2002). A significant degree of progress has been achieved on
the practical modelling of flocculation (e.g. Winterwerp et al., 2006; Baugh and Manning,
2007, Soulsby and Manning, 2012). In terms of general modelling applicability, these floccu‐
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 151
[Link]

lation advancements are still limited to the modelling of solely pure cohesive sediment
estuaries. Due to the complexity of the mixed sediment flocculation process (as demonstrated
in this chapter), statistical relationships between floc properties acquired from direct empirical
observations can be used to quantify the response of flocculation to different environmental
conditions (Manning and Dyer, 2007). Fig. 14 shows a conceptual representation of the 25M:
75S data compared to pure mud suspensions, at a SPM concentration of 5000 mg.l-1. The mixed
sediment macrofloc settling curve, within a turbulent shear stress (τ) region of 0.06-0.6 Pa, can
be quantified by the following algorithm:

Ws macro = 0.259 + 5.76*t– 7.61*t2 + 0.000317*SPM (1)

Figure 14. Conceptual illustration of Wsmacro (blue lines) & Wsmicro (red lines) trends for a mixed sediment suspension of
ratio 25M:75S (solid lines) and a pure mud (dotted lines) suspension, all for a total concentration of 5 gl-1, plotted
against shear stress.

A parametric multiple regression was used to generate Eqn 1. For this particular type of multi-
regression derivation we are using non-homogeneous dimensions, therefore the units used
are as follows: Wsmacro = mm.s-1, τ = Pa, and SPM = mg.l-1. Demonstrating an R2 = 0.84, the
algorithm is a close approximation of the parameterised observations covering the 200-5000
mg.l-1 laboratory experimental SPM concentration range. Eqn 1 is just one form of algorithm
and others can be generated from the data depending upon the modelling input variables.

The general structure of Eqn. 1 is similar to the pure mud macrofloc settling velocity relation‐
ship derived by Manning (2004a) as part of the Estuary Processes Research Project – EstProc
(Estuary Process Consortium, 2005). The general shape of the Eqn. 1 curve is similar to the
flocculation schematic proposed by Dyer (1989), with an increase in settling velocity at low
stress due to flocculation enhanced by shear, and floc disruption at higher stresses for the same
concentration. Also, the combined influence of concentration and turbulent shear on the
control of the macrofloc properties, as listed in Eqn. 1, agrees with the hypotheses offered by
both Puls et al. (1988) and Kranck and Milligan (1992).
152 Sediment Transport Processes and Their Modelling Applications

However, the relative magnitudes and peaks in the mixed sediment conceptual curves
(illustrated in Fig. 14) differ from the pure mud representations in a number of ways. The
microflocs in the 25M:75S mixed suspension microflocs settle at a maximum velocity of 4.7
mm.s-1; this is 380% quicker than the equivalent pure mud and nearly double the Wsmicro for a
75% mud suspension. Interestingly, the mixed suspension Wsmicro is virtually the same as the
macrofloc settling velocity for pure mud. The peak 25M:75S Wsmicro occurred at a shear stress
of 0.6 Pa, which falls within the “moderately-high” shear stress zone (Manning, 2004a); 0.2-0.3
Pa above the shear stress region typically recognised as producing optimum stimulation for
pure mud flocculation (Manning, 2004a).

Manning and Dyer (2007) demonstrated that, for varying levels of suspended concentration,
mud microfloc settling in turbulent flows could be represented by a single algorithm curve.
In contrast, mixed sediment microfloc settling velocities appear to be dependent upon both
concentration and shear stress variations, as well as the proportion of mud and sand. This is
indicated by different curves representing Wsmicro throughout a shear stress range at varying
concentration levels, even when the mud:sand ratio is constant. From this we can deduce that
the mixed sediment Wsmicro parameter is far more sensitive to changes in SPM concentration,
compared to pure mud microflocs whose dynamics only seem to vary with turbulent shear
stress.

If we now examine the macrofloc settling for the conceptual curve for a 5000 mg.l-1 25M:75S
mixed suspension (Fig. 14), one can observe that the maximum Wsmacro of 3.5 mm.s-1 occurs at
a shear stress of about 0.35-0.4 Pa; the same stress range as pure mud macroflocs. However,
the 25M:75S macroflocs are settling 1.2 mm.s-1 (or 25%) slower than both the Wsmicro peak for
the mixed sediments and the Wsmacro for pure mud.

If we consider the mixed sediment settling velocity variations in terms of Krone’s (1963) classic
hierarchical order of aggregation theory, the smaller microflocs (D < 160 µm) are generally
considered to be the building blocks from which the macroflocs are composed. The microflocs
tend to display a much wider range in effective densities and settling velocities than the
macrofloc fraction. It is highly plausible that for mixed sediments, the microfloc fraction
samples may comprise both flocculated mud and some unflocculated sand grains depending
on mud:sand ratio, concentration and shear stress. This could account for the faster microfloc
settling velocities with rising sand content and concentration.

The macroflocs are deemed to be composed of microflocs, so this fraction will also contain both
cohesive and non-cohesive particulates. The intra-bonding of microfloc to microfloc is usually
far weaker than the closer internal particle bonds of individual microflocs. This means that
macrofloc bonding relies heavily on the sediment cohesional properties (primarily those from
extra-cellular polymeric substances), and these will exponentially decrease with muddy
sediments being replaced by non-cohesive sands.

The parameterisation of biological process for inclusion in numerical sediment transport

models is notoriously difficult, and algorithms such as Eqn. 1 do not include a specific
“biological” term. However, where the algorithms are based on data derived from natural
sediments which would include some of the biological effect. A limitation of many mixed
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 153
[Link]

sediment laboratory studies, is that the mud:sand matrix is over-simplified through the use of
a pure clay mineral (e.g. kaolinite) devoid of any biology. As clay minerals only flocculate
through electrostatic (i.e. salt) flocculation, at best a segregated environment may be simulated
if the water is brackish, but resultant mixed sediment flocculation effects will never be
observed.

9. Modelling implications of mixed sediment flocculation

The prediction and modelling of mud:sand segregation effects on processes such as deposition
are very useful from an estuarine management perspective. Numerical models are typically
the chosen tools with which estuarine management groups attempt to predict sediment
transport rates. In order for these models to provide sufficiently accurate results, a good
scientific understanding of the flocculation process and interactions between mud and sand
is required (e.g. Chesher and Ockenden, 1997; van Ledden, 2002; Waeles et al., 2008), and these
processes need to be adequately described mathematically.
The complexity of mud:sand suspensions and a general lack of suitable experimental data
which can describe the resultant dynamics of different mixtures of mud and sand, means that
most numerical sediment transport models treat mud and sand as entirely separate entities.
These conditions may exist for a segregational environment. However, if the mud:sand
particles interact as a combined matrix, it has the potential to flocculate (as demonstrated in
this chapter). This research has indicated that when mud and sand are mixed in different ratios
and interact, the level of inter-particle cohesion can also vary and this is reflected in the
macrofloc:microfloc mass settling flux distributions. Therefore it may be important for
modellers to consider potential flocculation effects when parameterising mixed sediment
deposition in turbulent flows that are conducive to flocculation.
When faced with a potential mixed sediment regime, an estuarine sediment transport modeller
has two initial basic choices. The first and most simple option, is to assume that the mud:sand
mixtures act solely as one sediment type when suspended, thus entirely demonstrating either
cohesive or non-cohesive settling characteristics. If all sediment is assumed to be non-cohesive,
e.g. pure sand grains devoid of any cohesive matter, the SPM would behave as inert particles
as their dynamic settling spectrum would not alter greatly with increasing concentration as
they do not flocculate. Similarly pure sand grain dynamics are not affected by shear stresses
in the same way muddy sediments are. Thus, the settling properties of pure sand suspensions
are similar over the SPM concentration range (200-5000 mg.l-1) encompassed by the flume
experimental data reported in this chapter; this is also because the influence of hindered
settling is not important in this range of concentration. In contrast, if all SPM present is deemed
to be pure mud, flocculation will completely dominate the settling process.
The second option acknowledges the presence of a mud:sand mixed environment; the issue is
then how this is handled. For example, Van Ledden’s (2002) mixed sediment model employed
the segregational criteria for low concentration depositional simulations in which flocculation
effects are ignored. However, if it is assumed that the mixed suspensions are acting in a
154 Sediment Transport Processes and Their Modelling Applications

segregated manner, when in fact they are demonstrating a degree of flocculation, a wide range
in predicted settling flux errors may arise from the modelling output.
To illustrate the potential pitfalls of solely using either a sand or mud settling parameterisation,
when there is actually a flocculating mud:sand mixture present, we compare fraction-
maximum settling velocities for: pure mud, pure sand and a 50M:50S ratio suspension, all at
an SPM concentration of 200 mg.l-1. For the 100% mud condition, the respective macrofloc and
microfloc settling velocities are 2.4 mm.s-1 and 0.9 mm.s-1. The contrasting pure sand settling
velocity values are Wsmacro_sand = 20.1 mm.s-1 and Wsmicro_sand = 7.4 mm.s-1; this was a comparative
7 to 8 -fold settling velocity rise for the two respective pure sand fractions, over the pure mud.
An equal division of mud and sand resulted in an observed mixed sediment macrofloc settling
velocity of 1.6 mm.s-1, which was more than twelve times slower than the pure sand macrofloc-
equivalent sized fraction and two thirds the velocity of the pure mud macroflocs. However,
the observed 50M:50S microflocs fell at 2.2 mm.s-1, which was three-times slower than pure
sand, and twice as fast as pure mud suspensions. This example demonstrates the importance
of obtaining high quality temporal and spatial settling velocity data of mixed sediments in
suspension. It is anticipated that the effects of mixed sediment flocculation on numerical
sediment transport modelling, will be the topic of future research and publication.

10. Conclusion

The aim of this chapter was to provide an overview of mixed sediment flocculation dynamics
and how they can influence sediment transport. It has drawn on key literature and new data
to address this aim. The theoretical aspects relating to the flocculation of mud:sand mixtures
include flocculation processes, segregation versus flocculating suspensions, and biological
influences on mixed sediment flocculation.
In order to demonstrate the flocculation potential and characteristics of mud:sand mixtures,
the second part of the chapter has drawn on the findings from recently completed laboratory
studies that examined the flocculation dynamics for mud:sand (M:S) mixtures primarily using
Tamar estuary mud and silica sand at different concentrations and shear rates in a mini-
annular flume. Turbulent shear stresses during the experimental runs ranged from 0.06-0.9 Pa
(±5%), with maximum flow speeds in the annular flume of about 0.7 m.s-1, for three total
suspended sediment concentrations representative of estuarine concentrations, namely 200,
1000 and 5000 mg.l-1. The video-based LabSFLOC instrument was used to determine floc
properties including size, settling velocity, density, and mass.
The experiments showed that as mud content decreased, the particle cohesion efficiency
reduces which can limit the growth potential of the macrofloc fraction (sizes > 160 µm). For a
75M:25S suspension, the settling velocity Wsmacro was slightly quicker than the microflocs at
200 mg.l-1, but almost three-times as fast at the higher suspended concentration (5000 mg.l-1).
Parameterised data indicated that by adding more sand to a mud:sand mixture, the settling
velocity of the macrofloc fraction slows and the settling velocity of microflocs (sizes < 160 µm)
increases.
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 155
[Link]

In terms of floc composition, effective density data of many of the microfloc fractions ranged
between 800-1200 kg.m-3. This would suggest that the finer sand grains tended to interact and
bond better with the smaller floc structures, accounting for the quicker microfloc settling
velocities observed.

The general trends revealed by the pre-determined (Tamar mud and silica) mixtures were also
observed with independent tests on naturally mixed Portsmouth Harbour sediments. How‐
ever, compositionally, the Portsmouth sediment matrix produced differences in the absolute
settling velocities of the macrofloc and microfloc fractions from those of the Tamar mixtures.
Both fractions of the Portsmouth sediment tended to fall quicker than their Tamar mixed
sediment equivalents. It is proposed that this could be a result of a different sand grain size
distribution combined with stronger bio-film coatings producing added cohesion in the
Portsmouth sediment mixtures. This would permit a greater uptake of the sand grains within
the macrofloc fraction, whilst also potentially forming the faster settling microflocs observed.

The data showed that the greater the sand content of a mixed suspension, the higher the total
mass settling flux (MSF). As the microflocs have been seen to be more conducive at flocculating
with the finer sand grains, and the Wsmicro rose with rising sand content, one can see that this
smaller size fraction is extremely important in terms of the total MSF for less muddy suspen‐
sions. By averaging the MSF over the entire concentration and shear stress ranges for a constant
ratio of mud (M) and sand (S), the data revealed that for a predominantly sandy suspension
(25M:75S), the microflocs represented the majority of the total MSF. In contrast, the microflocs
contributed less than half of the settling flux for a much muddier mixture (75M:25S).

Biology is considered to be extremely important in the mixed sediment flocculation process.

For example, the presence of sticky extracellular polymeric substances (EPSs) produced by
epipelic and epipsammic diatoms could significantly enhance particle bonding. Energy
dispersive spectroscopy analysis confirmed the presence of both clay minerals and quartz
mineral fragments within a natural microfloc. A high resolution transmission electron
microscopy (TEM) image revealed EPS fibrils linking the biological and inorganic particles
within a micro-structural framework of a microfloc matrix.

Since estuaries may have mixed or segregational mud:sand environments and numerical
models are used to inform management decisions, some issues relating to the parameterisation
of mud:sand flocculation and their implementation in sediment transport models have been
discussed. It is anticipated that these two topics will be the subject of future research and
publication on mixed sediment flocculation.

Acknowledgements

The mini-annular flume experiments were primarily funded by the HR Wallingford Company
Research Programme as part of the ‘Mud:Sand Transport’ projects (DDD0301 and DDD0345),
and completed during the ‘Sediment in Transitional Environments’ – SiTE project (DDY0427).
156 Sediment Transport Processes and Their Modelling Applications

Author details

Andrew J. Manning1,2, Jeremy R. Spearman1, Richard J.S. Whitehouse1, Emma L. Pidduck2,

John V. Baugh1 and Kate L. Spencer3

1 HR Wallingford, Howbery Park, Wallingford, Oxfordshire, UK

2 School of Marine Science & Engineering, University of Plymouth, Plymouth, Devon, UK

3 Department of Geography, Queen Mary – University of London, Mile End Road, London,
UK

References

[1] Alvarez-Hernandez, E. (1990). The influence of cohesive sediment on sediment

movement in channels of circular cross-section. PhD thesis, University of Newcastle-
upon-Tyne.

[2] Bass, S.J., Manning, A.J. and Dyer, K.R. (2006). Preliminary findings from a study of
the upper reaches of the Tamar Estuary, UK, throughout a complete tidal cycle: Part
I. Linking sediment and hydrodynamic cycles. In: J.P.-Y. Maa, L.P. Sanford and D.H.
Schoellhamer (eds), Coastal and Estuarine Fine Sediment Processes - Proc. in Marine
Science 8, Amsterdam: Elsevier, pp. 1-14, ISBN: 0-444-52238-7.

[3] Baugh, J.V. and Manning, A.J. (2007). An assessment of a new settling velocity pa‐
rameterisation for cohesive sediment transport modelling. Continental Shelf Re‐
search, doi:10.1016/[Link].2007.03.003.

[4] Berlamont, J.E. (2002). Prediction of cohesive sediment transport and bed dynamics
in estuaries and coastal zones with integrated numerical simulation models. In: Win‐
terwerp, J.C., Kranenburg, C. (eds), Fine Sediment Dynamics in the Marine Environ‐
ment - Proc. in Mar. Sci. 5., Elsevier, Amsterdam, pp. 1-4.

[5] de Brouwer, J.F.C., Wolfstein, K., Ruddy, G.K. (2005). Biogenic stabilization of interti‐
dal sediments: The importance of extracellular polymeric substances produced by
benthic diatoms. Microbial Ecology, 49, 501-512.

[6] Buffle, J. and Leppard, G.G. (1995). Characterisation of aquatic colloids and macro‐
molecules. 2. Key role of physical structures on analytical results. Environmental Sci‐
ence and Technology, 29, 2176-2184.

[7] Burban, P.Y. (1987). The flocculation of fine-grained sediments in estuarine waters.
MSc. thesis, Dep. of Mech. Eng. Univ. of Calif., Santa Barbara, USA.
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 157
[Link]

[8] Cahoon, L.B. (1999). The role of benthic microalgae in neritic ecosystems. Oceanogra‐
phy and marine biology: an annual review, 37, 47-86.

[9] Chesher, T.J. and Ockenden, M.C. (1997). Numerical modelling of mud and sand
mixtures. In: N. Burt, R. Parker and J. Watts (Eds), Cohesive Sediments – Proc. of IN‐
TERCOH Conf. (Wallingford, England), Chichester: John Wiley & Son, pp. 395-406.

[10] Dankers, P.J.T., Sills, G.C. and Winterwerp, J.C. (2007). On the hindered settling of
highly concentrated mud-sand mixtures. In: T. Kudusa, H. Yamanishi, J. Spearman
and J.Z. Gailani (Eds), Sediment and Ecohydraulics - Proc. in Marine Science, INTER‐
COH 2005, Amsterdam: Elsevier, pp. 255-274.

[11] Dearnaley, M.P. (1996). Direct measurements of settling velocities in the Owen Tube:
a comparison with gravimetric analysis. Journal of Sea Research, Vol. 36, Nos. 1-2, 36,
41-47.

[12] Dyer, K.R. (1986). Coastal and Estuarine Sediment Dynamics. Wiley & Sons, Chiches‐
ter, 342p.

[13] Dyer, K.R. 1989. Sediment processes in estuaries: future research requirements. J. Ge‐
ophys. Res., 94 (C10): 14,327-14,339.

[14] Dyer, K.R., Cornelisse, J.M., Dearnaley, M., Jago, C., Kappenburg, J., McCave, I.N.,
Pejrup, M., Puls, W., van Leussen, W. and Wolfstein, K. (1996). A comparison of in-
situ techniques for estuarine floc settling velocity measurements. Journal of Sea Re‐
search 36, 15-29.

[15] Dyer, K.R. and Manning, A.J. (1998). Observation of the size, settling velocity and ef‐
fective density of flocs, and their fractal dimensions. Journal of Sea Research 41,
87-95.

[16] Edzwald, J.K. and O’Melia, C.R. (1975). Clay distributions in recent estuarine sedi‐
ments. Clays and Clay Minerals, 23:39-44.

[17] Eisma, D. (1986). Flocculation and de-flocculation of suspended matter in estuaries.

Neth. J. Sea Res., 20 (2/3), 183-199.

[18] Estuary Process Consortium (2005). Final Report of the Estuary Process Research
Project (EstProc) – Algorithms and Scientific Information. Integrated Research Re‐
sults on Hydrobiosedimentary Processes in Estuaries, R & D Technical Report pre‐
pared by the Estuary Process Consortium for the Fluvial, Estuarine and Coastal
Processes Theme, co-funded by Defra & Environment Agency, Report FD1905/TR3,
140p.

[19] Feates, N.G. and Mitchener, H.J. (1998). Properties of dredged material: measure‐
ment of sediment properties of dredged material from Harwich Harbour. HR Wall‐
ingford Report TR 46.
158 Sediment Transport Processes and Their Modelling Applications

[20] Fennessy, M.J., Dyer, K.R. and Huntley, D.A. (1994). Size and settling velocity distri‐
butions of flocs in the Tamar Estuary during a tidal cycle. Netherlands Journal of
Aquatic Ecology, 28: 275-282.

[21] Fennessy, M.J., Dyer, K.R., Huntley, D.A. and Bale, A.J. (1997). Estimation of settling
flux spectra in estuaries using INSSEV. In: N. Burt, R. Parker and J. Watts (Eds), Co‐
hesive Sediments – Proc. of INTERCOH Conf. (Wallingford, England), Chichester:
John Wiley & Son, pp. 87-104.

[22] Fitzpatrick, F. (1991). Studies of sediments in a tidal environment. Ph.D. Thesis, De‐
partment of Geological Sciences, University of Plymouth, 221p.

[23] Förstner, U. and Wittmann G. T. W. (1983). Metal Pollution in the Aquatic Environ‐
ment. Springer Verlag, Berlin, Heidelberg et New York, 2nd revised edition, 486p.

[24] Gerbersdorf, S.U., Bittner, R., Lubarsky, H., Manz, W. and Paterson, D.M. (2009). Mi‐
crobial assemblages as ecosystem engineers of sediment stability. J Soils Sediments,
9, 640–652.

[25] Grabowski, R.C., Droppo, I.G. and Wharton, G. (2011). Erodibility of cohesive sedi‐
ment: The importance of sediment properties. Earth-Science Reviews, 105, 101-120.

[26] Glasgow, L.A. and Lucke, R.H. (1980). Mechanisms of deaggregation for clay-poly‐
mer flocs in turbulent systems. Ind. Eng. Chem. Fundam., 19: 148-156.

[27] Gratiot, N. and Manning, A.J. (2004). An experimental investigation of floc character‐
istics in a diffusive turbulent flow. In: P. Ciavola and M. B. Collins (Eds), Sediment
Transport in European Estuaries, Journal of Coastal Research, SI 41, 105-113.

[28] Harlow, D. (1980). Sediment Processes, Selsey Bill to Portsmouth, PhD thesis, Depart‐
ment of Civil Engineering, University of Southampton.

[29] Harper, M.A. and Harper, J.F. (1967). Measurements of diatom adhesion and their re‐
lationship with movement. British Phycological Bulletin, 3, 195-207.

[30] Hickman, M. and Round, F.E. (1970). Primary production and standing crops of epi‐
psammic and epipelic algae. British Phycological Journal, Vol 5 (2), pp. 247-255.

[31] HR Wallingford (1998). SandCalc: Marine Sands Calculator Interface. Version 2.0 for
Windows. Software by Tessela & HR Wallingford.

[32] Hydraulics Research (1959). Portsmouth harbour investigation, parts i and ii. Reports
213b and 214, Technical report, Hydraulics Research.

[33] Kamphuis, W and Hall, K.R. (1983). Cohesive material erosion by unidirectional cur‐
rent. J. Hyd. Eng., ASCE, 109, 49-61.

[34] Kennish, M.J. (1986). “Ecology of estuaries Volume 1: Physical and chemical aspects.”
Boca Raton Florida, CRC Press, 1.
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 159
[Link]

[35] Koglin B. (1977). Assessment of the degree of aggregation in suspension. Powder

Technology 17, 219-227.

[36] Klimpel R.C. and Hogg R. (1986). Effects of flocculation conditions on agglomerate
structure. Journal of Colloid Interface Science 113, 121-131.

[37] Kolmogorov, A.N. (1941a). The local structure of turbulence in incompressible vis‐
cous fluid for very large Reynolds numbers. C. R. Acad. Sci. URSS, 30: 301.

[38] Kolmogorov, A.N. (1941b). Dissipation of energy in locally isotropic turbulence. C. R.

Acad. Sci. URSS, 32: 16.

[39] Kranck, K. and Milligan, T.G. (1992). Characteristics of suspended particles at an 11-
hour anchor station in San Francisco Bay, California. Journal of Geophysical Re‐
search, 97, 11373-11382.

[40] Krone, R.B. (1962). Flume studies of the transport of sediment in estuarial shoaling
processes. Final report. Hyd. Eng. Lab. and Sanitary Eng. Lab., University of Califor‐
nia, Berkeley.

[41] Krone, R. B. (1963). A study of rheological properties of estuarial sediments. Report

No. 63-68, Hyd. Eng. Lab. and Sanitary Eng. Lab., University of California, Berkeley,
63-68.

[42] Little, C. (2000). The biology of soft shores and estuaries. Oxford University Press
(UK), 252p., ISBN: 978-0-19850-426-9.

[43] Lonsdale, B. J. (1969). A sedimentary study of the eastern Solent, Master’s thesis, De‐
partment of Oceanography, University of Southampton.

[44] Manning, A.J. (2001). A study of the effects of turbulence on the properties of floccu‐
lated mud. Ph.D. Thesis. Institute of Marine Studies, University of Plymouth, 282p.

[45] Manning, A.J. (2004a). The observed effects of turbulence on estuarine flocculation.
In: P. Ciavola and M. B. Collins (eds), Sediment Transport in European Estuaries,
Journal of Coastal Research, SI 41, 90-104.

[46] Manning, A.J. (2004b). Observations of the properties of flocculated cohesive sedi‐
ment in three western European estuaries. In: P. Ciavola and M. B. Collins (Eds),
Sediment Transport in European Estuaries, Journal of Coastal Research, SI 41, 70-81.

[47] Manning, A.J. (2006). LabSFLOC – A laboratory system to determine the spectral
characteristics of flocculating cohesive sediments. HR Wallingford Technical Report,
TR 156.

[48] Manning, A.J., Bass, S.J. and Dyer, K.R. (2006). Floc Properties in the Turbidity Maxi‐
mum of a Mesotidal Estuary During Neap and Spring Tidal Conditions. Marine Ge‐
ology, 235, 193-211.
160 Sediment Transport Processes and Their Modelling Applications

[49] Manning, A.J., Baugh, J.V., Spearman, J. and Whitehouse, R.J.S. (2009). Flocculation
Settling Characteristics of Mud:Sand Mixtures. Ocean Dynamics, PECS2008 SI, DOI:
10.1007/s10236-009-0251-0.

[50] Manning, A.J. and Dyer, K.R. (1999). A laboratory examination of floc characteristics
with regard to turbulent shearing. Marine Geology 160, 147-170.

[51] Manning, A.J. and Dyer, K.R. (2002a). The use of optics for the in-situ determination
of flocculated mud characteristics. J. Optics A: Pure and Applied Optics, Institute of
Physics Publishing, 4, S71-S81.

[52] Manning, A.J. and Dyer, K.R. (2002b). A comparison of floc properties observed dur‐
ing neap and spring tidal conditions. In: J.C. Winterwerp and C. Kranenburg (Eds),
Fine Sediment Dynamics in the Marine Environment - Proc. in Marine Science 5, Am‐
sterdam: Elsevier, pp. 233-250, ISBN: 0-444-51136-9.

[53] Manning, A.J. and Dyer, K.R. (2007). Mass settling flux of fine sediments in Northern
European estuaries: measurements and predictions. Marine Geology, 245, 107-122,
doi:10.1016/[Link].2007.07.005.

[54] Manning, A.J., Spearman, J. and Whitehouse, R.J.S. (2007). Mud:Sand Transport –
Flocculation & Settling Dynamics within Turbulent Flows, Part 1: Analysis of labora‐
tory data. HR Wallingford Internal Report, IT 534, 32p.

[55] Manning, A.J. and Whitehouse, R.J.S. (2009). UoP Mini-annular flume – operation
and hydrodynamic calibration. HR Wallingford Technical Report, TR 169.

[56] McCave, I.N. (1984). Size spectra and aggregation of suspended particles in the deep
ocean. Deep-Sea Res., 31: 329-352.

[57] Mehta, A.J., Jaeger, J.M., Valle-Levinson, A., Hayter, E.J., Wolanski, E. and Manning,
A.J. (2009). Resuspension Dynamics in Lake Apopka, Florida. Final Synopsis Report,
submitted to St. Johns River Water Management District, Palatka, Florida, June 2009,
Report No. UFL/COEL-2009/00, 158p.

[58] Mehta, A.J. and Lott, J.W. (1987). Sorting of fine sediment during deposition. Proc.
Speciality Conf. Advances in Understanding Coastal Sediment Processes. Am. Soc.
Civ. Eng., New York, pp. 348-362.

[59] Mietta, F. (2010). Evolution of floc size distribution of cohesive sediments. Ph.D. The‐
sis, Delft University of Technology, Faculty of Civil Engineering and Geosciences,
The Netherlands, 169p.

[60] Migniot, C. (1968). Study of the physical properties of various very fine sediments
and their behaviour under hydrodynamic action. La Houille Blanche, 23 (7). (Trans‐
lation of French text).

[61] Mitchener, H.J., Torfs, H. and Whitehouse, R.J.S. (1996). Erosion of mud/sand mix‐
tures. Coastal Engineering, 29, 1-25 [Errata, 1997, 30, 319].
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 161
[Link]

[62] Mory, M., Gratiot, N., Manning, A.J. and Michallet, H. (2002). CBS layers in a diffu‐
sive turbulence grid oscillation experiment. In: J.C. Winterwerp and C. Kranenburg
(Eds.), Fine Sediment Dynamics in the Marine Environment - Proc. in Mar. Science 5,
Amsterdam: Elsevier, pp.139-154, ISBN: 0-444-51136-9.

[63] Nowell, A.R.M., Jumars, P.A. and Eckman, J.E. (1981). Effects of biological activities
on the entrainment of marine sediments. Mar. Geol., 42, 133-153.

[64] Ockenden, M.C. and Delo, E.A. (1988). Consolidation and erosion of estuarine mud
and sand mixtures – an experimental study. HR Wallingford Report, SR 149.

[65] Owen, M.W. (1976). Determination of the settling velocities of cohesive muds. Hy‐
draulics Research, Wallingford, Report No. IT 161, 8p.

[66] Panagiotopoulus, I., Voulgaris, G. and Collins, M.B. (1997). The influence of clay on
the threshold of movement of fine sandy beds. Coastal Eng., 32, 19-43.

[67] Parker, D.S., Kaufman, W.J. and Jenkins, D. (1972). Floc break-up in turbulent floccu‐
lation processes. J. Sanitary Eng. Div., Proc. Am. Soc. Civil Eng., 98 (SA1): 79-97.

[68] Paterson, D.M. (1989). Short-term changes in the erodibility of intertidal cohesive
sediments related to the migratory behaviour of epipelic diatoms. Limnol. Oceanogr.
34: 223-234.

[69] Paterson, D.M., Crawford, R.M. and Little, C. (1990). Subaerial exposure and changes
in the stability of intertidal estuarine sediments. Estuarine Coastal and Shelf Science,
30, 541-556.

[70] Paterson, D.M. and Hagerthey, S.E. (2001). Microphytobenthos in contrasting coastal
ecosystems: Biology and dynamics. In: Ecological comparisons of sedimentary shores
(K. Reise, Ed.), Ecological studies, pp. 105-125.

[71] Pidduck, E.L. and Manning, A.J., in prep. A Laboratory Examination of Flocculation
Properties Exhibited by Natural Sediment Mixtures from Portsmouth Harbour, UK.
HR Wallingford Technical Report, TR 182.

[72] Puls, W., Kuehl, H. and Heymann, K. (1988). Settling velocity of mud flocs: results of
field measurements in the Elbe and the Weser Estuary. In: J. Dronkers, and W. van
Leussen, (eds), Physical Processes in Estuaries. Berlin: Springer-Verlag, pp. 404-424.

[73] Raudkivi, A.J. (1998). Loose boundary hydraulics. 3rd Edition. Balkema, Rotterdam.

[74] Reid, G.K. and Wood, R.D. (1976). “Ecology of inland water and estuaries.” New
York: D. Van Nostrand Company.

[75] Soulsby, R.L. and Manning, A.J. (2012). Cohesive sediment settling flux: settling ve‐
locity of flocculated mud. Technical Note DDY0409-01, HR Wallingford, Walling‐
ford, UK.
162 Sediment Transport Processes and Their Modelling Applications

[76] Spearman, J.R., Manning, A.J. and Whitehouse, R.J.S. (2011). The settling dynamics of
flocculating mud:sand mixtures: Part 2 – Numerical modelling. Ocean Dynamics, IN‐
TERCOH 2009 special issue, DOI: 10.1007/s10236-011-0385-8.

[77] Spencer, K.L., Manning, A.J., Droppo, I.G., Leppard, G.G. and Benson, T. (2010). Dy‐
namic interactions between cohesive sediment tracers and natural mud. Journal of
Soils and Sediments, Volume 10 (7), doi:10.1007/s11368-010-0291-6.

[78] Tambo, N. and Hozumi, H. (1979). Physical characteristics of flocs – II. Strength of
flocs. Water Research, 13, 409-419.

[79] Tambo, N. and Watanabe, Y., (1979). Physical characteristics of flocs-I. The floc densi‐
ty function and aluminium floc. Water Research 13, 409-419.

[80] ten Brinke, W.B.M. (1993). The impact of biological factors on the deposition of fine
grained sediment in the Oosterschelde (The Netherlands). Ph.D. Thesis, University of
Utrecht, The Netherlands, 252p.

[81] Tolhurst, T.J., Gust. G. and Paterson, D.M. (2002). The influence on an extra-cellular
polymeric substance (EPS) on cohesive sediment stability. In: J.C. Winterwerp and C.
Kranenburg (Eds), Fine Sediment Dynamics in the Marine Environment - Proc. in
Marine Science 5, Amsterdam: Elsevier, pp. 409-425, ISBN: 0-444-51136-9.

[82] Tomi, D.T., Bagster, D.F. (1978). The behaviour of aggregates in stirred vessels: Part I
- Theoretical considerations on the effects of agitation. Trans. Inst. Chem. Eng., 56,
1-8.

[83] Torfs, H. (1994). Erosion of layered sand-mud beds in uniform flow. Proc. 24th Int.
Conf. Coastal Eng., Kobe, Japan, 23-28 October 1994.

[84] Torfs, H., Mitchener, H.J., Huysentruyt, H. and Toorman, E. (1996). Settling and con‐
solidation of mud/sand mixtures. Coastal Engineering, 29, 27-45.

[85] Tsai, C.H., Iacobellis, S. and Lick, W. (1987). Flocculation of fine-grained sediments
due to a uniform shear stress. J. Great Lakes Res., 13: 135-146.

[86] Uncles, R.J., Stephens, J.A. and Harris, C. (1998). Seasonal variability of subtidal and
intertidal sediment distributions in a muddy, macrotidal estuary: the Humber-Ouse,
UK. In: Sedimentary Processes in the Intertidal Zone, Black, K.S., Paterson, D.M. and
Cramp, A. (Eds), Geological Society, London, Special Publications, 139, 211-219.

[87] Underwood, G.J.C. and Kromkamp, J. (1999). Primary production by phytoplankton

and microphytobenthos in estuaries. Advances in ecological research, 29, 93-153.

[88] Underwood, G.J.C. and Paterson, D.M. (2003). The importance of extracellular carbo‐
hydrate production by marine epipelic diatoms. Advances in Botanical Research (in‐
corporating Advances in Plant Pathology), Vol. 40, Elsevier, Amsterdam, pp.183-240,
ISBN: 0-12-005940-1.
 Flocculation Dynamics of Mud: Sand Mixed Suspensions 163
[Link]

[89] Van de Ven, T.G. and Hunter, R.J. (1977). The energy dissipation in sheared coaggu‐
lated soils. Rheologica Acta, 16, 534-543.

[90] van Ledden, M. (2002). A process-based sand-mud model. In: J.C. Winterwerp and C.
Kranenburg (Eds.), Fine Sediment Dynamics in the Marine Environment - Proc. in
Mar. Science 5, Amsterdam: Elsevier, pp.577-594, ISBN: 0-444-51136-9.

[91] van Ledden, M. (2003). Sand-mud segregation in estuaries and tidal basins. Ph.D.
Thesis, Delft University of Technology, The Netherlands, Report No. 03-2, ISSN
0169-6548, 217p.

[92] van Leussen, W. (1988). Aggregation of particles, settling velocity of mud flocs: a re‐
view. In: Dronkers, J., van Leussen, W. (Eds), Physical Processes of Estuaries, Berlin:
Springer, pp. 347-403.

[93] van Leussen, W. (1994). Estuarine macroflocs and their role in fine-grained sediment
transport. Ph.D. Thesis, University of Utrecht, The Netherlands, 488p.

[94] van Wijngaarden, M., Venema, L.B., De Meijer, R.J., Zwolsman, J.J.G., Van Os, B. and
Gieske, J.M.J. (2002a). Radiometric sand-mud characterisation in the Rhine-Meuse es‐
tuary, Part A: Fingerprinting. Geomorphology, 43, 87-101.

[95] van Wijngaarden, M., Venema, L.B., and De Meijer, R.J. (2002b). Radiometric sand-
mud characterisation in the Rhine-Meuse estuary, Part B: In situ mapping. Geomor‐
phology, 43, 103-116.

[96] Waeles, B., Le Hir, P. and Lesueur, P. (2008). A 3D morphodynamic process-based

modelling of a mixed sand/mud coastal environment : the Seine Estuary, France. In:
T. Kudusa, H. Yamanishi, J. Spearman and J.Z. Galiani, (eds.), Sediment and Ecohy‐
draulics - Proc. in Marine Science 9, Amsterdam: Elsevier, pp. 477-498, ISBN:
978-0-444-53184-1.

[97] Whitehouse, R.J.S., Soulsby, R., Roberts, W. and Mitchener, H.J. (2000). Dynamics of
Estuarine Muds. Thomas Telford Publications, London, 232p.

[98] Widdows, J., Blauw, A., Heip, C.H.R., Herman., P.M.J., Lucas, C.H., Middelburg, J.J.,
Schmidt, S., Brinsley, M.D., Twisk, F. and Verbeek, H. (2004). Role of physical and
biological processes in sediment dynamics of a tidal flat in Westerschelde Estuary,
SW Netherlands. Mar. Ecol. Prog. Series, 274, 41-56.

[99] Williamson, H.J. (1991). Tidal transport of mud / sand mixtures: Sediment distribu‐
tions – a literature review. HR Wallingford, Report SR 286.

[100] Williamson, H.J. and Ockenden, M.C. (1993). Laboratory and field investigations of
mud and sand mixtures. In: Sam S.Y Wang (Ed.), Advances in Hydro-science and
Engineering, Proceedings of the First International Conference on Hydro-science and
Engineering, Washington D.C. (7-11 June 1993), volume 1, pp. 622-629.
164 Sediment Transport Processes and Their Modelling Applications

[101] Winterwerp, J. C. (1998). A simple model for turbulence induced flocculation of co‐
hesive sediment, J. Hyd. Eng., 36 (3), 309-326.

[102] Winterwerp, J.C. and van Kesteren, W.G.M. (2004). Introduction to the physics of co‐
hesive sediment in the marine environment. Developments in Sedimentology, 56,
van Loon, T. (Ed.), Amsterdam: Elsevier, 466p.

[103] Winterwerp, J.C., Manning, A.J., Martens, C., de Mulder, T., and Vanlede, J. (2006). A
heuristic formula for turbulence-induced flocculation of cohesive sediment. Estuar‐
ine, Coastal and Shelf Science, 68, 195-207.

[104] Wolanski, E. (2007). Estuarine Ecohydrology. Elsevier (Amsterdam, The Nether‐

lands), 157p, ISBN: 978-0-444-53066-0.
 Chapter 7

The Gravel-Bed River Reach Properties Estimation in

Bank Slope Modelling

Levent Yilmaz

Additional information is available at the end of the chapter

[Link]

1. Introduction

Church and Kellerhals [2] point out the difficulty of adequately characterizing a gravel bed by
a single grain size distribution for a relatively long river reach. Bray [1] indicated that the
initiation of motion calculations gave as a result in which the gravel bed is immobile or at least
not highly mobile at flows by flooding boundary layers. The basic data for each gravel-bed
river reach are directly applied to a specified equation to compute the average velocity.

The knowledge about the hydraulic geometric parameters, width, depth and area of the river
at the bankful discharge are required for solving a variety of problems related to rivertraining,
location of river constructions and navigation. To predict the average velocity of flow, the
resistance offered to the flow by the boundary and air-water interface needs to be known. In
methods for the prediction of width, depth, area and the flow velocity or resistancecoefficient
the results of the analysis of the available gravel-bed river data will be given.

2. Method

The resistance characteristics and the study of hydraulic geometry for gravel-bed rivers is the
main method for finding all the hydraulic characteristics. The hydraulic geometry refers to the
geometrical characteristics of the cross-section such as the average width w, average depthh
and area A (=wh) at the bankful discharge Q.

The basic data for each gravel-bed river reach are directly applied to a specified equation to
compute the average velocity. Then for each reach the percent deviation (PDEV) of the
computed average velocity from the “observed” average velocity is computed. The distribu‐
166 Sediment Transport Processes and Their Modelling Applications

tion of the percent deviations associated with a specified equation is then determinedfor the
different gravel-bed rivers reaches. A summary of the parameters to describe the distribution
of the percent deviations for each of the specified equations is given in Table. 1.

Equation Mean Standard Minimum Median Value Maximum Value

(1) (2) Deviation (3) Value (4) (5) (6)

Manning’s Eq. -3.3 29.6 -50.0 -7.0 83.2

n by modified Cowan
44.9 43.7 -18.6 31.8 181.9
n by Strickler

n=0.41 D1/6 50 37.5 40.9 -23.1 25.0 156.9

n = 0.038 D1/690
2.5 28.8 -41.8 -3.1 74.4
n by Limerinos

Keulegan’s Eq. 54.2 46.1 -12.7 40.4 195.3

k3 = D50 47.0 42.7 -17.3 35.2 169.2

k3 = D65 32.9 38.3 -23.9 23.0 136.4

k4 = D90
8.6 29.4 -26.6 -0.7 116.1
Lacey’s Eq.

Table 1. Statistics for Gravel-bed River Reaches [1]

Some of the characteristics which differentiate gravel-bed rivers from the alluvial rivers are:

a. much steeper slope (0.001 – 0.02 )

b. resistance is higher than the alluvial rivers

There is scope of using all the available gravel-bed river data and develop non dimensional
relationships for the hydraulic geometry. In the analysis of river and channel problems it must
be given a relationship between the average velocity U, the depth h or the hydraulic radius R,
channel slope S and some coefficient which is related to the channel boundary. This is known
as the resistance relationship [3]. The work of Lacey [4] about the sand-bed rivers has shown
that for such rivers depth h or hydraulic radius R ~ (Q/f1)1/3, width W or wetted perimeter P~
Q0.50, Area A ~ Q5/6/ f1 1/3 where f1 is Lacey’s silt factor and is given by f1 = 1,76 (d)0.5, d being the
median size of bed material in mm. As regards the gravel-bed rivers Kellerhals and Bray [5]
have related W, h, A to Q and sediment size d as

W = 2.08 Q 0.528d -0.70 (1)

h = 0.256 Q 0.331d -0.25 (2)

 The Gravel-Bed River Reach Properties Estimation in Bank Slope Modelling 167
[Link]

All such equations are based on the analysis of limited amount of data and are not
dimensionally homogeneous. Only the dimensionless parameters W/d, h/d and U/(Δγs d/
ρf)0/5 and related them to the dimensionless discharge Q/ d2 (Δγs d/ρf)0/5 are given [3]. Hence
Δγs is the difference in specific weights of sediment and water and ρf is the mass density
of water. Hence there is scope of using all the available gravel-bed river data and devel‐
op non-dimensional relationships for the hydraulic geometry. In the analysis of river and
channelproblems we need also a relationship between the average velocity U, the depth h
or the hydraulic radius R, channel slope S and some coefficient which is related to the
channelboundary. This is known as the resistance relationship [3]. The resistance relation‐
ship isexpressed in dimensionless form as [3],

U h1/6
in Manning’s Equation : = (3)
ghS n g

U C
in Chezy’s Equation : = (4)
ghS g

U 8
in Darcy-Weisbach Equation: = (5)
ghS f

U æhö
= Fç ÷ (6)
ghS èdø

In the above equations n is Manning’s roughness coefficient, C is Chezy’s discharge coefficient,

f is Darcy-Weisbach resistance coefficient and F is a function. These coefficientsdepend on the
resistance, offered to the flow by the channel boundary and air-water interface [3]. The
available data in Turkey at East Black Sea Basin have been analysed in a unified manner to
obtain dimensionally homogeneous relationships for W, h, A and U.

3. Data

A summary of data were classified as bankful discharge and variable discharge. The bankful
discharge data were used to study the hydraulic geometry. The variable discharge data
pertainto discharges other than the bankful in any stream. In order to study the effect of bed
condition, each set of data were further subdivided into those with mobile bed, and those
withpaved bed. There is no need to subdivide the data, because both sets of data behaved in
similar manner.
168 Sediment Transport Processes and Their Modelling Applications

4. Analysis of hydraulic geometry

The dependent variables can be any two of the four variables average width W, average depth
h, area of flow A = Wh and the average velocity U. The independent variable related to the
flow is bankful discharge Q. The sediment representing the bed and the banks will be described
by the median size of the bed material d, its geometric standard deviation σg andthe difference
in the specific weights of sediment and water Δγs [3]. It is known that for a given stream the
channel slope is related to the bankful discharge Q, the slope decreasingas Q increases in the
downstream direction [3]. If we deal with the data from different basins,S and Q will not be
related and hence S should be taken as an independent variable. If we ignore QB, because the
gravel-bed rivers carry a small amount of sediment load, we cananalyse as [3],

W, h, A, U = F (Q, d,σg,Δγs ρf, µ, S) (7)

With simplifications Garde [3] gave the dimensionless relationship for hydraulic geometry of
different river basins as,

é ù
ê ú
W h A U ê Q ú
, , , = Fê ,Sú (8)
d d d 2 Dg s ê d 2 Dg s d ú
d ê ú
rf rf
ë û

If studied regime types of relations, we must plot W, h, and A against Q on log-log scale which
yielded straight lines giving equations by [3],

W = 4.547 Q0.507 (9)

h = 0.293 Q0.332 (10)

A = 1.330 Q0.839 (11)

By comparing this equations, also the North Anatolian River Reaches will be investigated. The
exponents of Q obtained in Eqs. [9], [10], [11] are very close to those obtained by Lacey [4].
Similar investigation was carried out using W/d, h/d and A/d2 and determining their variation
W h A Q
with , , , =F by plotting on log-log scales [3].
d d d2 Δγs
2
d d
ρf
 The Gravel-Bed River Reach Properties Estimation in Bank Slope Modelling 169
[Link]

The relationships given by Garde [3] are,

W/h = 7.675 Q0.448 (12)

h/d = 0.504 Q0.373 (13)

A/ d 2 = 3.872 Q0.821 (14)

5. Method

From [Link] the program for landslides can be estimated by TALREN 4 which is
ideal for checking the stability of natural slopes, cut or fill slopes, earth dams and dikes. It takes
into account various types of reinforcements, such as: anchors and soil nails, piles and
micropiles, geotextiles and geogrids, steel and polymer strips. There is another new user-
friendly graphical interface with:
a. In the program, definition of the profile using a mouse, rulers and a grid. Other features
include pop up menus and choice of soil colours.

b. Ability to load background drawings (.jpg [Link] formats) and adjust to scale.

c. Several construction stages and calculation alternatives can be handled in the same file.

d. Tables illustrating main soil, load and reinforcement data.

e. Various output options for graphical display and tables (shadings, forces in reinforce‐
ments, detailed results for each failure surface, etc.)

f. Wizards and databases to help produce the best model and choice of input data ( partial
safety factors).
New calculation functionalities:
a. Automatic search option for circular failure surfaces (no need to define a manual grid).

b. No limit on the number of elements you can define ( points, layers, reinforcements,
hydraulic mesh, etc.)

c. Future upgrade option for TALREN 4 users: calculation method based on limit analysis
theory.

d. TALREN still benefits from extensively used methods as limit equilibrium calculation
along potential failure surfaces using the Fellenius, Bishop or perturbations methods.

e. Ability to take into account hydraulic conditions.

170 Sediment Transport Processes and Their Modelling Applications

f. Seismic loads are taken into account by the pseudostatic method.

6. Data uncertainties

For estimation of landslides condition we require precipitation, streamflow, evapotranspira‐

tion and watershed morphology. The effects of data uncertainties must be considered in
different ways:
1. whether the model parameters are determined from calibration or from physical meas‐
urements and principles,

2. whether the model is used to estimate real events (landslide forecasting), or to estimate
synthetic events ( design storms and generation of synthetic flows which reasoned the
landslides. These issues are considered separately.
In Turkey the landslides can be seen in the Karst environment. Karst is a term applied to topogra‐
phy fdrmed in regions of limestone or dolomite bedrock by the vigorous solution work of
groundwater. One recognizes karst topography by the presence of large numbers of sinkholes,
solution valleys, disappearing streams, and landslides. The development of karst topography is
enhanced by the presence of well-jointed carbonates or evaporites near the surface. It is also
enhanced by rainfall. And sufficient relief to insure continuous movement of groundwater that
will carry away dissolved matter. The term karst comes from a limestone plateau in Yugoslavia
where solution features are well developed. Similar topography can be found in Turkey,
Kentucky, Tennessee, Indiana, northern Florida, and Puerto [Link] of mass wasting Earth
materials on slopes shows the movement where it shows as the result of landslides.

Rate of Movement Amount of Water Present

Flow (movement distributed throughout material)

Creep Slow Water not necessary

Rock glacier, rock, stream Slow Water not necessary

Solifluction Fast Water-saturated

Mudflow, debris flow Slow or fast Much water

Earthflow Slow or fast Much water

Slide (movement as one Mass on a slip surface)

Debris avalanche Fast Wet or dry

Slump Fast Wet or dry

Landslide, rockslide Fast Wet or dry

Fall (free fall of rock or soil)

Table 2. Types of mass movement (Levin, 1986)

 The Gravel-Bed River Reach Properties Estimation in Bank Slope Modelling 171
[Link]

Creep is a small form of land movement where the amount of water is not necessary and its
measure is only a few centimeters in one year (Watkins et all., 1975). Creep can decrease if we
can follow it through the earth surface and is a form of small earth flow. There are two types
of creep, soil and rock creep which can be observed.
Landslide prevention
Simple engineering techniques have been used to prevent the landslide, for example, by
flattening the cut-slope angle the landslide movement of erosion can improve in an easy way
by construction of infrastructure (Levin, 1986). Meandering environment shows us another
way of landslides as an example, polygonal ground on the flood plain of the Kogosukruk River,
Alaska. Scale of air photograph 1: 20.000 (Courtesy of U.S. Geological Survey)

7. Relation for meander tortuosity

The relationship between the tortuosity ratio and other parameters can be expressed as,

LR/LV= f ( W, D, S, m ) (15)

This can be reduced to the dimensionless equation,

LR/LV=f ( W/D, S, m/D ) (16)

If mean velocity and discharge per length of channel width are assumed as two more relevant
parameters, Eq.(3) can further be modified as,

LR/LV=f ( W/D, S, m/D, R, F ) (17)

in which R=q/v= the Reynolds number; F= V/ (gD)1/2= the Froude number; q= discharge per
unit length width; v= kinematic viscosity; V= mean velocity; and g= gravitational acceleration.
To investigate the actual relationship and its validity, river data or laboratory data for
meandering flumes were needed for all the parameters involved. The study of the effect of
parameters W/D, S and m/D individually on meander tortuosity, plots of LR/LV against these
three parameters indicate that channels with low tortuosity ratio, i.e., more or less straight
channels, have wide and shallow cross sections, steeper slopes, and relatively coarser bed
material. A value of LR/LV equal to one indicates straight channels. With gradual reduction
in the value of all three parameters W/D, S, and m/D, the tortuosity ratio increases, indicating
that meanders become more and more acute.
Flow curvature creates superelevation and transverse flow across the section of a channel bend.
The strength of the transverse current depends on boundary friction. In wide and shallow
172 Sediment Transport Processes and Their Modelling Applications

channels the ratio of roughness elements to flow depth is higher than in deep channels because
of coarser bed material as well as shallower depth, the higher roughness ratio results in more
frictional resistance and hence weaker transverse flow than in narrow and deep channels. In
considering river patterns, channels can logically be divided into two main groups, single
channel streams and multichannel streams, with a transition range between the two. Single
channel streams can be further subdivided into meandering channels and straight channels
with a transition between them. Meanders can be classified as regular or irregular, simple or
compound, acute or flat, and sine, parabolic, circular or sine-generated curves.

Meandering channels are formed if the flow dynamics corresponds with the channel mor‐
phology. Braided channels occur if flow dynamics and channel morphology are incompatible.
Alluvial channels are unstable because the stability criteria for the channel bed and for the
channel banks are different.

Meander flow takes place in one single channel which oscillates more or less regularly with
meandering river amplitudes that tend to increase with time. Meanders are found in beds of
fine sediment with gentle slopes.

x/L Elevation from Run.1 Run. Run. Run.

(Distance) bottom(mm) (τmax=100mN/m2) 2(τmax=239mN/m2) 3(τmax=300mN/m2) 4(τmax=390mN/m2)

0 -5.00 0.63 1.06 1.27 1.5

0.05 -4.00 0.64 1.17 1.32 1.58

0.1 -3.00 0.655 1.30 1.40 1.77

0.15 -2.00 0.84 1.40 1.50 1.89

0.20 -1.00 0.92 1.5 1.67 1.96

0.25 0.00 0.97 1.58 1.78 2.06

0.30 1.00 1.09 1.70 1.93 2.2

0.35 2.00 1.19 1.80 2.00 2.3

0.40 3.00 1.25 1.85 2.08 2.3

0.45 4.00 1.25 1.92 2.1 2.4

0.50 5.00 1.27 1.94 2.1 2.4

0.55 4.00 1.27 1.95 2.1 2.4

0.60 3.00 1.14 1.87 2.0 2.4

0.65 2.00 1.00 1.76 2.0 2.36

0.70 1.00 0.79 1.67 1.97 2.29

0.75 0.00 0.71 1.50 1.87 2.2

0.80 -1.00 1.0 1.58 1.80 2.1

0.85 -2.00 1.0 1.40 1.58 1.9

0.90 -3.00 0.72 1.25 1.48 1.7

0.95 -4.00 0.62 1.10 1.20 1.5

1.00 -5.00 0.556 1.10 1.19 1.4

Table 3. Shear Stress Distribution by landslides at mendering channels experimental set-up

 The Gravel-Bed River Reach Properties Estimation in Bank Slope Modelling 173
[Link]

8. Model application

The results of model applications were carried out for the same situations as the mathematical
model at the Technical University of Berlin, Institute Wasserbau and Wasserwirtschaft
(Yilmaz, 1990), started with flat bed, continued until ∂z/ ∂t=0. Then the beds were solidified,
and precise measurements of the bed configuration and the velocity were performed. Plan
geometries of runs consist of a sine-generated curve and an asymmetrical meander loop,
respectively. The latter is derived by a Fourier series analysis on several typical bends. The
meso-scale bed configuration in alluvial streams is highly dependent on the width-depth ratio
of the channel. The velocity measurements were made with small mechanical current meters
fixed to a 1m high frame that rested on the bottom while measuring the lower points on the
profile. The frame was suspended at different levels above the bottom to collect the data
represented by the higher points. Velocity profiles are plotted semi-logarithmically with the
dots representing field data and the smooth lines showing model predictions. The mean
velocity was calculated from a fit to the entire data set, not for each profile. Smaller dunes (0.50
m high, 2 m long) were superimposed on the large sand wave. Smith and McLean (1977)
estimated the roughness parameter for both the skin friction and the form drag due to the
smaller dunes to be 0.141 cm3. Three different perturbations are recognized:

1. Alternating bars: The bed configuration reached an equilibrium state after one hour,
and the quantitative and qualitative agreements are given. Sensitivity analyses of each
term in Eqs. 2 and 3 into the development of alternating bars were also carried out.
The term in the Eq. (3) ∂v/∂s was found to play the most important role in develop‐
ing alternating bars.
2. Braided bars : The calculated velocity vectors and bed configurations were given after
one hour. Divergence and convergence of flow streamlines in a wide straight channel and
the meso-scale bed configuration of braided bars can be clearly seen.
3. No bars: The calculated velocity vector and bed configuration were given after one hour.
The velocity distributions are almost uniform and the bed configuration is two-dimen‐
sional with less scour and fill than in the case of alternating bars and braided bars.

Numerical calculations are performed using the hydraulic conditions as listed in Table.4.

9. Observations

If the sediment transport behaves as bed load, the sediment surface at meandering channels
will deform into transverse waves. These bed forms can have a variety of scales ranging from
ripples through small dunes to fully developed dunes or sand waves. Smith (1970) gave that,
under pure bed load transport, a flat sand bed is unstable at all wavelengths to small pertur‐
bations in boundary topography so that with sufficient time all infinitesimal undulations will
grow in height. His analysis predicts that, for bed features of finite wave number, a fastest
growing wave exists only when there is a lag between the boundary shear stress and the
174 Sediment Transport Processes and Their Modelling Applications

Width Of Size of Bed

Run Number Average Bed Flow rate Q x Average Water
Channel B(m) Material Froude number (7)
(1) Slope (4) 103 (m3/s) (5) Depth h (m) (6)
(2) Dx10-2 (3)

10 0.50 0.30 1/75 1.00 0.70 "/>1

8 0.50 0.30 1/100 1.25 0.80 ”

5 0.50 0.30 1/120 1.15 0.69 ”

9 0.50 0.30 1/60 1.20 0.75 ”

7 0.50 0.30 1/80 1.00 0.70 ”

9 0.50 0.30 1/90 1.10 0.73 ”

11 0.50 0.30 1/110 1.20 0.75 ”

5 0.50 0.30 1/70 1.00 0.70 ”

8 0.50 0.30 1/50 1.10 0.74 ”

Table 4. Experimental Condition for Alternating Bars

sediment transport rate; this is the ripple instability. The tendency for larger bed forms to have
a seemingly discrete wavelength distribution, and a wavelength associated with fastest
growth, is not explained by such a primitive stability model, so Smith (1970) suggested that
wake affects also had to be taken into account.

It was showed that once perturbations are of finite amplitude, the larger stresses at the crests
cause the crests to propagate faster than the troughs, thus imparting asymmetrical shapes to
the waves. When the asymmetry is strong enough, the flow will separate, which creates a
momentum deficit downstream of the wave crest much like that found in the wake of a circular
cylinder. At the point of reattachment, the near-bottom velocity and stress are both zero.
Downstream from this point, an internal boundary layer must develop beneath the momentum
defect, or wake, region. The internal boundary layer adjusts to the velocity of the wake region
above it, which increases downstream due to the flux of momentum into the wake from the
interior. This produces two competing processes that are critical to determination of the
boundary shear stress. They are : accelerating effect of an outward diffusing velocity defect;
and the decelerating effect of a thickening boundary layer. In the near-field, spatial acceleration
of the fluid in the wake dominates the decelerative effects of the internal boundary layer, but
in the mid-field the opposite is true, and the net result is a decrease in the near-bed velocity in
this region. In the far-field the boundary layer ultimately engulfs the wake entirely, and the
boundary shear stress asymptotically approaches equilibrium. The essential features of this
response to separation are preserved over an upsloping surface such as the stoss side of a bed
form. Consequently, the resulting maximum in the stress profile has important consequences
for both bed deformation and bed-form growth.
 The Gravel-Bed River Reach Properties Estimation in Bank Slope Modelling 175
[Link]

10. Conclusions

The flow resistance in a meander bend is considerably increased due to the form resistance of
the patterns about which much is not known. It depends on a number of factors including
grain friction, form resistance of two- and three dimensional patterns, skin friction of the non-
separated oscillatory component and the sediment transport rate.

Author details

Levent Yilmaz

Address all correspondence to: lyilmaz@[Link]

Hydraulic Division, Civil Engineering Department, Technical University of Istanbul, Mas‐

lak, Istanbul, Turkey

References

[1] Bray, D. I. Generalized Regime-Type Analysis of Alberta Rivers, thesis presented to the
University of Alberta, at Edmonton, Canada, in partial fulfilment of the requirements
for the degree of Doctor of Philosophy, (1972).

[2] Church, M, & Kellerhals, R. On the Statistics of Grain Size Variation Along a gravel
River, Canadian Journal of Earth Sciences, 15(7), (1978).

[3] Garde, R. J. Hydraulic Geometry and Resistance of Gravel-Bed Rivers, Proceedings of

the Indian National Science Academy, 67 (A), (2001). , 2001, 597-609.

[4] Lacey, G. A General Theory of Flow in Alluvium, Journal of the Institution of Civil
Engineers, 27 (5518), (1947).

[5] Levin, H. L. Contemporary Physical Geology, Saunders College Publishing, New York,
(1986). , 558.

[6] Kellerhals, R, & Bray, D. I. Sampling Procedures for Coarse Fluvial Sediments, Journal
of the Hydraulics Division, 97 (HY8), (1971).

[7] Watkins, J. S, Bottino, M. L, & Morisawa, M. Our Geological Environment, Philadelphia,

Saunders College Publishing, (1975). , 335.
 Chapter 8

Scour Caused by Wall Jets

Ram Balachandar and H. Prashanth Reddy

Additional information is available at the end of the chapter

[Link]

1. Introduction

Scour that results directly from the impact of the hydraulic structure on the flow and which
occurs in the immediate vicinity of the structure is commonly called local scour. It is very
important to reduce local scour caused by impacting jets. A thorough understanding of the
erosion of the bed due to local scour remains a challenge since it is associated with a highly
turbulent flow field. The size, shape and density of sand particles, flow velocity, flow depth,
turbulent intensity and shape of scour hole influence the complex processes of entrainment,
suspension, transportation and deposition of sediment. This chapter briefly discusses trans‐
port and local scour mechanism of cohesionless sand caused by plane, two-dimensional and
three-dimensional wall jets. The aim of the chapter is to facilitate the reader to understand scour
characteristics, scour profile measurements, effects of tailwater depth, sand grain size, width of
the channel, laboratory test startup conditions and scour that occurs under ice cover conditions.

2. Scour hole characteristics

The scour hole formation by various forms of jets (e.g., circular, two-dimensional (2D) and
three-dimensional (3D)) can destabilize a hydraulic structure in the immediate vicinity of the
jet. A typical visual example of the scour process caused by a 2D jet is presented herein to
illustrate the complexity of the flow. As the flow is commenced, the jet exits the nozzle or the
sluice gate, interacts with the bed and scouring action takes place. This is usually called the
digging phase. During digging, the jet is directed towards the bed, a hole is formed and the
excavated sand is deposited as a mound just downstream of the scour hole. The scouring
process is very rapid in this phase. The location of the impingement point changes in the
longitudinal direction during the digging process. Following the digging period, the jet flips
towards the free surface and a refilling process is commenced. In this phase, the water surface
178 Sediment Transport Processes and Their Modelling Applications

tends to be wavy. The surface jet impinges on the mound region and some of the sediment
deposited on the mound falls back and refills the scour hole. In time, the jet suddenly flips
back towards the bed and once again causes rapid digging of the bed with lifting of the bed
material into suspension. The free surface is not wavy during the rapid digging process. An
intermediate hump is formed in the scour hole. In time, the jet is once again directed towards
the free surface and refilling occurs. The alternate digging and refilling occurs only at low
tailwater depths. The asymptotic condition (i.e., when changes to the bed profile is minimal)
is attained after a very long time period. The above process is illustrated in Fig. 1 (Balachandar
et al. 2000). Fig. 1(a) shows the end of a refilling cycle prior to the commencement of the digging
process. Fig. 1(b) shows the beginning of digging phase, with a recirculating flow region
rotating in a clockwise direction being formed upstream of impingement and a counter
clockwise roller formed in the downstream section. Fig. 1(c) shows sand is primarily trans‐
ported by advection of suspended particles. Fig. 1(d) shows a relatively calm water surface
profile during the digging phase. The redirection of the jet towards the water surface (surface
jet) after the end of digging phase causes the water surface to become wavy (Figs. 1e, 1f and
1g). Refilling takes place in Figs. 1(f) and 1(g). Fig. 1(h) shows the reoccurrence of digging
phase. Studying the effects of tailwater depth, flow properties and sediment properties on the
scour hole dimensions will help in better predicting the scour hole geometry.

The definition sketch of the scour hole geometry is shown in Fig. 2 (Sui et al. 2008). Several
researchers including Dey and Sarkar (2006a, b), Pagliara et al. (2006) and Balachandar et al.
(2000) studied the role of the following geometrical and hydraulic parameters governing the
scour process:

Figure 1. Demonstration of local scour cycle: (a) scour begins; (b) digging phase commences; (c) digging continues;
(d) maximum digging; (e) filling phase begins; (g) maximum fill; (h) reoccurrence of digging phase. (Balachandar et al.
2000, copyright permission, NRC Research Press)
 Scour Caused by Wall Jets 179
[Link]

Figure 2. Definition of scour parameters (Sui et al. 2008, copyright permission of Science Direct)

i. Nozzle hydraulic radius (defined as bo / 4)

ii. Jet exit velocity (U o ) - the undiminished mean velocity in potential core

iii. Grain size - generally, the median grain size (d50) of the cohesionless bed is considered
as the representative grain size

iv. Densimetric Froude number ( F o ) -F o = U o / gd50Δρ / ρ, Δρ is submerged density of

bed material, and ρ is density of bed material.

v. Tailwater depth (H or yt ) - defined as depth of water over the original bed

vi. Channel width (W) - the width of downstream channel into which the jet is exiting

vii. Expansion ratio (ER ) - channel width to jet thickness,

viii. Time evaluated from the start of the flow (t )

ix. Submergence ratio - tailwater depth to the thickness of the jet at its origin.

The variables of interest include:

i. Maximum scour depth (εm)

ii. Location of maximum scour depth ( xm or δ ) from the nozzle exit

iii. Volume of scour (V ) - measured by determining volume of the water needed to fill
the scour hole.

iv. Scour hole length (L s ) - maximum extension of the scour hole along the midsection.

v. Scour hole width (w )- maximum extension of the scour hole measured perpendicular
to the flow direction.
180 Sediment Transport Processes and Their Modelling Applications

vi. Ratio of the length of the scour hole at asymptotic conditions to jet exit velocity
(L s / U o)

Rajaratnam and Berry (1977), Lim (1995), Aderibigbe and Rajaratnam (1996), Ade and
Rajaratnam (1998) and Sui et al. (2008) studied wall jet scour and concluded that the key
parameter that effects the scour hole dimensions is F o . Sediment uniformity also influences
the scour geometry. The formation of the armor layer in non-uniform beds leads to smaller
scour geometry when compared to that formed in uniformly distributed sediment (Aderibigbe
and Rajaratnam 1996, Mih and Kabir 1983). In general, it has been assumed that the effect of
the sediment size on scour hole dimensions can be absorbed by the densimetric Froude number
(Rajaratnam (1981), Mazurek and Rajaratnam (2002) and Rajaratnam and Mazurek (2003)). The
influences of the various factors are described in the forthcoming sections. Sui et al. (2008)
investigated the effect of the sediment size on scour hole dimensions in detail.
The effect of tailwater depth on the scour hole dimensions due to impinging and free falling
jets have been investigated by Ghodsian et al. (2006) and Aderibigbe and Rajaratnam (1996).
Critical values of the tailwater depth were introduced by Aderibigbe and Rajaratnam (1996)
and Ghodsian et al. (2006) such that beyond the critical value (either by increasing or decreasing
the tailwater depth), the scour hole dimensions decreased. However, Ali and Lim (1986),
Faruque (2004), Sarathi et al. (2008) and Mehraein et al. (2010) have stated that scour hole
dimensions due to 2D and 3D wall jets increase either by increasing or decreasing the tailwater
depth from the critical value.

3. Scour due to wall jets

3.1. Characteristics of two–dimensional wall jets

Mohamed and McCorquodale (1992) investigated the local scour downstream of a rectangular
opening with a swept-out hydraulic jump and identified two stages of local scour develop‐
ment.

i. An initial stage of local scour which occurs rapidly (short-term scour).

ii. A progressive stage which approaches equilibrium after a very long time (long-term
scour).

Mohamed and McCorquodale (1992) observed that the equilibrium depth for short-term scour
established rapidly in less than 1% of the time to reach the long-term scour depth. The short-
term scour although not as deep as the long-term scour occurs much closer to the apron; the
bed is more highly fluidized than in the regime that governs the long-term scour. They have
also stated that short-term scour due to plane horizontal supercritical jets under low tailwater
conditions is related to the energy dissipation regime that dominates the flow. They identified
seven different jet forms:

1. Attached jet: Bottom boundary of jet conforms to the bed and top boundary of jet forms
the free surface from beginning of jet exit.
 Scour Caused by Wall Jets 181
[Link]

2. Moving jump: Hydraulic jump formed by the jet is propagating downstream.

3. Wave jump: Standing wave type of hydraulic jump is formed downstream of the jet exit.

4. Surface jet: Top boundary of jet forms the free surface and bottom boundary of jet is not
confined.

5. Plunging jump: Jet is plunging down along the slope of the scour hole. Specifically, the
bottom boundary is along the slope of scour hole and top boundary of jet is submerged.

6. Inverted jump: Top and bottom boundaries of jet try to attain the free surface and bed,
respectively, beyond scour hole.

7. Classical jump (as in stilling basins)

The more rapid short-term scour was associated with regimes 1, 2, 3 and 6, while the deeper
long-term scour was associated with regimes 4 and 5 indicated above.

Rajaratnam (1981) found that maximum depth and length of scour hole increased linearly with
the logarithm of time. He noted that the scour hole reached an asymptotic stage. Rajaratnam
(1981) concluded that maximum depth and length of scour are largely dependent on the
densimetric Froude number. Johnston (1990) investigated three different scour hole regimes
created in shallow tailwater conditions. Two scour hole regimes were formed when the jet
permanently attaches itself to either the bed or free surface boundary whilst the third was
formed when the jet periodically flips between the free surface and channel bed. Johnston
(1990) found that the scour hole development in deep water conditions is orderly and invari‐
ably reaches a well-defined asymptotic state while in shallow conditions such a state is
sometimes not reached. The depth of flow has a considerable influence on the near bed flow
field and may promote the flipping of the jet from one boundary to another. Balachandar et
al. (2000) suggested that tailwater depth was a key parameter when tailwater depth was less
than 16 times the nozzle thickness or sluice gate opening (bo ). Chatterjee et al. (1994) found that
maximum scour depth is a function of Froude number based on nozzle thickness
(F = U o / gbo ). Kells et al. (2001) concluded that scour geometry is dependent on sediment bed
grain size and initial jet exit velocity (U o ). Ali and Lim (1986) have developed the following
power law relationship for time evolution of scour:

V
R3
= 187.72( )εm 2.28
R

where, V is the volume of scour and R is hydraulic radius and εm is maximum scour depth.

3.2. Scour characteristics of three–dimensional wall jets

Scour caused by 3D and circular wall jets have been studied by Meulen and Vinje (1975), but
to a lesser degree than that caused by plane and impinging jets. Rajaratnam and Berry (1977)
studied the scour caused by circular wall jets on cohesionless soils. They concluded that the
geometric characteristics of the scour hole mainly depend on the densimetric Froude number.
Rajaratnam and Berry (1977) also found that the jet expands in an unconfined manner from
182 Sediment Transport Processes and Their Modelling Applications

the nozzle exit to the point of maximum scour depth. Ali and Lim (1986) concluded that for
3D jets, mean flow velocity for any section decreases continuously in the flow direction,
whereas for two-dimensional flows, the mean velocity increases as the flow develops. In three-
dimensional jet scour, no reverse flow was observed near the bed. It was observed in 3D flows
that there were occasional turbulent bursts near the bed which moved sand particles from the
upstream slope of the hole to the region of maximum scour. Ali and Lim (1986) have suggested
the following power law relationship for time evolution of volume of scour:
V
R3
= 49.36 ( ) εm 1.89
R

Hoffmans and Pilarczyk (1995) developed a semi-empirical relation for the upstream scour
slope and verified the equation using experimental results. Lim (1995) studied the effect of
channel width (expansion ratio) on scour development of 3D jets and concluded that there was
no effect of channel width on scour development for expansion ratios greater than 10. The
width of scour hole was affected by the width of downstream channel only when normal
diffusion of the 3D jet flow is restricted in transverse direction. Rajaratnam and Diebel (1981)
found that relative tailwater depth and relative width of downstream channel affect the
location of maximum scour depth. Chiew and Lim (1996) developed the following empirical
equations by using the densimetric Froude number ( F o ) as the main characteristic parameter
to estimate scour dimensions caused by circular wall jet:
εm w L
bo = 0.21F o bo = 1.90F o0.75 bo = 4.41F o0.75

Ade and Rajaratnam (1998) further emphasized the use of F o as the main parameter to analyze
scour caused by circular wall jets. However, they noted that asymptotic dimensions of scour
hole are dependent on tailwater depth for F o > 10. Ade and Rajaratnam (1998) also found that
the eroded bed profile near the nozzle attains asymptotic state earlier than locations away from
nozzle.

4. Role of fluid structures on two–dimensional scour

Hogg et al. (1997) have pointed out that a comprehensive understanding of the scour remains
elusive because of the complex nature of the flow field. Rajaratnam (1981), Rajaratnam and
Macdougall (1983), Wu and Rajaratnam (1995) and Rajaratnam et al. (1995) have made
significant contributions despite the complexity resulting from the hydrodynamic character‐
istics of the jet and the concave shape of eroded bed. Hopfinger et al. (2004) proposed new
scaling laws relating time and the attainment of the quasi-steady scour depth. They suggested
that turbulence created by Gortler vortices cause sediment transport and the associated
scouring mechanism due to the destabilization of the turbulent wall layer by the concave
curvature of the water sediment interface. Bey et al. (2007) studied the velocity field during
asymptotic conditions in the scoured region to understand the role of turbulent flow structures
that influence scour and evaluated higher-order velocity moments. Experimental findings of
Bey et al. (2007) are presented here.
 Scour Caused by Wall Jets 183
[Link]

Bey et al. (2007) carried out two tests at exit velocities of 1.0 m/s and 1.27 m/s (Tests A and B)
and at a tailwater depth ( H ) corresponding to 20 bo continuously for 12 days duration. These
tests can be classified under the high submergence flow regime. The high and low submergence
is not clearly defined in literature, however high submergence can be defined as a state where
no alternate flicking of the jet between the free surface and bed occurs as in low submergence
cases (Balachandar et al. 2000). Generally, a value of H / bo > 10 has been considered as high
submergence. The time to reach asymptotic conditions was found to be as low as t = 24h
depending on scour hole and ridge geometric parameter chosen. For example, the total length
of the scour affected region ( L T ) attains an asymptotic state at t = 24h, however the change in
the maximum depth of the scour hole (εm) is less than 5% after t = 48h. However, it should be
noted that beyond 72h, there is the presence of the turbulent bursts which cause local changes
in the location of the maximum depth of scour profile ( xm) but no significant changes in the
mean scour profile. Sectional and plan view of the scour region along with the definition of
various variables are shown in Fig. 3(a).

Bey et al. (2007) divided the entire test duration from the start of flow to the attainment of
asymptotic conditions into five time zones to study the influence of the different flow struc‐
tures. A time scale T = L s / U o is used to non-dimensionalize the five time zones, where L s is
length of scour hole at asymptotic conditions. Fig 3(b) shows variation of important scour
parameters with time in five time zones. Each of the time zones had certain dominant flow
features. Flow was characterized by the presence of longitudinal vortices and turbulent bursts
at the start of the test and during early time periods, and movement of the jet impingement
point during the later stages. Scour was very rigorous during first time zone (t / T ≤ 850). A 2D
hole was formed downstream of the nozzle and the bigger size particles were deposited further
downstream as a 2D mound. In second time zone (850 ≤ t / T ≤ 15x103), the large-scale suspen‐
sion of the bed material seen in the earlier time zone was reduced significantly. No major scour
happened in the negative slope region (Fig. 3a). Two or three longitudinal streaks and very
prominent concave depressions were observed. Hopfinger et al. (2004) reported that vortices
due to Gortler instability caused scour on the positive bed slope. During the third time zone
(15x103 ≤ t / T ≤125x103), there appeared a “scoop-and-throw” like scouring action on either
side of the flume axis, which caused longitudinal concave shaped depressions. A lifting
spiraling motion of bed particles occurred near the end of concave shaped depression due to
vortex activity. Dye injection confirmed the presence of spiral motions during the fourth time
zone (125x103 ≤ t / T ≤ 375x103) and also scoop-and-throw like scouring action slowed down
after 24 h. The scour hole attained an asymptotic state in the fifth time zone after 72 hours of
scour process (t / T ≥375x103). In the asymptotic stage, turbulent bursts were noticed to occur
in the near bed region across the section of the flume. In addition, two prominent scour
mechanisms occurred on either side of the flume axis causing the bed particles to be spiraled
toward the sidewalls from midsection and from sidewalls to the flume axis. However, no
particle movement was observed along the nozzle axis.

At asymptotic conditions, scour profile in the scour hole region was nominally two-dimension‐
al across the width of the flume. It should be remarked that the backward and forward move‐
184 Sediment Transport Processes and Their Modelling Applications

ment of the jet contact point with bed, the frequent but random turbulent bursts, and the two
prominent scour mechanisms (digging and refilling), all occurred at one time or the other.
However, in the mound region, the lateral profile was not two-dimensional and had two distinct
peaks. These peaks occurred closer to the sidewalls leaving a trough in center portion of the
mound. Jet impingement point on the bed moved backward and forward, and the mean location
of point of contact of jet with bed was close to the deepest point of scour hole. Wide range of
variation in instantaneous velocity from positive to negative values was a clear indication of
back and forth movement of jet. It was observed that most of the turbulence activity occurred in
the region of near-zero slope of scour hole. Analysis of third-order moments and quadrant
decomposition indicated sweep type events in the near bed region contributed to scour. Ejection
events near the bed caused suspension of the bed particles to be carried away by the average
flow velocity. Measured velocities at asymptotic conditions were extrapolated to other time
periods to conclude that the sweep and ejections contributed significantly to the scour process.

Figs. 4(a) and (b) show the flow fields for the tests A and B in the scour hole region at asymptotic
conditions. Velocity vectors show that the jet expands vertically to interact with the bed. A
large-scale recirculating region was found in the region above the jet axis, which extended to
about 35bo downstream of the jet. The center of this region was located at about 16bo along the
x-axis and about 10bo along the y-axis. It is important to recognize that this recirculation region
caused significant negative vertical velocities immediate vicinity of nozzle. Impact of the jet
on the bed also generated a flow separation near the region close to the nozzle. A similar flow
field was also observed in test B, which is qualitatively similar to that obtained in test A, and
is shown in Fig. 4(b). The mean impingement point in test B occurs farther from the nozzle and
consequently has a larger near-bed recirculation zone.

Figure 3. a) Definition sketch, (b) Variation of scour parameters with time (Bey et al. 2007, copyright permission of
ASCE)
 Scour Caused by Wall Jets 185
[Link]

Figure 4. Velocity vector plot for (a) Test A; (b) Test B (Bey et al. 2007, copyright permission of ASCE)

5. Measurements using video image analysis

Figures 5(a) and 5(b) show a set of water surface profiles at about 840 seconds from the start
of scour process. The profiles were obtained using a software called PROSCAN developed by
Balachandar and Kells (1998). The time to refers to the start of the cycle. Figure 5(a) shows the
186 Sediment Transport Processes and Their Modelling Applications

water surface profile during the digging phase and Fig 5(b) shows the water surface profiles
during the refilling phase. Figure 5(c) shows bed scour profiles corresponding to the water
surface profiles shown in Figs. 5(a) and 5(b). The bed profiles reflect the alternate digging and
refilling phases as explained ealier. At t = to, the bed surface profile is similar to that provided
by Rajaratnam (1981) where the water surface profiles are similar during digging and refilling
phases. In the study of Balachandar and Kells (1998), the bed profile at t = to+2 s is quite different
from initial bed profile at t = to. It is observed that at t = to +2 s, the dynamic flow field indicates
the digging phase near the sluice gate and piling up of sand (hump) on downstream side. After
t = to + 24 s, during the refilling phase, the sand particles in the hump part moved upstream
and eventually refilled the trough. The refilling phase occurred over a longer period of time
and the refilling process was nearing completion at t = to+179 s and profile eventually became
similar to the profile shape recorded at t = to.

5.1. Computational stereoscopy

Ankamuthu et al. (1999) developed a stereoscopy scheme to measure the depth of scour in
three-dimensional flow fields. The stereoscopy scheme makes use of an epipolar constraint
and a relaxation technique to match corresponding points in two images. A correlation
technique was developed to eliminate false matches. The depth of scour was calculated using
the parallax between the matched points. However, the method is yet to be used to measure
scour in a practical flow field.

6. The effect of tailwater depth on local channel scour

Ali and Lim (1986) studied scour caused by 3D wall jets at shallow tailwater conditions and
noted that tailwater had an influence on the maximum depth of scour at asymptotic conditions.
Tailwater depth which is under or above a critical condition causes an increase in maximum
depth of scour (Ali and Lim, 1986). Further analysis of their data indicates that the critical value
of tailwater depth increases with increasing densimetric Froude number. Ade and Rajaratnam
(1988) stressed the use of Fo as the characteristic parameter to describe scour caused by circular
wall jets and noted that the maximum depth of scour was found to be larger at high ratio of
tailwater depth to nozzle width and higher values of Fo, which is consistent with the meas‐
urements of Ali and Lim (1986). However, they noted that the asymptotic dimensions of the
scour hole were dependent on the tailwater conditions for only Fo >10. Rajaratnam and Diebel
(1981) concluded that the relative tailwater depth and relative width of the downstream
channel do not affect the maximum depth of scour, whereas the location of the maximum scour
was affected.

Faruque et al. (2006) presented the results of clear water local scour generated by 3D wall jets
in a non-cohesive sand bed at low tailwater depths. They indicated that extent of scour of 3D
wall jets is collectively influenced by the densimetric Froude number, tailwater depth, and
grain size-to-nozzle width ratio. Each parameter has a dominant influence compared to other
parameters at different flow conditions. For Fo < 5, tailwater depth has no effect on the
 Scour Caused by Wall Jets 187
[Link]

Figure 5. Variation of water surface profile and bed profile (after t = 823 s) (Balachandar and Kells, 1998, copyright
permission of NRC Research Press)
188 Sediment Transport Processes and Their Modelling Applications

maximum depth and width of scour. The effect of tailwater depth at higher densimetric Froude
numbers appears to be important at larger values of non-dimensional sediment diameter (d50/
bo). Previous observations have indicated that for Fo > 10, the effect of the tailwater depth was
significant. However, the results of Faruque et al. (2006) clearly indicate that the effect of
tailwater can be important at lower values of Fo depending on the value of d50/bo. At asymptotic
conditions, comparing the volume of scour at H/bo = 6 with that at H/bo =4, it was observed that
the scour volume is greater at the higher tailwater depth. Faruque et al. (2006) also indicated
that the maximum depth of scour is not necessarily deeper at higher values of tailwater depth
in the lower range of submergences.
Experimental results of Faruque et al. (2006) are further discussed here. Three different
tailwater depths (Tests A, B, and C) corresponding to 2bo, 4bo, and 6bo where bo = 26.6 mm and
a jet exit velocity = 1.31 m/s (in the three tests) were considered in the experiments. The
corresponding Reynolds number in the tests (Re = U o bo / ν) was 1.0 x 105 which shows preva‐
lence of fully turbulent conditions. The flow Froude number based on exit velocity and nozzle
width was 1.5, grain size was 0.85 mm and corresponding densimetric Froude number was
six. Velocity measurements were obtained using a single component fibre-optic laser Doppler
anemometer and scour profiles at different time intervals were obtained using a point gauge
with an electronic display unit. Fig. 6(a) illustrates the time development of scour profiles along
the nozzle axis for tests A, B and C. The profiles gradually attain an asymptotic state around
t = 48 h. Fig. 6(a) also illustrates the maximum depth of scour (εm) and the location of maximum
depth of scour from the nozzle exit ( xm) increase with increasing time. It was also found that
the difference between two consecutive profiles at a fixed distance from the nozzle increases
with increasing x . It should be noted that Ade and Rajaratnam (1988) found that maximum
depth of scour to be larger at deeper submergences. In Figs. 6(b-d), the extent of scour is
consistently larger at H/bo = 6 at all t, and therefore the volume of scour at asymptotic conditions
was greater for the larger submergence. The corresponding top view of the perimeter of the
scour hole is shown in Figs. 7(a-e) which indicates that the perimeter of the sour hole is
consistently larger at H/bo= 6 as compared to H/bo= 4. However, the hole perimeter expands
laterally as depicted in Fig. 7(e) at H/bo= 4 as asymptotic conditions are reached.

7. Effect of grain size on local channel scour

Scour occurs when the high velocity jet produces bed shear stresses that exceed the critical
shear stress to initiate motion of the bed material. Critical shear stress of the bed material is a
function of grain size and therefore, scour is a function of grain size. This section summarizes
the effect of grain size on local scour. Breusers and Raudkivi (1991) discussed achievement of
equilibrium state of scour and the similarity of scour profiles for various sizes of bed material
and jet velocity. Balachandar and Kells (1997) studied the scour profile variation with time
using a video imaging process on uniformly graded sediments caused by flow past a sub‐
merged sluice gate. Kells et al. (2001) investigated the effect of varying the grain size and, to a
lesser extent, the grain size distribution of the erodible bed material on the scour characteristics.
Figs. 8 and 9 are drawn from their study.
 Scour Caused by Wall Jets 189
[Link]

Figure 6. a) Development of scour profile along nozzle axis (b) comparison of scour profiles along nozzle centerline at
t = 3 h; (c) t= 48 h; and (d) t = 72 h (Faruque et al. 2006, copyright permission of ASCE)
190 Sediment Transport Processes and Their Modelling Applications

Figure 7. Comparison of plan view of perimeter of scour hole at various instances of time (a) t = 1 h; (b) t = 3 h; (c) t = 6
h; (d) t = 48 h; and (e) t = 72 h (Faruque et al. 2006, copyright permission of ASCE)
 Scour Caused by Wall Jets 191
[Link]

Figure 8. Effect of discharge and grain size on Phase-B maximum scour depth, εm, at t= 24h: (a) H/bo = 4.0; (b) H/bo =
6.3 and (c) H/bo= 8.8 (Kells et al. 2001, copyright permission of NRC research press)
192 Sediment Transport Processes and Their Modelling Applications

Figure 9. Effect of discharge and grain size on Phase-B location of point of maximum scour depth, xm, at t= 24h: (a)
H/bo= 4.0; (b) H/bo = 6.3 and (c) H/bo= 8.8 (Kells et al. 2001, copyright permission of NRC research press)
 Scour Caused by Wall Jets 193
[Link]

The figures show the effect of grain size with increasing discharge and tailwater depth.
Analysis of Figs. 8a, 8b, and 8c tell us that the magnitude of the maximum depth of scour (εm)
increases with an increase in flow rate for a given grain size and maximum depth of scour
decreases with increasing grain size for a given flow rate. It is also observed from Fig. 8 that
maximum scour depth increases with increasing tailwater depth for a given grain size.
Figs. 9a, 9b, and 9c demonstrate that the location of maximum scour depth (δ) moves upstream
with an increase in the grain size and moves downstream with increase in discharge. This is
due to the fact that shear stress increases with increase in flowrate or critical shear stress
decreases with decrease in grain size. The location of maximum scour depth (δ) moves
downstream with increasing tailwater depth.
In summary, it was found that the grain size had a significant influence on the extent of scour,
with more scour occurring with the smaller-sized material. Less scour occurred for graded
sand (indicated as mixed sand in the figure) than a uniform one having a similar median grain
size. Amount of scour increases with an increase in the discharge, hence the velocity of flow.
For any given discharge and grain size, the greater the tailwater depth, the greater is the scour
depth, extent, and volume of scour. It appears that the tailwater serves to slow the rate of jet
expansion, thus increasing the length of bed exposed to high velocity, hence high shear stress
conditions. It was also found that the tendency toward the dynamic alternating dig-fill cycling
was lessened with a reduction in the discharge, or an increase in the grain size or the tailwater
depth.

8. Effect of test startup conditions on local scour

A cursory evaluation of the test startup conditions in several of the studies mentioned in this
chapter indicates that scour pattern is quite varied and depends on how the flow is initially
commenced. For example, Kells et al. (2001) had the nozzle outlet plugged, whereas Bala‐
chandar et al. (2000) and Mohamed and McCorquodale (1992) had the sluice gate closed until
proper head and tailwater conditions were established. Following this, the nozzle was
unplugged or the sluice gate opened to a predetermined extent to generate the jet flow. This
requires a certain amount of time before a steady jet discharge can be established. Rajaratnam
(1981), Chatterjee et al. (1994), Mazurek (2002), and Aderibigbe and Rajaratnam (1998) created
a suitable constant head difference between the downstream and upstream sections prior to
generating the jet.
Johnston (1990) and Ali and Lim (1986) used a suitably sized aluminum sheet to cover the
leveled bed in order to prevent it from being disrupted on commencement of the flow. The
inflow was started; the flow and tailwater depths were then set to desired values, following
which the sheet was slowly removed. It is thus clear that even for a seemingly simple flow
emanating from nozzles or sluice gates, the flow and the corresponding scour pattern can
become complex due to various influences and one among them is the startup condition.
Deshpande et al. (2007) investigated the effect of test startup conditions on plane turbulent
wall jet behavior and the resulting scour profiles. The changing startup conditions also reflect
194 Sediment Transport Processes and Their Modelling Applications

practical situations where the discharge and tailwater conditions change during regular
operations. Deshpande et al. (2007) study also investigated the jet behavior and the different
regimes of flow for a range of submergences (both low and high) using a laser Doppler
anemometer (LDA). Furthermore, the effects of three different startup conditions on scour
include an instantaneous startup condition, a gradual startup condition, and a stepwise startup
condition were studied. Results of Deshpande et al (2007) are discussed below.

Figure 10. Effect of test startup conditions for H/bo = 4 (Deshpande et al. 2007, copyright permission of Taylor and
Francis)
 Scour Caused by Wall Jets 195
[Link]

Figure 11. Effects of test startup conditions on the scour hole profiles (Deshpande et al. 2007, copyright permission of
Taylor and Francis)
196 Sediment Transport Processes and Their Modelling Applications

Fig. 10 shows the velocity-time history near the nozzle exit for three startup conditions at a
low submergence (H/bo = 4). Velocity data clearly indicate that there is no large scale scour upto
about 150 s for stepwise and gradual change in flow. Velocity profile shows a significant dip
in the mean velocity accompanied by large scale increase in turbulence after 150 s which
indicates large scale scour. Further increase in valve opening increases velocity and large scale
scour and causes higher turbulent intensities. For t > 400s, there is the presence of low frequency
fluctuations. It was also found that a steady state velocity was attained after t = 2000 s in the
case of stepwise startup condition and after t = 600 s in the case of gradual startup condition.

Fig. 11(a) shows the effect of the startup conditions on scour at H/bo = 4 at 3h from the com‐
mencement of the flow. There are no significant differences in the profiles that can be attributed
to startup conditions. However, the step wise and gradual startup conditions influence the
mound to be slightly bigger and longer. It was observed that the digging phase continued for
a longer time for these two startup conditions as compared to instantaneous condition. As a
result, there has been more digging action and this is reflected in the profiles. This is an
important aspect to note while comparing scour profiles with different startup conditions. Fig.
11(b) illustrates the profiles along the nozzle centerline for three startup conditions at 24h from
the commencement of the test for H/bo= 4. No significant differences can be found in the profiles
and any minor differences that were noticed earlier have vanished. Figure 11(c) shows the
scour profiles at the three startup conditions at t = 36 h at H/bo =20. Clearly the effects of test
startup conditions are not evident at t = 36h. It was concluded that the effects of test startup
conditions did not influence the long term scouring process.

9. Effect of channel width on local scour

Lim (1995) pointed out that the scour profile is not affected by the channel sidewalls when the
expansion ratio is ten or greater. He also noted that the downstream channel would affect the
lateral development of the scour hole if it becomes too narrow and restrict the normal diffusion
of the three-dimensional jet. However, Faruque et al. (2006) have noted that scour hole
dimensions are affected by the width of the downstream channel even for an expansion ratio
(ER) as high as 14.5. They reported that the occurrence of the secondary ridges along the wall
should be an effect of the jet expansion ratio. Faruque et al. (2006) and Sarathi et al. (2008) found
that no secondary effects were observed for an ER of 41.4.

The effect of channel width on the extent of scour with varying tailwater conditions has been
documented by Bey et al. (2008). They investigated the effect of jet exit velocity, tailwater depth
and channel width concurrently on the scour characteristics. To this end, Bey et al. (2008)
studied four groups of tests (denoted as A, B, C and D) using LDA and scour profile meas‐
urements to characterize the flow field. Tests were conducted at four widths (w = 0.25 W , 0.5W,
0.75 W and W) and three different jet exit velocities (U o = 0.75, 0.90 and 1.16 m/s). Here, W
denotes a width of 0.4 m. In group A, for a given jet velocity and channel width, the submer‐
gence was varied in a stepwise fashion from a low tailwater (yt = 2bo) to a high tailwater
condition (yt = 20bo). The tests in group B were commenced from a high submergence condition
 Scour Caused by Wall Jets 197
[Link]

(yt = 20bo) with a stepwise reduction towards low submergence. Group C tests were chosen
such that the jet would continue to flip alternately between the bed and the free surface, and
yet be close to the high submergence range at a given exit velocity. Group D tests were carried
out at a fixed tailwater depth (yt = 20bo) for a period of 24 hours and velocity measurements are
obtained at various time intervals. The results of varying channel width on bed scour profile
variation are illustrated in Figs. 12 and 13.
Figure 12(a) and (b) shows the scour profiles for group C and group D for Uo = 1.16 m/s at
various channel widths at t = 24 h. It can be observed from Fig. 12 that the scour profiles are
dependent on channel width. The maximum scour depth decreases as the channel width
increased from w= 0.25W to 0.75W. However, the maximum scour depth at w= 1.0 W is more
than at w = 0.25W and 0.75W. From Fig. 12(b), it is observed that the maximum scour depth
decreases with decreasing channel width. For all other conditions maintained constant, at
higher submergence, it can be seen from the results that as the channel width increases, the
maximum scour hole depth decreases, followed by an increase in maximum scour hole depth
with further increase in width, whereas, at lower submergence, the scour hole depth decreases
with decreasing channel width. The elongated mound and the formation of an intermediate
hump in the scour hole near the nozzle (Fig. 12b) have been also observed in previous studies
of Kells et al. (2001) and Deshpande et al. (2007).
Figure 13(a) and 13(b) shows the variation of maximum depth (ε) of scour at t = 24 h for test
groups A and C, respectively. It can be noted that maximum depth of scour is a function of jet
exit velocity and channel width. Available data from literature with comparable values of
densimetric Froude number (Fo) are also shown in Fig. 13. It is observed from Fig. 13 that scour
depth is different for the group A and C for any given channel width.
Sui et al. (2008) studied scour caused by 3D square jets interacting with non-cohesive sand
beds to further understand the effects of channel width, tailwater conditions and jet exit
velocity. The tailwater ratio was varied from 2 to 12 times the nozzle width, while the channel
width was 31.6 and 41.4 times the nozzle width. Three different jet exit velocities were adopted
and two different bed materials fine sand (d50= 0.71 mm) and coarse sand (d50= 2.30 mm) was
used. Fig. 14 presents the variation of the different scour parameters with varying expansion
ratio and tailwater ratio for both fine (Fig. 14a-c) and coarse sand (Fig. 14d-f). As shown in Fig.
14a, it can be noted that the maximum scour width increases with increasing ER and attains a
maximum value at a tailwater ratio (TWR) = 4 for the tests with ER = 41.4. Increase of TWR
results in a decrease of maximum scour width. For TWR > 3, length of scour hole ( L s ) decreases
with increasing ER. From Fig. 14b, it can be observed that maximum ridge width increases
with increasing ER. It is also observed that the maximum length of scour hole is shorter for the
higher ER for TWR > 3. Figs. 14d-f show the variation of the scour geometry parameters for
the coarser sand at the two expansion ratios. Fig. 14d shows that the maximum width of scour
is smaller for the higher expansion ratio for 3 ≤ TWR ≤ 4 and the maximum width of scour is
higher for higher expansion ratio for TWR > 4.5. It is observed from Fig. 14d that the effect of
ER on maximum scour width decreases with increasing sand grain size. Fig. 14e shows that
there is no effect of ER on ridge width for coarse sand. Length of scour hole increases with
decreasing ER (Fig. 14f). One can conclude the following from the study of Sui et al. (2008):
198 Sediment Transport Processes and Their Modelling Applications

(a)

(b)

Figure 12. a) Scour profile variation with channel width with Uo = 1.16 m/s for group D, (b) Scour profile variation
with channel width with Uo = 1.16 m/s for group C (Bey et al. 2008, copyright permission of Taylor and Francis

(a)

(b)

Figure 13. a) Plots of maximum scour depth with channel width for various velocities in group A, (b) Plots of maxi‐
mum scour depth with channel width for various velocities in group C (Bey et al. 2008, copyright permission of Taylor
and Francis)
 Scour Caused by Wall Jets 199
[Link]

i. For the fine sand and for TWR > 3, scour hole was wider but shorter in length at the
higher expansion ratio. For the coarse sand, a similar trend was observed for TWR > 4.
ii. For the coarse sand and for 3 ≤ TWR ≤ 4, scour hole is strongly dependent on the
tailwater ratio, especially at the lower expansion ratio.
iii. The effect of expansion ratio reduces as densimetric Froude number increases.
iv. The extent of difference for different scour parameter due to expansion ratio reduces
with increasing TWR irrespective of the sand grain size.
v. Effect of ER reduces with increasing sand grain size.
vi. Results indicate that different expansion ratios create different scour profiles even for
TWR ≥ 12.
vii. In an effort to provide useful, but a simplified scour prediction equation at asymptotic
conditions, a relationship for the maximum scour depths has been proposed in terms
of densimetric Froude number, tailwater ratio and expansion ratio.
viii. Proposed relationship predicted scour depths correctly for a wide range of test
conditions.

10. Influence of densimetric Froude number on local scour

To better understand the scouring process, jets interacting with sand beds have been studied
by many researchers and empirical relations involving densimetric Froude have been pro‐
posed to predict local scour. Rajaratnam and Berry (1977) studied the scour produced by
circular wall jets and concluded that the main geometric characteristics of the scour hole are
functions of the densimetric Froude number. Rajaratnam and Diebel (1981) concluded that the
relative tailwater depth and width of the downstream channel only affect location of the
maximum scour, whereas densimetric Froude number affects maximum scour depth. Ali and
Lim (1986) indicated that the value of the critical tailwater condition increases with increasing
densimetric Froude number and the effect of tailwater becomes insignificant when H/bo is
beyond 16. Chiew and Lim (1996) studied local scour by deeply submerged circular jets of both
air and water and concluded that the densimetric Froude number was the characteristic
parameter in describing the scour hole dimensions. Ade and Rajaratnam (1998) noted that the
maximum depth of scour was larger at higher values of Fo. It was also noted that to attain an
asymptotic state at higher values of Fo, a longer time was required. Faruque et al. (2004)
concluded that the densimetric Froude number, tailwater depth and nozzle size-to-grain size
ratio, all have an influence on the extent of scour caused by 3-D jets. They speculated about
the dominance of each parameter at different flow conditions.
Sarathi et al. (2008) studied effect of the densimetric Froude number, tailwater depth and
sediment grain size on scour caused by submerged square jets. Results of their study with
respect to densimetric Froude number are presented in this section. Figures 15(a-c) show the
200 Sediment Transport Processes and Their Modelling Applications

Figure 14. Variation of different scour parameters with respect to tailwater ratio (TWR) for Fo= 10, fine sand (d50= 0.71
mm) and coarse sand (d50= 2.30 mm) (Sui et al. 2008, copyright permission of Science Direct)

asymptotic scour profiles along the centerline of the nozzle for different Fo values. The
asymptotic shape of the scour bed profiles for different Fo at H/bo = 18 are shown in Fig. 15(a).
It is observed that the maximum ridge height and its location, maximum scour depth and the
distance of the maximum depth of scour hole from the nozzle ( xm) increases with increasing
Fo. Figure 15(b) shows the scour profiles at three different values of Fo at lower tailwater
conditions ( H/bo=2). The ridge crest is sharper at the lower value of Fo and at higher Fo the ridge
is flat and directly related to the prevailing local velocity. Figure 15(c) shows the profiles at
different Fo for H/bo = 1. Ridge crests are flat and they are of constant height at different
densimetric Froude numbers. It is clear that the size and shape of the ridge is clearly dependent
on Fo and tailwater depth. Figure 15(d) shows the variation of the ridge height for different
densimetric Froude numbers and tailwater depth. Ridge height is higher with higher densi‐
metric Froude number. Ridge height increases with increasing tailwater depth for a given Fo
however, beyond H/bo = 6 for a given Fo, the ridge height attains a near constant value. Figure
16(a-c) shows the dependence of the normalized asymptotic scour parameters on Fo for a range
of tailwater ratio. Figure 16(a) shows asymptotic maximum scour depth (εm) is higher for higher
densimetric Froude number irrespective of the tailwater depths. Figure 16(b) and 16(c) show
 Scour Caused by Wall Jets 201
[Link]

similar trends of width of scour hole and length of scour hole. Scour hole dimensions are higher
for higher densimetric Froude numbers. Finally, Sarathi et al. (2008) concluded that the role of
grain size was completely absorbed by densimetric Froude number.

Figure 15. Scour geometry at asymptotic conditions at different densimetric Froude numbers (Sarathi et al. 2008,
copyright permission of Taylor and Francis)
202 Sediment Transport Processes and Their Modelling Applications

Figure 16. Variation of the asymptotic scour parameters with densimetric Froude number (Sarathi et al. 2008, copy‐
right permission of Taylor and Francis)

11. Local scour under ice cover

Ice covers in rivers effect the resistance to flow, flow velocity and depth. Ice cover on top of
the river with high sediment transport influences the development, size and shape of bed‐
forms. This section reviews ice cover influences on the relationships between flow and local
scour.

Ettema (2002) published an in-depth review on extent to which alluvial channels respond to
ice-cover formation, presence, and breakup. An imposed ice cover results in an increased
 Scour Caused by Wall Jets 203
[Link]

composite resistance to the flow. Ice cover thickness affects only a comparatively narrow region
near the upper level of flow in deep flows. On the other hand, in shallow flows, an ice cover
chokes the flow, which decreases the velocity resulting in the bed shear stress to be less than
the critical shear stress, and thereby decreases sediment transport. The studies also confirm
that sediment-transport rate increases with decreasing water temperature. The location of the
maximum velocity is dependent on the relative magnitudes of the channel bed and cover
underside resistance coefficients (Gogus and Tatinclaux, 1981). The bed elevation in an ice
covered stream must rise locally due to decrease in bulk velocity.
Asymptotic scour in the presence of smooth and rough ice cover was investigated by Sui et al
(2009) and arrived at the following conclusions:
i. The variation of maximum scour depth with tailwater ratio under covered flow was
different from that under open-water flow.
ii. Under open-water flow conditions, maximum scour depth decreased with increasing
tailwater depth, and beyond a certain tailwater depth, the values tend to increase and
attain constancy.
iii. Under covered flow, maximum scour depth increased with increasing tailwater
depth, and beyond a certain tailwater depth, the values tend to decrease and attain
constancy.
iv. Ice cover conditions do not influence maximum scour depth at large values of
submerged conditions the.
v. Maximum scour depth in the fine sand bed under ice covered flow was always less
than that noticed in the coarse bed material.
vi. Impact of ice cover on scour depth was also less obvious for finer sand.
vii. There was a negligible effect of ice cover on the scour parameters at lower densimetric
Froude number.
viii. Scour width increased with increasing tailwater depth but decrease with increasing
tailwater depth beyond a critical value and finally attained an asymptotic width in
free surface and with fine sand bed.
ix. In the case of the coarse sand, scour length was larger under rough ice cover condition,
but in the case of the fine sand, scour length was larger under open-flow condition.
x. Ridge height was not affected by type of bed material and with or without ice covered
flow.

12. Numerical modeling of scour caused by jets

Numerical modeling is gaining momentum since experimental investigations are time

consuming and expensive. Li and Cheng (1999) proposed a mathematical model based on
204 Sediment Transport Processes and Their Modelling Applications

potential-flow theory for simulating the equilibrium scour hole underneath offshore pipelines.
Karim and Ali (2000) tested effectiveness of a commercial software (FLUENT) to predict 2-D
flow velocity distribution and the bed shear stress generated by a turbulent water jet impinging
on rigid horizontal and scoured beds. Minimal and deeply submerged water jet simulations
were carried out. The close agreement between selected various experimental and computed
results were noted. The standard k − ε and the RNG k − ε models described the flow at the
boundary better than the Reynolds stress model (RSM).
Neyshabouri et al. (2003) attempted to obtain numerical predictions of the scour hole geometry
created by a free falling jet. The two-dimensional momentum equations, the continuity
equation were solved using a k − ε turbulence model. First, the turbulent flow due to a free
falling jet was computed, then the distribution of the sand concentration was determined based
on the convection-diffusion equation and the scoured bed was computed based on sediment
continuity equation. The above-mentioned steps were repeated until the equilibrium scour
hole was reached. Sediment transport calculations required specification of the sediment
concentration near the bed at the start of simulation. Stochastic and deterministic expressions
for sediment concentration near the bed proposed by van Rijn (1987) were used. The scour
profiles obtained using stochastic expression was most realistic as compared to using of
deterministic expression.
Adduce and Sciortino (2006) numerically and experimentally investigated local scour in clear
water scour conditions downstream of a sill followed by a rigid apron. A mathematical one-
dimensional model was developed which uses measured velocity fields (obtained using a
ultrasonic Doppler velocimeter) to simulate the scour hole evolution. The dune profiles
predicted by the model were similar to the measured profiles for large discharges, while when
the discharges were smaller, the dunes predicted by the model were always longer than those
measured.
Adduce and La Rocca (2006) studied different scour developments caused by a submerged jet,
the surface wave jet and the oscillating jet developed downstream of a trapezoidal drop
followed by a rigid apron. They have highlighted that as flow separation takes place at the
edge of the rigid bed, reverse circulation motion along the longitudinal section develops near
the bed downstream of the rigid bed itself. A stability analysis of the surface wave jet with
both a flat and a scoured bed was performed by using modified Saint-Venant equations with
correction terms accounting for the curvature of the streamlines. They concluded that the
stability of the surface wave jet is weakened by the presence of the scoured bed.
Boroomand et al. (2007) mathematically modeled an offset jet entering a domain with sediment
bed using ANSYS FLUENT with a two phase model in which water was primary phase and
sediment bed was the secondary phase. In multiphase models, setting up of initial conditions
is very important. These conditions include initial grid generation, initial phases and their
properties and volume fractions of each phase. The model determines the interphase exchange
coefficients, lifting force, virtual mass force, and interaction force between phases and solves
the continuity and momentum equations for each phase. Scour profiles calculated by the model
agreed fairly with the measurements. Also, the computed concentration profiles agreed fairly
with measurements except near the bed zone.
 Scour Caused by Wall Jets 205
[Link]

Liu and Garcia (2008) studied turbulent wall jet scour in which bed evolution was modeled by
solving the mass balance equation of the sediment. The free surface was modeled by the VOF
method while the scour process was modeled by the moving mesh method. The modeling
effort yielded a good agreement with measurements (velocity field, the maximum scour depths
and local scour profile). Further research is needed to investigate the effect of the turbulence
model for free surface waves (especially for near breaking and breaking waves) and to study
the possibility of using an Eulerian approach for morphological modeling.

Abdelaziz et al. (2010) developed a bed load sediment transport module and integrated into
FLOW-3D. This model was tested and validated by simulations for turbulent wall jet scour in
an open channel flume. Effects of bed slope and material sliding were also taken into account.
The hydrodynamic module was based on the solution of the three-dimensional Navier-Stokes
equations, the continuity equation and k − ε turbulence closure scheme. The rough logarithmic
law of the wall equation was iterated in order to compute shear velocity u*. The predicted local
scour profile fit well with the experimental data, however the maximum scour depth was
slightly under estimated and the slope downstream of the deposition dune was over estimated.

13. Scope for future research

The review of literature and our current understanding of scour by jets indicate that the
following need to be considered in the future:

i. Extend studies to include cohesive soils.

ii. Extend studies to higher range of densimetric Froude number (Fo > 10).

iii. Non-uniform sand beds need to be studied to determine the effect of different
gradations.

iv. To study the role of fluid structures with varying submergence, expansion ratio and
nozzle size to grain size ratio.

v. Developing analytical equations for three-dimensional scour volume.

vi. To understand the effect of the removal of the mound at the asymptotic state, on scour
depth, and 3-D scour hole development with time.

vii. To investigate separation, recirculation and eddies during digging and refilling
phases at various submergence ratios.

viii. Obtain instantaneous velocity fields during scour hole development to enhance
modeling efforts

ix. To study jet scour together with sediment transport by considering continuous
sediment influx through the jet and in the ambient flow.
206 Sediment Transport Processes and Their Modelling Applications

Author details

Ram Balachandar and H. Prashanth Reddy

Department of Civil and Environmental Engineering, University of Windsor, Canada

References

[1] Abdelaziz, S, Bui, M. D, & Rutschmann, P. Numerical simulation of scour develop‐

ment due to submerged horizontal jet", River Flow (2010). Dittrich, Koll, Aberle and
Geisenhainer (eds), 978-3-93923-000-7, 2010.

[2] Adduce, C. La Rocca, M. ((2006). Local scouring due to turbulent water jets down‐
stream of a trapezoidal drop: Laboratory experiments and stability analysis. Water
Resour. Res. 42(2), W, 02405, 1-12.

[3] Adduce, C, & Sciortino, G. (2006). Scour due to a horizontal turbulent jet: Numerical
and experimental investigation. [Link] Res. 44(5), 663-673.

[4] Ade, F, & Rajaratnam, N. (1998). Generalized study of erosion by circular horizontal
turbulent jets, J. Hydraulic Res. 36(4), 613-635.

[5] Aderibigbe, O, & Rajaratnam, N. (1998). Effect of sediment gradation of scour by

plane turbulent wall jets.” J. Hydraul. Eng. 10.1061/
(ASCE)0733-9429(1998)124:10(1034), 124 (10), 1034-1042.

[6] Aderibigbe, O, & Rajaratnam, N. (1996). Erosion of loose beds by submerged circular
impinging vertical turbulent jets, J. Hydraulic Res, 34(1), 19-33.

[7] Ali, K. H. M, & Lim, S. Y. (1986). Local scour caused by submerged wall jets, Proc. Ins
Civil Engineers. 81(2), 607-645.

[8] Ankamuthu, S, Balachandar, R, & Wood, H. (1999). Computational Steroscopy for

Three-dimensional Scour Depth Measurement in Channels. Can. J. Civ. Eng., , 26,
698-712.

[9] Balachandar, R, & Kells, J. A. (1997). Local channel scour in uniformly graded sedi‐
ments: The time-scale problem.” Can. J. Civ. Eng. 10.1139/cjce-24-5-799, 24 (5),
799-807.

[10] Balachandar, R, & Kells, J. A. (1998). Instantaneous water surface and bed scour pro‐
files using video image analysis.” Can. J. Civ. Eng. 10.1139/cjce-25-4-662, 25 (4),
662-667.

[11] Balachandar, R, Kells, J. A, & Thiessen, R. J. (2000). The effect of tailwater depth on
the dynamics of local scour.” Can. J. Civ. Eng., 27 (1), 138-150.
 Scour Caused by Wall Jets 207
[Link]

[12] Bey, A, Faruque, M. A. A, & Balachandar, R. (2007). Two Dimensional Scour Hole
Problem: Role of Fluid Structures”. J. Hydraul. Engng. 133(4), 414-430.

[13] Bey, A, Faruque, M. A. A, & Balachandar, R. (2008). Effects of varying submergence

and channel width on local scour by plane turbulent wall jets”. [Link]. Res. 46(6),
764-776.

[14] Boroomand, M, Neyshabouri, R, & Aghajanloo, S. , A. , S. K., ((2007). Numerical sim‐

ulation of sediment transport and scouring by an offset jet", Can. J. Civ. Eng., , 34

[15] Breusers, H. N. C, & Raudkivi, A. J. (1991). Scouring, Hydraulic Structure Design

Manual. A. A. Balkema, Rotterdam. The Netherlands.

[16] Chatterjee, S. S, Ghosh, S. N, & Chatterjee, M. (1994). Local scour due to submerged
horizontal jet.” J. Hydraul. Eng., 120 (8), 973-992.

[17] Chiew, Y. M, & Lim, S. Y. (1996). Local Scour by a Deeply Submerged Horizontal Cir‐
cular Jet”. J. Hydraul. Engng 122(9), 529-532.

[18] Day, R. A, Liriano, S, & White, R. W. (2001). Effect of tailwater depth and model scale
on scour at culvert outlets, Proceedings of the Institution of Civil Engineers Water
and Maritime Engineering, 148(3), 189-198.

[19] Deshpande, N. P, Balachandar, R, & Mazurek, K. A. (2007). Effects of submergence

and test startup conditions on local scour by plane turbulent wall jets Journal of Hy‐
draulic Research, 45 (3) (2007), , 370-387.

[20] Dey, S, & Sarkar, A. (2006a). Scour downstream of an apron due to submerged hori‐
zontal jets.” J. Hydraul. Eng., 132(3), 246-257.

[21] Dey, S, & Sarkar, A. (2006b). Scour response of velocity and turbulence in submerged
wall jets to abrupt change from smooth to rough beds and its application to scour
downstream of an apron.” J. Fluid Mech., , 556, 387-419.

[22] Ettema, R. (2002). Review of alluvial-channel responses to river ice.” J. Cold Reg.
Eng., 16(4), 191-217.

[23] Faruque, M. A. A, Sarathi, P, & Balachandar, R. (2006). Clear Water Local Scour by
Submerged Three-Dimensional Wall Jets: Effect of Tailwater Depth,” J. Hydraul.
Engng. 132(6), 575-580.

[24] FaruqueMd. ((2004). Transient local scour by submerged three dimensional wall jets,
effect of the tailwater depth. Msc thesis, University of Windsor.

[25] Ghodsian, M, Melville, B, & Tajkarimi, D. (2006). Local scour due to free overfall jet,
Ins Civil Engineers, Water Management. 159(4), 253-260.

[26] Hoffmans, G. J. C. M. (1998). Jet scour in equilibrium phase", Journal of Hydraulic

Engineering, , 124(4), 430-437.
208 Sediment Transport Processes and Their Modelling Applications

[27] Hoffmans, G. J. C. M, & Pilarczyk, K. W. (1995). Local scour downstream of hydraul‐

ic structures", Journal ofHydraulic Engineering, , 121(4), 326-340.

[28] Hogg, A. J, Huppert, H. E, & Dade, W. B. (1997). Erosion by planar turbulent wall
jets.” J. Fluid Mech., , 338, 317-340.

[29] Hopfinger, E. J, Kurniawan, A, Graf, W. H, & Lemmin, U. (2004). Sediment erosion

by Götler vortices: The scour problem.” J. Fluid Mech., , 520, 327-342.

[30] Johnston, A. J. (1990). Scourhole Developments in Shallow Tailwater”. J. Hydraul.

Res., IJHR 28(3), 341-354.

[31] Karim, O. A, & Ali, K. H. M. (2000). Prediction of flow patterns in local scour holes
caused by turbulent water jets. J. Hydraul. Res. , 38, 279-287.

[32] Kells, J. A, Balachandar, R, & Hagel, K. P. (2001). Effect of grain size on local channel
scour below a sluice gate.” Can. J. Civ. Eng. 10.1139/cjce-28-3-440, 28 (3), 440-451.

[33] Li, F, & Cheng, L. (1999). A numerical model for local scour under offshore pipelines
J. Hydraul. Eng., 125 (4), , 400-406.

[34] Lim, S. Y. (1995). Scour below unsubmerged full flowing culvert outlets. Proc. Ins
Civil Engineers, , 112, 136-149.

[35] Liu, X, & Garcia, M. H. (2008). Three-dimensional numerical model with free water
surface and mesh deformation for local sediment scour. Journal Waterway, Harbour,
Coastal and Ocean Engineering, ASCE 134 (4), 203-217.

[36] Mazurek, K. A, & Rajaratnam, N. (2002). Erosion of a polystyrene bed by obliquely

impinging circular turbulent air jets, J. Hydraulic Res. 40(6), 709-716.

[37] Mehraein, M, & Ghodsian, M. and Salehi Neyshaboury, S. A.A ((2010). Local scour
due to an upwards inclined circular wall jet, Proc. Ins Civil Engineers, , 164, 111-122.

[38] Melville, B. W, & Chiew, Y-M. (1999). Time scale for local scour at bridge piers. ASCE
Journal of Hydraulic Engineering, , 125(1), 59-65.

[39] Meulen, V, Vinje, T, & Three-dimensional, J. , J. local scour in noncohesive sedi‐

ments", Proc., 161th International Association for Hydraulic Research (IAHR)-
Congr., San Paulo, Brazil, , 263-270.

[40] Mih, W. C, & Kabir, J. (1983). Impingement of Water Jets on Non uniform S tream‐
bed, J. Hydraulic Eng. 109 (4), 536-548.

[41] Mohamed, M. S, & Mccorquodale, J. A. (1992). Short-term local scour", Journal of Hy‐
draulic Research, , 30(5), 685-699.

[42] Neyshabouri, S. A. A., Ferreira Da Silva, A. M. and Barron, R. (2003). Numerical Sim‐
ulation of Scour by a Free Falling Jet. J. Hydraul. Res., , 41(5), 533-539.
 Scour Caused by Wall Jets 209
[Link]

[43] Neyshabouri, S. A. A, & Barron, R. Ferreira da Silva, A.M. ((2001). Numerical Simula‐
tion of Scour by a Wall Jet”. Water Engng Res. 2(4), 179-185.

[44] Pagliara, S, Hager, W. H, & Minor, H. E. ((2006). Hydraulics of plane plunge pool
scour. J. Hydr. Engng. 132(5), 450-461.

[45] Rajaratnam, N. (1981). Erosion by plane turbulent jets, J. Hydraulic Res. 19(4),
339-358.

[46] Rajaratnam, N, & Berry, B. (1977). Erosion by circular turbulent wall jets, J. Hydraulic
Res. 15(3), 277-289.

[47] Rajaratnam, N, & Mazurek, K. A. (2003). Erosion of sand by circular impinging water
jets with small tailwater, J. Hydraulic Eng. 129(3), 225-229.

[48] Rajaratnam, N, Aderibigbe, O, & Pochylko, D. (1995). Erosion of sand beds by obli‐
que plane water jets.” Proc. Inst. Civ. Eng., Waters. Maritime Energ., 112(1), 31-38.

[49] Rajaratnam, N, & Diebel, M. (1981). Erosion below Culvert-Like Structure”. Proceed‐
ing of the 5th Canadian Hydrotechnical Conference, 26-27 May, CSCE, , 469-484.

[50] Rajaratnam, N, & Macdougall, R. K. (1983). Erosion by Plane Wall Jets with Mini‐
mum Tailwater”. J. Hydrual. Engng., 109(7), 1061-1064.

[51] Sarathi, P, Faruque, M. A. A, & Balachandar, R. (2008). Influence of tailwater depth,

sediment size and densimetric Froude number on scour by submerged square wall
jets, J. Hydraulic Res. 46(2), 158-175.

[52] Sui, J, Faruque, M. A. A, & Balachandar, R. (2009). Local scour caused by submerged
square jets under model ice cover, J. Hydraulic Eng., 135(4), 316-319.

[53] Sui, J, Faruque, M. A. A, & Balachandar, R. (2008). Influence of channel width and
tailwater depth on local scour caused by square jets, J. Hydro-environment Res, 2(1),
39-45.

[54] Van Rijn ((1987). Mathematical Modeling of Morphological Processes in the Case of
Suspended sediment transport. PhD. Dissertation, Delft University of Technology,
The Netherlands.

[55] Wu, S, & Rajaratnam, N. (1995). Free jumps, submerged jumps, and wall jets.” J. Hy‐
draul. Res., 33(2), 197-212.
 Section 3

Numerical Modelling of Sediment Transport

 Chapter 9

Modelling Cohesive Sediment Dynamics in the Marine

Environment

Katerina Kombiadou and Yannis N. Krestenitis

Additional information is available at the end of the chapter

[Link]

1. Introduction

The inflow of fine sediments in the marine environment affects various processes and has
significant socioeconomic and environmental impacts, the most straight-forward of which
is related to sedimentation in deltaic and coastal areas. However, fine sediments have the
ability to affect solar radiation in the water column and to absorb contaminants on their
surface [1,2]. Therefore, fine sedimentary plumes also affect primary productivity and the
distribution of pollutants in the column and the seabed. The distinction between fine (<63µm)
and coarse-grained (>63µm) sediments is rather general and varies with the type of the sedi‐
mentary matter, but it is a fact that the dominance of inter-particle cohesion forces over grav‐
itational forces varies inversely with the floc diameter. Thus, the effect of cohesion is much
more pronounced on the behaviour of clays (diameter <2µm) than of silts (2-63µm); in fact,
the development of cohesion in silty clays is mainly due to the clay fraction in the sedi‐
ment [1]. Adhesion diversifies the behaviour of cohesive and sandy sediments in the aquat‐
ic domain, adding further complexities to modelling efforts for fine sedimentary plumes.
Their movement is the result of the combined action of various physical processes and
forces (Figure 1).

The processes that control sediment transport in coastal microtidal areas can be divided in
the ones that take place in the water column (pelagic processes):

• advection

• dispersion

• coagulation – deflocculation
214 Sediment Transport Processes and Their Modelling Applications

• settling

• effects of seawater stratification

and the ones that take place at the boundary of the seabed (benthic processes):

• deposition

• erosion

• self-weight consolidation – resuspension

Figure 1. The main processes that control sediment transport in coastal, deltaic areas.

These processes are analyzed in the following sections in terms of physical interpretation
and mathematical parameterization, along with the three basic modelling approaches used
for the description of sediments in the aquatic domain. Finally, results from a particle-track‐
ing model applied in a micro-tidal shelf sea are presented and discussed.

2. Sediment transport models

Models describing transport and dispersion of sediments in the aquatic domain can be div‐
ided in Eulerian (e.g. [2]), that describe the evolution of suspended matter concentration us‐
ing finite differences and the mass conservation principle, Lagrangian (e.g. [3]) in which the
advection and dispersion of a specific amount of sediment mass is being traced with compu‐
tational time and in mixed Euler-Lagrange models (e.g. [4]), in which transport is expressed
by particles and dispersion through finite differences. The differences between these model‐
ling approaches and their advantages and disadvantages are given further down.

2.1. Eulerian models

The evolution of sediment concentration in an Eulerian Model (hereafter EM), assuming

mass conservation and incompressible flow, is written:
 Modelling Cohesive Sediment Dynamics in the Marine Environment 215
[Link]

¶c ¶cu ¶cv ¶c (w + w s ) ¶ æ ¶c ö ¶ æ ¶c ö ¶ æ ¶c ö
¶t
+
¶x
+
¶y
+
¶z
=
¶x çè
KH + ç KH
¶x ÷ø ¶y è
÷+
¶y ø ¶z çè
KV
¶z ÷ø
+S -L (1)

In the equation, c is the mass concentration of suspended sediments, u, v and w are the fluid
velocities along the x, y and z directions, ws the sediment settling velocity, KH and KV the tur‐
bulent dispersion coefficients in the horizontal and vertical direction and S and L sediment
source and loss terms, where applicable. The last three terms of the left-hand side express mat‐
ter advection, while the first three terms of the right-hand side express matter dispersion. Re‐
garding the source and loss terms, these can be mass exchanges in the bottom boundary, as
well as sediment inflow (for example due to riverine discharges). Considering the free sur‐
face as the reference for the vertical axis (z=0), the bottom boundary condition reads:

¶c
KH
¶z
= Rer - Rdep (2)
z =- H

where Rer και Rdep are the erosion or resuspension and deposition rates, respectively.

2.2. Lagrangian models

Largangian (or particle-tracking) Models (hereafter: LM) are stochastic models that describe
suspended matter in the column not as mass concentration in each of the cells of the compu‐
tational grid, but as passive particles of specific mass that are being advected and dispersed
by the flow. Thus the movement of a particle along direction i (i=1, 2 and 3 for directions x, y
and z, respectively) is given by formulas of the form:

dxi 6K i
= ui + ui ¢ , ui ¢ = x , x Î [ -1, +1] (3)
dt dt

In equation 3, the first term expresses the advective part of the particle movement, whose
range is defined deterministically by the velocity field, and the second the dispersive part that
is defined stochastically using a random number ξ that ranges from -1 to +1; the amplitude of
the stochastic part of the motion is determined by the corresponding mass dispersion coeffi‐
cient Κi. It is noted that for the case of the vertical transport (i=3), the deterministic particle ve‐
locity is the summation of the vertical flow velocity and the settling velocity of the particle:

u3 = w + w s (4)

In a tracer model particles in the domain are traced with simulation time and bear informa‐
tion (like the mass or the characteristic diameter of the particle, its origin etc) that follows the
particle throughout its movement, remaining constant or varying with time. Each particle ac‐
counts for a specified mass of sediment and, thus, a computational particle represents a cumu‐
lus of actual sedimentary particles that are considered to have the same properties and fate.
216 Sediment Transport Processes and Their Modelling Applications

2.3. Mixed Euler-Lagrange models

Euler-Lagrange Models (hereafter: ELM) are derived from the classic Eulerian advection-
dispersion equation, which, applying a simplified temporal discretization is written [4-6]:

¶c c n - c n -1
+ [u Ñ c ] = [ Ñ K Ñ c ]
n -1
(5)
n
+ u Ñc = ÑK Ñc Þ
¶t Dt

Considering an arbitrary, auxiliary value, cf, the former equation can be divided into two
parts. One that describes advection:

c f - c n -1
+ [u Ñ c ] = 0
n -1
(6)
Dt

and one expressing dispersion:

cn - cf
= [ ÑK Ñc ] (7)
n
Dt

The advection equation states that concentration remains constant along lines of equal con‐
centration that are defined by:

dxi i =1,2,3
= dt × ui (8)

Thus, these lines are being ‘traced’ backwards in time for each note of the computational
grid and the values for the temporal point n-1 are calculated for the corresponding ones of
step n. The cf concentrations for step n are determined using spatial interpolation. Finally,
the dispersion equation is solved using finite differences or elements, through which the val‐
ues of the concentration are updated for the next temporal step.

2.4. Common features and differences of the 3 models

One of the main disadvantages of the EM is that the advection-dispersion equation is solved
in all of the grid points of the domain, which, combined with the necessity to adhere to the
Courant-Friedrichs-Lewy (CFL) condition, increases the computational load. Contrastingly,
a LM performs the necessary calculations only in the locations of particles while an ELM sim‐
ulates the processes in a decoupled manner. It should be noted that, due to the stochastic na‐
ture of dispersion, simulating the same conditions with a random walk model can produce
different results regarding the suspended matter in the domain. This effect is lower, the more
the particles taken into account by the simulation. However, the number of particles in the
domain is a parameter that should be taken into consideration, since increasing the number
of particles increases the accuracy of the simulation but also increases computational load.
Therefore, it is important to optimize the number of particles selected so that the simula‐
 Modelling Cohesive Sediment Dynamics in the Marine Environment 217
[Link]

tion is accurate and also realizable within reasonable computational time. Even though ELMs
present numerical advantages, their most significant disadvantage is that they do not inher‐
ently conserve mass, either locally or globally [6] due to tracking errors and non-conserva‐
tive flow fields; these errors can be eliminated using very fine grids. Using concentrations, in
both EM and ELM, provides erroneous results in the location of the sources and does not ac‐
count for dispersion phenomena in a particular point, unlike LM [5] that describe the proc‐
esses at the level of the cohesive floc. Furthermore, the resolution of a model that uses
concentration to describe suspended mass in the domain is, unavoidably, defined by the dis‐
cretization of the grid. Contrastingly, a particle-tracking model can describe the particulate
matter distribution in scales much smaller than that of the spatial step. The transition from
particle distribution to concentrations is easily performed, when and where this is required,
from the mass of each particle and the number of particles in each grid cell. Finally, one of
the most important and interesting advantages in using LM is the ability to assign informa‐
tion to particles (like origin, kind, state, values of characteristic parameters etc) in the form
of indicators and to trace their evolution with time. Namely, knowing the ‘history’ of each
sedimentary parcel allows the visualization the trajectory of each particle and of the changes
it has undergone, as well as the segregation of particles according to common properties and
the investigation of transport patterns related to sedimentary origin.

3. Physical processes for cohesive sediments in the marine environment

3.1. Pelagic processes

The pelagic processes are the ones that take place in the water column and can generally be
described as advection and dispersion. However, flocculation and the corresponding
changes in the particles’ settling rate and stratification affect advective and dispersive matter
transport in the column; these processes are described following.

3.1.1. Turbulent dispersion

It is important to clarify the distinction between diffusion and dispersion of suspensions.

Mass diffusion can be molecular in the case of laminar flow, which is the case for example in
the stochastic Brownian motion of particles, or turbulent, in the case that the flow in the field
presents turbulent eddies. For the movement of particles to be considered diffusive, the exis‐
tence of a concentration gradient in suspended matter is required. In the opposite case,
when the concentration in the field is homogenous, dispersion theoretically has no effect
since the lateral expansion of the plume takes place in the same rate in both directions. The
mass diffusion coefficient ranges from 10-9m2/s to 102m2/s for molecular and turbulent diffu‐
sion, respectively. Contrastingly, the case of mass dispersion requires mixing of the flow
field due to velocity gradients and the corresponding coefficient ranges from 10-3m2/s to
102m2/s. In coastal and marine systems the seawater velocities are variable with depth and
along the direction of the flow and, therefore, turbulent dispersion is the main driving
mechanism. The scale of turbulent eddies is important to the evolution of a suspended sedi‐
mentary plume. In cases that the eddy is much larger than the plume, the movement is es‐
sentially advective and the shape of the plume remains unchanged, whereas for
intermediate eddies the plume is deformed and its boundaries are stretched; when turbulent
218 Sediment Transport Processes and Their Modelling Applications

scales are much smaller than the plume itself, the main effect is dispersive and leads to eve‐
ning of the shape of the plume [7].

Richardson [8] studied the evolution of sedimentary plumes from a point, non-constant
source considering the segregation between the particles of the plume. Defining standard
deviation of particles from their average position as the characteristic length-scale l, he
reached a correlation with the dispersion coefficient of the form:

K = 0.2 × l
4
3 (9)

where l is the length scale of the system. This equation is known as the ‘four-thirds power
law’. One of the most widely applied relationships for the calculation of the turbulent dis‐
persion coefficient from the hydrodynamic conditions is the Smagorinski formula [9,10]:

2 2 2
æ ¶u ö 1 æ ¶v ¶u ö æ ¶v ö
K H = C × dx × dy ç ÷ + ç + ÷ +ç ÷ (10)
è ¶x ø 2 è ¶x ¶y ø è ¶y ø

dz 2
KV = KH (11)
dx × dy

C, in equation 10, is a constant that depends on the horizontal discretization step and varies
from ~0.01 for steps of a few meters to 0.2 for steps of the order of kilometres, while the sub‐
scripts H and V (eq. 10-11) denote horizontal and vertical directions, respectively.

The dispersion coefficients can also be estimated from the corresponding values of turbulent
viscosity of seawater (vti) using relationships of the form [11]:

K i = b i Fmi v ti (12)

The introduction of coefficients βi and Fmi is due to the separation of seawater and sediment
mixing and to the reduction of vertical mixing, respectively. They are usually taken equal to
one, apart from the case of stratified environments, where the coefficient Fmi is estimated as a
damping function of the vertical transport; the effect of stratification is analyzed further down.

3.1.2. Flocculation - Settling

A basic distinction of fine-grained from sandy sediments is the development of cohesion

that enables the formation of bonds between particles and eventually the formulation of co‐
agulants of larger diameter than their primary constituents. These bonds may be due to at‐
tractive van der Waals or electrochemical forces. Van der Waals forces are particularly
strong but weaken significantly with distance, which means that they are effective only in
the case that the particles are at very low distances. Suspended oblate argillaceous particles
 Modelling Cohesive Sediment Dynamics in the Marine Environment 219
[Link]

are usually negatively charged in their surface and bear positive charge in the edges; the
overall electrostatic charge in the surface is negative and therefore repulsive forces exist be‐
tween particles. Free ions and cations in the marine environment form a cloud in the surface
of clay particles, known as (Guoy) double layer [12], that balances the electrostatic charge of
the particles and allows them to come closer together, collide and form bonds.
The flocculation process is mainly governed by three mechanisms [13-16]:

1. Random Brownian movement causes collisions between particles and their attachment
to a larger coagulant.
2. Turbulence in flows causes particles to collide and aggregate, on one hand, while turbu‐
lent shear can induce the breakup of an aggregate, when the shear overcomes the shear
strength of the bond between the primary particles.
3. Larger particles settling through a suspension can drift smaller particles, which attach
to them, thus forming larger aggregates (differential settling).

It should be noted that the contribution of Brownian motion to flocculation is significant on‐
ly for small particle diameters (~1-2µm) and results to aggregates of smaller strength, com‐
pared to the other two mechanisms [13]. As the diameter of the particle increases, shear
becomes the main criterion for the evolution of the process [17].
Other parameters that affect the formation of clayey aggregates, favouring the collision of
particles are [18]:

1. Salinity; it has been established that transition to a higher salinity environment leads to
increase of the diameter of cohesive suspended matter. The critical salinity value for co‐
agulation, S 0, related to the clay minerals contained the clay fraction are 0.6, 1 and 2.4
for kaolinite, illite and montmorillonite, respectively [16].
2. The existence of cations increases the attractive forces between particles and, therefore,
the aggregate’s strength.
3. The increase of temperature.
4. The existence of organic matter and detritus, bacteria, microphytes and macrophytes. It
is essentially a form of biostabilization of the aggregate’s structure by extracellular se‐
cretions of the microorganisms.
5. The increase of suspended matter concentration, which increases the frequency of colli‐
sions between particles and, therefore, the possibility of adhesion to a larger coagulate.

Dyer [19] established the theoretical background for the evolution of aggregate diameter un‐
der the effects of flow shear and suspended concentrations. Concentration increases the di‐
ameter of the floc, due to higher availability of matter, while shear stress increases
flocculation at low values, favouring inter-particle collisions, and causes floc-breakup and
corresponding decrease of the particle diameter for values higher than the strength of the
aggregate. Mathematical expressions that can describe this behaviour are exponential for‐
mulas with Munk-Anderson type damping functions, of the form [20]:
220 Sediment Transport Processes and Their Modelling Applications

1 + b1G
Dag = a1 c d2 + a2 (13)
1 + b2G d1

The constant parameters of equation 13 depend on the concentration of the suspension, c.

For high concentrations (c=3.7g/l) these constants have been experimentally estimated at
a1=0.08, a2=0.86, b1=29, b2=125, d1=4.6 and d2=-1.1, while for lower concentrations (c=0.8g/l) the
same constants read a1=0.06, a2=0.88, b1=33, b2=138, d1=3.8 and d2=-3.5 [21]. The turbulent dissi‐
pation parameter G expresses the effects of shear to the diameter evolution and depends on
the energy dissipation rate ε and kinematic viscosity v: G = ε / ν

A method widely applied for the parameterization of flocculation is based on the assign‐
ment of fractalic dimensions to the cohesive flocs, considering that every aggregate consists
of self-similar fractalic entities. Accepting that the linear dimension of the characteristic par‐
ticle is unitary, it follows that the particle volume is proportional to the number of primary
particles. Thus, if nf is its fractalic dimension, N the number of primary particles that form
the aggregate and L its linear dimension, it follows that [15]:

ln éëN ( L ) ùû
nf = lim
L ®¥ ln [L ]
(14)

Thus, the number of particles in an aggregate is proportional to the diameter of the primary
particle, Dp:

N ~ Dp nf (15)

The fractalic dimension for cohesive particles ranges from 1.4 for very fragile flocs, like ma‐
rine snow, to 2.6-2.8 for dense, consolidated beds. It is generally accepted that the fractalic
dimension increases from marine snow (1.4-1.8) to estuarine sediments (1.7-2.2) and to oce‐
anic sediments (1.94-2.14) [22,23]. An average fractalic dimension for estuarine and coastal
areas is nf=2 [15].

Applying the fractal theory to the evolution of cohesive aggregate diameter it follows that
there is an analogy between the ratio of aggregate and primary particle dimensions and the
corresponding volume concentration, ϕ [24]:

1
Dag æ f ö 3 - nf
=ç ÷ (16)
Dp ç fp ÷
è ø

The relationship between volume concentration and aggregate mass is written [15]:
 Modelling Cohesive Sediment Dynamics in the Marine Environment 221
[Link]

rs - rw c
f= = f ND 3 (17)
rf - rw rs s ag

ρs, ρw and ρf are solid phase, water and aggregate densities and fs is a shape constant which,
for the case of spherical particles equals to π/6. The increase in aggregate diameter also in‐
volves decrease in its density, ρf, compared to the primary particles, ρp, due to increase in
interstitial water [12-14,16,17]. Considering the density difference of solid and liquid phase,
the density of the aggregate can be calculated by [24]:

3 - nf
æD ö
rf = rw + ( rs - rw ) ç p ÷ (18)
ç Dag ÷
è ø

Based on the fractal theory, the evolution of the aggregate diameter in turbulent flow is [15]:

dDag
( ) (19)
p
= k AcGDag 4 - nf - kBG q +1 Dag - Dp Dag 2q +1
dt

The first term of the right hand-side of equation 19 expresses the effect of coagulation and
the second the effect of floc break-up. The corresponding dimensional flocculation and de‐
flocculation parameters (kA [m2/kg] and kB [sec1/2/m2]) are defined as:

q
Dp nf -3 Dp - p æ m ö
k A = k A¢ kB = kB¢ ç ÷ (20)
nf fs rs nf çè Fy ÷
ø

In equations 18-20, c is the suspended sediment concentration, Dp the diameter of the pri‐
mary flocs, q and p are constants (q~0.5 and p=3-nf), μ is the dynamic viscosity of water, Fy is
the strength of the sediment (Fy=O{10-10}) and kA’ and kB’ are non-dimensional flocculation
and deflocculation coefficients (kA’=O{10-1} and kΒ’=O{10-5}) [15].
The corresponding changes to the settling velocity of the aggregate can be calculated by the Sto-
kes velocity, taking into account shape factors α and β (α=β=1 for perfectly spherical particles):

a ( rf - rw ) g
ws = Dag 2 (21)
b 18 m

The shape factors can be simplified considering an irregularity parameter for the flocs, airr,
(airr≤1) equal to the ratio of minimum to maximum aggregate dimensions. Following [25]
and after some manipulations:
222 Sediment Transport Processes and Their Modelling Applications

ïìa = airr ïü
í ý (22)
îï b = 1 - l1 (1 - airr ) þï

where λ1 is a coefficient related to the orientation of the particle during settling. It equals 0.2
or 0.4 when the particle settles with its maximum axis parallel or perpendicular to depth.
Considering parallel orientation as the most frequent one, the settling velocity reads:

airr ( rf - rw ) g D 2
ws =
0.8 + 0.2airr 18 m
ag (23)

The relation between settling velocity, flow shear stress and suspended sediment concentra‐
tion (c in mg/l) for microflocs (<160µm) and macroflocs (>160µm), defined after experimen‐
tal measurements from estuarine areas is [26-29]:

ì0.644 + 0.000471c + 9.36t - 13.1t 2, 0.04Pa £ t £ 0.65Pa

ïï
w sma cro = í3.960 + 0.000346c - 4.38t + 1.33t 2, 0.65Pa < t £ 1.45Pa (24)
ï
ïî1.18 + 0.000302c - 0.49t + 0.057t , 1.45Pa < t £ 5.00Pa
2

ïì0.244 + 3.25t - 3.71t , 0.04Pa £ t £ 0.52Pa

2
w sm icro = í (25)
ïî0.65t 0.52Pa < t £ 10.0Pa
0.541
,

The predominance of macroflocs versus microflocs can be defined from their concentration
quotient, which was found to depend on the overall suspended sediment concentration
through a relationship of the form [26]:

cmacro
= 0.815 + 0.00318c - 0.14 × 10-6 c 2 (26)
cmicro

The settling velocity of particles can be distinguished in three ‘phases’, depending on the
suspended matter concentration [14]. For low concentrations, C1=0.1-0.3g/lt, free (Stokes’)
settling dominates. For concentrations higher than the limit for free settling and lower than
C2=2-20g/lt settling is highly affected by flocculation and involves increase of the settling ve‐
locity due to increase of the particle diameter. The settling velocity can be thus expressed as
an exponential function of concentration:

w s = k1c n1 , c1 £ c £ c2 (27)

where k1 (~4/3) and n1 Ο{10-1} are experimentally determined constants. Taking into account
the effect of shear stress, the settling velocity can be determined by the equation [30]:
 Modelling Cohesive Sediment Dynamics in the Marine Environment 223
[Link]

1 + aG
w s = w s,r (28)
1 + bG 2

In equation 28, ws,r is the settling velocity of a single particle in still liquid and a and b (O{1})
are empirical coefficients related to the sediments.

For concentrations higher than C2 the settling velocities reduce with increasing suspended
matter, phenomenon known as hindered settling. Hindered settling essentially expresses the
effect of neighbouring particles to the settling velocity of a specific particle, which is impor‐
tant especially for high concentration suspensions. This hindering effect is caused by eddy
separation around settling particles, causing reduction to the settling rates of particles with‐
in this buoyant flow, but also due to inter-particle collisions and electrochemical interac‐
tions. The hindered settling velocity related to the volumetric concentration is:

w s = w s,r (1 - f ) (29)
n

n is a parameter that varies inversely with the particle Reynolds number within the range
2.5<n<5.5. A corresponding expression [15] reads:

(1 - f* )m (1 - fp )
w s = w s,r , f* = min {1,f} (30)
1 + 2.5f

where ϕp is the volumetric concentration of primary particles and m is a constant for the in‐
clusion of possible non-linear effects. The expression that relates the mass concentration in
the field to the hindered settling velocity is similar [14]:

w s = w so ëé1 - k2 ( c - C2 ) ûù (31)
n2

For highly concentrated suspensions (~80-100g/lt) the settling velocity is practically zero.

A general formulation that can describe the evolution in conditions of flocculation and hin‐
dered settling is [14]:

a × cn
ws =
(32)
(c )
m
2
+ b2

The coefficients are determined experimentally; from experimental studies in various es‐
tuarine and riverine areas [31] the range of their values is: a=0.01-0.23, b=1.3-25.0, n=0.4-2.8
224 Sediment Transport Processes and Their Modelling Applications

and m=1.0-2.8. The maximum settling velocity in this case (for concentration equal to C2) is
[14]:

m - 2n
æ 2m ö
ç n - 1÷
è ø b
w so = a × b n - 2m , C2 = (33)
2m 2m
-1
n n

3.1.3. Effects of stratification

One of the most widely applied methods for the determination of the stability of the water
column is the use of the gradient Richardson number [32]:

N2 1 g ¶rw
Ri = > , N= -
æ ¶U ö
2
4 rw ¶z (34)
ç ¶z ÷
è ø

Ν is the Brunt-Väisälä frequency. In flows with Ri values lower that ¼ Kelvin-Helmholtz in‐
stabilities can occur. Based on this criterion, the effect of stratification to mass dispersion,
based on the mixing-length theory, is often expressed using Munk-Anderson type velocity
and mass dispersion damping functions (Fm) [7]:

a
Fm = (1 + b Ri ) {a ; -0.5, b ; 10} (35)

The flux Richardson number, which is the ratio of the buoyant forces to the turbulent kinetic
energy production [7], is also frequently used; the mass dispersion coefficient Km is [33]:

Rf e Rf ³ Rf ,crit ;0.15 e
Km =
1 - Rf N 2
¾¾¾¾¾¾¾ ® K m £ 0.2 (36)
N2

The overbar in equation 36 indicates time averaging. Regarding the mixing of freshwater
plume, expressed as entrainment velocity to the underlying layer (We), it can be estimated as
[34]:

e1Uo
We =
Ri
(37)

where Uo is the velocity of the underlying layer and e1 (O{10-3}) an empirical coefficient.
 Modelling Cohesive Sediment Dynamics in the Marine Environment 225
[Link]

Another indicator for the strength of the stratification is the Peclet number (Pe) that express‐
es the predominance of settling over dispersion (h: width of the mixing zone):

wsh
Pe = =1 (38)
Km

It has recently been established that, even in the case that density increases with depth, the
hydrostatic stability of the system is not guaranteed [35], since inhomogeneities in the densi‐
ty field can lead to transporting water masses that can induce a process of matter removal
from suspension. The prognosis of such instabilities cannot be described using a Richardson
number [33]. Double diffusion is the development of instable density gradients due to dif‐
ferential molecular diffusion of the properties of the fluid, which are capable of changing its
density. Given that the thermal diffusion coefficient (KT≈10-3m2/s) is two orders of magni‐
tude larger than the salinity coefficient (KS≈10-5m2/s), it follows that temperature diffuses
faster; this can lead to areas in the flow with higher density than the underlying fluid, thus
causing mixing of the water column. Double diffusive instabilities can form as salt fingers,
in the case that salinity is the destabilizing parameter (T and S decreasing with depth), or as
diffusive layers, when temperature is the destabilizing factor (T and S increasing with
depth). Salt fingering instability takes the form of tightly packed blobs of sinking salty and
rising fresher water masses near the thermocline, which quickly develops to transport away
from the interface. Diffusive layering convection involves a buoyant thermal diffusion flux
through the pycnocline that is higher (in terms of density) than the corresponding salinity
flux, thus leading to a downward density flux that causes transport from the stratified zone
to the homogenous layer. The susceptibility of the column to double diffusive phenomena
can be estimated using the stability ratio (Rp), which is the ratio of the stabilizing parameter
gradient to the destabilizing one, expressed in terms of density:

ì ¶T ¶S ì 1 ¶rw ü
a b ® salt fingering ïa = - r ¶T ï
ïï ¶z ¶z ï ï
Rp = í (39)
w
,í ý
ïb ¶S
a
¶T
® diffusive layering ï b = 1 ¶rw ï
îï ¶z ¶z ïî rw ¶S ïþ

In equation 39, α is the thermal expansion coefficient and β is the haline contraction coeffi‐
cient. The gravitational stability of the column increases with the stability ratio; it has been
established that double diffusive instabilities develop within the range 1<Rp<10 [36,37],
while areas where Rp≥10 are highly stable, with low settling rates of fine sedimentary
plumes (~cm/h) [37]. Thus, Rp=10 can be used as a threshold value for the inhibition of dou‐
ble diffusive instabilities and, therefore, for the indication of a highly stable column.
226 Sediment Transport Processes and Their Modelling Applications

These effects can be taken into account in sediment transport modelling, using damping
functions to the vertical advection (Fw) and dispersion (FKv) of matter [38]:

ì1 , Rp < 10
ï
Fw = í (40)
(
ïî 1 + Rp ) , Rp ³ 10

ìï1 , Rp < 10
FKv = í (41)
( )
ïîexp - × Rp , Rp ³ 10

3.2. Benthic processes

Modelling the transport of cohesive sediments in coastal and deltaic areas requires the de‐
scription of the processes of mass exchange between the column and the seabed through the
processes of erosion and deposition. In practice, formulating this module is one of the most
challenging tasks in modelling fine sediment transport, mainly due to the lack of general
and widely applicable formulae between erosion resistance and bed shear stress. However,
even if the main questions regarding mass exchange quantification were answered, the fact
that the characteristic parameters vary within the sediment with time and space (horizontal‐
ly and vertically) still remains [39].
There are three main distinctions regarding fine sediment transport in the boundary of the
bed, compared to coarse-grained sediments [40]:

1. Cohesive matter is transported exclusively in suspension, while coarser matter is also

transported in semi-contact with the bed, as bed load.
2. Cohesive sediments are not transported as dispersed particles. Flocculation increases
settling rates and is responsible their deposition.
3. Cohesive sediment beds undergo self-weight settling and consolidation. In cases of rap‐
id sedimentation the deposited sediments are light and present minimal resistance to
shear stress. Cohesive beds can be homogenous or vertically stratified regarding densi‐
ty and strength.

Various modules have been proposed for the exchange processes between seabed and water
column, which are mainly based on the transition of the matter to different states. Four char‐
acteristic states for a sediment-water mixture are defined in [1]:

1. Horizontally advected mobile suspension.

2. Concentrated benthic suspension (CBS), horizontally static but with vertical mobility.
3. Consolidating soft deposit (fluid mud).
4. Stable, consolidated part of the seabed.
The static suspension is formed by the deposition of the transported suspension, especially
under low hydrodynamic conditions, and presents minimal mechanical strength. As deposi‐
 Modelling Cohesive Sediment Dynamics in the Marine Environment 227
[Link]

tion progresses, a CBS is formed; CBSs are suspensions with strong interactions with the tur‐
bulent flow field through buoyant effects, which however continue to behave as Newtonian
fluids [41]. In time, consolidation and the related physicochemical changes lead to strength
development and, eventually, to the formation of a fluid mud layer with lower interstitial
water, higher shear strength and more stable structure. It must be noted that the transition
from one phase to another is gradual and that a change in the flow conditions may lead to
phenomena comparable to ones that an increase in concentration would produce. Whether
these phenomena take place or not depends on the temporal scales of the processes; for ex‐
ample the time scale of deposition and flocculation define if fluid mud is formed under low
flow conditions and the temporal scale of consolidation determines if re-entrainment or ero‐
sion phenomena partake during accelerated flow. Regarding the concentration/density (dry
and bulk, ρd and ρb) ranges of the aforementioned suspensions and deposits, the following
values are cited [42]:
• Dilute suspension (low concentration C=0.1-10kg/m3)

• CBS (high concentration C=10-100kg/m3)

• Fluid mud (C=ρd=100-250kg/m3, ρb=1050-1150kg/m3)

• Partially consolidated bed (ρd=250-400kg/m3, ρb=1150-1250kg/m3)

• Fully consolidated bed (ρd=400-550kg/m3, ρb=1250-1350kg/m3)

The most straight-forward treatment of the mass-exchange in the boundary of the bed is de‐
rived from the balance between settling, deposition and erosion, through conditions of the
form [39]:

¶c
es
¶z
- w s c = SE - SD (42)

In the former relationship, the right hand-side is the balance between turbulent dispersion
and settling and the right is the balance of erosion and deposition rates.

3.2.1. Bottom shear stress

The shear stress velocity in the bed can be related to the depth-averaged flow velocity (U)
through a Chezy coefficient, Ch [11]:

u* = U g Ch , Ch = 18 log (12h kS ) (43)

where h is the flow depth and ks an equivalent roughness height. The direct dependence if
the shear stress to the flow velocity at a small distance from the bed can also be used [7]:

t b = r × Cd × u(1m )2 (44)
228 Sediment Transport Processes and Their Modelling Applications

In equation 44, u(1m) is the velocity at a distance of 1m from the bed and Cd the bottom fric‐
tion coefficient (Ο{10-3}). This coefficient can be estimated assuming logarithmic velocity
profile through equations of the form [2]:

ì é 1 ln ( H + z ) ù -2 ü
ï ï
Cd = max í ê (45)
b
ú , 0.0025 ý
ïî ë k zo û ïþ

zo is the roughness length of the bed (Ο{10-3m}). High density gradients near the bed result
to reduction of the apparent roughness with corresponding increase in transport for the
same bottom shear stress. Thus, the actual erosion rates for the same flow conditions are
lower, compared to the ones calculated not taking into account these buoyancy effects in the
bottom boundary conditions. A significant modelling improvement is achieved using damp‐
ing factors in the bottom boundary conditions; the calculation of the bottom shear can be
performed through [43]:

¶U -1
u* = Ft k z , Ft = (1 + 100Ri ) 3
(46)
¶z

where Ft is the velocity damping factor, κ the von Karman constant, ∂U/∂z the velocity gra‐
dient at the marginal grid of the bed and Ri the gradient Richardson number. The damping
function essentially expresses the ratio of the actual turbulence to the turbulence in the ab‐
sence of suspended matter. Based on this approach, the gradient at the bed is calculated ex‐
plicitly from the velocity profile and not using the log-law. Therefore, the ‘corrected’
velocity is:

u* æ z ö 1 z ¶a
U= ln ç ÷,
k z è a zo ø Ft
= 1-
a ¶z
(47)

where α is a bottom roughness correction coefficient. The reduction in bottom roughness

corresponds to drag reduction, observed both in nature and in the laboratory. Series of nu‐
merical experiments [43] suggest that α can be estimated through exponential functions of
the form:

ë ( ( û))
a = exp éê - (1 + b w s u* ) 1 - exp -bRi m ùú , {b = 7.7, b = 1.25, m = 0.6} (48)

In coastal areas, where the activity of both waves and currents is significant, their combined
effect to the range of the shear stresses can be addressed through equations of the form [2]:

1
(49)
2
tb = fcw rw uc + uw
2
 Modelling Cohesive Sediment Dynamics in the Marine Environment 229
[Link]

In the former relationship uc and uw are the current and wave amplitude velocities near the
bed and fcw the wave-current friction factor.

3.2.2. Deposition

The sediment deposition rate (S) is typically defined using the Krone formula [44]:

ì0 , t b ³ t cr ,dep
ïï
S=í æ tb ö (50)
ïcb × w s çç - 1÷ , t b < t cr ,dep
÷
ïî è t cr ,dep ø

The particulate matter that can be transported by turbulent flow may be expressed by the
equilibrium concentration Ce [15,41]:

rb rs - rw u*3
Ce = K s
g rs hWs
(51)

where Κs is a first-order proportionality coefficient (Κs~0.7) and Ws is the characteristic

(mean) settling velocity. It can be assumed that, in the case that the system is in balance, this
velocity equals the entrainment velocity from a CBS, with the sediment suspension to be in
equilibrium. At the fluid mud-water interface little or no turbulence is produced. Therefore,
as turbulent mixing reduces, Ce is further reduced. This results to a cumulative effect and
the complete breakdown of the vertical suspended concentrations’ profile. For cohesive
sediments the equilibrium concentration can also be considered as saturation concentration.

The critical shear stress for deposition can be related to the strength of the cohesive floc, Fc
(Ο{10-8Ν}), through relationships of the form [24]:

1 Fc F
t cr ,dep = ; 0.06 c2 (52)
5p D 2 D

Critical shear velocity for deposition can also be expressed with respect to the settling veloc‐
ity of the particles [44]:

ì0.008 , w s £ 5 × 10-5
ïï
u*cr ,dep = í0.023 , w s ³ 3 × 10-4 (53)
ï
ïî0.094 + 0.02log10 (w s ) , 5 × 10 < w s < 3 × 10
-5 -4
230 Sediment Transport Processes and Their Modelling Applications

3.2.3. Erosion

The sediment that enters suspension during erosion of the seabed is typically quantified us‐
ing the erosion rate, which is the eroded mass per unit surface and unit time [ΜL-2Τ-1]. For
homogenous, consolidated beds, the well known Partheniades formula applies [46]:

æ t ö
e = e o ç b - 1÷
çt ÷ (54)
è cr ,er ø

The critical erosion shear stress (τcr,er) and the erosion rate (εο) constant are considered to be
constant with time and within the sediment. However, resistance to erosion depends on var‐
ious parameters, among which sediment composition, porosity and degree of consolidation
of the deposit [47]. The seabed may be soft, partially consolidated, with high water content
(>100%), or may be a denser, more stable deposit. The way in which erosion takes place also
varies with the range of the applied shear stress and the stress history of the deposit. Three
erosion processes have been identified [14]:

1. floc-to-floc erosion, or surface erosion of the bed,

2. high concentration fluid mud entrainment and

3. mass erosion of the bed.

A soft bed that usually consists of freshly deposited clayey matter that is under consolida‐
tion presents non-uniform characteristics and is therefore vertically stratified. The erosion
rate in this case is [14]:

e = e f × exp éa (t b - t cr ,er ( z ) ) ù
b
ëê úû (55)

In equation 55, εf is the floc erosion constant [kg/m2/s] (equals the erosion rate in the case
that the shear stress equals the threshold value), τcr,er(z) is the erosion resistance profile with
depth z [Pa] and β (β~0.5) and α [Pa-β] are constants. The values of εf (10-4-10-7 gcm-2min-1)
and α (5-20mN-1/2) depend on the sediment-water mixture.

It is clear that one of the most significant parameters in modelling erosion of cohesive bed is
the determination of the critical shear threshold. Various experimentally determined formu‐
las have been proposed, among which many relate the erosion threshold to the density of
the bed through relationships of the form:

t cr ,er = x × rd z (56)
 Modelling Cohesive Sediment Dynamics in the Marine Environment 231
[Link]

There is a high variability in the values of the constants involved. We indicatively mention
that [48] estimated them at ξ=5.42 10-6 and ζ=2.28, while [49] defined ξ=1.2 10-3 and ζ=1.2. It
is noted that, alternatively to the use of the sediment dry density, bulk density is also used
in self-similar expressions. From insitu measurements in a macrotidal mudflat [50] a posi‐
tive correlation of the critical erosion shear stress was found with the sediment water con‐
tent (W), the bulk density (ρb) and the mass loss on ignition (LOI):

ì -0.0012W + 0.51
ï
t cr ,er = í 0.0013 rb - 1.4 (57)
ï-0.041LOI + 0.65
î

It follows that the erosion rate constant and the critical shear threshold are decisive parame‐
ters for the quantification of eroding masses. These parameters are usually experimentally
determined; however the values proposed in literature present high variability. The main
reason for this variability is the dependence of the erosion process on various factors, like:

• Density: the resistance to erosion increases with the density of the bed.

• Floc diameter – fine fraction content: the effect of the percentage of fines in the sediment
to the critical erosion stress is strongly non-linear.

• Clay composition: the lithological composition of clay affects the behaviour of the sedi‐
ment under shear.

• Stress history: the consolidation degree of the sediment is a significant parameter since
clayey sediments change their mechanical behaviour in cases of preloading and overcon‐
solidation.

• Organic content: experimental measurements have shown that the erosion threshold is
significantly lower for the case of organic-rich beds.

• Biostabilisation: bacteria and benthic diatoms increase the cohesion of the bed through se‐
cretion of extracellular polymeric substances, while the existence of macrophytes in the
surface of the bed prohibits erosion reducing the flow velocities through their canopy and
at the same time stabilizing the sediment through their root system.

• Bioturbation: the existence of micro-fauna in the surface layer of the sediment reduces the
resistance to erosion through loosening of the sediment structure during grazing and the
production of faecal pellets.

• Measurement method (laboratory or insitu): the measurement method itself introduces

uncertainties regarding the accuracy and the general applicability of the values, given
that there is high variability for similar areas using different instrumentation. The differ‐
ences are even higher between insitu and laboratory experiments, with the former to pro‐
vide significantly higher erodiblity parameters than the latter.
232 Sediment Transport Processes and Their Modelling Applications

3.2.4. Self-weight consolidation – Resuspension

A fresh mud deposit has the form of a sediment-water mixture with sediment concentra‐
tions of the order of a few tens of grams per liter. This mixture evolves in three stages
[12,42]:

1. Deposition of aggregates during the first hours forming fluid mud

2. Depletion of the interstitial water in a period of one to two days

3. Depletion of the intra-particle pore water (very slow consolidation)

Typical consolidation time-scales for a deposit with bulk density of 1150-1250kg/m3 are of
the order of 1 week, 1 month, 3 months and 0.5-1year for a layer thickness of 0.25m, 0.5m,
1m and 2m, respectively [42].

The resistance of the deposit to resuspension increases with consolidation time, process that
can be parameterised considering a simplified vertical structure of the bed. In its simplest
form, only two layers are taken into account, a surface, non-consolidated, low-strength layer
and a deeper, fully consolidated one; different densities and erodibility parameters are as‐
signed to each layer. This approach can be extended to a polystromatic representation of the
vertical composition of the bed and of the consolidation process.

One of the most widely applied parameterizations for the process of consolidation is the
Gibson equation [51]:

¶f rs - rw ¶ 1 ¶ æ ¶s ¢ ö
¶t
-
rw ¶z
( )
kf 2 -
rw g ¶z çè
kf
¶z ÷ø
=0 (58)

Equation 58 is written in Eulerian form and using the solid volume concentration [51], while
k and σ’ are the permeability and the effective stress, respectively. Under the reasoning that
the critical failure stress for the sediment comes as a result of failure of the bonds between
and within the aggregates, a Mohr-Coulomb type criterion applies:

t cr ,er = c ¢ + tan js ¢ (59)

The first term of equation 59 expresses the effect of cohesion (c’), which is the shear strength
in zero effective stress, and φ is the friction angle.

Exponential functions that relate the compactibility of the sediment through parameters like
the void ratio, e, to effective stress are also used [52]:

e = ( eo - e¥ ) exp ( -ls ¢ ) + e¥ (60)

 Modelling Cohesive Sediment Dynamics in the Marine Environment 233
[Link]

Another empirical relationship [53] between the critical erosion shear velocity and the yield
strength of muddy deposits reads:

ì
t y = b rd 3 < 3 N m 2 üï
1
ïu*cr ,er = 0.013t y ,
4
í 1
ý (61)
ïu t y = b rd 6 > 3 N m 2 ïþ
î *cr ,er = 0.009t y ,
2

The first of the equations refers to fluid mud and the second to mud that has exceeded the
plasticity limit. Accepting that the critical shear stress threshold for resuspension (τcr,res) of
deposited sediments ranges between the corresponding values for deposition (τcr,dep) and
erosion (τcr,er), at deposition (td=0) and after full consolidation (td=tfc), respectively, its value
can be estimated through exponential equations of the form [38]:

t cr ,res = t cr ,dep + (t cr ,er - t cr ,dep ) × é1 - e- n ×td ù , n ; 7 tfc [T -1] (62)

ë û

3.3. The fine sediment transport model

Based on the preceding analysis, the formulation of a sediment transport module involves
the mathematical description of all the aforementioned pelagic and benthic processes and
the interactions between them. A typical outline of such a module is presented in figure 2,
showing the transition of the sedimentary plume to various conditions, the governing proc‐
esses at each stage and the interactions between the processes. These considerations were
taken into account in formulating the Fine Sediment Transport Model (FSTM); the model is
outlined in the present work and characteristic results from simulations in Thermaikos Gulf
(NW Aegean Sea) with point (rivers, mechanical erosion) and distributed (atmosphere)
sources are presented and discussed.

Figure 2. Schematic representation of a cohesive sediment transport module. Dashed vectors denote transition from
one state (ovals) to another and solid vectors indicate effect of a process (rectangles) to another.
234 Sediment Transport Processes and Their Modelling Applications

FSTM [38] was formulated using the particle-tracking method (LM), thus describing sedi‐
ments in the marine environment as particles of specific mass and characteristic properties
(assigned to each particle in the form of indicators) that are being passively advected and
dispersed by the currents. FSTM accounts for all the pelagic and benthic processes that were
described in the previous section. More specifically, the position of the particle in the next
temporal step is calculated from eq. [3-4], using the dispersion coefficients for the amplitude
of the stochastic particle displacement defined in eq. [10-11]. The evolution of the cohesive
particle diameter is defined using the fractal method (eq. 19-20), while the changes in the
bulk density of the aggregate are calculated from equation 18. These parameters are used to
determine the settling velocity of the particle through the modified Stokes law (eq. 23). The
effects of stratification to the vertical propagation of the sedimentary plume are taken into
account using the double diffusion theory and assigning damping functions (eq. 40-41) to
the vertical deterministic and stochastic displacements. The constants γ, δ and κ were esti‐
mated at 0.18, -0.9 and 0.03 [38]. Regarding near-bed processes, the benthic shear stress is
defined explicitly, from the horizontal velocity profiles (eq. 46), and the critical shear veloci‐
ty for deposition of a particle is determined from its settling velocity, using equation 53. Af‐
ter the deposition of a particle onto the bed, the critical shear stress for resuspension of the
particle is considered to increase with depositional time (eq. 62). After full consolidation
(td=tfc) the particle is considered to be part of the seabed, which is assumed homogenous and
fully consolidated (eq. 54). Regarding the boundary conditions of FSTM, radiation is applied
at the open boundaries of the domain (particles are allowed to cross the boundary and are
excluded from following computational cycles) and reflection is applied at the surface (par‐
ticles are not allowed to escape to the atmosphere), at the bed (in cases that a particle does
not deposit it is reflected to its previous vertical position) and at ‘dry’ grid cells (along one
or both horizontal directions). The sources of particles may be the topmost grid cell in the
vertical sense, in cases of river-borne or aeolian transported particles, or the bottom cell (for
physically or mechanically eroded matter).

4. Characteristic applications

4.1. The application domain: Thermaikos Gulf

Thermaikos Gulf (Figure 3), a micro-tidal, elongated shelf in the NW Aegean Sea, is an area
of high socio-economic and environmental significance. Numerous anthropogenic activities
take place in the vicinity of the gulf, forcing the marine environment with residues from ag‐
ricultural, urban and industrial residues. Significant marine transport loads also exist, since
the port of Thessaloniki is the second largest harbor in Greece and a gateway to the Balkans.
Due to its trophic state (mesotrophic, compared to the oligotrophic open Aegean Sea) and its
mild-sloped bottom relief, Thermaikos is also one of the key marine sites for aquaculture
and fishing (trawling and otherwise) of the Hellenic region. Sources of freshwater and fine
sediments by rivers and smaller streams are distributed along the greater part of its north‐
ern and western coastlines; four are the main rivers discharging in the area, Axios, Pinios,
Aliakmon and Loudias (Figure 3), of which the former two are the largest. It should be not‐
ed that the deltaic area of Axios, Aliakmon and Loudias is protected by the Ramsar conven‐
tion (Wetlands of International Importance) and is included in the EU network of protected
areas Natura 2000 (GR1220010).
 Modelling Cohesive Sediment Dynamics in the Marine Environment 235
[Link]

FSTM was applied in the domain of Thermaikos, covering the aquatic area that extends
southwards to longitude 39.6°N and eastwards to latitude 23.5°E (Figure 3). The grid is hori‐
zontally curvilinear (dx=dy=1/60°) and vertically Cartesian (dz=2m). The necessary input hy‐
drodynamic and physical parameters (seawater velocities, salinity and temperature fields)
were derived by the North Aegean Sea (NAS) model [54]. Time-series of sediment fluxes
from the 4 main rivers, used in the simulations, were recorded during the METRO-MED
project [55]. Results from three applications are presented further down, for river-borne
sediments, aeolian transported dust and mechanically (trawling) eroded matter. It is noted
that the hydrodynamics for the former case are after results of NAS with climatological forc‐
ing, thus expressing the typical circulation features in the domain.

Figure 3. Bathymetry and geomorphological parts Thermaikos Gulf and its location in the Mediterranean Sea

4.2. Point source (river-borne sediment) simulation

4.2.1. Transport patterns and particulate matter distribution

One of the major advantages of formulating FSTM using the EL method is the ability to
track the history of each particle and to segregate sediment in the domain according to com‐
mon properties. An example of visualizing such information is given in figure 4; the figure
shows the evolution of suspended and surface (top 2m) particles, 30, 120, 210 and 300 days
from the beginning of the simulation (corresponding to late January, April, July and Octo‐
ber, respectively). The chromatic encoding denotes the riverine origin of the particles. These
representations give information regarding the seasonal variability and the transport pat‐
terns of sediments by each of the 4 rivers.
236 Sediment Transport Processes and Their Modelling Applications

It can be noted that, during winter (Figure 4b), Suspended Particulate Matter (SPM) enter
the enclosed inner (northmost) part of the gulf, matter that mainly originates from Axios. At
the same time, stratification is strong in the northern river-dominated area, leading to high
presence of SPM in the surface layer (Figure 4a). The northward expansion of sediment from
Axios is probably due to anticyclonic movement of the surficial waters in the gulf during the
wintry months [56]. This movement is also responsible for the northward expansion of the
Pinios plume, ascertainment that is in accordance with [57]. Matter from Aliakmon is mainly
transported towards the south along the western coastline, together with part of the sedi‐
ment supply of Axios. During the vernal period (Figure 4d) a distinct cyclonic transport pat‐
tern is detected in the outer part that forces sediments to move southwards along the west
coast, while SPM from Axios are still present in the inner gulf. The same transport pattern is
observed for surficial sediments (Figure 4c) with a strong effect of freshwater plumes along
the western coastline, effect that keeps sediments in the surface layer at high distances from
the outflows. The dispersion of particles is high near the northerly river-outflow system,
forcing particles (mainly from Axios) to move eastwards in addition to the prevailing cy‐
clonic movement. The presence of SPM during summer is low (Figure 4f) due to reduced
river-borne sediment fluxes; suspended particles appear mainly in the northern part of the
outer gulf, with low SPM in the surface layer (Figure 4e). During autumn (Figure 4h), the
increase of sediment supplies causes related increase to SPM and a predominant cyclonic
transport pattern exists in the northern part of the gulf; a small part of the Axios sediment
supply, again, spreads eastwards and towards the inner gulf. Stratification is strong not only
for the northern part, but also near the outfalls of Pinios (Figure 4g), with a significant
spreading of the surficial plume, which is the most profound of the whole year.

Figure 4. Variation of surficial and suspended particles at the 30th (mid-winter) (a, b), 120th (mid-spring) (c, d), 210th
(mid-summer) (e, f) and 300th day of the simulation (mid-autumn) (g, h). The chromatic encoding denotes the differ‐
ent riverine origin of the particles.
 Modelling Cohesive Sediment Dynamics in the Marine Environment 237
[Link]

4.2.2. The contribution of the rivers to the sedimentation of the gulf

The existence of transport and sedimentation patterns related to the riverine origin of SPM
and the geomoprhological parts of the gulf becomes evident. The following table (Table 1)
presents the percentages of deposition in each region for each of the four rivers, to facilitate
the investigation of the spatial distribution of the sediments with respect to its riverine origin.

Part of Thermaikos River

Axios Loudias Aliakmon Pinios

Inner part 28.2% 19.8% 5.8% 0.0%

Outer part 43.5% 44.7% 61.5% 2.4%

Extended part 28.3% 35.5% 32.7% 97.6%

Table 1. Sedimentation percentages of the rivers in each part of the gulf

Figure 5. Simulated sedimentation rate [mm/y] (a) and sediment types after [58] (b).

Sediments from Aliakmon present a constant, throughout the year, transport pattern, with
the prevalence of advection against dispersion, and predominant movement along the low-
depth areas of the western coast. The majority of the sediment supply of Aliakmon deposits
in the outer gulf, while a very small part of the supply reaches the enclosed inner Thermai‐
kos. Contrastingly, Axios appears as the main sediment supplier for the inner gulf, along
with a much smaller contribution from the Loudias. Matter from Axios presents a great dis‐
persal in the outer part of the gulf, accumulating sediments along the eastern and western
coasts; nearly half of the sediment fluxes of Axios settle in the outer gulf (Table 1) while the
238 Sediment Transport Processes and Their Modelling Applications

remaining half equally supplies the coastal areas of the inner and extended Thermaikos.
Similar transport and sedimentation patterns are exhibited by Loudias, mainly due to the
proximity of the two river outflows. Particles from Pinios supply the coastal zone along the
western coastline, primarily northwards and secondarily southwards from the outflow, re‐
maining almost exclusively in the extended part of the domain. The northward deflection of
the Pinios plume is due to the strong bottom slope near the mouth of the river [57] and to
the effect of the Sporades eddy, which frequently turns anti-cyclonic [56].

The simulated sedimentation rates (Figure 5a) are in good correlation with corresponding
values defined after core-sampling in the area [58]. Furthermore, the areas of high river-
borne sediment accumulation coincide with the locations in the gulf where the sediment
ranges from mud to sandy mud (Figure 5b), validating the modelling approach. In the coast‐
al zone of the eastern outer Thermaikos, where the benthic material grades from sandy mud
to muddy sand, the simulated depositing trends are reducing towards the south.

4.3. Distributed source (aeolian transported dust) simulation

Results of the SKIRON/Eta forecasting system and the Eta/NCEP regional atmospheric mod‐
el [58] were used for the determination of dust inflow in the domain. The model provides
prognoses for total dust deposition over the Mediterranean Sea, parameter that was used af‐
ter two-linear interpolation to transfer data from the domain of the Mediterranean to the do‐
main of Thermaikos.

The time-series of the total dust mass deposited onto the aquatic surface for the period
13/04-13/06/2005 is depicted in figure 5 b. This specific period was selected on account of an
intense dust storm that occurred on April 17, 2005. During this incident thick bands of Sa‐
haran dust covered the whole Hellenic area, as depicted in the satellite image of figure 5 a. It
can be noted that the dust storm of April 17th is predicted by the atmospheric model (Figure
5b) and that three more episodes of lower intensity followed in May (7, 12 and 21). The total
mass introduced to the domain during the period in question is 0.72 10-6t, value that exceeds
the corresponding annual contribution of the rivers of the area, estimated at 0.65 10-6t/y [55].
It is noted that the simulation was extended by 1 month (up to 13/07).

4.3.1. Dust distribution

Investigating the temporal evolution of total suspended and deposited dust masses (Figure
6a) it can be noted that the peak-values of the suspended dust masses, as expected, follow
the related peaks of dust inflow (Figure 5b), while deposition increases with time. Maximum
total suspended dust masses are recorded in mid-May, while almost all of the matter has
settled on the bed at the end of the simulation. The average suspended concentrations at the
surface (Figure 6b) follow the morphology of the input time-series, which is expected since
the surface is the source of the particles, with values ranging from 8 to 16mg/l at peak ‘load‐
ing’. The depth-averaged concentrations (Figure 6b) show a small temporal hysteresis in
peak values compared to the inflow time-series, of the order of 1-2 days, and range between
0.05 and 1.5mg/l. The maximum suspended concentrations in the field are of the order of
 Modelling Cohesive Sediment Dynamics in the Marine Environment 239
[Link]

21mg/l and 50mg/l for the water column and the surface layer. The average suspension
times of dust in the domain were estimated at 17 days. At the same time, the residence time
of dust in deposition increases almost exponentially with time, indicating that deposition
prevails over resuspension; near the end of the simulation, the majority of particles settled
mainly in the low-depth areas along the coastlines.

Figure 6. MODIS satellite imagery of the dust storm on April 17, 2005 (a) (the image is a property of NASA) and tem‐
poral evolution of the total deposited mass in the domain during the period 13/04-13/06/2005 (b).

Figure 7. Temporal evolution of total suspended and deposited mass (a) and of suspended depth-averaged and abso‐
lute concentrations in the surface layer (b) and horizontal trajectories of randomly selected dust particles (c): the loca‐
tion of entrance in the field is denoted as a dot.
240 Sediment Transport Processes and Their Modelling Applications

4.3.2. Dust particle trajectories

As in any simulation using a LM, it is possible to track the movement and changes in char‐
acteristics of sediments with time using FSTM. Such an example is given in figure 6c as hori‐
zontal trajectories of randomly selected dust particles; the location of entrance of the particle
in the surface layer of the domain is noted with a dot. Some of these particles escaped the
domain through the open boundaries (orange and mauve curves), while others settled on
the bed after performing particularly arbitrary movements and presenting very different
suspension times. Emphasising on the particle with the maximum residence time in the field
(blue curve in figure 6c), it can be noted that it entered the field on 19/04(18:00) and initially
moved cyclonically for 4 days, period during which its characteristics remained stable. Fol‐
lowing, the particle entered an anti-cyclonic motion for 5 days, at the end of which
28/04(17:30) it started to move cyclonically; this motion lasted for 5 days, during which the
particle performed a full circle. Afterward, the particle transported westwards for 3.3 days,
period after which its movement became particularly arbitrary. In the vertical sense (data
not shown) the particle retained an almost stable settling rate, with an average value of
50mm/hr. The particle finally settled onto the bed 38.5 days after its entrance in the field.

4.4. Mechanically eroded matter

FSTM was also applied for the investigation of mechanically eroded (trawled) masses in
Thermaikos, application whose main findings are described briefly in the present chapter;
the reader is referred to [60] for further information. The mechanical erosion rates were esti‐
mated at 430gr/m 2, value directly comparable to similar coastal areas, while the sedimentary
matter mobilized due to the activity of the benthic trawlers during one trawling season was
found significantly higher than the respective contribution of the rivers in the gulf. The sedi‐
ments, after mobilization by the trawls, continued to move at low distances from the bed,
forming Benthic Nefeloid Layers (BNLs), finding that supports the claim that part of the
BNLs found in Thermaikos originates from the trawling activity in the area. The redistribu‐
tion of sediments due to trawling was found generally low, with the matter to redeposit at a
small distance from its generation, typically lower than 5km and with average suspension
times of 1 to 5 days. The mass export rates from the gulf were also found to be small, espe‐
cially for the initial part of the trawling period.

5. Conclusions

The present work concentrates on the physical interpretation and modelling approaches
used for the main processes that drive sediment transport on coastal and shelf seas. The
analysis focused on typical modelling methods and on the physical processes that take place
in the water column and near the bed, including the corresponding mass-exchanges. More
specifically, advection-dispersion, flocculation and settling mechanisms were analyzed as
the processes that take place in the column, along with the effects of stratification to the ver‐
tical movement of fine sediments. Regarding near-bed processes, the aspects addressed in‐
 Modelling Cohesive Sediment Dynamics in the Marine Environment 241
[Link]

clude shear stress conditions, particle deposition and gradual consolidation and erosion of
the bed. In order to accurately describe the fate of cohesive sediments in the marine environ‐
ment, a numerical tool must generally include all of these processes. However, there are cas‐
es in which some of the processes have minimal or no effect; for example, hindered settling
does not partake for low concentration environments (<20mg/l) and can therefore be ignor‐
ed, while stratification becomes significant only in the presence of strong density gradients,
typical in estuarine areas. Likewise, wind-generated wave impacts on bottom shear become
significant in low-depth (<10m), coastal areas and need to be accounted for only in high res‐
olution regional scale models.

The FSTM model, formulated based on the particle-tracking method and the considerations
analyzed in the chapter, was briefly described and indicative results from applications in
Thermaikos Gulf, a microtidal shelf sea of the Northern Aegean Sea, regarding point and
distributed sedimentary sources were presented and discussed; transport and sedimentation
patterns in the gulf were identified for the river-borne sediments of the area. The prevailing
cyclonic circulation of the area appears as the main mechanism controlling sedimentation
pathways in the domain and forcing higher sediment accumulation along the shallow areas
of the western coast. Sediment from Aliakmon is the most characteristic example of this be‐
haviour, with consistent, throughout the year, cyclonic movement and deposition; the great‐
er part of the Aliakmon sediment influx (~⅔) accumulates in the outer gulf, while a
significant amount (~⅓) reaches the southern part of the shelf. The geomorphology of Ther‐
maikos also plays a significant part in the phenomena; the location of the Axios delta to the
centre of the northern coastline exposes its sediments to stronger currents of alternating di‐
rections and, thus, its sediments present much higher dispersion in the field, compared to
the other rivers of the area. This behaviour is enhanced by seasonal stratification, which
forces sediment to remain suspended longer at periods of high freshwater supplies from the
rivers. Nearly half of the sediment from Axios settles in the outer part of the gulf, along the
shallow areas of both the western and eastern coasts. The remaining part of its outflow co‐
equally replenishes the sediment of the extended and inner parts. In fact, the mathematic in‐
vestigation showed that Axios is the main sediment supplier of the enclosed inner
Thermaikos. The steep bottom relief near the outflows of Pinios, combined with the hydro‐
dynamics of the area, forces its particles to settle along the western coast close to the delta,
with a higher northward spreading. The mass export towards the deep Sporades basin was
estimated by the simulation at approximately 10% of the total riverine sediment supply,
which corresponds to around 85 103t/yr; this high sediment accumulation in the shelf is con‐
sistent with the characterisation of Thermaikos as a sediment trap.

Results from the three-month simulation of aeolian transported dust during the dust storm
reported in mid-April of 2005, showed concentrations of suspended dust between 8 and 16mg/
l at maximal loading and corresponding depth-averaged values ranging from 0.05 to 1.5mg/l.
The mean suspension times of dust in the water column were around 17 days, while dust dep‐
osition prevailed over resuspension. These high concentrations and suspension times reveal
the significance of the process of aeolian transport to the suspended masses in the gulf that
can exceed the contribution of river-borne matter considerably in cases of dust storms.
242 Sediment Transport Processes and Their Modelling Applications

Author details

Katerina Kombiadou* and Yannis N. Krestenitis

Laboratory of Maritime Engineering and Maritime Works, Civil Engineering Department,

Aristotle University of Thessaloniki, 54124, Thessaloniki

References

[1] Mehta, A. J., Hayter, E. J., Parker, W. R., Krone, R. B., & Teeter, A. M. (1989). Cohe‐
sive sediment transport I: Process Description. Journal of Hydraulic Engineering,
115(8), 1076-1093.
[2] Wang, X. H., & Pinardi, N. (2002). Modeling the dynamics of sediment transport and
resuspension in the northern Adriatic Sea. Journal of Geophysical Res, 107, 3225.
[3] Rodean, H. C. (1996). Stochastic Lagrangian Models of Turbulent Diffusion. American
Meteorological Society, 84.
[4] Barros, A. P., & Baptista, A. M. (1990). An Eulerian-Lagrangian Model for Sediment
Transport. In: Spaulding M., editor. Estuaries, in Estuarine and Coastal Modeling.
ASCE, 102-112.
[5] Suh, S-W. (2006). A hybrid approach to particle tracking and Eulerian-Lagrangian
models in the simulation of coastal dispersion. Environmental Modelling and Software,
21(2), 234-242.
[6] Oliveira, A., & Baptista, A. M. (1998). On the role of tracking on Eulerian-Lagrangian
solutions of the transport equation. Advances in Water Resources, 21, 539-554.
[7] Lewis, R. (1997). Dispersion in estuaries and coastal waters. John Wiley & Sons Ltd.
England, 312.
[8] Richardson, L. F. (1926). Atmospheric diffusion shown on a distance-neighbour
graph. Proceedings of the Royal Society, A110, 709-737.
[9] O’Brien, J. J. (1986). The diffusive problem. In: O’ Brien JJ, editor. Advanced physical
oceanographic numerical modelling. ASI series. D. Reidel Publ. Comp., 127-144.
[10] Mellor, G. L. (1996). Introduction to physical oceanography Princeton University,
New Jersey ., 260.
[11] O’Connor, B. A. (1993). Three-dimensional sediment-transport models. In: Abbott
MB, Price WA, editors. Estuarial and harbour engineer’s reference book, Chapman and
Hall Pubs. London, 199-214.
[12] Partheniades, E. (2009). Cohesive sediments in open channels. Properties, transport and
applications, Elsevier, 358.
 Modelling Cohesive Sediment Dynamics in the Marine Environment 243
[Link]

[13] Krone, R. B. (1972). A field study of flocculation as a factor in estuarial shoaling proc‐
esses. Technical Bulletin 19. U.S. army Corps of Engineers, 32-62.

[14] Mehta, A. J. (1993). Hydraulic behaviour of fine sediment. In: Abbott MB, Price WA,
editors. Coastal, Estuarial and Harbour Engineer’s Reference Book, Chapman and Hall.
London, 577-584.

[15] Winterwerp, J. C. (1999). On the dynamics of high-concentrated mud suspensions.

Judels Brinkman & Ammerlaan. Delft. , 172.

[16] McAnally, W. H., & Mehta, A. J. (2001). Collisional aggregation of fine estuarial sedi‐
ment. In: McAnally WH, Mehta AJ, editors. Coastal and Estuarine Sediment. Proc. in
Marine Science 3, Elsevier, 19-39.

[17] McAnally, W. H., & Mehta, A. J. (2002). Significance of aggregation of fine sediment
particles in their deposition. Estuarine Coastal and Shelf Science, 54, 643-653.

[18] Eisma, D. (1986). Flocculation and de-flocculation of suspended matter in estuaries.

Netherlands Journal of Sea Research, 20(2-3), 183-199.

[19] Dyer, K. R. (1989). Sediment processes in estuaries: future research requirements.

Journal of Geophysical Research, 94(C10), 14327-14339.

[20] Spearman, J. R., & Roberts, W. (2002). Comparison of flocculation models for applied
sediment transport modelling. In: Winterwerp JC, Kranenburg C editors. Fine Sedi‐
ment Dynamics in the Marine Environment. Proc. in Marine Science 5, Elsevier, 277-293.

[21] Bale, A. J., Widdows, J., Harris, C. B., & Stephens, J. A. (2006). Measurements of the
critical erosion threshold of surface sediments along the Tamar Estuary using a mini-
annular flume. Continental Shelf Research, 26, 1206-1216.

[22] Chen, S., & Eisma, D. (1995). Fractal geometry of in situ flocs in the estuarine and
coastal environments. Netherlands Journal of Sea Research, 32(2), 173-182.

[23] Logan, B. E., & Wilkinson, D. B. (1990). Fractal geometry of marine snow and other
biological aggregates. Limnology and Oceanography, 35(1), 130-136.

[24] Kranenburg, C. (1999). Effects of floc strength on viscosity and deposition of cohesive
sediment suspensions. Continental Shelf Research, 19(13), 1665-1680.

[25] Sternberg, R. W., Berhane, I., & Ogston, A. S. (1999). Measurement of size and set‐
tling velocity of suspended aggregates on the northern California continental shelf.
Marine Geology, 154, 43-53.

[26] Manning, A. J. (2004). Observations of the properties of flocculated cohesive sedi‐

ment in three Western European Estuaries. Journal of Coastal Research, SI41, 70-81.

[27] Manning, A. J., & Dyer, K. R. (2002). The use of optics for the in-situ determination of
flocculated mud characteristics. Journal of Optics A: Pure and Applied Optics, 4, S71-
S81.
244 Sediment Transport Processes and Their Modelling Applications

[28] Baugh, J. V., & Manning, A. J. (2007). An assessment of a new settling velocity pa‐
rameterisation for cohesive sediment transport modelling. Continental Shelf Research,
27(13), 1835-1855.
[29] Manning, A. J., & Dyer, K. R. (2007). Mass settling flux of fine sediments in Northern
European estuaries: measurements and predictions. Marine Geology, 245, 107-122.
[30] Van Leussen, W. (1994). Estuarine macroflocs and their role in fine grained sediment
transport. PhD thesis. University Utrecht, 488.
[31] McAnally, W. H., & Mehta, A. J. (2007). Fine grained sediment transport. In: Garcia
MH, editor. Sedimentation engineering: processes, management, modeling, and
practice. ASCE manuals and reports on engineering practice [110], ASCE, 253-306.
[32] Turner, J. S. (1973). Buoyancy Effects in Fluids. Cambridge University Press, 91-126.
[33] Gargett, A. E. (2003). Differential diffusion: an oceanographic primer. Progress in Oce‐
anography, 56, 559-570.
[34] Burrows, R., & Ali, K. H. M. (2001). Entrainment studies towards the preservation
(containment) of ‘freshwaters’ in the marine environment. Journal of Hydraulic Re‐
search, 39(6), 591-599.
[35] Kunze, E. (2003). A review of oceanic salt-fingering theory. Progress in Oceanography,
56, 399-417.
[36] Kelley, D. E., Fernando, H. J. S., Gargett, A. E., Tanny, J., & Özsoy, E. (2003). The dif‐
fusive regime of double-diffusive convection. Progress in Oceanography, 56, 461-481.
[37] Schladow, S. G., Thomas, E., & Koseff, J. R. (1992). The dynamics of intrusions into a
thermohaline stratification. Journal of Fluid Mechanics, 236, 127-165.
[38] Kombiadou, K., & Krestenitis, Y. N. (2012). Fine sediment transport model for river
influenced microtidal shelf seas with application to the Thermaikos Gulf (NW Ae‐
gean Sea). Continental Shelf Research, 36, 41-62.
[39] Dearnaley, M. P., Roberts, W., Jones, S., Leureur, K. C., Lintern, D. G., Merckelbach,
L. M., Sills, C. G., Toorman, E. A., & Winterwerp, J. C. (2002). Measurement and
modelling of the properties of cohesive sediment deposits. In: Winterwerp JC, Kra‐
nenburg C, editors. Fine Sediment Dynamics in the Marine Environment. Proc. in
Marine Science 5. Elsevier, 57-89.
[40] Teeter, A. M., Johnson, B. H., Berger, C., Stelling, G., Scheffner, N. W., Garcia, M. H.,
& Parchure, T. M. (2001). Hydrodynamic and sediment transport modelling with em‐
phasis on shallow-water vegetated areas (lakes, reservoirs, estuaries and lagoons).
Hydrobiologia, 444, 1-23.
[41] Winterwerp, J. C., Bruens, A. W., Gratiot, N., Kranenburg, C., Mory, M., & Toorman,
E. A. (2002). Dynamics of concentrated benthic suspension layers. In: Winterwerp JC,
Kranenburg C, editors. Fine Sediment Dynamics in the Marine Environment. Proc. in
Marine Science 5, Elsevier, 41-55.
 Modelling Cohesive Sediment Dynamics in the Marine Environment 245
[Link]

[42] Van Rijn, L. C. (1998). Mud Coasts. Principles of coastal morphology, Aqua Pubs. Am‐
sterdam. Netherlands, 6.1-6.51.
[43] Toorman, E. A., Bruens, A. W., Kranenburg, C., & Winterwerp, J. C. (2002). Interac‐
tion of Suspended Cohesive Sediment and Turbulence. In: Winterwerp JC, Kranen‐
burg C, editors. Fine Sediment Dynamics in the Marine Environment. Proc. in Marine
Science 5. Elsevier., 7-23.
[44] Krone, R. B. (1962). Flume studies of the transport of sediment in estuarial shoaling
processes. Final Report. Hydraulic Engineering Laboratory and Sanitary Engineering
Research Laboratory. University of California. Berkeley. USA. , 110.
[45] Pohlmann, T., & Puls, W. (1994). Currents and transport in water. In: Sündermann J,
editor. Circulation and contaminant fluxes in the North Sea. Springer Verlag. Berlin,
345-402.
[46] Partheniades, E. (1965). Erosion and deposition of cohesive soils. Journal of the Hy‐
draulics Division, 91(HY2), 105-138.
[47] Parchure, T. M., & Mehta, A. J. (1985). Erosion of Soft Cohesive Sediment Deposits.
Journal of Hydraulic Engineering, 111(10), 1308-1326.
[48] Thorn, M. F. G., & Parsons, J. G. (1980). Erosion of cohesive sediments in estuaries:.
An engineering guide, 3rd International Symposium on Dredging Technology. Paper F1.
BHRA Fluid Engineering, 349-358.
[49] Ockender, M. C., & Delo, E. A. (1988). Consolidation and erosion of estuarine mud
and sand mixtures-An experimental study. HR Wallingford. Report No SR 149.
[50] Mitchener, H. J., & O’Brien, D. J. (2001). Seasonal variability of sediment erodibility
and properties on a microtidal mudflat, Peterstone Wentlooge, Severn Estuary, UK.
In: McAnally WH, Mehta AJ, editors. Coastal and Estuarine Fine Sediment Processes.
Proc. in Marine Science 3. Elsevier, 301-322.
[51] Merckelbach, L. M., Kranenburg, C., & Winterwerp, J. C. (2002). Strength modeling
of consolidating mud beds. In: Winterwerp JC, Kranenburg C, editors. Fine Sediment
Dynamics in the Marine Environment. Proc. in Marine Science 5. Elsevier, 359-373.
[52] Hawlader, B. C., Muhunthan, B., & Imai, G. (2008). State-dependent constitutive
model and numerical solution of self-weight consolidation. Géotechnique, 58(2),
133-141.
[53] Migniot, C. (1968). A study of the physical of different very fine sediments and their
behaviour under hydrodynamic action. La Houille Blance, 7, 591-620.
[54] Kourafalou, V. H., & Tsiaras, K. (2007). A nested circulation model for the North Ae‐
gean Sea. Ocean Science, 3, 1-16.
[55] Karamanos, H., & Polyzonis, E. (2000). Thermaikos Gulf: the adjacent land area and
its potentiality for the freshwater, sediment and pollutant supply. In: Final Scientific
Report of the EU-MAST-III ELOISE Project METRO-MED. Part B. CT960049, 55-67.
246 Sediment Transport Processes and Their Modelling Applications

[56] Kourafalou, V. H., & Barbopoulos, K. A. (2003). High resolution simulations on the
North Aegean Sea seasonal circulation. Annales Geophysicae, 21(1), 251-265.

[57] Kontoyannis, H., Kourafalou, V. H., & Papadopoulos, V. (2003). Seasonal characteris‐
tics of the hydrology and circulation in the northwest Aegean (eastern Mediterra‐
nean): Observations and modelling. Journal of Geophysical Research, 108(C9), 3302.

[58] Karageorgis, Α., & Anagnostou, Ch. (2001). Particulate matter spatial-temporal distri‐
bution and associated surface sediment properties: Thermaikos Gulf and Sporades
Basin, NW Aegean Sea. Continental Shelf Research, 21, 2141-2153.

[59] Nickovic, S., Kallos, G., Papadopoulos, A., & Kakaliagou, O. (2001). A model for the
prediction of desert dust cycle in the atmosp here. Journal of Geophysical Research, 106,
18113-18129.

[60] Kombiadou, K., & Krestenitis, Y. N. (2011). Simulating the fate of mechanically erod‐
ed masses in the Thermaikos Gulf. Continental Shelf Research, 31, 817-831.
 Chapter 10

Quasi-3D Modeling of Sediment Transport for Coastal

Morphodynamics

Yun-Chih Chiang and Sung-Shan Hsiao

Additional information is available at the end of the chapter

[Link]

1. Introduction

Sand transport plays a very important role in many aspects of coastal and marine engineering.
The balance of moving sands influences the construction of harbours, coastal defence, offshore
wind turbine and oil rig, offshore platform and pipeline and many other engineering. Coastal
sands may be carried by currents (such as tidal currents, wind-driven-currents, wave-driven-
currents, storm surge driven currents), or by waves (monsoon waves or typhoon waves), or
influenced by bedform changes, or all of them acting together and interacting in general sea
state.

We can easily consider a sediment budget for a coastal area where a control section or a control
volume is selected. The changing rate of net accretion or erosion of the coastal area of sea bed-
level depends on the net transport rates at which sediments are entering or leaving the control
section or the control volume. If the sum of the inflow sediment transport rates is larger than
that of the outflows, the bed-level will tend to accrete; if the sum of the inflow sediment
transport rates is smaller than that of outflow rates, the bedform will erode.

Consequently, accurate prediction of sediment transport rates is an important element in

coastal engineering, foundations of offshore structures and morphological studies for the
coastal environment. The procedure of the coastal morphological modeling system is shown
as Fig. 1. The prediction of net sediment transport rates is a subject of great importance to
coastal engineers and morphological modellers concerned with mediumand long-term
shoreline changes. The aim of this chapter is to provide models for calculating the
hydrodynamics and dynamic quantities of sediment transports in coastal zone, especially for
the applications in surf zone.
248 Sediment Transport Processes and Their Modelling Applications

Initial topography

Wave simulation

Q-3D Current simulation Bed-slope

updating
Sediment transport

Bed evolution calculation

Yes Season change or

New forcing checking

No
Final topography

Fig. 1: The procedure of the coastal morphological modeling system

Figure 1. The procedure of the coastal morphological modeling system
1.1 Threshold of sediment motion

In order to estimate the changes of sediment budget, a quantitative evaluation of the net
1.1. Threshold oftransport
sediment sediment motion
rates are required. The sediment transport rate is defined as the amount
of sediment per unit time entering or leaving a control volume, which is the vertical plane of
In order unit width perpendicular to the sea mean water-level. The mechanics of sediment motion is
to estimate the changes of sediment budget, a quantitative evaluation of the net
depended on the effects of sediment dynamics while the frictions exerted on the sea bed by
sediment the
transport
hydrodynamicsare
rates required.
forcing Theas
agents, such sediment
waves andtransport
currents. rate is defined as the amount of
sediment per unit time entering or leaving a control volume, which is the vertical plane of unit
While the friction exerted on the sea bed by waves and currents, the sediment ‘entrainment’
width perpendicular to the sea mean water-level. The mechanics of sediment motion is
is the sand grains are carried up from bed. The ‘bed load’ sediment transport is the
dependedentraining
on the effects
sand grainsof sediment dynamics
rolling, hopping whilealong
and sliding the frictions
the bedform,exerted
and ison the sea bed by
dominant
the hydrodynamics forcing
with the inertial agents,
force and suchonasgrains
drag force waves lessand
thancurrents.
gravity force for slow flow or large
grains. Suspended load sediment transport is the portion of the entraining sediment that is
While thecarried by exerted
friction the largeronflow
the(orsea
thebed
wavebylarge
wavesenough)
and which
currents,settles slowly
the enough‘entrainment’
sediment and
moves with the stream.
is the sand grains are carried up from bed. The ‘bed load’ sediment transport is the entraining
sand grains
To rolling,
estimate hopping
the sedimentandtransport
sliding rate
along the bedform,
is more difficult, itand
mayis be
dominant withtwo
divided into the inertial
force andcomponents
drag forceby oncross-shore
grains less andthan
longshore direction.
gravity The slow
force for cross-shore
flow sediment
or large transport is
grains. Suspended
mainly carried by the skewness and asymmetry wave orbital motion and cycle-mean water
load sediment transport
level around is the
the surf [Link] of thesediment
The longshore entraining sediment
transport that is
is primarily carried with
dominated by the larger
flow (or the wave large
wave-driven enough)
currents. which
In order settlesthe
to quantitate slowly enough
sediment andrate,
transport moves with
included thebed
both stream.
load and suspended load, in each direction, an empirical relationship has been derived
To estimate the ‘transport
between sediment transport
rate’ rate is morecomponents,
and the energetic-based difficult, such
it may
as bedbe shear-stress,
divided into two
wave-driven currents, wave orbit velocity….. and so on. The agreement between
components by cross-shore and longshore direction. The cross-shore sediment transport is
measurements (or experiments in hydraulic laboratory) and calculations associated with the
mainly carried by the skewness and asymmetry wave orbital motion and cycle-mean water
level around the surf zone. The longshore sediment transport is primarily dominated with
wave-driven currents. In order to quantitate the sediment transport rate, included both bed
load and suspended load, in each direction, an empirical relationship has been derived
between ‘transport rate’ and the energetic-based components, such as bed shear-stress, wave-
driven currents, wave orbit velocity….. and so on. The agreement between measurements (or
experiments in hydraulic laboratory) and calculations associated with the relationship has
been widely applied and predicts accurately sediment transport along long, straight-like
beaches.
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 249
[Link]

The total load sediment transport models are widely applied with coastal engineering in last
three decades. Although the models are relatively simple and easy to use, but there are some
weaknesses when applied around the surf zone. We will discuss in next section.

1.2. Interaction between sediment motion and bed features for morphodynamics

Nearshore sandbars are the important and popular feature of natural beaches morphody‐
namics. The cross-shore location of sandbars changes by the interactions between the
sandbar and the sediment transport fluxes from waves and wave-driven-currents. Hoefel
and Elgar (2003) indicated the mechanics of wave-induced sediment transport and sand‐
bar migration: large waves breaking on the sandbars caused offshore mean currents,
which maximum near the sandbar crest, will lead sandbars moved offshore; small waves
pitching forward on the sandbars made the onshore acceleration skewness of wave orbi‐
tal velocities, which maximum near the sandbar crest, will lead the sandbar moved on‐
shore.

The nearshore sandbars could protect shorelines from wave attack by dissipating wave en‐
ergy offshore through sandbar-crest-induced wave breaking. In general coasts, the dynam‐
ic behavior of nearshore sandbars are similar to quasi-cycle for storm and seasons waves
alternated. However, the formation and evolution of sandbars are very important to coast‐
al planners and engineers. Prediction of the dynamic behaviour of nearshore sandbar sys‐
tems could be of great importance, there are many studies about the evolutions and
migrations of nearshore sandbars by numerical simulations in last decade (Hsu et al.,
2006; Long et al., 2006; Ruessink et al., 2007; Drønen and Deigaard, 2007; Houser and
Greenwood, 2007; Ruessink and Kuriyama, 2008; Ruessink et al., 2009; Pape et al., 2010;
Almar et al., 2010).

Figure 2. The mechanics of offshore sandbar migration. (Hoefel and Elgar,2003)

250 Sediment Transport Processes and Their Modelling Applications

Figure 3. The mechanics of onshore sandbar migration. (Hoefel and Elgar,2003)

According to Hoefel and Elgar (2003), two mechanisms are commonly used in the explanation
of morphodynamics of sandbars migration. The first mechanism type is the migration of
offshore sandbars. The offshore sandbars migrated seaward observed during storms were
driven primarily by a maximum in the offshore mean current (Under highly energetic storm
conditions, breaking waves cause near bottom seaward flows, also called “undertow”) near
the sandbar crest. Offshore sandbar migration during storms results from feedback and
interaction between breaking waves driven the “undertow” and bathymetric evolution (Elgar
et al., 2001).

The second mechanism associates the migration of onshore sandbars. There are many studies
have suggested mechanisms that could drive sandbars migration shoreward. Trowbridge and
Young (1989) and Trowbridge and Madsen (1984) demonstrated that nonlinear wave
boundary layer processes might play a role. Onshore sandbars migration might also be derived
by the systematic changes in wave kinematics when passing over nearshore sandbars. As
waves shoaling, their shapes are often described as “skewed” and “asymmetric” (Elgar,
1987), the mean water elevation depressed (the “Wave set-down”) leads mean currents been
weak. The non-breaking wave caused sediment transport over sandbars is driven
predominately by wave asymmetric orbital velocities. Under the steep skewed and asymmetric
waves, the water particle velocity is accelerated strongly as the asymmetric orbital velocity
rapidly changes from maximum offshore to maximum onshore (e.g., Elgar et al., 1988).

In order to describe morphodynamics sandbars well, a key parameter for cross-shore sediment
transport under breaking and near-breaking waves is well performed the near-bed skewed
and asymmetric wave orbital velocity. Therefore, the three dimensionality of the
hydrodynamic system should be considerable and must be taken into account. Most sediment
transport models are based on phase-averaged wave models, depth-integrated hydraulic
models (currents and wave driven currents) and sediment transport formula. In the hydraulic
models, nearshore currents have previously been predicted by using two-dimensional models
in the horizontal plane (2DH model). However, in the surf zone, the direction of current vectors
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 251
[Link]

near the water surface is different from that at the sea-bottom because of the effect of undertow
velocities. Nearshore currents have spiral profiles in the vertical direction. Undertows also
play an important role in the morphodynamical changes on a littoral beach such as the cross-
shore migration of longshore bars. In order to accurately predict the changes of sandbars
migration, it is very important that the three-dimensional distribution of nearshore currents
is determined. Therefore, well predicted nonlinear wave dynamics, vertical current and
sediment transport models can be a good tool for nearshore sandbar morphodynamics,
especially for cases where cross-shore transport mechanisms over sandbars are important.
Drønen and Deigaard (2007) compared 2D horizontal depth-integrated approach and quasi-3D
numerical model with formations of alongshore bars on gradual slope beach by normal and
oblique incident waves, and the quasi-3D model produces a crescentic bar while the depth
integrated model predicts almost straight sections of the bar interrupted by rip channels.
Consequently, considering the accuracy and efficiency, a nonlinear waves model with
quasi-3D sediment transport model can be applied as morphodynamics sandbar models.

1.3. The importance of quasi-3D sediment transport modeling

The numerical simulation of hydrodynamic and sediment transport processes form a powerful
tool in the description and prediction of morphological changes and sediment budgets in the
coastal zone. One of the key elements in a morphodynamics model is the correct quantification
of local sand transport. Most sediment transport models are based on phase-averaged wave
models, depth-integrated hydraulic models (currents and wave driven currents) and sediment
transport formula. In the hydraulic models, nearshore currents have previously been predicted
by using two-dimensional models in the horizontal plane (2DH model). However, in the surf
zone, the direction of current vectors near the water surface is different from that at the sea-
bottom because of the effect of undertow velocities. Nearshore currents have spiral profiles in
the vertical direction. Undertows also play an important role in the morphological changes on
a littoral beach such as the cross-shore migration of longshore bars. In order to accurately
predict the changes of beach profile, it is very important that the three-dimensional distribution
of nearshore currents is determined. Therefore, well predicted vertical current and sediment
transport models can be a good tool for coastal area morphological modelling, especially for
cases where cross-shore transport mechanisms are important. Considering the accuracy and
efficiency, a quasi-3D model can be applied as a coastal profile model or a coastal area model.

Some models for determining the vertical distribution of nearshore currents have previously
been proposed. de Vriend et al. (1987) presented a semi-analytical model and suggested that
a 3D model is required when the sediment transport in the cross-shore direction becomes
important; and then Svendsen and Lorenz (1989) proposed an analytical model composed of
cross-shore and longshore current velocities. In recent years, many quasi-3D numerical models
have been developed by extending 2DH model with one-dimensional velocity profile model
defined in the vertical direction (1DV model), have also been proposed (Sanchez et al., 1992;
Briand and Kamphuis, 1993; Okayasu et al., 1994; Elfrink et al., 1996; Rakha, 1998; Kuroiwa et
al,. 1998; Drønen and Deigaard, 2000; Davis and Thorne, 2002; Fernando and Pan, 2005; Drønen
and Deigaard, 2007; Li et al., 2007). In these models, the mean flow is determined by the 2DH
252 Sediment Transport Processes and Their Modelling Applications

model, and the velocity profiles across water column in the vertical direction are resolved by
using a 1DV model. While surface wave field and 3D flow field have been analyzed, sediment
transport vectors in the horizontal plane can be calculated with the sediment transport profile
across water column in the vertical direction.

The aim of this chapter is to develop an accurate model for estimation of local sediment
transport rate of the nearshore both inside and outside of the surf zone. A two-dimensional
2D fully nonlinear Boussinesq wave module is combined with a quasi-3D hydrodynamic
module (2DH and extended 1DV module). The 1DV hydrodynamic modules similar to those
described by Elfrink et al. (1996) with surface-roller concept and a one-equation turbulence
model are developed. The calculation of sediment transport rates is based on the formula with
wave asymmetric and ripple-bed effects developed by Lin et al. (2009). The quasi-3D
hydrodynamic modules are validated and compared, for regular waves over fixed beds. The
local sediment transport rates is also calculated and validated with experimental data.

2. 2DH waves and nearshore currents models

In this section, the two-dimensional wave and nearshore current models are described as
below:

2.1. Wave model

The wave model is based on the fully nonlinear Boussinesq equation developed by Wei et al.
(1995); the equation is expressed by velocity with an arbitrary water depth. Bottom friction,
wave breaking and subgrid lateral turbulent mixing as proposed by Kennedy et al. (2000), are
also expressed by equations. The governing equations are shown as below:

2
¶h z h2 h
+ Ñ × [( h + h )ua ] + Ñ × {( a - )hÑ(Ñ × ua ) + ( za + )hÑ[Ñ × ( hua )]} = 0 (1)
¶t 2 6 2

¶ua z ¶u ¶u
+ gÑh + (ua × Ñua ) + za { a Ñ(Ñ × a ) + Ñ[Ñ × ( h a )]} + R f - R b - R s = 0 (2)
¶t 2 ¶t ¶t

In the above equations, x=(x,y) are the horizontal coordinates coincident with the still water
surface and z is the vertical coordinate; t is the time; ∇ is the horizontal gradient operator,
defined as (∂/∂x,∂/∂y); η(x,t) is water surface elevation; g is the gravitational acceleration.
h=h(x) represents the water depth, uα=(up,vp) is the particle velocity vector at z=zα. Rf, Rb and Rs
are the effects of bottom friction, wave breaking and subgrid lateral turbulent mixing,
respectively. The detail mathematical operations are shown below,
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 253
[Link]

K
Rf = u |u | (3)
h +h a a

K is the friction coefficient. The wave breaking term, Rb=(Rbx, Rby) is represented as

1 1
Rbx = {[n (( h + h )ua )x ]x + [n (( h + h )ua )y
h +h 2 (4)
+ n (( h + h )va )x ]y }

1 1
Rby = {[n (( h + h )va )y ]y + [n (( h + h )ua )y
h +h 2 (5)
+ n (( h + h )va )x ]x }

The eddy viscosity (ν ) is defined as

n = Bd b2 ( h + h )ht (6)

Kennedy et al. (2000) proposed the mixing length, δb, is 1.2. The parameter B controls the
occurrence of energy dissipation is defined as

ì 1, ht ³ 2ht*
ï
ïh
B = í t* - 1, ht* < ht £ 2ht* (7)
ïht
ï
î 0, ht £ ht*

The onset and cessation of wave breaking using the parameter, ηt*, is represented as

ì h (F) , t - t0 > T *
t
ï
ht* = í (I ) t - t
0 (F) (I ) * (8)
ïht + * (ht - ht ), 0 £ t - t0 < T
î T

ht( I ) = 0.65 gh ,ht( F ) = 0.15 gh , T * = 5 h / g (9)

where T* is the transition time, t0 is the time when wave breaking occurs, and t-t0 is the age of
the breaking event. The subgrid lateral mixing terms Rs=(Rsx, Rsy) is displayed as follow
254 Sediment Transport Processes and Their Modelling Applications

1 1
Rsx = {[n (( h + h )ua )x ]x + [n s (( h + h )ua )y
h +h s 2 (10)
+ n s (( h + h )va )x ]y }

1 1
Rsy = {[n (( h + h )va )y ]y + [n s (( h + h )ua )y
h +h s 2 (11)
+ n s (( h + h )va )x ]x }

The parameter νs is the eddy viscosity due to the subgrid [Link] can be calculated by:

1
n s = cm DxDy[(U x )2 + (Vy )2 + (U y + Vx )2 ]1/2 (12)
2

where cm is the mixed coefficient.

2.2. 2DH nearshore current model

Based on computed characteristics of wave fields, the radiation stress terms can then be found
and input into the depth integrated (2DH) nearshore current module, which solves the depth-
and-wave-period averaged continuity and momentum equations at each local point on
horizontal plane, for calculating wave driven current:

∂η ∂ ∂
+ U (h + η ) + V (h + η ) = 0 (13)
∂t ∂x ∂y

∂U
∂t
+U
∂U
∂x
+V
∂U
∂y
= fV − g +
∂x ρ ∂x
(
∂ η 1 ∂ τxx ∂ τ yx
+
∂y
+ )
(14)
1 1
ρ h + η ) ( sx bx ) ρ (h + η ) ∂ x
+ ( τ −τ − (
∂ S xx ∂ S yx
+
∂y
)
∂V
∂t
+U
∂V
∂x
+V
∂V
∂y
= − fV − g +
∂y ρ ∂x
(
∂ η 1 ∂ τxy ∂ τ yy
+
∂y
+ )
(15)
1 1
ρ h + η ) ( sy by ) ρ (h + η ) ∂ x
+ ( τ −τ − (
∂ S xy ∂ S yy
+
∂y
)
where U and V are depth-integrated nearshore current velocities in x and y direction
respectively, Sxx, Sxy and Syy are radiation stress tensor, g is acceleration due to gravity, ρ is
water density, h is water depth, η is water surface elevation, τxx, τxy and τyy are Reynolds stress
tensor, τs and τb are shear stress on surface and bottom. The friction factor for combined wave-
current flow in bottom shear stresses and the mixing coefficient in Reynolds stresses are
suggested by Chiang et al. (2010).
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 255
[Link]

3. Quasi-3D extended: 1DV velocity model

Fig. 4 depicts the coordination of the quasi-3D hydraulic system. The x coordinate is defined
in the cross-shore direction towards shore. The y coordinate denotes the long shore direction.
The Z coordinate is toward from sea bed to surface in depth-direction.

3.1. Numerical formulation

The distributions of the velocity profiles in the long-shore and cross-shore direction at each
local point along the vertical water column are found through the following momentum
equations:

¶u 1 ¶p ¶ æ ¶u ö 1 ¶t x
=- + n + (16)
¶t r ¶x ¶z çè t ¶z ÷ø r ¶z

¶v 1 ¶p ¶ æ ¶v ö 1 ¶t y
=- + n + (17)
¶t r ¶y ¶z çè t ¶z ÷ø r ¶z

where u and v are wave-period-averaged velocities across the water column in the cross-shore
¯
and long-shore direction respectively, p is the pressure, νt is the turbulence viscosity, τ x and
¯y
τ are wave-period-averaged wave-induced shear stress through water column. In equation
(16) and (17), the convection terms in the left hand side of the equations have been neglected
which follows the conventional 1DV type of models by assuming the spatial gradient at the
interested site is small. This is due to the fact that the convection terms have been counted in
the 2DH nearshore current model and the 1DV model is only used to resolve the vertical profile
of the flow velocities and sediment concentration in suspension.

Figure 4. The coordination of the quasi-3D hydraulic system

256 Sediment Transport Processes and Their Modelling Applications

The pressure gradient term may be divided into two components by inside/outside boundary
layer. The pressure gradient term can be easily calculated from the variation of cycle-mean
water free surface. Within the boundary layer, we can calculate from the time differential of
velocity at boundary layer:

y
¶p dU wx 0 ¶p dU w 0
= , = (18)
¶x dt ¶y dt

x y
where U w0 and U w0 is the wave orbit velocity outside the boundary layer.

The cycle-mean wave-induced shear stress under waves is consist of wave motion component
(τw), wave breaking surface roller component (τr), boundary layer streaming component (τb)
and mean water surface changed (wave set-up/set-down) component (τsu), as shown in Fig.
5. The Boundary layer streaming components can be neglected, because it is small than the
others. The total shear stress is defined as

t = t w + t r + t su (19)

The shear stress distribution due to wave motion (τw) is in accordance with the derivations of
Deigaard and Fredsoe (1989):

-1 dE f æ h-yö
tw = ç1 + ÷ (20)
c dx è 2h ø

where Ef is cycle-mean energy flux due to wave motion (Svendsen, 1984):

2
E f = r gch = Br gcH 2 (21)

The coefficient B is 1/12 while wave breaking and 1/8 in general.

The shear stress due to wave breaking (τw) can be calculated by the concept of surface roller.
According to Svendsen (1984) and Deigaard et al. (1986), the shear stress of surface roller is
assumed to be constant by experiment:

r d( Ac )
tr = - (22)
T dx

where c is wave celerity, T is wave period, and A is the area of surface roller (as shown in
Fig. 7) can be easily calculated by A=0.09H (Deigaard et al., 1986).
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 257
[Link]

Figure 5. Total shear stress and its component from surface to bed in vertical direction under wave

The shear stress due to variation of mean water level (τsu) can be easily calculated by

æ z ö ¶h
t su = - r gh ç 1 - ÷ (23)
è h ø ¶x

The turbulence viscosity should be specified through certain turbulence models. In the present
study, the one-equation k-closure is adopted as follows:

¶k ¶ æ n tb ¶k ö éæ ¶u ö 2 æ ¶v ö 2 ù P
r
= çç ÷÷ + n tb êç ÷ + ç ÷ ú + - e (24)
¶t ¶z è s k ¶z ø êëè ¶z ø è ¶z ø úû r

where k is the turbulent kinetic energy, constant σk=1.0, Pr is the turbulent production due to
surface wave breaking by surface-roller concept and ε is dissipation rates which is given as:

e = c1 k 3/2 / l (25)

in which l is the turbulence length scale computed as (Deigaard et al., 1991):

ìïc 1/ 4k z
l=í 1
( )
, z £ lmax / k c11/ 4
(26)
îï lmax , z > lmax / (k c )
1/ 4
1

where lmax=0.1h, c1=0.09.

258 Sediment Transport Processes and Their Modelling Applications

Figure 6. The concept of surface roller due to wave breaking (Deigaard et al., 1986)

Figure 7. The turbulence generated and energy dissipation due to wave breaking (Deigaard et al., 1986)

The turbulence generated and energy dissipation at the water surface due to wave breaking
is computed following Deigaard et al. (1991), as shown in Fig. 7 :

Pr = a p DISS (27)

where constant αp = 0.33, the energy dissipation term DISS is suggested by Deigaard (1989):

DISS = r gcd r tan f0 (28)

where c is wave celerity, φ0 is wave initial breaking angle as 10 degree, δr is thickness of surface
roller head suggested by Deigaard et al. (1986):

2 ( h + H / 2)
dr = (29)
L
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 259
[Link]

3.2. Boundary condition and key parameters

3.2.1. Eddy viscosity

According to Brøker et al. (1991), the eddy viscosity is calculated by assuming the total kinetic
energy could be the sum of three contributions: the oscillatory near bed boundary layer (νtw ),
wave breaking (νtb), and the time-averaged currents (νtU ) respectively. The eddy viscosity
outside and inside the wave boundary layer is calculated by eq. (30), and eq. (31).

n t = n tU 2 + n tb 2 (30)

n t = n tw 2 + n tb 2 (31)

The component of oscillatory near bed boundary layer (νtw ) is calculated from (Okayasu et al.,
1988)

n tw = zk U f (32)

where κ=0.4 is von karman constant, U f = τb / ρ is the frictional velocity under wave.

The component of wave breaking (νtb) is calculated from (Rakha, 1998):

n tb = l k (33)

where k is turbulence kinetic energy, and l is turbulence length scale.

The contribution of the time-averaged currents (νtU ) is calculated also from (Rakha, 1998):

2 2
2 æ du ö æ dv ö
n tU = lmax ç ÷ +ç ÷ (34)
è dz ø è dz ø

where lmax is the same as eq. (26).

3.2.2. Thickness of wave boundary layer

The thickness of wave boundary layer (δw) is calculated from (Soulsby et al., 1993):
260 Sediment Transport Processes and Their Modelling Applications

u* maxT
dw = (35)
2p

where u*max = τmax / ρ, τmax is the maximum shear stress by waves and currents.

4. Sediment transport formula

The total sediment transport is consisted of bed load and suspended load suggested by Chiang
et al. (2011):

qtotal = qb + qs (36)

where qb is the bed load sediment transport rate, and the qs is the suspended load sediment
transport rate.

4.1. Bed load sediment transport

Following Chiang et al. (2011), the instantaneous bed load transport rate due to wave
asymmetric and ripple-bed effects is given as:

1/2
qb = F b é g ( s - 1) d50
3 ù
(37)
ë û

1/2
æ 2 2 ö
(38)
F b = ç F bx + F by ÷
è ø

F bx = 11 étc (q cx - q c )
1.65 1.65 ù
- tt (qtx - q c ) (39)
êë úû

( )
1.65
F by = 11 q y - q c (40)

where g being the gravity acceleration, s the relative density (s=ρs/ρ, with ρs the density of
sediment), d50 is median diameter, Φb is dimensionless sediment transport rates, subscript x, y
indicates along wave propagating and perpendicular direction. The wave-crest-half period,
wave-trough-half period, and perpendicular wave-period averaged equivalent Shields stress
are suggested by Chiang et al. (2011) as following:
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 261
[Link]

1 ( fcw )cx ucx

2
qcx = (41)
2 ( s - 1) gd50

1 ( fcw )tx utx

2
qtx = (42)
2 ( s - 1) gd50

1 fcU0 sin (f )
2 2
qy =
2 ( s - 1) gd50
(43)

where fw, fc, and fcw are the friction coefficient of wave, current and wave-current interaction,
ucx and utx are equivalent phase-averaged near-bed velocity under crest and trough half period.

ì é -0.19 ù
ï0.00251 ´ exp ê 5.21 ´ exp çæ aw ÷ö ú,
aw
> 1.57
ïï ê çK ÷ ú Ks
è s ø
fw = í ë û (44)
ï aw
ï 0.3, £ 1.57
ïî Ks

2
é 0.4 ù
fc = 2 ê ú (45)
ëê ln ( h / z0 ) - 1 ûú

( fcw )i = e i fc + (1 - e i ) fwi (46)

where i = c(crest) or t(trough), and weigh coefficient given as:

U0
ec = (47)
U w max + U0

U0
et = (48)
U w min + U0

In eq. (44), aw is half wave orbit closure:

U wT
aw = (49)
2p
262 Sediment Transport Processes and Their Modelling Applications

According to Soulsby (1997), the total roughness (Ks) of sea bed during wave passing is
consisted of grain related component (Kss), form drag component (Ksf) and sediment transport
component (Kst):

K s = K ss + K sf + K st (50)

In eq. (37), the grain related component is given as (Nielson, 1992 and Soulsby, 1997),

K ss = 2.5 ´ d50 (51)

The form drag component associated with sandy ripples is defined (Davis and Villaret, 2003):

K sf = 25hr (hr / lr ) (52)

where the λr and ηr is the wave length and wave height of full-developped sandy ripples, the
empirical relationship formula is shown as Nielson (1992). The sediment transport component
of friction roughness (Kst) is given as (Wilson, 1989):

K st = 5q s d50 (53)

In eq. (40), d50 is the sediment median grain size, and θs is the entraining bed shear-stress.

4.2. Suspended load sediment transport

The suspended sediment transport can be calculated by integrated sediment concentration (C)
of vertical water column from bottom to surface. The sediment concentration C in the water
column is found through the mass conservation equation:

¶C ¶C ¶ æ ¶C ö
= ws + ç e sd ÷ (54)
¶t ¶z ¶z è ¶z ø

where ws is settling velocity, εsd is the coefficient of sediment diffusion. According to Fredsøe
and Deigaard (1992), εsd is equivalent to turbulent eddy viscosity νt. The settling velocity is
defined as (Soulsby, 1997):

n é
( ) ù
1/2
ws = 10.36 2 + 1.049 D*3 - 10.36 ú (55)
d50 êë û
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 263
[Link]

where D* is the dimensionless grain diameter:

1/3
é g ( s - 1) ù
D* = ê ú d50 (56)
2
ëê n ûú

After the suspended sediment concentration and vertical velocity profiles obtained, the
instantaneous suspended transport rate can be evaluated as:

h
qs = ò c ( z )u ( z ) dz (57)
za

where za is the The reference height a is specified as 2.5d50 (Soulsby, 1997), h is the water depth.

5. Model validation and discussion

The quasi-3D sediment tranport model is validated against wave flume tests and existed
numerical models.

5.1. Model validation with Cox and Kobayashi (1996)

The wave and quasi-3D model system described above was firstly tested by regular wave with
uniform sloped bed, and compared with experiment (Cox and Kobayashi, 1996). The test
conditions are given: wave height 13.2cm, wave period 2.2 sec, the length of wave flume is
14.0m, width is 1.5m, depth is 30.0cm, slope-1:35; sand medium diameter is 1.0mm, and bed
roughness height ks=1mm. The numerical results of wave height and layouts of wave flume is
shown as Fig. 8. In Fig. 8, the black triangles indicate experimental data by Cox and Kobayashi
(1996), the red line is the results of present wave model, and the black line is the numerical
results from Rakha (1998). The wave nonlinear effects and wave breaking and regenerating
can be observed well, it’s shown good agreement with experiment.

Figure 8. The validation of wave height and the layout of wave flume with results from Cox and Kobayashi (1996)
264 Sediment Transport Processes and Their Modelling Applications

The comparisons of the vertical velocity profile at various position used by Cox and Kobayashi
(1996) are also adopted here to assess the present model’s accuracy for wave propagating at a
slope in wave flume. The velocity profiles of various position are shown as Fig. 9 ~ Fig. 14. The
black triangles also indicate experimental data by Cox and Kobayashi (1996), the blue line is
the results of present 1DV model, and the black line is the numerical results from Rakha
(1998). Compared with experiments, they are shown good performance for the validation of
distribution of the vertical velocity profile before and after wave breaking.

Figure 9. The validation of vertical velocity profile at wave flume x=4.2m from Cox and Kobayashi (1996)

Figure 10. The validation of vertical velocity profile at wave flume x= 6.6m from Cox and Kobayashi (1996)

Figure 11. The validation of vertical velocity profile at wave flume x=7.8m from Cox and Kobayashi (1996)
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 265
[Link]

Figure 12. The validation of vertical velocity profile at wave flume x= 9.0m from Cox and Kobayashi (1996)

Figure 13. The validation of vertical velocity profile at wave flume x= 10.2m from Cox and Kobayashi (1996)

Figure 14. The validation of vertical velocity profile at wave flume x= 11.4m from Cox and Kobayashi (1996)

5.2. Model validation with Ting and Kirby (1994)

The Boussinesq wave model and quasi-3D nearshore current model system described above
was tested by regular wave with uniform sloped bed, and compared with experiment (Ting
and Kirby, 1994, test 1). The test conditions are given: wave height 12.5cm, wave period 2.0
sec, the length of wave flume is 13.0m, width is 1.5m, depth is 30.0cm, slope-1:35; sand medium
diameter is 1.0mm, and bed roughness height ks=1mm. The numerical results of wave height
and layouts of wave flume is shown as Fig. 15. In Fig. 15, the black triangles indicate
266 Sediment Transport Processes and Their Modelling Applications

experimental data by Ting and Kirby (1994), the red line is the results of present wave model,
and the black line is the numerical results from Rakha (1998). The wave nonlinear effects and
wave breaking and regenerating can be observed well, it’s also shown good agreement with
experiments.

Figure 15. The validation of wave height and the layout of wave flume with results from Ting and Kirby (1994), test 1

The comparisons of the vertical velocity profile at various position used by Ting and Kirby
(1994) are also adopted here to assess the present model’s accuracy for wave propagating at a
slope in wave flume. The velocity profiles of various position are shown as Fig. 16 ~ Fig. 21.
The black triangles also indicate experimental data by Ting and Kirby (1994), the blue line is
the results of present 1DV model, and the black line and dotted line are the numerical results
from Rakha (1998) model A and model B. Compared with experiments, they are shown good
performance for the validation of distribution of the vertical velocity profile before and after
wave breaking. Because of full-nonlinear Bossinesq wave model and accuracy wave breaking
dissipation terms in 1DV model, the present model performed well than Rakha (1998).

5.3. Validation for sediment transport calculations

To assess the present quasi-3D sediment transport model’s ability of prediction for local
sediment transport under combined waves and currents for a range conditions, a series of

Figure 16. The validation of vertical velocity profile at wave flume x=5.945m with results from Ting and Kirby (1994)
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 267
[Link]

Figure 17. The validation of vertical velocity profile at wave flume x=6.665m with results from Ting and Kirby (1994)

Figure 18. The validation of vertical velocity profile at wave flume x=7.275m with results from Ting and Kirby (1994)

Figure 19. The validation of vertical velocity profile at wave flume x=7.885m with results from Ting and Kirby (1994)

Figure 20. The validation of vertical velocity profile at wave flume x=8.495m with results from Ting and Kirby (1994)
268 Sediment Transport Processes and Their Modelling Applications

Figure 21. The validation of vertical velocity profile at wave flume x=9.110m with results from Ting and Kirby (1994)

experiments in the wave flume have been observed by Dibajnia and Watanabe (1998), Dibajnia
and Kioka (2000), and Dibajnia et al. (2001). All of experimental data are shown as Table 1.

The comparison has been carried out between the period-averaged net sediment transport and
experimental results for different wave-current conditions and the results are reasonable
accurate within a factor of 2, as shown in Fig. 22. There are 66% numerical results greater than
experiments, and 8% numerical results out of compared accurate factor of 2. They are shown
good performance to predict the local sediment transport rates in nearshore region.

D50(cm) 2Uw(cm/s) T(sec) umax/2Uw 2Tpc/T U0(cm/s) q(cm2/s)

0.02 163.8 3.9 0.67 0.29 0 0.133

0.02 148.2 3.9 0.68 0.29 0 0.094

0.02 136.1 3.9 0.68 0.29 0 0.079

0.02 126.9 3.9 0.67 0.29 0 0.042

0.02 103.9 3.7 0.59 0.41 0 0.014

0.02 119.4 3.7 0.59 0.41 0 0.055

0.02 128.5 3.7 0.59 0.41 0 0.048

0.02 136.5 3.8 0.58 0.42 0 0.051

0.02 140.5 3.5 0.68 0.32 0 0.08

0.02 129 3.5 0.68 0.32 0 0.049

0.02 121.2 3.6 0.68 0.31 0 0.061

0.02 108.4 3.6 0.6 0.39 0 0.031

0.02 144.8 3.8 0.58 0.42 0 0.072

0.02 122.8 3.6 0.68 0.31 0 0.041

0.02 117.1 3.5 0.59 0.41 0 0.032

0.02 119.8 3.5 0.59 0.41 0 0.046

0.02 114.6 3.6 0.60 0.4 0 0.044

0.02 109.5 3.9 0.67 0.29 0 0.062

 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 269
[Link]

D50(cm) 2Uw(cm/s) T(sec) umax/2Uw 2Tpc/T U0(cm/s) q(cm2/s)

0.02 126.3 3.5 0.59 0.41 0 0.032

0.02 105.6 3.6 0.68 0.31 0 0.043

0.02 137.3 3.5 0.67 0.33 0 0.086

0.02 122.9 3.9 0.67 0.29 0 0.066

0.02 112.7 3.7 0.59 0.41 0 0.046

0.02 135.9 3.5 0.59 0.41 0 0.069

0.02 119.2 3.9 0.68 0.3 16.3 0.117

0.02 114.7 3.8 0.58 0.42 14.3 0.095

0.02 116.1 3.7 0.59 0.41 5.6 0.063

0.02 118.8 3.9 0.68 0.29 11 0.1

0.02 189.5 4 0.65 0.31 0 0.21

0.02 184.2 3.8 0.64 0.35 0 0.171

0.02 167.5 3.8 0.65 0.33 0 0.134

0.02 151.7 3.8 0.67 0.31 0 0.109

0.02 183.2 3.6 0.65 0.34 0 0.142

0.02 183.2 3.6 0.65 0.35 0 0.139

0.02 179.3 3.6 0.65 0.34 0 0.123

0.02 175.6 3.6 0.65 0.34 0 0.103

0.02 165.5 3.6 0.65 0.33 0 0.125

0.02 163.4 3.6 0.65 0.34 0 0.119

0.02 155.6 3.6 0.66 0.33 0 0.082

0.02 146.4 3.6 0.65 0.33 0 0.075

0.055 275.3 4.2 0.57 0.43 0 1.184

0.055 265 4.1 0.59 0.41 0 0.773

0.055 239.1 4 0.62 0.37 0 0.683

0.055 208.3 4 0.64 0.36 0 0.439

0.055 264.8 3.6 0.59 0.41 0 0.634

0.055 260.1 3.6 0.61 0.39 0 0.534

0.055 250.1 3.6 0.62 0.38 0 0.56

0.055 225.3 3.6 0.63 0.37 0 0.459

0.08 280.5 4.2 0.57 0.43 0 1.373

0.08 276.3 4.1 0.57 0.43 0 1.44

0.08 270.2 3.6 0.57 0.43 0 1.137

0.08 264.2 4 0.58 0.42 0 0.8

Table 1. Experiments from Dibajnia and Watanabe (1998), Dibajnia and Kioka(2000), and Dibajnia et al. (2001).
270 Sediment Transport Processes and Their Modelling Applications

Figure 22. Measured against calculated sediment transport rates with present model and experimental results from
Dibajnia and Watanabe (1998), Dibajnia and Kioka(2000), and Dibajnia et al. (2001).

6. Conclusion

In this chapter, a quasi-3D numerical model has been developed to predict sand transport in
the coastal region. The whole model system is consisted of a fully nonlinear Bossinesq wave
module, 2HD depth-integral wave-driven current module, 1DV velocity profile module, and
the sediment transport formula. Numerical results indicate that the wave breaking and
regenerating are good agreement with experiments. The phenomenon of undertow for wave
breaking or not is performed well, and the vertical velocity profiles are shown good accuracy
with experiments. The numerical results of sediment transport have been compared with
experiment and obtained are reasonably accurate within a factor of 2. The quasi-3D sediment
transport model system was then used to simulate several laboratory studies to test its ability
to reproduce the important nearshore morphodynamic processes.
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 271
[Link]

Acknowledgements

The work was partially financially supported by the National Science Council (Taiwan, R.O.C.)
Project: NSC99-2218-E-320- 002-MY3.

Author details

Yun-Chih Chiang1 and Sung-Shan Hsiao2

1 Tzu Chi University, Taiwan

2 National Taiwan Ocean University, Taiwan

References

[1] Almar, R, Castelle, B, Ruessink, B. G, Sénéchal, N, Bonneton, P, & Marieu, V. (2010).

Two- and three-dimensional double-sandbar system behaviour under intense wave
forcing and a meso-macro tidal range”, Continental Shelf Research, , 30, 781-792.

[2] Briand, M, & Kamphuis, J. W. (1993). Waves and currents on natural beaches: a quasi
3d numerical model,” Coastal Eng. 20, , 101-134.

[3] Brøker, I, & Stive, M. J. F. (1991). Validation of hydrological models, phase 1, Danish
Hydraulic Institute, Horsholm, Denmark.

[4] Brøker, I, Deigaard, R, & Fredsøe, J. (1991). Onroffshore sediment transport and
morphological modelling of coastal profiles,” Proc. ASCE Specialty Conf. Coastal
Sediments’ 91, Seattle, WA, , 643-657.

[5] Chiang, Y. C, Hsiao, S. S, & Lin, M. C. D Sediment Transport Modeling for Coastal
Morphodynamics”, Proc. of the 21 (2011) International Offshore and Polar Eng.
Conference, Maui, Hawaii, USA, June 19-24, , 1053-1058.

[6] Cox, D, & Kobayashi, N. (1996). Undertow profiles in the bottom boundary layer under
breaking waves.” 25th Int. Conf. Coastal Eng., ASCE , 3, 3194-3206.

[7] Davies, A, & Thorne, P. (2002). DV modelling of sand transport by waves and currents
in the rippled bed regime,” In: Coastal Eng. Conf. 2002 (ICCE). ASCE, New York, ,
2600-2611.

[8] Davis, A. G, & Villaret, C. (2003). Sediment Transport Modelling for Coastal
Morphodynamics,” Proc. of the International Conference on Coastal Sediments, 2003,
Florida, USA, , 1-14.
272 Sediment Transport Processes and Their Modelling Applications

[9] Deigaard, D, Fredsøe, J, & Hedegaard, I. B. (1986). Suspended sediment in the surf
zone,” J. of Waterway, Port, Coastal and Ocean Eng., ASCE, , 112, 115-128.

[10] Deigaard, R, & Fredsøe, J. (1989). Shear stress distribution in dissipative water waves,”
Coastal Eng., , 13, 357-378.

[11] Deigaard, R, Justesen, P, & Fredsøe, J. (1991). Modelling of undertow by a one-equation

turbulence model.” Coastal Eng., , 15, 431-458.

[12] Deigaard, R. (1993). A note on the three-dimensional shear stress distribution in a surf
zone.” Coastal Eng. , 20, 157-171.

[13] De Vriend, H. J, & Stive, M. J. F. (1987). Quasi-3D Modelling of Nearshore Currents,”

Coastal Eng., , 11, 565-600.

[14] Dibajnia, M, & Watanabe, A. (1996). A transport rate formula for mixed-size sands,”
Proc. 25th Int. Conf. on Coastal Eng., , 3791-3804.

[15] Dibajnia, M, & Watanabe, A. (1998). Transport rate under irregular sheet flow
conditions,” Coastal Eng. , 35, 167-183.

[16] Dibajnia, M, & Kioka, W. (2000). Long waves and the change in cross-shore sediment
transport rates on a sheet flow dominated beach,” Coastal Eng. Jpn., JSCE & World
Scientific, , 42-1, 87-110.

[17] Dibajnia, M, & Watanabe, A. (2001). A representative wave model for estimation of
nearshore local transport rate,” Coastal Eng. Jpn. , 43, 1-38.

[18] Drønen, N, & Deigaard, R. (2000). Three dimensional near-shore bar morphology.”
Proc. 27th Int. Conf. on Coastal Eng. ASCE, Sydney, , 3205-3217.

[19] Drønen, N, & Deiggard, R. (2007). Quasi-three-dimensional modelling of the

morphology of longshore bars,” Coastal Eng. , 54, 197-215.

[20] Elfrink, B, Broker, I, Deigaard, D, Hansen, E, & Justesen, P. (1996). Modelling of q3d
sediment transport in the surf zone,” In: Edge, B.L. (Ed.), Coastal Eng. Conf. ASCE,
Orlando, Florida, USA, , 3805-3817.

[21] Elgar, S. (1987). Relationships involving third moments and bispectra of a harmonic
process,” IEEE Trans. on Acoustics, Speech, and Signal Processing, 35, , 1725-1726.

[22] Elgar, S, Guza, R. T, & Freilich, M. H. (1988). Eulerian measurements of horizontal

accelerations in shoaling gravity waves,” J. Geophys. Res., , 93, 9261-9269.

[23] Elgar, S, Gallagher, E. L, & Guza, R. T. (2001). Nearshore sandbar migration,” J. of

Geophysical Research, , 106(C6), 11623-11627.

[24] Fernando, P. T, & Pan, S. (2005). Modelling wave of hydrodynamics around a scheme
of detached leaky breakwaters,” In: Smith, J.M. (Ed.), Proceeding of the 29th International
Conference on Coastal Eng.. World Scientific, Lisbon, Portugal, , 830-841.
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 273
[Link]

[25] Fredsøe, J, & Deigaard, R. (1992). Mechanics of Coastal Sediment Transport, World
Scientific, Singapore.

[26] Hoefel, F, & Elgar, S. (2003). Wave-induced sediment transport and sandbar
migration”, Science, , 299, 1885-1887.

[27] Houser, C, & Greenwood, B. (2007). Onshore Migration of a Swash Bar During a Storm”,
J. of Coastal Research, , 23, 1-14.

[28] Hsu, T, Elgar, J, Guza, S, & Wave-induced, R. T. sediment transport and onshore
sandbar migration”, Coastal Eng., , 53, 817-824.

[29] Kuroiwa, M, Noda, H, & Matsubara, Y. (1998). Applicability of a quasi-three

dimensional numerical model to nearshore currents,” Proc. of 26th ICCE, , 815-828.

[30] Kennedy, A. B, Chen, Q, Kirby, J. T, & Dalrymple, R. A. (2000). Boussinesq Modeling

of Wave Transformation, Breaking, and Runup. I: 1D," Journal of Waterway, Port, Coastal
and Ocean Eng., ASCE, , 126(1), 39-47.

[31] Li, M, Fernando, P. T, Pan, S, Connor, O, & Chen, B. A. D., ((2007). Development of a
quasi-3d numerical model for sediment transport prediction in the coastal region.”
Journal of Hydro-environment Research, , 1, 143-156.

[32] Lin, M. C, Kuo, J. C, Chiang, Y. C, & Liou, J. Y. (1996). Numerical Modeling of

Topography Changes in Sea Region,” Proc. 18th Conf. on Ocean Engineering in Republic
of China, Nov. 1996, , 627-637.

[33] Lin, M. C, Chiang, Y. C, & Hsiao, S. S. (2009). A study of the sediment transport formula
under combined wave and current conditions.” J. of Coastal and Ocean Eng., , 9(2),
177-205.

[34] Long, W, Kirby, J. T, & Hsu, T. J. ((2006). Cross shore sandbar migration predicted by
a time domain Boussinesq model incorporating undertow”, Coastal Eng. Conf., ,
2655-2667.

[35] Nielsen, P. (1992). Coastal Bottom Boundary Layers and Sediment Transport, Advanced
Series on Ocean Engineering. Singapore: World Scientific Publication., 4

[36] Nwogu, O. (1993). An Alternative Form of the Boussinesq Equations for Modeling the
Propagation of Waves from Deep to Shallow Water," J. of Waterway, Port, Coastal and
Ocean Eng., ASCE, , 119(6), 618-638.

[37] Okayasu, A, Shibayama, T, & Horikawa, K. (1988). Vertical variation of undertow in

the surf zone,” Proc. 21st Int. Conf. on Coastal Eng., ASCE, Malaga, , 478-491.

[38] Okayasu, A, & Katayama, H. (1992). Distribution of undertow and long-wave

component velocity due to random waves,” Proc. of 23rd Int. Conf. on Coastal Eng.,
ASCE, , 883-893.
274 Sediment Transport Processes and Their Modelling Applications

[39] Okayasu, A, Hara, K, & Shibayama, T. (1994). Laboratory experiments on 3-D nearshore
currents and a model with momentum flux by breaking wave,” Proc. 24th Int. Conf.
Coastal Eng., , 2461-2475.

[40] Pape, L, Plant, N. G, & Ruessink, B. G. (2010). On cross-shore migration and equilibrium
states of nearshore sandbars,” J. of Geophysical Research, , 115, 1-15.

[41] Rakha, K. A. d phase resolving hydrodynamic and sediment transport model.” Coastal
Eng. , 34, 277-311.

[42] Roelvink, J. A, & Reniers, A. J. H. M. (1995). LipllD Delta Flume Experiments, a profile
dataset for profile model validation. Report H2130, January 1995, Delft Hydraulics, The
Netherlands.

[43] Ruessink, B. G, Kuriyama, Y, Reniers, A. J. H. M, Roelvink, J. A, & Walstra, D. J. R.

(2007). Modeling cross- shore sandbar behavior on the timescale of weeks”, J. Geophys.
Res., 112, F03010.

[44] Ruessink, B. G, & Kuriyama, Y. (2008). Numerical predictability experiments of cross-

shore sandbar migration”, Geo. Research Letters, , 35, 16-35.

[45] Ruessink, B. G, Pape, L, & Turner, I. L. (2009). Daily to interannual cross-shore sandbar
migration: Observations from a multiple sandbar system”, Continental Shelf Research, ,
29, 1663-1677.

[46] Sanchez-arcilla, A, Collado, F, & Rodriguez, A. (1992). Vertically varying velocity field
in Q-3D nearshore circulation," Proc. of Int. Coastal Eng. Conference, , 2811-2824.

[47] Soulsby, R. L. (1997). Dynamics of marine sands, Thomas Telford Publications, London.

[48] Soulsby, R. L, Hamm, L, Klopman, G, Myrhaug, D, Simons, R, & Thomas, G. (1993).

Wave-current interaction within and outside the bottom boundary layer,” Coastal
Eng. , 21, 41-69.

[49] Splinter, K. D, Holman, R. A, & Plant, N. G. dynamic model for sandbar migration and
2DH evolution”, J. of Geophysical Research, , 116, 10-20.

[50] Svendsen, I. A. (1984). Wave heights and set-up in a surf zone.” Coastal Eng. , 8, 303-329.

[51] Svendsen, I. A, & Lorenz, R. S. (1989). Velocities in combined undertow and longshore
currents,” Coastal Eng., , 13, 55-79.

[52] Ting, F. C. K, & Kirby, J. T. (1994). Observation of undertow and turbulence in a

laboratory surf zone.” Coastal Eng. , 24, 51-80.

[53] Trowbridge, J, & Madsen, O. S. (1984). Turbulent Wave Boundary Layers 1. Model
Formulation and First- Order Solution,” J. Geophys. Res., C5), , 89, 7989-7997.

[54] Trowbridge, J, & Young, D. (1989). Sand transport by unbroken waves under sheet flow
conditions,” J. Geophys. Res., , 94, 10.
 Quasi-3D Modeling of Sediment Transport for Coastal Morphodynamics 275
[Link]

[55] Wai, O, Chen, Y, & Li, Y. driven coastal sediment transport model.” Coastal Eng.
Journal, , 46, 385-424.

[56] Wei, G, Kirby, J. T, Grilli, S. T, & Subramanya, R. (1995). A Fully Nonlinear Boussinesq
Model for Surface Waves. Part 1. Highly Nonlinear Unsteady Waves," J. Fluid Mech., ,
294, 71-92.

[57] Wilson, K. C. (1989). Friction of wave induced sheet flow,” Coastal Eng. , 12, 371-379.
 Chapter 11

Derivation of Sediment Transport Models for

Sand Bed Rivers from Data-Driven Techniques

Vasileios Kitsikoudis,
Epaminondas Sidiropoulos and Vlassios Hrissanthou

Additional information is available at the end of the chapter

[Link]

1. Introduction

Hydraulic engineers and geologists have studied sediment transport in natural streams and
rivers for centuries due to its importance in understanding river hydraulics. Erosion and
deposition of sediment alters the hydraulic geometry of the channel and may cause increase
of flood frequency as well as navigation problems from excessive deposition. Moreover, dis‐
charge of industrial and agricultural residuals sets the sediment particles to be the primary
transporters of toxic substances that contaminate aquatic systems. High sediment discharge
peaks may be destructive for fish habitats and ecosystems, and long-term sediment yield af‐
fects the design and function of constructions such as dams and reservoirs, as well as the
coastal erosion at the basin outlet.

Sediment transport in sand bed rivers and natural streams is a complex process. For its
quantification, numerous sediment transport functions have been introduced in the past
years based on different concepts. There are four basic approaches used in the derivation
of sediment transport formulae (Yang, 1977): 1) The deterministic approach, which obeys
the laws of physics and usually is based on an independent variable like slope, shear
stress, stream power, unit stream power etc. 2) The regression approach, which has
emerged from the thought that sediment transport is such a complex phenomenon that
cannot be described by a single dominant variable. 3) The pioneering probabilistic ap‐
proach of Einstein (1942), which highlighted the complexity and the stochastic nature of
the sediment transport in a rather laborious way for common usage in engineering, and 4)
The regime approach, which was developed as a result of long-term measurements in
equilibrium conditions.
278 Sediment Transport Processes and Their Modelling Applications

The emerging results from all these concepts usually differ drastically from each other and
from the measured data. Consequently, none of the published sediment transport equations
has gained universal acceptance in confidently predicting sediment transport rates, especial‐
ly in rivers. An alternative approach may be the usage of data-driven modeling, which is
especially attractive for modeling processes, in which knowledge of the physics of the prob‐
lem is inadequate. The scope of this chapter is the utilization of some widely used data-driv‐
en techniques, namely artificial neural networks (ANNs) and symbolic regression based on
genetic programming (GP) in order to determine the dominant dimensionless variables that
can be used as inputs in such schemes and generate sediment transport models for natural
streams and rivers that are based solely on the data without presuming anything about their
structure and their degree of nonlinearity.

For the proper training of a data-driven scheme, data of good quality are needed. Since field
measurements accommodate the peculiarities of the considered streams and the inclusion of
noise in the measurement process is inevitable, the training data comprise solely laboratory
flume measurements. The testing data, however, comprise exclusively field measurements
in order to implement the models in actual applications. Based on this concept, the approach
of the basic trend of the function is feasible and the derived model will be applicable to the
data range for which it will be trained. Regarding the efficiency of scaling in the sediment
transport context, model-prototype comparisons have shown that correspondence of behav‐
ior is often well beyond expectations, as has been attested by the successful operation of
many structures designed from model tests (Pugh, 2008). This study exhibits the potential of
machine learning in capturing functions with physical meaning since the training and test‐
ing sets have significant differences in their statistical distributions. The determination of the
input variables that best define the problem is accomplished by the assessment of some
common independent dimensionless variables based on their correlation with the sediment
concentration and the aid of ANNs on the basis of a tentative trial-and-error procedure. Sub‐
sequently, ANNs and symbolic regression are utilized in order to derive equations from the
selected input combinations.

2. Data mining and data-driven techniques in the context of sediment

transport

The recorded observations of a system can be further analyzed in the search for the informa‐
tion they encode. Such automated search for models accurately describing data constitutes a
direction that can be identified as that of data mining. Data mining and knowledge discov‐
ery aim at providing tools to facilitate the conversion of data into a number of forms, such as
equations. The latter provide a better understanding of the process generating or producing
these data. These models combined with the already available understanding of the physical
processes result in an improved understanding and novel formulations of physical laws and
improved predictive capability (Babovic, 2000).
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 279
[Link]

Data-driven modeling (DDM) and machine learning techniques used for predictions are es‐
sentially modernized regression schemes with the significant advantage over the classical
regression schemes that they do not have to presume the structure of the nonlinear model,
which they attempt to fit. They are based on simple ideas, usually inspired from the way
nature works, and their only prerequisite is a good, although usually large, data set. The da‐
ta are usually divided into three sets, namely the training, validation and testing set. The
training set trains the scheme on the basis of a minimization criterion and the validation set
is used as a stopping criterion for training to avoid overfitting to the data used for training.
The test set is used to evaluate the generated model. The minimization criterion, on the basis
of which the training process takes place, is usually a sum of errors between the computed
outputs and the actual measured data. The optimization model that is used for the minimi‐
zation depends on the data-driven scheme and may be deterministic as well as stochastic.

Inferring models from data is an activity of deducing a closed-form explanation based solely
on observations. These observations, however, represent a limited source of information.
The question emerges as to how this, a limited flow of information from a physical system
to the observer, can result in the formation of a model that is complete in the sense that it
can account for the entire range of phenomena encountered within the physical system in
question and describe even the data outside the range of previously encountered observa‐
tions. The present efforts are characterized by the search for a model that is capable of ac‐
quiring semantics from syntax. Clearly, every model has its own syntax. Artificial neural
networks have the syntax of a network of interconnected neurons, whereas genetic pro‐
gramming has the syntax of treelike networks of symbolic expressions in reverse Polish no‐
tation. The question is whether such a syntax can capture the semantics of the system it
attempts to model (Babovic, 2000). Witten et al. (2011) argued that the universal learner is an
idealistic fantasy since experience has shown that no single machine learning scheme is ap‐
propriate to all data mining problems. Certain classes of model syntax may be inappropriate
as a representation of a physical system. One may choose the model whose representation is
complete, in the sense that a sufficiently large model can capture the data’s properties to a
degree of error that decreases with an increase in the model size. For example, one may de‐
cide to expand Taylor or Fourier series and decrease the error by adding terms in a series.
However, in these cases, semantics almost certainly would not be caught (Babovic, 2000).

2.1. Artificial neural networks

ANN is the most widely used data-driven method. Since abundant information on ANNs is
available in the literature [e.g. Haykin (2009)], only a brief description of ANNs is provided,
with regard only to the methodology applied herein. ANN is a broad term covering a large
variety of network architectures and structures. The most common of them, and the one uti‐
lized herein, is the multilayer feedforward network. This type of network is a parallel dis‐
tributed information processing system that consists of the input layer, the hidden layer(s),
and the output layer, and the information goes only in a forward direction. Each layer com‐
prises a number of neurons, each one of which is connected with those in the successive lay‐
er with synaptic weights that determine the strength of the connections. The hidden and
280 Sediment Transport Processes and Their Modelling Applications

output layer neurons have an inherent activation function, which accommodates the nonlin‐
ear transformation of the input data to the targets. In this study, the neurons of the hidden
layer(s) will have the hyperbolic tangent activation function, which squashes the data be‐
tween (-1, 1), and the single neuron of the output layer will have the linear activation func‐
tion, which simply returns the value that is passed to it. The input data are scaled to the
range (-0.9, 0.9) because, if the values are scaled to the extreme limits of the transfer func‐
tion, the size of the weight updates is extremely small and flat-spots in training are likely to
occur (Maier and Dandy, 2000).
The training process of an ANN may be viewed as a “curve fitting” problem and the net‐
work itself may be considered simply as a nonlinear input-output mapping (Haykin, 2009).
Supposing that a deterministic relation between sediment load concentration and some spe‐
cific independent variables exists, a multilayer feedforward ANN is able to approximate this
function, if it includes at least one hidden layer with a sufficient number of neurons (Hornik
et al., 1989). However, this universal approximation theorem does not specify if a single hid‐
den layer is optimal in the sense of learning time, ease of implementation, or (more impor‐
tantly) generalization (Haykin, 2009). As a result, several network architectures are tested in
order to determine the optimal one.
Although the implementation of ANNs is extensive and successful in water resources appli‐
cations [e.g. Maier and Dandy (2000)] and in the prediction of daily suspended sediment da‐
ta [e.g Cigizoglu (2004)], it is quite sparse in the prediction of sediment concentration from
other independent hydraulic variables. Nagy et al. (2002) reviewed some widely used sedi‐
ment discharge equations and selected some of the dominant dimensionless variables of the
problem as input neurons for an ANN that was trained and tested with field data. Bhatta‐
charya et al. (2005) used dimensionless parameters obtained from the Engelund and Hansen
(1967) formula in order to train and test an ANN with a mixture of flume and field data,
whilst in similar studies Bhattacharya et al. (2004, 2007) scrutinized further the possible in‐
put parameters based on the same data. Yang et al. (2009) chose as input variables combina‐
tions of dimensional quantities and applied them to field data. All of these works used the
back-propagation algorithm (Rumelhart et al., 1986) for training the ANNs and compared
the results with some of the most popular sediment transport formulae. For all the cases, the
ANNs generated superior results.

2.2. Symbolic regression based on genetic programming

Many seemingly different problems in artificial intelligence, symbolic processing and ma‐
chine learning can be viewed as requiring discovery of a computer program that produces
some desired outputs for particular inputs. The process of solving these problems can be re‐
formulated as a search for a highly fit individual computer program in the space of possible
ones. GP extends the concept of genetic algorithms and provides a way to search for this fit‐
test individual computer program (Koza, 1992).
GP works by randomly generating a population of computer programs (represented by tree
structures) and each individual program in the population is measured in terms of how well
it performs in the particular problem environment. This measure is called the fitness meas‐
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 281
[Link]

ure (Koza, 1992) and usually is a sum of errors between the outputs predicted by the pro‐
gram and the actual ones. Initially, the generated computer programs will have exceedingly
poor fitness. Nonetheless, some individuals in the population will turn out to be somewhat
fitter than others. These differences in performance are subsequently exploited. The Darwin‐
ian principle of reproduction and survival of the fittest and the genetic operations of sexual
recombination (crossover) and mutation are used to create a new offspring population of in‐
dividual programs from the current population. The reproduction principle involves the se‐
lection, in proportion to fitness, of a computer program from the current population that
survives from the generation by being copied into the new population. The genetic process
of sexual recombination is used to create new offspring programs from two parental pro‐
grams selected in proportion to fitness. The parental programs are typically of different sizes
and shapes. The offspring programs are composed of subexpressions from their parents and
are, typically, of different sizes and shapes as well. Intuitively, if two programs are some‐
what effective in solving a problem, then some of their parts probably have some merit. By
recombining randomly chosen parts of somewhat effective programs, the result may be the
production of new programs that are even fitter in solving the problem (Koza, 1992). Muta‐
tion serves the potentially important role of restoring lost diversity in a population by re‐
placing random subtrees of variable length with other random ones. Its purpose is to
prevent premature convergence to unsatisfactory solutions. After the operations of repro‐
duction, crossover and mutation are performed on the current population, the offspring
population replaces the old one. Each individual in the new population of programs is then
measured for fitness and the process is iterated for a predetermined number of generations.
This algorithm will produce populations of programs, which over many generations tend to
exhibit increasing average fitness in dealing with their environment. The individual com‐
puter program that performs best in the evolved generations is considered to be the fittest.

A multigene individual consists of multiple genes, each of which is a GP evolved tree. In multi‐
gene symbolic regression, each prediction ŷ of the output variable y is formed linearly by the
weighted output of each of the genes plus a bias term (Searson, 2009). Each tree is a function of
the input variables. Mathematically, a multigene regression model can be written as:

yˆ = d0 + d1 ´ tree1 + ... + dM ´ treeM (1)

where d0=bias (offset) term; d1, …, dM are the gene weights and M is the number of genes
comprising the current individual. The gene weights are automatically determined by a
least squares procedure for each multigene individual. The number and structure of the
trees is evolved automatically during a run (subject to user defined constraints) using the
training data. Hence, multigene symbolic regression combines the power of classical linear
regression with the ability to capture nonlinear behavior without needing to pre-specify the
structure of the nonlinear model. During a run, genes are acquired and deleted using a tree
crossover operator called two-point high level crossover. This allows the exchange of genes
between individuals and it is used in addition to the “standard” GP recombination opera‐
tors (Searson et al., 2010).
282 Sediment Transport Processes and Their Modelling Applications

GP has been implemented in hydraulic engineering in the last years with very good re‐
sults. Babovic and Abbott (1997) applied GP to some representative problems, while Ba‐
bovic and Keijzer (2000) highlighted the usage of GP as a data mining tool in which the
human expert interprets models suggested by the computer, aiming at knowledge discov‐
ery. Minns (2000) suggests that the symbolic expressions obtained from GP may be less
accurate than the ANN in mapping the experimental data. However, these expressions
may be more easily examined in order to provide insights into the processes that created
the data. In the context of sediment transport, Zakaria et al. (2010) applied gene-expres‐
sion programming, which is similar to multigene symbolic regression, to predict the total
bed material load for rivers using dimensional quantities from field data, and outper‐
formed some of the traditional sediment load formulae. Azamathulla et al. (2010) utilized
GP in order to predict the scour depth at bridge piers and obtained results superior to
those of ANNs and regression equations.

3. Sediment transport

Sediment load is the material being transported, and it can be divided into wash load and
bed material load. The wash load is the fine material of sizes, which are not found in appre‐
ciable quantities on the bed, and is not considered to be dependent on the local hydraulics of
the flow, but instead is dependent on the upstream supply. As a practical definition, the
wash load is considered to be the fraction of the sediment load finer than 0.062 mm. The bed
material load is the material of sizes, which are found in appreciable quantities on the bed
and it can be conceptually divided into the bed load (the portion of the load that moves near
the bed) and the suspended load (the portion of the load that moves in suspension), al‐
though the division is not precise. The consequent difficulty, however, to separate bed load
from turbulence dominated suspended load leads to a total load definition for the quantifi‐
cation of sediment transport in sand bed rivers. A dimensionless, commonly used measure
for sediment quantification is concentration by weight in parts per million (ppm), which is
the ratio of the sediment discharge to the discharge of the water-sediment mixture, both ex‐
pressed in terms of mass per unit time, here called Ct. This can be given as

r sQst
Ct = 106 (2)
rQ + r sQst

For practical reasons, the density of the water-sediment mixture is taken to be approximate‐
ly equivalent to the density of water. This approximation will cause errors of less than one
percent for concentrations less than 16000 ppm (Brownlie, 1981a).

The parameters governing a sediment transport process can be described by (Yalin, 1977)

qt = f ( V , D , d , S , g , r s , r ,n ) (3)
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 283
[Link]

Since the data-driven schemes are trained and validated with flume data but tested with
field data and in order to ensure dimensional consistency in the derived models, the input
and output variables should be dimensionless. Instead of applying dimensional analysis
and Buckingham’s π theorem, the independent variables of Eq. (3) will be introduced by
some common and well-known dimensionless variables that have physical meaning and
have been utilized for the creation of various sediment transport formulae. These variables
are directly related to quantities the engineer can readily visualize and measure; they are
listed as follows and summarized in Table 1.

Froude number, which gives a measure of the ratio of inertial forces to gravitational forces
of the flow. For the flume data, the depth will be the hydraulic radius of the bed which is
equivalent to the mean depth of an infinitely wide channel with the same slope, velocity and
bed friction as the flume, and is calculated according to the sidewall correction of Vanoni
and Brooks (1957). This elaboration is due to the fact that in flume experiments the sand cov‐
ered bed will generally be much rougher than the flume walls, and thus will be subjected to
higher shear stresses.

V
Fr = (4)
gD

Reynolds number, which gives a measure of the ratio of inertial forces to viscous forces of
the flow

VD
Re = (5)
n

Shear Reynolds number, the physical meaning of which, is the ratio of particle size to the
thickness of the viscous sublayer δ, because δ is proportional to v/U*.

U * d50
Re * = (6)
n

Dimensionless shear stress or Shields number

t
t* = (7)
(g s - g ) d50

Dimensionless grain diameter. It is a dimensionless expression for grain diameter that can
be derived by eliminating shear stress from the two Shields parameters (Shields, 1936); or
from the drag coefficient and Reynolds number of a settling particle, by eliminating the set‐
tling velocity; or dimensionally, with immersed weight of an individual grain, fluid density,
and viscosity as the variables (Ackers and White, 1973). The dimensionless grain diameter
284 Sediment Transport Processes and Their Modelling Applications

is, therefore, generally applicable to coarse, transitional, and fine sediments and is the cube
root of the ratio of immersed weight to viscous forces. Thus

13
é g ( g s g - 1) ù
dgr = d50 ê ú (8)
ëê n2 ûú

Dimensionless stream power. The power equation appears first to have been applied to
sediment transport by Rubey (1933) and later by Velikanov (1955). It was again suggested
by Knapp (1938), and was later introduced by Bagnold (1956) in a paper wherein the flow‐
ing fluid was regarded as a transporting machine. The available power supply, or time rate
of energy supply, to unit length of a stream is the time rate of liberation in kinetic form of
the liquid’s potential energy as it descends the gravity slope S. Denoting this power by Ω,
Bagnold (1966) derived the formula

W = r gQS (9)

The mean available power supply to the column of fluid over unit bed area, to be denoted
by ω, is therefore

W r gQS
w= = = r gDSV = t V (10)
W W

In order to define a dimensionless transport parameter that encapsulates Bagnold’s view of

sediment transport as a stream power related phenomenon, Eaton and Church (2011) devel‐
oped the following formula

w
w* = 3 2 (11)
r éë g ( g s g - 1) d ùû

Dimensionless unit stream power. Yang (1972) reviewed the basic assumptions used in the
derivation of conventional sediment transport equations. He concluded that the assumption
that sediment transport rate could be determined from water discharge, average flow veloci‐
ty, energy slope, or shear stress is questionable. Consequently, the generality and applicabil‐
ity of any equation derived from one of these assumptions is also questionable. The rate of
energy per unit weight of water, available for transporting water and sediment in an open
channel with reach length x and total drop of Y, is

dY dx dY
= = VS (12)
dt dt dx
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 285
[Link]

Yang (1972) defines the unit stream power as the velocity-slope product and argues that
the rate of work being done by a unit weight of water in transporting sediment must be
directly related to the rate of work available to a unit weight of water. Thus, total sedi‐
ment concentration or total bed material load must be directly related to unit stream pow‐
er. While Bagnold (1966) emphasized the power that applies to a unit bed area, Yang
(1972, 1973) emphasized the power available per unit weight of fluid to transport sedi‐
ments. The fact that sediment discharge or concentration is dominated by the unit stream
power has been confirmed by Vanoni (1978) as well. While Yang divided unit stream
power VS by fall velocity ωs to obtain a dimensionless variable, Vanoni (1978) divided the
product VS by (gd50)1/2. Both d50 and ωs are commonly used for describing the size of sedi‐
ment particles. However, d50 can only reflect the physical size of sediment particles, while
ωs can also reflect the interaction between sediment particles and water, which is affected
by particle shape, water viscosity and temperature. On the other hand, the computation of
fall velocity is problematic and a common source of errors. The emerging variables ex‐
pressing dimensionless unit stream power according to Yang and Vanoni are, respective‐
ly, the following

VS
ws (13)

and

VS
gd50
(14)

No Dimensionless variables

1 Froude number, Fr
2 Reynolds number, Re
3 Shear Reynolds number, Re*
4 Dimensionless shear stress, τ*
5 Dimensionless grain diameter, dgr
6 Dimensionless stream power, ω*
7 Yang’s dimensionless unit stream power, VS/ωs
8 Vanoni’s dimensionless unit stream power, VS/(gd50)1/2

Table 1. Dimensionless variables assessed for the determination of the dominant ones

Yang (1977, 2003) argued that total sediment discharge correlates best with unit stream pow‐
er based on the plots of Figure 1. Nonetheless, equations based on the other hydraulic varia‐
bles have been used successfully as well.
286 Sediment Transport Processes and Their Modelling Applications

Figure 1. Relationships between total sediment discharge and (a) water discharge, (b) velocity, (c) slope, (d) shear stress, (e)
stream power, and (f) unit stream power, for 0.93 mm sand in an 8 ft wide flume [obtained from Yang (2003)]

4. Data preparation and determination of the inputs

Since data-driven techniques require a large number of quality data that represent a wide
spectrum of the considered problem in order to be trained efficiently, the database assem‐
bled by Brownlie (1981b) is utilized. Brownlie’s (1981b) database contains 7027 records (5263
laboratory records and 1764 field records) in 77 data files. These data were subjected to a
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 287
[Link]

screening process similar to the one Brownlie (1981a) used for the derivation of his formula.
Firstly, the measurements that were not verified by Brownlie, were incorrect or incomplete,
were removed. Secondly, because only flows with sand beds were considered, median parti‐
cle sizes were limited to values between 0.062 mm and 2.0 mm. To avoid samples with large
amounts of gravel or fine, cohesive material, geometric standard deviations were restricted
to values smaller than 5, and some other constraints were imposed in order to reduce side‐
wall effects, eliminate shallow water effects, and overcome accuracy problems associated
with low sediment concentration. In addition to these, only flume measurements with uni‐
form flows were considered and supercritical flows were removed due to the subcritical
flows that usually prevail in nature, in sand bed rivers. Finally, the measurements with spe‐
cific gravity outside the quartz density range were neglected as well as measurements that
had extreme temperature values. Wherever the temperature was missing, a value of 15 oC
was used for the calculation of kinematic viscosity. For the laboratory data, the sidewall cor‐
rection of Vanoni and Brooks (1957) was utilized to adjust the hydraulic radius to eliminate
the effects of the flume walls. If sediment concentration is correlated with velocity, however,
the sidewall correction will be of little use. These restrictions are shown in Table 2.

Restriction Reason
0.062 mm ≤ d50 ≤ 2.0 mm Sand only

σg ≤ 5 Eliminate bimodal distributions

W/D > 4 Reduce sidewall effects (only for laboratory data)

R/d50 > 100 Eliminate shallow water effects

C > 10 ppm Accuracy problems associated with low concentration

Fr < 1 Subcritical flows

2.57 ≤ Specific gravity ≤ 2.68 Natural sediments

Table 2. Restrictions imposed on data

Since measurements in natural streams and rivers are notoriously difficult, and sometimes
inaccurate, and the inclusion of field data to the training set would result in a model applica‐
ble only to rivers similar to those the data were obtained from, field data are excluded from
the training set. Consequently, the training set consists solely of laboratory flume data so
that the noise embedded in the training set is minimized. The testing set, however, compris‐
es exclusively field data in order to test the derived mathematical models in actual problems
that occur in nature. With this technique, the generated models will have general applicabil‐
ity to the data range for which they are trained. The final database consists of 984 laboratory
records and 600 field records that lie within the range of the laboratory records that consti‐
tute the training set, due to the data sensitive nature of DDM.
Further pruning of the outliers in the training dataset and the subsequent increase of data
homogeneity would be beneficial for the training procedure, however, this would be at the
expense of the amount of training data, which are already significantly reduced from the
screening process. Since most DDM methods perform well when the data has a distribution
that is close to normal (Bhattacharya et al., 2005), a log-transformation of the input and out‐
288 Sediment Transport Processes and Their Modelling Applications

put variables of all datasets was applied so that the distributions of the transformed varia‐
bles were closer to normal. Figure 2 depicts the distribution of the flume sediment
concentrations for the original and the log-transformed values.

Figure 2. Distribution of flume sediment concentration in ppm (a) before log-transformation and (b) after log-trans‐
formation

For the creation of training and validation sets the available 984 laboratory measurements
were placed in descending order with respect to sediment concentration and for every three
successive measurements that were picked for the training set, the fourth one was selected
for the validation set. This procedure was iterated for all the laboratory data and the
emerged training and validation sets comprise 739 and 245 measurements, respectively. The
600 field measurements constitute the test set. Table 3 shows some statistical measures of the
potential variables of these sets. Table 4 shows the datasets from which the data used in this
study were obtained and some representative values of each set. The abbreviations used in
Table 4 are the same with those Brownlie (1981b) used in his data compilation; consequent‐
ly, all the references to the original datasets may be obtained from that study.

Statistical measures dgr Fr Re* VS/ωs VS/(gd50)0.5 ω* τ* Ct

Minimum value 2.405 0.166 2.468 0.0010 0.0014 0.093 0.0323 10.2
Maximum value 38.800 0.999 158.292 0.5350 0.2818 134.6 5.828 47300
train set

Mean value 10.069 0.490 18.356 0.0297 0.0183 5.065 0.450 977.9
Laboratory data

Standard deviation 6.337 0.188 17.476 0.0425 0.0198 9.263 0.412 2575.8
Skewness coefficient 1.456 0.943 3.219 5.127 5.633 6.598 4.618 11.758
Minimum value 2.508 0.202 2.938 0.0017 0.0016 0.227 0.0578 11.3
validation set

Maximum value 38.505 0.968 146.416 0.2984 0.1150 81.56 2.555 10630
Mean value 9.708 0.482 17.220 0.0292 0.0173 5.397 0.459 861.4
Standard deviation 6.760 0.173 17.467 0.0337 0.0153 9.128 0.374 1412.2
Skewness coefficient 1.590 0.912 3.579 3.200 2.429 4.212 1.671 3.372
Minimum value 2.531 0.166 6.165 0.0017 0.0015 0.365 0.094 11
Maximum value 33.931 0.992 131.127 0.1528 0.0857 118.7 5.805 11400
Field data

test set

Mean value 10.654 0.375 27.910 0.0202 0.0156 14.13 0.943 1239.3
Standard deviation 6.968 0.150 18.994 0.0194 0.0121 16.73 0.648 1472.2
Skewness coefficient 1.705 1.027 2.105 2.676 1.893 2.350 1.726 2.554

Table 3. Statistical measures of the train, validation and test sets

 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 289
[Link]

Range of field variables

Velocity (m/s) Depth (m) Slope (‰) Ct (ppm)

Code No.
min max min max min max min max
BAL 25 0.226 1.093 0.091 0.256 0.44 2.1 19 3776

BEN 1 0.205 0.205 0.038 0.038 0.5 0.5 10.2 10.2

BRO 6 0.372 0.616 0.047 0.060 2.4 3.5 1200 5300

CHY 7 0.423 0.586 0.066 0.101 1.11 2 99.4 345

COS 12 0.403 0.503 0.140 0.156 0.45 1.01 10.954 102.08

DAV 69 0.244 0.792 0.076 0.305 0.248 2.67 11.3 1760

EPA 16 0.440 0.706 0.088 0.300 0.6 3.68 32 1017

EPB 19 0.265 0.762 0.148 0.304 0.262 1.6 45 1810

FOL 6 0.388 0.599 0.036 0.047 3.74 4.02 845 1848

FRA 11 0.361 0.450 0.129 0.161 0.938 1.693 39.979 166.34

GKA 27 0.302 0.635 0.032 0.124 1.8 6.401 205 3160

GUY 145 0.225 1.321 0.058 0.405 0.37 9.5 12 47300

JOR 7 0.401 0.557 0.070 0.105 1.12 1.67 95.8 306.7

KEN 6 0.412 0.799 0.047 0.109 1.7 4.2 550 2070

KNB 9 0.277 0.674 0.070 0.168 0.56 2.5 14 1740

LAU 10 0.326 0.671 0.076 0.221 0.8 2.1 550 4240

MCD 11 0.480 0.660 0.082 0.146 1.11 1.67 151.2 615.8

MPR 15 0.426 0.835 0.112 0.490 0.42 4.066 14.357 1091.1

MUT 17 0.131 0.505 0.029 0.102 0.5 7.5 11 10630

NOR 27 0.524 1.802 0.256 0.585 0.47 5.77 33 8870

OBR 45 0.214 0.953 0.088 0.165 0.57 3.23 17 1332.5

OJK 14 0.338 0.586 0.075 0.135 1.09 2.67 66.791 3355.7

PRA 25 0.254 0.701 0.076 0.305 0.282 2.87 11.63 560

SAT 1 0.332 0.332 0.193 0.193 0.44 0.44 66.877 66.877

SIN 58 0.277 0.597 0.066 0.117 1 4 35.7 1105

SON 1 0.465 0.465 0.043 0.043 6.7 6.7 6300 6300

STE 27 0.514 1.364 0.091 0.302 2.01 4.03 640 4615

STR 15 0.345 0.835 0.047 0.223 0.950 4.62 417 6300

TAY 11 0.348 0.878 0.077 0.160 0.89 2.09 13.979 2269.7

VAB 12 0.234 0.772 0.071 0.169 0.7 2.8 37 2500

VAH 6 0.319 0.558 0.176 0.238 0.642 1.303 31 1490

WLM 5 0.538 0.669 0.204 0.223 0.912 2.14 31.125 196.1

WLS 61 0.358 1.360 0.110 0.302 0.269 1.98 102 11700

WSA 195 0.165 0.555 0.034 0.170 1 2 11.3 587.19

WSB 36 0.444 0.578 0.108 0.176 1 2 55.8 379

WSS 13 0.377 0.388 0.073 0.075 1 1 53.8 94.6

ZNA 13 0.224 0.783 0.05 0.783 1.66 4.7 150 1975

Total 984 0.131 1.802 0.029 0.585 0.248 9.5 10.2 47300

(a)
290 Sediment Transport Processes and Their Modelling Applications

Range of field variables

Velocity (m/s) Depth (m) Slope (‰) Ct (ppm)

Code No.
min max min max min max min max
AMC 5 0.473 0.739 0.796 1.009 0.237 0.33 52 448

ATC 6 1.739 2.028 10.881 14.112 0.038 0.0513 102.374 567.343

CHO 26 0.846 1.597 2.103 3.414 0.115 0.254 149.826 1316.9

COL 58 0.617 1.266 1.134 3.371 0.107 0.407 35.6 768.7

HII 34 0.186 0.930 0.025 0.732 0.84 10.7 116.311 5638.6

MID 35 0.593 1.125 0.247 0.412 0.928 1.572 437.760 2269.2

MIS 5 1.756 2.423 11.400 17.282 0.082 0.134 178.001 511.707

MOU 91 0.366 1.350 0.040 0.438 1.36 3.15 26.763 2600.6

NIO 40 0.625 1.271 0.398 0.588 1.136 1.799 392 2750

RGC 8 0.805 1.518 0.923 1.512 0.53 0.8 674 2695

RGR 254 0.295 2.384 0.159 2.326 0.69 2.31 11 11400

RIO 38 0.624 2.384 0.332 1.463 0.74 0.89 463.65 4544.38

Total 600 0.186 2.423 0.025 17.282 0.038 10.7 11 11400

(b)

Table 4. (a) Range of laboratory variables, (b) Range of field variables

Data-driven techniques can be used for data mining since the only prerequisite for their function
is the determination of the input parameters without the need to predefine the structure of the
model and the degree of nonlinearity. The determination of the input parameters for the data-
driven schemes will be made with a tentative assessment through a trial-and-error procedure.
The correlation coefficient r has been employed in order to reveal any existing linear depend‐
ence in log-log plots between sediment concentration and any of the variables listed in Table 1

N
å (Yi - Y ) ( Xi - X )
(15)
i =1
r=
N 2 N
å (Yi - Y ) å ( Xi - X )
2

i =1 i =1

where Y denotes sediment concentration and X denotes the independent variable. Table 5
shows the correlation coefficient for log-log plots for the flume and field data of Tables 4a and 4b.
From the techniques proposed, the trial-and-error process will be accomplished with the aid of
ANNs, due to their speed, and after the determination of the most promising combinations that
may serve as an input layer, the other data-driven techniques will be implemented as well.

VS/ω VS/(gd50)1/2 ω* τ* Fr Re Re* dgr

Flume data r 0.862 0.885 0.754 0.681 0.759 0.463 -0.014 -0.314
Field data r 0.730 0.687 0.601 0.587 0.492 0.148 -0.144 -0.405

Table 5. Correlation between sediment concentration and independent dimensionless variables of the flume and
field data of Table 4 in log-log plots
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 291
[Link]

The findings shown in Table 5 partially agree with the diagrams depicted in Figure 1, since
sediment discharge is best correlated with unit stream power and stream power both for
laboratory and for field data.
After the tentative assessment based on ANNs, of several input combinations, the most po‐
tent ones, which will be applied to the data-driven schemes, seem to be those listed in Table
6. These combinations include the independent variables of Eq. (3) and others that are rela‐
tively easily measured and commonly used in engineering. It is noteworthy that all combi‐
nations comprise dimensionless grain diameter and Froude number among others. Whilst
Froude number gives a measure of the ratio of inertial forces to gravitational forces of the
flow and is a commonly used variable in hydraulic engineering, the potential usage of di‐
mensionless grain diameter is twofold. Firstly, it introduces kinematic viscosity and median
grain diameter and secondly provides homogeneity in the input data. The necessity for the
provided homogeneity can be seen from combination (a) where shear Reynolds number,
which essentially includes dimensionless grain diameter, is included as well. The absence of
any of these two terms in combination (a) has detrimental effects in the predictive capability
of the generated model. The other variables for the combinations examined herein are those
that most sediment transport formulae rely heavily on, namely dimensionless unit stream
power, dimensionless stream power and dimensionless shear stress, and are best correlated
with sediment concentration as shown in Table 5. For combination (a) Yang’s dimensionless
unit stream power was preferred to Vanoni’s because, despite the fact that the calculation of
fall velocity may be problematic, it reduced significantly the sum of errors between calculat‐
ed and observed values. The other two combinations (b) and (c) comprise just three varia‐
bles because shear is embedded in dimensionless stream power and dimensionless shear
stress, respectively. Furthermore, it seems that there is no other potential input combination,
besides those listed in Table 6, since any other combination tested gave results that declined
by orders of magnitude.

Input combinations

a dgr, Fr, Re*, VS/ωs

b dgr, Fr, ω*
c dgr, Fr, τ*

Table 6. Input combinations that will be applied to the data-driven schemes

5. Applications and results

The potential of training a DDM scheme solely with flume data and subsequently applying
it to a test set comprising exclusively field data has been shown in Kitsikoudis et al. (2012a,
2012b) where ANNs and symbolic regression were utilized, respectively, for the prediction
of sediment concentration in sand bed rivers. In these studies, however, the data were not
subjected to elaboration and screening, in order to demonstrate the potential modeling abili‐
292 Sediment Transport Processes and Their Modelling Applications

ty of this technique with crude data. As a result, input data were kept in large numbers, and
the generated models yielded very good results, better than those obtained from the com‐
mon sediment transport formulae. However, it is known that the incorporation of knowl‐
edge can be proved beneficial to the predictive capability of DDM schemes as long as this is
accomplished by transformation and elaboration of the fundamentals. Sediment transport
and open channel hydraulics rely heavily on empirical equations and ideal flows; therefore,
data transformation based on such assumptions does not guarantee the enhancement of the
predictive capabilities of the DDM scheme. Nevertheless, the sidewall correction of Vanoni
and Brooks (1957) was applied for the proper calculation of the shear stress in flume meas‐
urements and additionally the restrictions of Table 2 were imposed to the data for the re‐
moval of various biases resulting to a significantly reduced data amount. On the contrary, a
criterion for the initiation of motion has been omitted, due to the stochastic character of tur‐
bulence, and was left up to the DDM scheme to define the effective portion of the flow that
quantifies the transport rate.
Since every data-driven technique has its own syntax, the three possible input combinations
of Table 6 are tested individually with the aid of both ANNs and symbolic regression. The
evaluation of the modeled results Pi with respect to the observed ones Oi will be made on
the basis of the root mean square error (RMSE),

N
å (Oi - Pi )
2

i =1
(16)
RMSE =
N

coefficient of determination (R2) or Nash-Sutcliffe model efficiency coefficient (E) (Nash and
Sutcliffe, 1970),

N
å (Oi - Pi )
2

R2 = 1 - i =1
N (17)
å (Oi - O )
2

i =1

and discrepancy ratio (DR). The latter is the percentage of calculated concentrations that lie
between one half and two times the respective measured concentrations.

5.1. ANNs application

This study was implemented in MATLAB with the aid of the neural network toolbox (Demuth
et al., 2009). Since the usage of Levenberg-Marquardt training function gave the best results in a
similar study in Kitsikoudis et al. (2012a), it was utilized for training in this application as well.
Due to the importance of the initial values of the synaptic weights in the search for local mini‐
ma of the error function, which is the mean square error between calculated and observed val‐
ues, a MATLAB code was written, which determines the most efficient ANN within 5000
training executions, for each network architecture, with random initial weights for every repe‐
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 293
[Link]

tition. The most efficient ANN is taken to be the one that yields only positive sediment concen‐
trations, in order for the results to have physical meaning, and after the training provides the
highest DR in the test set. For this evaluation, DR is preferred over RMSE, because the latter em‐
phasizes on large concentrations. Models that derived slightly worse results than others, but
had much simpler structure were preferred due to the principle of parsimony. Figures 3-5 de‐
pict the scatter plots of the best derived models, for each input combination of Table 6, for the
field data of the test set. These models that perform best are described in Table 7. Table 8 shows
the best models and their performance measures for the training, validation and test sets. Final‐
ly, Table 10 shows a comparison between the ANN induced models and some of the common‐
ly used sediment transport functions for the rivers data constituting the test set. It should be
mentioned that several of these formulae are calibrated with part of the data (especially the
Brownlie formula) that are used for the comparison and despite that significant advantage they
still generate inferior results to those of the ANNs.

Network architecture (neurons in input-hidden-

Input combination from Table 6
output layers)

ANN (a) a 4-5-1

ANN (b) b 3-6-1

ANN (c) c 3-11-2-1

Table 7. Best performing models for each possible input combination

Figure 3. Scatter plot for the field data of the test set, of measured sediment concentration and computed from ANN,
based on input combination (a)
294 Sediment Transport Processes and Their Modelling Applications

Figure 4. Scatter plot for the field data of the test set, of measured sediment concentration and computed from ANN,
based on input combination (b)

Figure 5. Scatter plot for the field data of the test set, of measured sediment concentration and computed from ANN,
based on input combination (c)
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 295
[Link]

DR 0.5-2 (%) DR 0.25-4 (%) RMSE R2

Training set 85.12 98.51 722.20 0.9213

ANN (a)

Validation set 82.86 97.55 936.72 0.5582

Test set 72.50 92.33 1168.76 0.3687

Training set 82.81 97.56 734.13 0.9187

ANN (b)

Validation set 85.71 95.92 884.41 0.6062

Test set 71.67 93.17 1202.69 0.3315

Training set 84.84 98.24 509.42 0.9608

ANN(c)

Validation set 84.08 98.37 872.36 0.6169

Test set 70.67 91.33 1221.72 0.3102

Table 8. Performance measures of the optima ANNs

From Table 8 can be inferred that any of the three combinations listed in Table 6 has its own
merit and that sediment transport can be quantified by physical quantities that can be either
vectors or scalars.

5.2. Symbolic regression application

The basic computation tool for the implementation of symbolic regression is provided by
GPTIPS (Searson, 2009), which is an open source MATLAB toolbox. Since every problem has
its own peculiarities, proper adjustments must be made to the GPTIPS parameters in order
to obtain good results. The most important parameters are the population size, the number
of generations, the using functions, the maximum number of genes and the maximum tree
depth. Searson et al. (2010) have found that enforcing stringent tree depth restrictions often
allows the evolution of relatively compact models that are linear combinations of low order
nonlinear transformations of the input variables. After several runs, only input combination
(b) gave results superior to those of the classical formulae. The GPTIPS derived formula for
this combination is the following

w*
( )
dgr Fr dgr + Fr - w * +
( )
0.4
*
Ct = 1542 w Fr 3
- 0.4794 Fr - 313.6
3 (18)
æ 6.218 ö
ç * - Fr + 1 ÷
è w ø

Figure 6 depicts the scatter plot of measured and calculated from Eq. (18) sediment concentra‐
tions for the field data of the test set, whilst Table 9 and Table 10 show its performance for the
training, validation and testing set, and the comparison with other formulae, respectively.
296 Sediment Transport Processes and Their Modelling Applications

Figure 6. Scatter plot for the field data of the test set, of measured sediment concentration and computed from sym‐
bolic regression, based on input combination (b)

DR 0.5-2 (%) DR 0.25-4 (%) RMSE R2

Training set 68.20 84.84 632.21 0.9397

Validation set 66.53 85.31 964.70 0.5315
Test set 71.00 91.33 1218.12 0.3143

Table 9. Performance measures of symbolic regression, based on combination (b)

DR 0.5-2 (%) DR 0.25-4 (%) RMSE R2

ANN (a) 72.50 92.33 1168.76 0.3687

ANN (b) 71.67 93.17 1202.69 0.3315
ANN (c) 70.67 91.33 1221.72 0.3102
Symb. regression (b) 71.00 91.33 1218.12 0.3143
Ackers & White 58.33 88.67 1405.38 0.0872
Brownlie 68.33 91.00 1274.44 0.2494
Engelund & Hansen 69.67 92.33 1244.83 0.2838
Karim & Kennedy 64.83 91.50 1341.34 0.1685
Molinas & Wu 52.83 84.83 1423.06 0.0641
Yang 49.50 83.33 1403.07 0.0902

Table 10. Comparison of ANNs of the input combinations (a), (b) and (c) and Eq. (18), derived from symbolic
regression for the combination (b), with sediment transport formulae based on the river data of the test set
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 297
[Link]

The results obtained from ANNs for all the combinations are superior to those of the classi‐
cal sediment transport formulae in terms of DR, RMSE and R2. Combination (a) performed
best in all evaluation measures, besides the second DR criterion in the range 0.25-4 where
combination (b) gave better results. The third combination came up third with respect to all
evaluation measures. However, these results by no means can be considered conclusive,
since it is essentially unknown whether they are the best results derived from the ANN or
just results obtained from the trapping in a local minimum of the minimization process in
the network’s training algorithm. From the results generated from symbolic regression, only
combination (b) managed to surpass the classical sediment transport functions. The other
two combinations gave results inferior to those of Engelund and Hansen and Brownlie for‐
mulae, but superior to those of the others. In addition, symbolic regression derived its best
results without utilizing the log-transformation of the input data. Regarding the other sedi‐
ment transport functions, the formula of Engelund and Hansen performed best. The small
values of the coefficient of determination R2 in Table 10 reflect the difficulty of predicting
sediment transport rates in natural streams and rivers, due to random turbulent bursts that
accentuate the stochastic nature and exacerbate the complexity of the problem.
Although these results cannot be considered conclusive, it seems that the ANNs yield better
results. GPTIPS sometimes (usually when only a few input variables are involved) lags be‐
hind a neural network model in terms of raw predictive performance, but the equivalent GP
models are often simpler, shorter and may be open to physical interpretation (Searson,
2009). This is partially due to the fact that ANNs are much faster than the time consuming
GP and for given time they can run multiple times comparing to GP. Moreover, since the
testing set comes from a database with different statistical distributions than the one from
which the training set originates, the exploration of as many as possible local minima of the
training function may prove beneficial to the training process. ANNs have this property,
whilst GP is based on a stochastic concept seeking the global minimum. This may be one
reason for the superiority of ANNs in this study, where the training data comprise flume
measurements, whilst the testing data consists of field measurements.

6. Conclusions

This study utilized two widely used data-driven techniques, namely ANNs and symbolic
regression, in a novel way since the data used for training and those used for testing came
from datasets with different statistical distributions. This difference is owned to the fact that
the training and validation set comprises exclusively laboratory flume data, while the test‐
ing set consists solely of field data. Based on this concept, the inclusion of noise emanated
from the field measurements will not be embedded in the training data and additionally the
generated models will have general applicability since the inclusion of field data in the
training set would confine them to the specific streams from which the data were obtained.
The determination of the input parameters was accomplished by a tentative assessment of
some of the widely used dimensionless parameters in sediment transport and open channel
hydraulics. This assessment showed that three combinations had the potential to serve as in‐
298 Sediment Transport Processes and Their Modelling Applications

puts and were involved in this application, in which they all yielded very good results, bet‐
ter than those obtained from the commonly used formulae on the basis of root mean square
error and the ratio of computed to measured transport rates. Unit stream power, stream
power, and shear stress were the dominant independent variables of the three combinations,
respectively, and the results have shown that each one, of these widely used variables in the
context of sediment transport, has its own merit. The results generated from the ANNs were
better from those obtained from symbolic regression; however, the explicit equation that
was derived from the latter can be more easily interpreted. Finally, the results obtained in
this study may enhance the confidence in using data-driven techniques, despite their black-
box nature, because, in order to perform well in a dataset from a different system from the
one they were trained, the induced equations must have physical meaning.

Notation

The following symbols are used in this chapter:

Ct = sediment concentration by weight in parts per million (ppm)

D = mean flow depth (m)

Q = water discharge (m3/s)

Qst = sediment discharge (m3/s)

R = hydraulic radius (m)

S = energy slope

T = water temperature (oC)

u* = shear velocity (m/s)

u*c = critical shear velocity (m/s)

V = mean flow velocity (m/s)

W = channel width (m)

d = grain diameter (m)

d50 = median grain diameter (m)

f = friction factor

g = gravitational acceleration (m/s2)

γ = specific weight of water (N/m3)

γs = specific weight of sediment (N/m3)

ν = kinematic viscosity of water (m2/s)

 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 299
[Link]

ρ = density of water (kg/m3)

ρs = density of sediment (kg/m3)

σg = geometric standard deviation of bed particles [(d84/d50 + d50/d16)/2]

τ = shear stress [kg/(m.s2)]

τ* = dimensionless shear stress

ω = stream power (kg/s3)

ω* = dimensionless stream power

ωs = settling or fall velocity of sediment (m/s)

Appendix A

In flume experiments, the sand covered bed will generally be much rougher than the flume
walls, and thus will be subjected to higher shear stress. Separation of the shear force exerted
on the bed from that on the lateral boundaries was first proposed by Einstein (1950). The
line of analysis pursued as follows is that proposed by Johnson (1942) and modified by Va‐
noni and Brooks (1957). The principal assumption is that the cross-sectional area can be div‐
ided into two parts, Ab and Aw, in which the streamwise component of the gravity force is
resisted by the shear force exerted in the bed and walls, respectively. It is further assumed
that the mean velocity and energy gradient are the same for Ab and Aw, and that the Darcy-
Weisbach relation can be applied to each part of the cross section as well as to the whole, i.e.

V 2 8 gA 8 gAb 8 gAw
S
=
fp
=
fb pb
=
f w pw
(19)

in which, p = the wetted perimeter; and the subscripts b and w refer to the bed and wall
sections, respectively. For a rectangular channel p=2D+W; pw=2D; pb=W. Introducing the
geometrical requirement A=Ab+Aw into Eq. (19) results in

2D
fb = f +
W
( f - fw ) (20)

The wall friction factor fw is further related to the ratio of Re/f, where Re=4VR/ν and f can be
calculated from the experimental data. This relationship, which was originally given as a
graph of fw against Re/f by Vanoni and Brooks (1957), can also be described by the function

-1
fw = é 20 ( Re f ) - 39 ù
0.1
êë úû (21)
300 Sediment Transport Processes and Their Modelling Applications

which is obtained by curve fitting (Cheng and Chua, 2005). Finally, fb is calculated from Eq.
(20) and Rb = Ab/pb from Eq. (19). Rb is consequently used for the calculation of the bed shear
velocity and bed shear stress.

Despite its several obvious deficiencies (division of the cross section into two noninteracting
parts, determination of friction factors for section components on the basis of a pipe friction
diagram, use of the same mean velocity for each subsection, etc.), the side-wall correction
procedure appears to yield fairly reliable estimates of the friction factors for flow over sand
beds with no flume walls present (Vanoni, 2006).

Appendix B

For the calculation of particle fall velocity in a clear, still fluid, van Rijn (1984) suggested the
use of the Stokes law for sediment particles smaller than 0.1 mm

1 rs - r d2
ws = g (22)
18 r n

For suspended sand particles in the range 0.1 to 1 mm, the following type of equation, as
proposed by Zanke (1977), can be used

ì 12 ü
n ïé ær ö gd 3 ù ï
ws = 10 í ê1 + 0.01ç s - 1 ÷ 2 ú - 1ý
d ï ëê r
(23)
è ø n ú
û ïþ
î

For particles larger than about 1 mm, the following simple equation can be used (van Rijn, 1982)

12
éæ r ö ù
ws = 1.1 êç s - 1 ÷ gd ú (24)
ëêè r ø ûú

Appendix C

Ackers and White formula: Ackers and White (1973) applied dimensional analysis to ex‐
press the mobility and transport rate of sediment in terms of some dimensionless pa‐
rameters. It has been shown that the transport of fine materials is best related to gross
shear, shear velocity being the representative variable, and that the transport of coarse
materials is best related to the net grain shear, mean velocity being the representative
variable. The following equations do not necessarily apply in an upper phase of trans‐
port. However, it was shown that the following relationships are not sensitive to bed
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 301
[Link]

form; they apply to plain, rippled, and duned configurations. Their mobility number for
sediment is

1- n
U *n é V ù
Fgr = ê ú (25)
gd ( g s g - 1) êë 32 log ( 10 D d ) úû

Coefficients C, A, m and n are related to the dimensionless grain diameter dgr based on best-
fit curves of laboratory data with sediment sizes greater than 0.04 mm and Froude numbers
less than 0.8. They are shown in Table 11.

13
é g ( g s g - 1) ù
dgr = d ê ú (26)
ëê n2 ûú

Finally, they related the bed material load to the mobility number as follows

n m
XD æ U * ö æ Fgr ö
Ggr = ç ÷ = C çç - 1÷
÷
(27)
dg s g è V ø è A ø

where X = rate of sediment transport in terms of mass flux per unit mass flow rate

dgr ≥ 60 1 < dgr < 60

n = 0.00 n = 1.00-0.56logdgr

A = 0.17 A = 0.23dgr -0.5+0.14

m = 1.50 m = 9.66dgr -1+1.34

C =0.025 logC = -3.53+2.86logdgr-(logdgr) 2

Table 11. Coefficients of the Ackers and White formula

Brownlie formula: The Brownlie (1981a) relations are based on regressions of over 1000 ex‐
perimental and field data points. For normal or quasi-normal flow, the transport relation
takes the form

-0.3301
æ R ö
( )
1.978
Ct = 7115c f Fg - Fgo S0.6601 çç ÷÷ (28)
è d50 ø

where

V
Fg = (29)
( rs r - 1) gd50
302 Sediment Transport Processes and Their Modelling Applications

( )
0.5293
Fgo = 4.596 t c* S-0.1405s g -0.1606 (30)

t c* = 0.22Y + 0.06 × 10 -7.7 Y (31)

-0.6
æ ( rs r - 1) gd50 d50 ö
Y =ç
ç v
÷
÷
(32)
è ø

cf = 1 for laboratory flumes and 1.268 for field channels.

Engelund and Hansen formula: Using Bagnold’s stream power concept and the similarity
principle, Engelund and Hansen (1967) established the following sediment transport formula

f 'f = 0.1q 5 2 (33)

where

2 gDS
f'= (34)
V2

qt
f= (35)
rs ( rs r - 1) gd50 3

t
q= (36)
(g s - g ) d50

where qt = total sediment discharge by weight per unit width. Strictly speaking, the Engelund
and Hansen formula should be applied to those flows with dune beds in accordance with the
similarity principle. However, many tests have shown that it can be applied to the upper flow
regime with particle size greater than 0.15 mm without serious deviation from the theory.

Karim and Kennedy formula: Karim and Kennedy (1990) applied nonlinear multiple regres‐
sion analysis to derive relations among flow velocity, sediment discharge, bed form geome‐
try, and friction factor of alluvial rivers. A database comprising 339 river flows and 608 flume
flows was used in their analysis. The obtained sediment load predictor is given by

qs V
log = -2.279 + 2.972 log
(g s g - 1) gd50
3
(g s g - 1) gd50
U* - U* c U* - U* c
(37)
V D
+1.060 log log + 0.299 log log
(g s g - 1) gd50 (g s g - 1) gd50 d50 (g s g - 1) gd50
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 303
[Link]

where qs =volumetric total sediment discharge per unit width.

Molinas and Wu formula: This empirical relation is based on Velikanov’s gravitational

power theory, which assumes that the power available in flowing water is equal to the
sum of the power required to overcome flow resistance and the power required to keep
sediment in suspension against gravitational forces. Molinas and Wu (2001) argued that
the predictors of Ackers and White, Engelund and Hansen, and Yang have been devel‐
oped with flume experiments representative of shallow flows and cannot be applied to
large rivers having deep flow conditions. Motivated by the need for having a total bed
material load predictor for application to large sand bed rivers, they used stream power
and energy considerations together with data from large rivers (e.g., Amazon, Atchafa‐
laya, Mississippi, Red River), to obtain an empirical fit for the total bed material load con‐
centration in ppm

Ct =
(
1430 0.86 + Y Y 1.5 ) (38)
0.016 + Y

where Ψ = universal stream power, which is defined as

V3
Y= 2 (39)
g ( r s r - 1) Dws éëlog ( D d50 ) ùû

One advantage of this approximation is that the energy slope does not have to be measured
directly, which is always a challenge in large alluvial rivers. On the other hand, since Moli‐
nas and Wu (2001) do not mention how the wash load was separated from the bed material
load and the same large river data were used both to develop and to test their formulation,
Eq. (38) might overestimate bed material load concentrations when applied to other large
rivers not included in the calibration (Garcia, 2008).

Yang formula: To determine total sediment concentration, Yang (1973) used Buckingham’s
π theorem and the concept of unit stream power, which is given by the product of mean
flow velocity and energy slope. The coefficients in Yang’s equation were determined by run‐
ning a multiple regression analysis for 463 sets of laboratory data. The equation obtained is

wsd50 U
log Ct = 5.435 - 0.286 log - 0.457 log *
n ws
æ wd U ö æ VS VcrS ö
(40)
+ çç 1.799 - 0.409 log s 50 - 0.314 log * ÷÷ log çç - ÷
è n ws ø è ws ws ÷ø

The critical dimensionless unit stream power VcrS/ω is the product of the dimensionless crit‐
ical velocity Vcr/ω and the energy slope S, where
304 Sediment Transport Processes and Their Modelling Applications

ì 2.5 U * d50
ï + 0.66 for 1.2 < < 70
ï log æ U * d50 ö - 0.06 n
ï ç ÷
è n ø
Vcr ï
=í
ws ï
(41)
U * d50
ï 2.05 for 70 £
ï n
ï
î

Author details

Vasileios Kitsikoudis1, Epaminondas Sidiropoulos2 and Vlassios Hrissanthou1

1 Department of Civil Engineering, Democritus University of Thrace, Xanthi, Greece

2 Department of Rural and Surveying Engineering, Aristotle University of Thessaloniki,

Thessaloniki, Greece

References

[1] Ackers P., White W. R. (1973). Sediment Transport: New Approach and Analysis.
Journal of the Hydraulics Division, ASCE; 99(HY11) 2041-2060.
[2] Azamathulla H. Md, Ab Ghani A., Zakaria N. A., Guven A. (2010). Genetic Program‐
ming to Predict Bridge Pier Scour. Journal of Hydraulic Engineering, ASCE; 136(3)
165-169.
[3] Babovic V. (2000). Data Mining and Knowledge Discovery in Sediment Transport.
Computer Aided Civil and Infrastructure Engineering; 15(5) 383-389.
[4] Babovic V., Abbott M. B. (1997). The Evolution of Equations from Hydraulic Data.
Part II: Applications. Journal of Hydraulic Research, IAHR; 35(3) 411-430.
[5] Babovic V., Keijzer M. (2000). Genetic Programming as a Model Induction Engine.
Journal of Hydroinformatics; 2(1) 35-60.
[6] Bagnold R. A. (1956). Flow of Cohesionless Grains in Fluids. Philosophical Transac‐
tions of the Royal Society [London], Series A; 249 235-297.
[7] Bagnold R. A. (1966). An Approach to the Sediment Transport Problem from General
Physics. Prof. Paper 422-I. U.S. Geological Survey.
[8] Bhattacharya B., Price R. K., Solomatine D. P. (2004). A Data Mining Approach to
Modelling Sediment Transport. In: Liong S., Phoon K., Babovic V. Proceedings of the
6th International Conference on Hydroinformatics, 21-24 June 2004, Singapore. World
Scientific.
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 305
[Link]

[9] Bhattacharya B., Price R. K., Solomatine D. P. (2005). Data Driven Modeling in the
Context of Sediment Transport. Physics and Chemistry of the Earth; 30(4-5) 297-302.

[10] Bhattacharya B., Price R. K., Solomatine D. P. (2007). Machine Learning Approach to
Modeling Sediment Transport. Journal of Hydraulic Engineering, ASCE; 133(4)
440-450.

[11] Brownlie W. R. (1981a). Prediction of Flow Depth and Sediment Discharge in Open
Channels. Report No. KH-R-43A. Pasadena, California: W. M. Keck Laboratory of
Hydraulics and Water Resources, California Institute of Technology.

[12] Brownlie W. R. (1981b). Compilation of Alluvial Channel Data: Laboratory and Field.
Report No. KH-R-43B. Pasadena, California: W. M. Keck Laboratory of Hydraulics
and Water Resources, California Institute of Technology.

[13] Cheng N. S., Chua L. H. C. (2005). Comparisons of Sidewall Correction of Bed Shear
Stress in Open-Channel Flows. Journal of Hydraulic Engineering, ASCE; 131(7)
605-609.

[14] Cigizoglu H. K. (2004). Estimation and Forecasting of Daily Suspended Sediment Da‐
ta by Multi-Layer Perceptrons. Advances in Water Resources; 27(2) 185-195.

[15] Demuth H. B., Beale M. H., Hagan M. T. (2009). Neural Network Toolbox: For Use
With MATLAB. The Mathworks Inc.; 2009.

[16] Eaton B. C., Church M. (2011). A Rational Sediment Transport Scaling Relation Based
on Dimensionless Stream Power. Earth Surface Processes and Landforms; 36(7)
901-910.

[17] Einstein H. A. (1942). Formulas for the transportation of Bed Load. Transactions,
ASCE; 107 (Paper No. 2140) 561-573.

[18] Einstein H. A. (1950). The Bedload Function for Sediment Transportation in Open
Channel Flows. Technical Bulletin No. 1026. Washington D.C.: U.S. Department of
Agriculture, Soil Conservation Service.

[19] Engelund F., Hansen E. (1967). A Monograph on Sediment Transport in Alluvial

Streams. Copenhagen: Teknisk Vorlag.

[20] Garcia M. H. (2008). Sediment Transport and Morphodynamics. In: Garcia M. H.

(ed.) ASCE Manuals and Reports on Engineering Practice No. 110, Sedimentation En‐
gineering: Processes, Measurements, Modeling, and Practice. Virginia, U.S.A.: ASCE,
p21-163.

[21] Haykin S. (2009). Neural Networks and Learning Machines, 3rd Edition. New Jersey:
Prentice Hall.

[22] Hornik K., Stinchcombe M., White H. (1989). Multilayer Feedforward Networks are
Universal Approximators. Neural Networks; 2(5) 359-366.
306 Sediment Transport Processes and Their Modelling Applications

[23] Johnson J. W. (1942). The Importance of Considering Side-Wall Friction in Bed-Load

Investigations. Civil Engineering, ASCE; 12(6) 329-331.

[24] Karim M. F., Kennedy J. F. (1990). Menu of Coupled Velocity and Sediment-Dis‐
charge Relations for Rivers. Journal of Hydraulic Engineering, ASCE; 116(8) 978-996.

[25] Kitsikoudis V., Sidiropoulos E., Hrissanthou V. (2012a). A New Approach for ANN
Modeling of Sediment Transport in Sand Bed Rivers. Proceedings of the 10th Interna‐
tional Conference on Hydroinformatics, CD-ROM format, 14-18 July 2012, Hamburg,
Germany.

[26] Kitsikoudis V., Sidiropoulos E., Hrissanthou V. (2012b). Implementation of Multi‐

gene Symbolic Regression in the Sediment Transport Quantification Problem for
Sand Bed Rivers. Proceedings of the 9th International Symposium on Ecohydraulics,
CD-ROM format, 17-21 September 2012, Vienna, Austria.

[27] Knapp R. T. (1938). Energy Balance in Stream Flows Carrying Suspended Load.
Transactions, American Geophysical Union; 501-505.

[28] Koza J. (1992). Genetic Programming: On the Programming of Computers by Means

of Natural Selection. Cambridge MA: MIT Press.

[29] Maier H. R., Dandy G. C. (2000). Neural Networks for the Prediction and Forecasting
of Water Resources Variables: A Review of Modeling Issues and Applications. Envi‐
ronmental Modeling and Software; 15(1) 101-124.

[30] Minns A. W. (2000). Subsymbolic Methods for Data Mining in Hydraulic Engineer‐
ing. Journal of Hydroinformatics; 2(1) 3-13.

[31] Molinas A., Wu B. (2001). Transport of Sediment in Large Sand-Bed Rivers. Journal
of Hydraulic Research, IAHR; 39(2) 135-146.

[32] Nagy H. M., Watanabe K., Hirano M. (2002). Prediction of Sediment Load Concentra‐
tion in Rivers Using Artificial Neural Network Model. Journal of Hydraulic Engi‐
neering, ASCE; 128(6) 588-595.

[33] Nash J. E., Sutcliffe J. V. (1970). River Flow Forecasting Through Conceptual Models,
Part I – A Discussion of Principles. Journal of Hydrology: 10(3) 282-290.

[34] Pugh C. A. (2008). Sediment Transport Scaling for Physical Models. In: Garcia M. H.
(ed.) ASCE Manuals and Reports on Engineering Practice No. 110, Sedimentation En‐
gineering: Processes, Measurements, Modeling, and Practice. Virginia, U.S.A.: ASCE,
p1057-1066.

[35] Rubey W. W. (1933). Equilibrium Conditions in Debris-Laden Streams. Transactions,

American Geophysical Union 14th Ann. Mtg; 497-505.

[36] Rumelhart D. E., Hinton G. E., Williams R. J. (1986). Learning Representations of

Back-Propagation Errors. Nature (London); 323(9) 533-536.
 Derivation of Sediment Transport Models for Sand Bed Rivers from Data-Driven Techniques 307
[Link]

[37] Searson D. (2009). GPTIPS: Genetic Programming & Symbolic Regression for MAT‐
LAB User Guide.

[38] Searson D. P., Leahy D. E., Willis M. J. (2010). GPTIPS: An Open Source Genetic Pro‐
gramming Toolbox for Multigene Symbolic Regression. In Ao S. I., Castillo O., Doug‐
las C., Feng D. D., Lee J. (eds.) Proceedings of the International MultiConference of
Engineers and Computer Scientists, Vol. I, IMECS 2010, 17-19 March 2010, Hong
Kong.

[39] Shields A. (1936). Application of Similarity Principles and Turbulence Research to

Bedload Movement. Transl. into English by Ott W. P. and Van Uchelen J. C. Pasade‐
na, California: California Institute of Technology; 1936.

[40] van Rijn L. C. (1982). Computation of Bed-Load and Suspended Load. Report S487-II,
Delft Hydraulics Laboratory, Delft, The Netherlands.

[41] van Rijn L. C. (1984). Sediment Transport, Part II: Suspended Load Transport. Jour‐
nal of Hydraulic Engineering, ASCE; 110(11) 1613-1641.

[42] Vanoni V. A. (1978). Predicting Sediment Discharge in Alluvial Channels. In: Water
Supply and Management. Oxford: Pergamon Press. p399-417.

[43] Vanoni V. A. (2006). ASCE Manuals and Reports on Engineering No. 54, Sedimenta‐
tion Engineering. Virginia, U.S.A.: ASCE.

[44] Vanoni V. A., Brooks N. H. (1957). Laboratory Studies of the Roughness and Sus‐
pended Load of Alluvial Streams. Sedimentation Laboratory Report No. E68. Pasade‐
na, California: California Institute of Technology.

[45] Velikanov M. A. (1955). Dynamics of Channel Flow – v.2. In: Sediments and the
Channel, 3rd Edition. Moscow: State Publishing House for Tech.– Theoretical Lit.;
p107-120 [in Russian].

[46] Witten I. H., Frank E., Hall M. A. (2011). Data Mining: Practical Machine Learning
Tools and Techniques, 3rd Edition. Burlington, MA: Morgan Kaufmann.

[47] Yalin M. S. (1977). Mechanics of Sediment Transport. Oxford: Pergamon Press.

[48] Yang C. T. (1972). Unit Stream Power and Sediment Transport. Journal of the Hy‐
draulics Division, ASCE; 98(HY10) 1805-1836.

[49] Yang C. T. (1973). Incipient Motion and Sediment Transport. Journal of the Hydraul‐
ics Division, ASCE; 99(HY10) 1679-1704.

[50] Yang C. T. (1977). The Movement of Sediment in Rivers. Surveys in Geophysics; 3(1)
39-68.

[51] Yang C. T. (2003). Sediment Transport: Theory and Practice. Original edition
McGraw-Hill; 1996. Reprint edition by Krieger Publication Company.
308 Sediment Transport Processes and Their Modelling Applications

[52] Yang C. T., Marsooli R., Aalami M. T. (2009). Evaluation of Total Load Sediment
Transport Formulas Using ANN. International Journal of Sediment Research; 24(3)
274-286.

[53] Zakaria N. A., Azamathulla H. Md, Chang C. K., Ab Ghani A. (2010). Gene Expres‐
sion Programming for Total Bed Material Load Estimation – A Case Study. Science of
the Total Environment; 408(21) 5078-5085.

[54] Zanke U. (1977). Berechnung der Sinkgeschwindigkeiten von Sedimenten. Mitt. des
Franzius – Instituts für Wasserbau, Heft 46, Seite 243. Technical University Hann‐
over.
 Provisional chapter
Chapter 12

Modelling of Sediment Transport in Shallow Waters

Modelling
by of Sediment
Stochastic Transport
and Partial in Shallow
Differential Waters by
Equations
Stochastic and Partial Differential Equations
Charles Wilson Mahera and Anton Mtega Narsis

Charlesinformation
Additional Wilson Mahera and
is available at Anton
the end of theMtega
chapter Narsis

Additional information is available at the end of the chapter

10.5772/52237

[Link]

1. Introduction
Sediment transport in coastal waters take place in near-shore environments due to the
motions of waves and currents resulting in the formation of characteristic coastal landforms
such as beaches, barrier islands, and capes, etc. Though sediment transport modelling has
been carried for over three decades[26], the study remains a challenging topic of research
since a unified description of the process is still to be achieved[25]. A brief review of sediment
transport modelling and the widely used classical as well as most recent methods and their
limitations in applications are given by [22, 34]. It is observed that sediment transport
studies (e.g. the underlying physics and methods used) and modes of sediment transport
are yet to be fully studied. Both experimental approaches and mathematical modelling
coupled with advanced numerical solutions are needed for better understanding of the
how fundamental sediment transport processes is significant for environmental researchers
to provide practical and scientifically sound solutions to hydraulic engineering problems.
According to [34], the choice of a method for solving a specific problem depends on the
nature and complexity of the problem itself, the capabilities of the chosen model to simulate
the problem adequately, data availability for model calibration and verification and overall
available time and budget for solving the problem. [34] also found that discrepancies between
hydrodynamic/sediment transport model predictions and measurements can be attributed
to different causes. They include over simplification of the problem by using an inappropriate
model 1D versus 2D or 2D versus 3D, the use of inappropriate input data, lack of appropriate
data for model calibration, unfamiliarity with the limitations of the hydrodynamic/sediment
transport equations used in developing the model, and computational errors in source codes
because of approximations in the numerical schemes used in solving the governing equations
(boundary condition problems/truncation errors because of discretization.
310 Sediment Transport Processes and Their Modelling Applications
2 Sediment Transport

Modelling sediment transport basically involves interaction of hydrological, hydrodynamic

and sedimentology processes. The interactions cause variability of the process parameters
which are partly deterministic (having a known structure in space/time, e.g. the yearly
storm/calm weather period) and partly stochastic (e.g. sediment pick-up or turbulent
viscosity, and also weather variations), both leading to slow spatial/temporal variations
in the sand wave-dynamics[31]. Therefore, stochastic characteristics of the governing
parameters (such as, suspended sediment inflow concentration) and parameters describing
the beginning and rate of erosion and sedimentation respectively are considered as stochastic
variables. The observations of the stochastic nature of bedload transport have given impetus
to a probabilistic formulation of bedload transport equations.
In stochastic modelling spatio-temporal behavior of phenomena is modelled with random
components. The random walk simulation model enables the observation of the phenomena
in scales much smaller than the grid size, as well as the tracing of the movement of individual
particles, thereby describing the natural processes more accurately. Concentrations of
particles are easily calculated from the spatial positions of the particles and, more
importantly, when and where required. Furthermore, errors due to numerical diffusion
observed in methods such as Finite differences or Finite elements, are avoided and there is
considerable reduction in computational time since the calculating load is restricted to the
domain parts where the majority of the parcels are gathered.
In a random walk model the displacement of an arbitrary particle, at each time step consists
of an advective, deterministic component and an independent, stochastic component. In
a simplified one-dimensional transport model the Brownian motion of a particle can be
described by the Langevin equation[32].
In order to investigate the fate of suspended sediment in coastal and estuarine waters as
well as the evolution of sea or river beds, sediment dynamics need to be represented at
a scale relevant to the numerical discretized solution, and significant effort is devoted to
parameterize sediment processes. Sediment diffusivities, settling velocities, and cohesive
processes such as flocculation all have an impact on suspended sediment throughout the
water column. The approaches implemented in these coastal models may present distinct
strengths and shortcomings with regard to some important issues for coastal zones, both
numerical and physical. While these detailed limitations need to be considered as part
of model assessment, more general issues also hinder present state of the art models. In
particular, sediment transport is inherently highly empirical, which is further compounded
by issues arising from turbulence closure schemes.
In this chapter we focus on deterministic (i.e., process based) coastal ocean models, which
are being increasingly used to study coastal sediment dynamics and coastal morphological
evolution[23, 27–29]. These models usually treat the short term (hours to days) to medium
term (days to months) evolution. Historically, they were first based on depth averaged
equations (two dimensional horizontal (2DH) models) and were applied both to riverine
see [33] and coastal[24] environments.
We also study the methods based on probability concept appeared to be superior for
predicting local transport of bedload. Although several deterministic methods show
comparable performance for predicting total sectional transport rate, their performances
are significantly reduced for predicting lateral variation of local transportation rates [22].
Development of a stochastic theory based model that can explicitly present the random term
 Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 311
Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 3
[Link]
10.5772/52237

of sediment concentrations could be achieved. With the development of better numerical

techniques, the stochastic differential equations can be solved using Itô’s integration
technique without the need to rely on analytical solutions under simplified conditions.
It is known that bedform changes have a significant effect on the flow dynamics contributed
by imbalance between sediments in and out from those areas. The imbalance can easily
be disturbed by the external factors, such as extreme storm events, mean sea-level rise,
changes in tidal regime, human interferences and so on. Therefore, a better prediction
of these bedforms is required to be able to understand their sensitivity to external
conditions. In this chapter we developed and describe a particle-based approach to simulate
entrainment,transport, and settling of non-cohesive sediments in shallow waters. Sediment
distributions are modelled as a set of particles that are tracked on an individual basis by
solving Lagrangian transport equations that account for the drift part by the mean flow,
settling, and random horizontal motions.
The rest of this chapter is organised as follows. The brief description of an Eulerian
transport model is done in Section 3. The particle model for sediment transport is discussed
in Section 4. The interpretation of the partial differential equation called Fokker-Planck
equation into the well-known Eulerian transport model for sediment transport is described
in Section 5. The description and discussion of the two dimensional channel for a test case
of sediment transport is carried out in Section 7.
In our work we do carry out the estimation of the change in bedforms. Nevertheless, we do
not yet recompute the flow velocities when a change in the shallow water depth occurs.

2. Shallow water flow equations

In order for particle models to describe transport problems in shallow waters, the inputs such
as water flow velocities [U ( x, y, t), V ( x, y, t)] T , water levels ξ, water depths H ( x, y, t) and so
forth are required. In our application, the inputs are often computed by the hydrodynamic
model, which can solve the depth-averaged shallow water equations or 3 dimensional
shallow water [15]. The generated results in this case are written into a matlab format that
can be loaded and read in the particle model for simulation of sediment transport. The
inputs are assumed to satisfy the shallow water equations. The momentum equations are
represented by the following equations:

1
∂U ∂U ∂U ∂ξ U (U 2 + V 2 ) 2
+U +V + g − fV + g =0 (1)
∂t ∂x ∂y ∂x (Cz )2 H
1
∂V ∂V ∂V ∂ξ V (U 2 + V 2 ) 2
+U +V + g + fU + g = 0. (2)
∂t ∂x ∂y ∂y (Cz )2 H

The velocity is uniform over the vertical, therefore, for that reason, the rise and fall of free
surface is given by equations of conservation of mass called the continuity equation:

∂H ∂(UH ) ∂ (V H )
+ + = 0, (3)
∂t ∂x ∂y
312 Sediment Transport Processes and Their Modelling Applications
4 Sediment Transport

where

H = h + ξ is the total depth;

ξ is the water-level with respect to a reference;
h depth of the water with respect to a reference;
Cz bottom friction coefficient (Chezy coefficient);
g acceleration of gravity;
f Coriolis parameter.

The shallow water equations are entirely described by equations (1)-(3), provided the closed
and open boundary conditions and initial fields are given [9].

3. Eulerian sediment transport model

In this section we briefly introduce the Eulerian model for sediment transport. We consider
noncohesive type of sediment particle. The dynamics of the suspended particles can be
described by well-known Eulerian transport model with the source and sink terms included.
The following Eulerian sediment transport model is similar to that in [20], for example:

∂( HC ) ∂( HUC ) ∂( HVC ) ∂ ∂HC ∂ ∂HC

+ + − (D ) − (D )
∂t ∂x ∂y ∂x ∂x ∂y ∂y
= −γHC + E (U, V ) · λs . (4)

Where C ( x, y, t) is depth averaged concentration, γ is the deposition coefficient, E (U, V ) =

(U 2 + V 2 )(m2 s−2 ) is a function of flow velocities and the term λs · E (U, V ) models erosion
of sediment particles. The particle pick up function is parameterized as λs · E (U, V ), where,
λs is the erosion coefficient, it can be related to sediment properties (grain size, grain shape).
This parameterisation is motivated by the analysis of field observations reported in [14]
and reference therein. Typically, λs ≈ 3 × 10−2 (kgsm−4 ) for fine sand. In this article λs =
0.0001(kgm−4 s) is within the range reported in literature (see e.g., [14, 16]). Note that the term
γHC models the deposition of sediment and γ is the deposition coefficient, it is reported that
γ ≈ 4 × 10−3 s−1 [14] for fine sand.

3.1. Determination of bedlevel changes by Eulerian transport model

In addition to suspension and deposition processes, the following equation is used to
determine the depth changes and therefore the change of bed-level in each grid cell i, j:

∂h 1
= ( d e − s e ). (5)
∂t (1 − po )ρs

Where se = γHC stands for deposition and de = λs · E (U, V ) stands for erosion (term
responsible for suspending particles). Sea bed porosity is represented by po, ρs is the density
of sediment particles.
 Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 313
Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 5
[Link]
10.5772/52237

In this section we have constructed a simplified transport model which is derived from the
Eulerian transport model (4). This simple model is then made consistent with the simplified
Lagrangian particle model. In this way it becomes easy to compare the bedlevel changes
to see if they are similar. We simplified Equation (4) by assuming that the deposition and
erosion processes balance:

γCH = (U 2 + V 2 ) · λs . (6)

Quite often, transport in water is defined as the product of the concentration of sediment
particles C and a velocity U or V as well the depth of water in that grid cell. Thus, transport
along x and y directions is respectively given by;

q x = UCH and qy = VCH,

where in vector form q = [q x , qy ] T , using equation (6), it follows that

q = [(U 2 + V 2 )U · f d , (U 2 + V 2 )V · f d ] T , (7)

where f d = λγs stands for the drag force. This depends on the properties of a particle for
example its size or its area. Thus, in order to determine how much mass exits or comes into
a given location, it is important to consider the divergence. The divergence determines the
average rate of how much mass comes into the cell(change of mass per second per area):

∂m
= −div(q),
∂t
(8)

RT
where q = T1 0 qdt, div stands for divergence. Consequently, Equation (8) represents the
rate of how much mass stays behind or leaves the cell by assuming the absence of destruction
or creation of a matter. Since we want to determine the effects of sediment transport on the
sea bedforms, the equation for the bed level is represented by;

∂h 1 ∂m
= . (9)
∂t (1 − po )ρs ∂t

In the present application the determination of the bed level change using finite difference
scheme is estimated by the following equations:

∂h 1
=− · div(q).
∂t ρs (1 − po )
314 Sediment Transport Processes and Their Modelling Applications
6 Sediment Transport

For cases where flow is in one direction for example when v = 0, transport is given by the
following equation:

∂h fd
=− · div(U 3 ).
∂t ρs (1 − po )

With the aid of Equation (8)–(9), accordingly the determination of bed level changes is now
done by using the following equation:

 
− fd T ∂um ∂vm
∆h ≈ · + . (10)
ρs (1 − po ) ∂x ∂y

RT RT
where um = T1 0 (U 2 + V 2 )Udt and vm = T1 0 (U 2 + V 2 )Vdt. Next, let us now discuss the
Lagrangian particle model in the following section.

4. A particle model for sediment transport in shallow waters

A particle model is a description of a transport process by means of random walk models.
Random walk model is defined as the stochastic differential equation that describes the
movement of a particle that subsequently undergoes a displacement, which consists of the
drift part and a stochastic(diffusive) part [4, 13].

4.1. Integration of particle movement

In this section, the following 2-dimensional stochastic differential equations is developed:

h   i √
Itô D ∂H ∂D
dX (t) = U + H ∂x + ∂x dt + 2DdW1 (t),
h   i (11)
Itô D ∂H ∂D ) √
dY (t) = V + H ∂y + ∂y dt + 2DdW2 (t),

where the Brownian process W1 (t) and W2 (t) are Gaussian [3], and D ( x, y, t) is the horizontal
dispersion coefficient for sediment transport. Typically, D = O(10 − 100)m2 /s [14]. Note that
U ( x, y) and V ( x, y) are the flow velocities along the x and y direction respectively given in
m/s, H ( x, y) is the averaged total depth plus relative water levels due to waves, dW1 (t) and
dW2 (t) are independent increments of Brownian motions with mean (0, 0) T and covariance
❊[dW1 (t)dW2 (t)T ] = Idt where I is an identity matrix ([3, 10]). The simulation of sediment
transport is initiated with zero number of particles.

4.2. Deposition of sediment particles

We associate with each sediment particle a binary state which at any time t is given by


1 particle is in suspension
St =
0 otherwise (particle is on the sea bed).
 Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 315
Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 7
[Link]
10.5772/52237

Given a particle in suspension, we are interested in the transition of state 1 to state 0. In

continuous form, this transition can be modelled by the following equation

dP(St = 1)
= − γ · P( St = 1), where initially P ( S0 = 1 ) = 1 (12)
dt

where γ( x, y, t) is the deposition coefficient, in this chapter γ = γ( x, y, t) is constant, P(St =

1) is the probability that the state of the particle at time t is 1. When the particle is in the
flow, its evolution is described by the following transition probability equation in discrete
form:
P(St+∆t = 1 | St = 1) = P(S0 = 1) · [1 − γ · ∆t]
(13)
= [1 − γ∆t]
Assuming the system state (e.g. flow field and turbulence patterns) to be constant during
the period of the time step, it follows that the probability that a particle will be sedimented
is given by

P(St+∆t = 0 | St = 1) = 1 − [P(St+∆t = 1 | St = 1)]. (14)

4.3. Suspension of sediment particles

Mass represents concentration of a group of particles at a certain location. A source term is
included in our particle model such that the expected number of suspended particles (enp)
in grid cell i, j at time t is given by

∆x · ∆y · ∆t · U 2 + V 2 · λs
enp(i,j,t) = , (15)
Mp

where M p is the mass of each particle, ∆x and ∆y are the width of the grid cells along x
and y directions respectively, ∆t is the time step size, and λs is the erosion coefficient. For
each grid cell i, j we use the expected number of particles from Equation (15) to determine
the actual number of particles to be suspended. This is done by drawing a number from a
Poisson distribution function.

5. The relationship between the Kolmogorov Forward Partial differential

and The Advectiion Diffusion transport model
In this section the relationship between the Kolmogorov Forward Partial differential
corresponding to the 2-dimensional Stochastic differential equations and the well known
two dimensional advection diffusion equations for sediment transport is discussed in detail.
Since we are interested in the particle being in suspension, we assume that the particle at
position ( x, y) at time t has expectation of their mass h·i defined by;

hm( x, y, t)i = f ( x, y, t) · P(St = 1). (16)

316 Sediment Transport Processes and Their Modelling Applications
8 Sediment Transport

This is known as mass density of particles per unit area of the grid box. The Kolmogorov
Forward Partial differential is known as the Fokker-Planck equation(FPE)[3]. In this section
we incorporate the processes such as suspension and sedimentation states of the particles in
the model for transport of sediments in shallow water. We let D be the diffusion coefficient,
P(St = 1), is the probability that particle is in suspension, (St = 1) denotes a state that
a particle is in suspension and (St = 0) denotes the state that the particle is deposited
on the sea bed or bed of the shallow water. The stochastic process ( Xt , Yt ) is a Markov
process. The probability density function of the particle position f ( x, y, t) based on the two
dimensional Stochastic Differential equations (SDEs) (11) evolves according to the following
Fokker-Planck equation [9].

  
∂ f ( x, y, t) ∂ ∂H ∂D
=− U+( D )/H + · f ( x, y, t)
∂t ∂x ∂x ∂x
  
∂ ∂H ∂D
− V+( D )/H + · f ( x, y, t)
∂y ∂y ∂y
1 ∂2 1 ∂2
+ 2
( f ( x, y, t) · 2D ) + ( f ( x, y, t) · 2D ) . (17)
2 ∂x 2 ∂y2

The resulting sediment transport model (11) is an extension of two dimensional particle
model for pollutant dispersion in the shallow waters developed by [9]. The extension in
the present model includes the erosion and deposition terms. It is possible to derive the
Fokker-Planck equation that describes the probabilistic transport of the sediments from one
location to another. The derivation of the Fokker-Planck equation is e done as follows, we
first differentiate equation (16) with respect to time t, to obtain

∂ ∂ ∂
hm( x, y, t)i = P(St = 1) f ( x, y, t) + f ( x, y, t) P(St = 1),
∂t ∂t ∂t

next with the aid of Equation (12), it follows that,

∂ ∂
hm( x, y, t)i = P(St = 1) f ( x, y, t) − γ f ( x, y, t) · P(St = 1). (18)
∂t ∂t

Therefore, we also add the erosion term to Equation (18) and come up with

∂ ∂
hm( x, y, t)i = P(St = 1) f ( x, y, t) − γ f ( x, y, t) · P(St = 1) + λs · E (U, V ) (19)
∂t ∂t

next we multiply on both sides of Equation (17) by P(St = 1) to obtain

  
∂ ∂ D ∂H ∂D
P(St = 1) f ( x, y, t) = − U+ + f ( x, y, t) · P(St = 1)
∂t ∂x H ∂x ∂x
  
∂ D ∂H ∂D
− V+ + f ( x, y, t) · P(St = 1)
∂y H ∂y ∂y
1 ∂2 1 ∂2
+ ( 2D · f ( x, y, t ) P ( S t = 1 )) + (2D · f ( x, y, t) · P(St = 1)) . (20)
2 ∂x2 2 ∂y2
 Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 317
Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 9
[Link]
10.5772/52237

The substitution of Equation (20) into equation (19) gives the following Fokker-Planck
equation which now have extra terms that model the deposition and erosion of sediments.

  
∂hm( x, y, t)i ∂ D ∂H ∂D
=− U+ + · hm( x, y, t)i
∂t ∂x H ∂x ∂x
  
∂ D ∂H ∂D
− V+ + · hm( x, y, t)i
∂y H ∂y ∂y
1 ∂2 1 ∂2
+ 2
(2D · hm( x, y, t)i) + (2D · hm( x, y, t)i)
2 ∂x 2 ∂y2
− γ · hm( x, y, t)i + λs · E (U, V ). (21)

The last two terms in the transport equation (21) represents the process of sediment
deposition and erosion of sediments respectively. An average mass hm( x, y, t)i per unit area
(kg/m2 ) of a particle is related to a particle depth averaged concentration C ( x, y, t) in mass
per unit volume (kg/m3 ) ([9, 18]).
The concentration of materials is given in kg/m3 , therefore, the expected mass of a particle
at position ( x, y) can be related by the concentration in that grid cell(location) as follows:

hm( x, y, t)i = H ( x, y, t) · C ( x, y, t). (22)

Substitution of equation (22) into the Fokker-Planck equation (21) leads to the two
dimensional Advection diffusion partial differential equation commonly known as Eulerian
sediment transport model (4). Consequently, the transport equation (4) which was discussed
in Section 3, is consistent with the particle model for sediment transport (11)-(26).
Therefore, after having constructed the particle model for sediment transport, it is now
necessary to develop the equations that cater for the bed level changes using the particle
model.
In the next section we shall briefly discuss the numerical approximation of our particle model.

6. Numerical approximation of the particle model

Euler scheme is used in the numerical implementation of the particle model. The scheme is
convergent in the weak sense with accuracy of order O(∆t) and it is half order accurate in
the strong sense. Higher order schemes for stochastic differential equations are described
in [11]. The discretisations of the hydrodynamic flow models is widely discussed by
[15], for example. The particle model (11)-(26) is discretised and uses the following Euler
scheme to approximate the numerical solutions. We discretise the two dimensional stochastic
differential equations for integrating the movement of the particle in similar way to that as
in [9] with the modifications by the inclusion of the sedimentation and deposition parts:

 
∂H ∂D √
X̄ (tk+1 ) = X̄ (tk ) + U + ( D )/H + ∆tk + 2D∆W1 (tk ) (23)
∂x ∂x
318 Sediment Transport Processes and Their Modelling Applications
10 Sediment Transport

 
∂H ∂D √
Ȳ (tk+1 ) = Ȳ (tk ) + V + ( D )/H + ∆tk + 2D∆W2 (tk ) (24)
∂y ∂y

Pk+1 (St = 1) = (1 − γ( x, y, t)∆k )Pk (St = 1). (25)

Where X̄ (tk+1 ) and Ȳ (tk+1 ) are the numerical approximations of X (t) and Y (t) respectively,
while X̄ (t0 ) = X (t0 ) = x0 and Ȳ (t0 ) = Y (t0 ) = y0 are initial locations of a particle. In
addition to Eqns. (23)-(25), we also use Eqns. (15)-(26) to make the simulation of sediment
transport complete. There are several schemes that can be used for simulation process for
instance, Euler, Heun, Milstein scheme and Runge kutta methods. Much detailed work on
numerical methods for stochastic models can be found in (e,g., [11]).
Numerical schemes such as the Euler scheme often show very poor convergence behaviour.
This implies that, in order to get accurate results, small time steps are needed thus requiring
much computation. Another problem with the Euler (or any other numerical scheme) is
its undesirable behaviour in the vicinity of boundaries; a time step that is too large may
result in particles unintentionally crossing boundaries. Therefore, the treatment of boundary
condition for particle Models is often done in the following section as follows.

6.1. Boundaries
One problem with numerical integration of particle positions arises in the vicinity of
boundaries. Given the current location, ( X (t), Y (t)), we may find that the new location,
( X (t + ∆t), Y (t + ∆t)), is on the other side of a boundary, i.e. the particle has crossed a
boundary. Depending on the type of boundary this may be physically impossible. We
consider two types of boundaries. The first type, closed boundaries, represents boundaries
intrinsic to the domain such as banks, sea bed, and coastal lines. The second type of
boundaries are open boundaries, which arise from the modeller’s decision to artificially
limit the domain because particles are not expected to reach any further or simply because
no domain information is available at those locations. It is clear that it is undesirable to have
particles cross the first type of boundary, whereas for the second type it is quite natural.
Based on this classification, we apply the following rules to particles crossing borders during
integration:

• In case an open boundary is crossed by a particle, the particle remains in the sea but is
now outside the scope of the model and is therefore removed.
• In case a closed boundary is crossed by a particle during the drift step of integration, the
step taken is cancelled and the time step halved until the boundary is no longer crossed.
However, because of the halving, say n times, the integration time is reduced to 2−n ∆t,
leaving a remaining (1 − 2−n )∆t integration time, which, at a constant step size, requires
at least another 2n − 1 steps in order to complete the full time-step ∆t. Note that at each of
these steps it may be needed to further reduce the step size. This further reduction applies
only to the current time step, leaving the step size of following sub-steps unaffected. This
method effectively models shear along the coastline.
 Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 319
Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 11
[Link]
10.5772/52237

• If a closed boundary is crossed during the diffusive part of integration, the step size
halving procedure described above is maintained with the modification that in addition
to the position, the white noise process is also restored to its state prior to the invalidated
integration step. The process of halving the time step and continuing integration with
the reduced step size is repeated until the full ∆t time step has been integrated without
crossing a boundary.

In addition, this chapter have also considered the pick up of sediment particles at the inflows
this will be discussed in the next section.

6.2. Particle flux at open boundaries

We now consider particle flux at open boundaries. This flux is the difference between
particles flowing into and out of the domain. The number of particles flowing out should
not be controlled as it is a natural consequence of the movement of a particle. As soon as a
particle crosses an open boundary it is considered gone and further integration is no longer
possible as no data outside the domain is given. For this very reason, however, we do need
to explicitly model the particles flowing in. We determine the expected number of particles
entering the domain as follows:




∆y·∆t·V ·(U 2 +V 2 )·λs
inflow parallel to y-axis
 γM p
enp(i,j,t) = (26)


 ∆x ·∆t·U ·(U 2 +V 2 )·λs
inflow parallel to x-axis,
γM p

where, γ is the deposition coefficient. The actual number of particles added at the domain
boundary at each iteration is obtained by drawing a value from the Poisson distribution
parameterised by the above expectation value.
To determine the actual number of particles to be suspended in a grid cell i, j we
draw a number from a Poisson distribution function with mean enp(i,j,t) determined by
Equation (15). We assume that particles are infinitely many on the sea bed. However, the
particle that is suspended is not the same as the one that is deposited.
Before implementing the particle model (11)-(26), we first required to show the consistence
between the Fokker-Planck equation and its the Eulerian transport model. This will be
described in the next Section.

7. Determination of bedlevel changes using particle models

Comparing with a simplified form of Equation (4), where in this case we assume for the local
change in mass ∂m ∂t ≈ de − se . We also assume that the deposition and erosion processes
balance (see Section 3.1). The approximation of the change in mass with respect to time, in
each grid cell i, j in the this particle model is determined by the following equations:

∂m ∆N p 1
= Mp. (27)
∂t ∆t ∆x∆y
320 Sediment Transport Processes and Their Modelling Applications
12 Sediment Transport

∂m
Using equation (5) and the fact that ∂t ≈ de − se , the equation for the bed level change can
be derived as follows,

∂h ∆N p 1
= Mp.
∂t ∆t ρs (1 − po )∆x∆y

Where, M p is the mass of each particle, ρs and po denote respectively the density of an
individual grain particle and the bed porosity. From [14], we find ρs = 2650kg/m3 and
po ≈ 0.5. ∆N p is the difference between the number of deposited and suspended particles at
each iteration in each grid cell i, j. Hence the cumulated (integrated) change of the level of
the sea ∆h in m for all time steps is determined by the following equation:

Z T
∂h
∆h = dt
0 ∂t
∆N p
∆h ≈ ∑ ρs (1 − po)∆x∆y M p . (28)

More information about the effect of parameters on the sea bed level changes can be found
in [16], for example.

7.1. Primary input of the model

The initial field is defined as

depth = h0 + h1 · exp(−(( x − 0.0).2 )/(2 · (wd)2 )),

where the initial amplitude of the disturbance is h1 = 10.0, width of disturbance (wd) =
2000m, the tidal period T = 720 minutes. Constant for sediment transport rate K =
0.16kgs3 /m6 , porosity p0 = 0.5, ρ = 2600[kg/m3 ], density of sediment, horizontal domain
length of x goes from −10000m to 10000m. The horizontal domain length of y goes from 2m
to 4500m.
Diffusion constant D = 10m2 /s, starting time Tstart = 0s, number of seconds in a year
Tyear = 365 · 24 · 60 · 60s/year. Final time s Tstop = 100 · Tyear, M = 50 is the number of grid
points across the channel, dt = 1.0 · year is the time-step s. Tidal mean discharge per unit
width m2 /s is given by q0 = h0 · u0 , tidal amplitude discharge per unit width m2 /s is given
by q1 = h0 · u1 . We then determine the changes in the bedforms as described in Section 3.1.
The rate of the changes in the level of the floor of the sea is described by the divergence
of transport in each grid cell. For comparison purposes, the divergence of transport is we
computed by using the original data from the hydrodynamic model as described in this
section.
Note that the two dimensional channel in a Matlab code is used whereby in the routine,
we first compute the flow fields U, V as well as its depths in each grid cell. Then compute
average load transport and finally compute the divergence using a finite difference methods
in a Matlab code. Where the divergence of the U-vector field along x and the V-vector field
 Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 321
Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 13
[Link]
10.5772/52237

along y is evaluated using central differences wherever possible and forward or backward
differences on the boundaries. Note that the average change of the bedforms in each grid
cell are estimated by using Equation (10) in Section 3.1. The results obtained are eventually
compared with those obtained when the estimations of the changes of bedforms are carried
out using Equation (28) with the aid of the two dimensional SDEs or sometimes in this
chapter we call it as the Lagrangian particle model. This stochastic model is discussed in
Section 7. Note that both results due to Equation (10) and Equation (28) are similar interms
of shapes, (see Fig. 2).
−4 −4 Rate of bedlevel
x 10 x 10
1.5
4
1

0.5
2
∆ h [m/year]

0
−0.5

−1
−2

−1.5
−1 −0.5 0 0.5 1 −1 −0.5 0 0.5 1
x 4 x−axis 4
x 10 x 10

(a) (b)

Figure 1. Change of bed level in m/year for a two dimensional channel (a) is due to the particle model, while (b) the result is
computed by using the Eulerian approach.

−4
x 10
−4
3
x 10
4
2
2
∆ h [m/year]

1
0

−2 0

−4
6000 −1
4000 1
2000 0 4
x 10 −2
y 0 −1 x

(a) (b)

Figure 2. Change of bed level in m/year for a two dimensional channel (a) is due to the particle model, while (b) the result is
computed by using the Eulerian approach.

The Figures 1 and 2 represent the results obtained by solving the same problem by using
two approaches. Fig. 1(a) and Fig. 2(a) are due to a simplified particle model with very
small effect of the diffusion. On the other hand, in Figures 1(b) and Fig. 2(b) are due to a
simplified Eulerian model, no diffusion is considered. We should expected to get deposition
at the retardation of the flow and erosion at the acceleration of the flow. Resulting in the
net migration of the channels in the direction of (U 3 ) as in [19]. The positive sign on the
colorbar in the figures imply that deposition is taking place while the negative sign implying
the occurrence of erosion of sediments.
Some results, in Fig. 3(a,b), represent the local depth change in two selected cells. Part (a)
shows a steady deposition in the grid cell at the location ( x, y) = (5km, 2km) whereas another
grid cell in the location ( x, y) = (−2km, 0.8km), part (b), there is also a steady deposition. The
diffusion coefficient in the test case for the particle model is 0.00001m2 /s, γ = 0.00013s−1 .
322 Sediment Transport Processes and Their Modelling Applications
14 Sediment Transport

−4
x 10 (x,y)=(−5000m, 800m)
10
0.04

8
0.03
6
∆ h [m/year]

∆ h [m/year]
0.02
4

0.01
2

0 0

−2 −0.01
Jan ’99 Feb ’99 Mar ’99 Apr ’99 Jan ’99 Feb ’99 Mar ’99 Apr ’99
time time

(a) Location at ( x, y) = 5000, 2000) (b) Location at ( x, y) = −5000, 800)

Figure 3. Cumulative local changes in depth by using data of 90 days for two selected grid cells in the ideal two dimensional
channel domain light particles

The parameters for the test case experiments using ideal domain with the PM are found in
(Table 1.)
Constant Unit Value Constant Unit Value
porosity - 0.50 γ s −1 0.00013
grid offset m (−10000, 2) λs kg·s·m−4 0.001
grid size - 22 × 101 D m2 · s −1 0.000010
cell size m 200 × 200 δ - 0
sand density kgm−3 2650 Mp kg 3000
initial location m (1800, 400) f = λγs kg · s2 · m−4 O(10−4 )

Table 1. Parameters used by particle model for the sediment transport in the test case

8. Transport of heavy particles in shallow water

As mentioned earlier, sediment transport is a complex process determined by various
properties of the sediment materials. Let us in this section consider that we deal with heavy
particles. Heavy particles unlike the lighter ones, tend to attain the equilibrium much faster
than the tidal cycle. The description of the derivation of the equations for the heavy particles
have followed similar lines using the same equations as those for the lighter particles.

∂(CH ) ∂( HUC ) ∂( HVC )

+ + = Se − Sd (29)
∂t ∂x ∂y

where in this case Se = λs U k is erosion term, −Sd = γCH, is the deposition term
where λs , γ, k are constants. In addition, an equation for the change of the bedlevel is
presented:

∂h 1
= · ( Se − Sd ). (30)
∂t ρs (1 − po )
 Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 323
Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 15
[Link]
10.5772/52237

However, the difficulty with the heavy (coarser) material is that the parameter γ becomes
larger (settling velocity) and this makes the equation (29) ’stiff’. The maximum allowed time
step thus becomes very small and since we want to make long simulation this is not very
convenient. Therefore, the solution is to make a first order approximation and assume that
the two source terms Se and Sd are much bigger than the right side terms of Equation (29).
Therefore, we can write equation (29):

 
∂(CH ) ∂( HUC ) ∂( HVC )
δ + + = Se − Sd , (31)
∂t ∂x ∂y

now we substitute a Taylor series expansion of the depth averaged concentration CH

CH = Q = Q0 + δQ1 + O(δ2 ),

into Equation (31) to get the following equations such that 0-order is given by :

Se − Sd = 0, (32)

while the 1st order is:

∂ ( Q0 ) ∂(UQ0 ) ∂(VQ0 )
+ + ) = −(γQ1 ). (33)
∂t ∂x ∂y

In other words, in Equation (33), we are looking for a source term (γQ1 ) that balances
the advective transport of the 0 -order solution. Although this is easy in finite difference
approach, however, in the particle model, Equation (33) can be approached as follows. The
best we can do so far is to solve:

∂Q ∂(UQ) ∂(VQ)
+ + = 0, (34)
∂t ∂x ∂y

with (32) as an initial condition for every time step separately and then set

∂Q
γQ1 ≈ . (35)
∂t

In other words, Equation (33) should be more or less balanced. If we now omit the source
term and measures the rate of changes, these should be approximately equal. Therefore for
the heavy particles, between the beginning and end of the integration time loop should do
the following

(i) First we remove all deposited particles.

(ii) Followed by generating particles according to (see Equation 32)
324 Sediment Transport Processes and Their Modelling Applications
16 Sediment Transport

(iii) Store net change in number of particles (concentration).

(iv) Next we do one time step integration of the particle using Equation (34).
(v) Compute differences over the previous step using (iii).

For now we have assumed that

∂Q0
| | ≪ |γQ1 |.
∂t

Note that

λs
transport vector q x = UHC = U (U 2 + V 2 ) (36)
γ
λs
transport vector qy = V HC = V (U 2 + V 2 ) , (37)
γ

since λs U 2 − γCH = 0, the equation for the rate change of the bedlevel due to the transport
coarse material is given by

 
∂h ∂q x ∂qy 1
=− + · , (38)
∂t ∂x ∂y ρs (1 − po )

λs
where f d = γ is the pick up function which depends into the characteristics of the
sediment/sand materials. For example in case of a mixture of larger sand of volume=l 3 will
have a mass=ρs l 3 . While a sand of double size whose volume is 8l 3 will have mass=ρs 8l 3 .
Therefore, the two particle require different value of the drag force particle f d . But in this
chapter we assume that all particles have the same mass, l=length.

For a fixed grid cell For a fixed grid cell

0 0.16

−0.02 0.14

−0.04 0.12
change in h meters
change in h meters

−0.06 0.1

−0.08 0.08

−0.1 0.06

−0.12 0.04

−0.14 0.02

−0.16 0
Jan ’99 Feb ’99 Mar ’99 Apr ’99 Jan ’99 Feb ’99 Mar ’99 Apr ’99
time (days) time (days)

(a) Local erosion for heavy particles (b) Local deposition for heavy particles

Figure 4. Cumulative local changes in depth by using data of 90 days for two selected grid cells in the ideal two dimensional
channel domain for heavy particles
 Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 325
Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 17
[Link]
10.5772/52237

9. Discussion and conclusions

In this chapter we have developed a two dimensional stochastic differential equations. This
type of Mathematical model is known as Particle model in some cases. In particular we have
developed a two dimensional particle model for sediment transport in which we have added
two more equations to model erosion and deposition processes. The description of Transport
of materials in shallow waters or in atmosphere can be described by Eulerian approach.
This is a deterministic approach in which Partial Differential Equations are used. Numerical
methods are usually implemented to approximate the solutions of the PDEs. In this case,
one can be faced with computation problems if the dimensional of the model is high. An
in most cases if you modify the model to include in more processes it may not be possible
to get a closed solution. An alternative approach called the Lagrangian approach can be
also used to describe the transport of sediments[35]. This is a probabilistic approach that
uses the transition probabilities to derive the state of the transported material. The transition
probabilities or density functions is the solution of the Fokker -Planck equation which is also
a type of PDEs.
The crucial part is to show that FPE is consistent with ADEs, once that is done then one
can derive the underlying SDEs. The derived SDEs can be used as a particle model for the
simulation of the transport of materials in the shallow or atmosphere.
In this chapter we have derived the Fokker-Planck equation and included the deposition
and erosion terms based on the developed particle model. Moreover, we have also shown
that by interpreting the Eulerian sediment transport as the Fokker Planck equation with
the additional terms, it becomes possible to derived the underlying particle model that is
consistent with the Eulerian sediment transport model. Furthermore, the results of sediment
transport due to particle model has been compared to that obtained by computing the
transport using Eulerian model in their simplest form. We have got some results for the
changes in the bed level for the data of 90 days for an idealized two dimension domain. We
have also used our model to test the prediction of bed-level changes by using the approach of
parallel computations of bedforms using the real data of the Dutch North sea[5]. Therefore,
at least for now we can say that we have solved the set of mathematical equations called
particle model for sediment transport. These equations have given us reasonable results
for the sea bedlevel changes. Nevertheless, the determination of the morphological changes
is a complicated process that depends on several factors such as waves, the size of sand,
mass and density[36]. The particle model in this work has been simplified, what we can
say is that the results are reasonable. But more factors will have to be taken into account.
For instance, in the particle model we have considered that all particles have equal mass
while in reality each particle has different mass. For better predictions of the complex
behaviour sediment trasnsport that vary with time and space a feedback among water
motion, sediment transport and bottom changes, we recommend the coupling of both the
hydrodynamic and the transport models. Moreover, make sure the grid mesh are of the
same form for the hydrodynamic and particle model. Sometimes you may also be required
to change the number of particles into concentration. That can be done by using a function
called point spread function(PSF).
326 Sediment Transport Processes and Their Modelling Applications
18 Sediment Transport

Author details
Charles Wilson Mahera and Anton Mtega Narsis
1 University of Dar-es-salaam, College of Natural and Applied Sciences, Department of
Mathematics, Dar-es-salaam, Tanzania
Department of General Studies, Dar es Salaam Institute of Technology, Dar es Salaam,
Tanzania

References
[1] P. S. Addison, Bo Qu, A. Nisbet, and G. Pender.(1997). A non-Fickian particle-tracking
diffusion model based on fractional Brownian motion. International Journal for numerical
methods in fluids, Vol, 25, No. 0, pp. 1373–1384.

[2] B.Øksendal (2005). Stochastic Differential Equations. An introduction with applications, 6th
Ed, Springer, New York,.

[3] Arnold. L. (1974). Stochastic differential equations: Theory and applications, Wiley, London.

[4] R. W. Barber & N. P. Volakos (2005). Modelling depth-integrated pollution dispersion

in the gulf of thermaikos using a lagrangian particle technique, Proceedings of Inter.
Confer. on Water Resources Management, pp. 173–184, ISBN, Portugal, April and 2005,
WIT Trans. on Ecol. and the Env., Vol 80, Algarve

[5] W. M. Charles, [Link] den Berg, H. X. Lin, A. W. Heemink, and M. Verlaan

(2008). Parallel and distributed simulation of sediment dynamics in shallow water
using particle decomposition approach. International Journal of Parallel and Distributed
Computing,Vol 68 717Ű728.

[6] H. B. Fischer, E.J. List, R.C.Y. Koh, J. Imberger, & N.H. Brooks (1979). Mixing in Inland
and Coastal Waters, 6th Ed,Academic Press, New York.

[7] C. W. Gardiner, (2004). Handbook of Stochastic Methods for physics, Chemistry and the
Natural Sciences, Springer-Verlag, Berlin.

[8] J. Fredsøe and R. Deigaard, (1992). Mechanics of Coastal Sediment transport, World
Scientific, Singapore.

[9] A. W. Heemink (1990). Stochastic modeling of dispersion in shallow water Stochastic

Hydrology and hydraulics, Vol. 4, :161-174

[10] A. H. Jazwinski, (1970). Stochastic Processes and Filtering Theory, Academic Press, New
York.

[11] P. E. Kloeden and E. Platen, (1999). Numerical solutions of Stochastic Differential equations.
Application of Mathematics,Stochastic Modelling and applied probability,Springer-Verlag,
Berlin Heidelberg.

[12] J. Fredsøe and R. Deigaard, (1992). Mechanics of Coastal Sediment transport, World
Scientific, Singapore.
 Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 327
Modelling of Sediment Transport in Shallow Waters by Stochastic and Partial Differential Equations 19
[Link]
10.5772/52237

[13] A. Matheja (2000). Sediment transport using the random walk method on top of cobra
Proc. of the 4th of Inter. Confer. on Hydroinformatics, Iowa City, USA, 1–7

[14] H. M. Schuttelaars and H. E. de Swart (1996). An idealized long-term morhpodynamic

model of a tidal embayment Eur. [Link]., B/Fluids Vol. 15(1), :55–80

[15] G. S. Stelling (1983). Communications on construction of computational methods for

shallow water flow problems PhD Dissertation, Delft University of Technology, The
Netherlands

[16] K. de Jong (1998). Tidally Averaged Transport Models PhD Dissertation, Delft University
of Technology, The Netherlands.

[17] A. F. B. Tompson and L. W. Grelhar(1990). Numerical simulation of the solute transport

in three-dimensional,randomly heterogeneous porous media Water Resources Research,
Vol. 26(10) :2541–2562.

[18] H. F. P. van den Boogaard, M. J. J . Hoogkamer, and A. W. Heemink (1993). Parameter

identification in particle models Stochastic Hydrology and hydraulics, Vol. 7(2), :109-130.

[19] J. van den Kreeke, S. E. Hoogewoning, and [Link],H. F. P. van den Boogaard, M. J.
J . Hoogkamer, and A. W. Heemink (1993). Analytical model for the morphodynamics
of a trench in the presence of tidal currents Continental Shelf Research, Vol. 22(11-12),
:1811-1820.

[20] L. C. van Rijn (1993). Principles of sediment transport in rivers, estuaries and

[21] Amoudry, L. O., and A. J. Souza, A. J. (2011) .Deterministic coastal morphological and
sediment transport modelling: A review and discussion", Rev. Geophys., 49, RG2002,
doi:10.1029/2010RG000341.

[22] Bhuiyan, F. 2011 Stochastic and Deterministic Methods of Computing Graded Bedload
Transport, Sediment Transport - Flow and Morphological Processes", Prof. Faruk
Bhuiyan (Ed.), ISBN: 978-953-307-374-3.

[23] Brown, J. M., & Davies, A. G. (2009) Methods for medium term prediction of the net
sediment transport by waves and currents in complex coastal regions", Cont. Shelf Res.,
29, pp.1502-1514.

[24] de Vriend, H. J. (1987)2DH mathematical modelling of morphological evolutions in

shallow water, Coastal Eng., Vol 11, pp. 1-27.

[25] Dewals, B, Rulot, F., Erpicum, S., Archambeau, P. & Pirotton, M. (2011) Advanced Topics
in Sediment Transport Modelling: Non-alluvial Beds and Hyperconcentrated Flows, Sediment
Transport, Dr. Silvia Susana Ginsberg (Ed.), ISBN: 978-953-307-189-3.

[26] Frey, P. & Church, M. (2009) How River Beds Move, Science, Vol 325(5947), pp. 1509 -
1510.
328 Sediment Transport Processes and Their Modelling Applications
20 Sediment Transport

[27] Harris, C. K., Sherwood, C. R., Signell, R. P., Bever, A. J., & Warner, J. C. (2008)
Sediment dispersal in the northwestern Adriatic Sea, J. Geophys. Res., 113, C11S03,
doi:10.1029/2006JC003868.

[28] Hu, K., Ding, P., Wang, Z., and Yang, S. (2009) A 2D/3D hydrodynamic and sediment
transport model for the Yangtze Estuary, China, J. Mar. Syst., Vol 77, pp. 114-136.

[29] Lumborg, U. (2005) Modelling the deposition, erosion, and flux of cohesive sediment
through Øresund, J. Mar. Syst., Vol 56, pp. 179-193.

[30] Man, C. (2007) Stochastic Modelling of Suspended Sediment Transport in Regular and
Extreme Flow Environments, PhD Dissertation, State University of New York at Buffalo,
p. 182.

[31] Németh, A. A. & Hulscher, S. J. M. H., (2003) Finite amplitude sand waves in shallow
seas, IAHR-symposium on River, Coastal and Estuarine Morphodynamics, Barcelona,
Spanje, editors Sánchez-Arcilla, S. en Bateman, A., pp. 435-444.

[32] Rodean, H.C. (1996) Stochastic Lagrangian Models of Turbulence Diffusion, American,
Meteorological Society, 84pp, Boston.

[33] Struiksma, N., Olesen, K. W., Flokstra, C., & de Vriend, H. J. (1985) Bed deformation
in curved alluvial channels, J. Hydraul. Res., Vol 23(1), pp. 57-79.

[34] Papanicolauo, A. N., Elhakeem, M., Krallis, G., Prakash, S., & Edinger, J., (2008)
Sediment Transport Modeling Review-Current and Future Developments, Journalof
Hydraulic Engineering, ASCE, 14pgs

[35] Hunter, J. R,Crais, P.D and Phillips H. E.(1993) On the use of random walk models
with spatially variable [Link] Physics, 106:366–376, 1993.

[36] H. M. Schuttelaars and H. E. de Swart (2000) Multiple morphodynamic equilibria in

tidal embayments.J. Geophysical Research, 105(c10):14,105–14,118.
 Chapter 13

Numerical Modeling Tidal Circulation and

Morphodynamics in a Dumbbell-Shaped Coastal
Embayment

Yu-Hai Wang

Additional information is available at the end of the chapter

[Link]

1. Introduction

Channel-shoal (ridge) system is a common morphological feature in wide, shallow coastal

bays and estuarine mouth where tidal flow is relatively stronger. Sediment transport and
morphological evolution is complex within such a system as constrained by interacting tidal
force, river current, sediment source and characteristics, shoreline configuration, etc.
As the deeper tidal channels are usually utilized as navigation courses or water supply
source for coal & nuclear power plants or other engineering purposes, it is vitally important
to maintain the stability of these tidal channels, that is, they should not be allowed to mi‐
grate, merge or perish by siltation.
This chapter chooses the dumbbell-shaped Qinzhou Bay as the study site to investigate the
sediment transport process and resultant morphological evolution of the channel-shoal sys‐
tem within the Bay using numerical simulations under the status quo situation. This is bene‐
ficial to the planned large-scale coastal engineering projects that might exert a profound
long-term influence upon the stability of the channel-shoal system.

1.1. Research question

Qinzhou Harbor is one of the most important sea harbors connecting southwestern China’s
inland and the southeast Asian countries, it delivered up to 47.162 million ton cargos in
2011. The harbor has been using waterways as its navigational channel. Meanwhile, coal &
nuclear power plants, industrial development zones, recreational parks, land reclamations
and many other coastal projects have been or planned to be constructed around the Qinzhou
Bay coast, all of which compete for the limited shoreline and water area resources.
330 Sediment Transport Processes and Their Modelling Applications

For the optimum planning of the coastal engineering projects regarding size, location and
the sustainable regional economic, social and environmental development, it is critically ur‐
gent to know how the stability of the channel-shoal system under the present coastal config‐
uration and bathymetry. This might be answered by investigating the sediment transport
processes and morphodynamics of the channel-shoal system.

1.2. Site description

Qinzhou Bay is located on Guangxi Province’s coast facing South China Sea (Figure 1).

Figure 1. Sketch map of Qinzhou Bay coastal configuration (the axis is local coordinate, unit: m).

Appearing as a dumbbell shape it consists of three parts: the inner enclosed bay, also
known as Maoweihai, the outer bay and the Yingling tidal inlet connecting two bays.
Two rivers, i.e. Maolingjiang and Qinjiang flow into the inner bay, delivering annually
27.73×109 m3 water and 86.4×103 t sediment; the outer bay is a show, trumpet-shaped bay,
its area is nearly 2.55×108 m2 with a mean depth of 4.67m (calculated by mean sea level).
A complex channel-shoal system is present within the outer bay, consisting of three domi‐
nant tidal waterways, i.e. the east waterway, middle waterway and west waterway and
 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 331
[Link]

sand ridges/shoals between them (Figure 1). The Yingling tidal inlet is 10.1 km long and
1.1~3.5 km wide with a water depth of 5~20 m; it is a rocky inlet with a total of 71 vari‐
ous-sized islands and 72 narrow, small waterways.

2. Dynamics

2.1. Tidal regime

The spring tide and middle tide in Qinzhou Bay are diurnal throughout most of the year but
become irregularly diurnal in March and September each year, while the neap tide is usual‐
ly semi-diurnal throughout the whole year. The mean tidal range is 2.51m, and the maxi‐
mum tidal range is 5.27m. The flood tides last longer than the ebb tides in spring, middle
and neap tides. It is 13 hours plus 14 minutes, 11 hours plus 18 minutes for spring tide; 14
hours plus 36 minutes and 10 hours plus 7 minutes for middle tide, and 6 hours plus 33 mi‐
nutes and 5 hours plus 40 minutes for neap tide, respectively [1].

Tidal flow in the outer bay demonstrates as reciprocating flow in parallel to the major water‐
ways; the mean flood velocity, ebb velocity of spring tide is 0.37m/s and 0.51 m/s, respec‐
tively, the mean flood velocity, ebb velocity of middle tide is 0.33m/s and 0.38 m/s,
respectively, the mean flood velocity, ebb velocity of neap tide is 0.22m/s and 0.18 m/s;
while flow velocity in the Yingling inlet becomes significantly larger, the mean flood veloci‐
ty, ebb velocity of spring tide is 0.67m/s and 0.90 m/s, the mean flood velocity, ebb velocity
of middle tide is 0.57m/s and 0.68 m/s, the mean flood velocity, ebb velocity of neap tide is
0.42m/s and 0.33m/s, respectively, and the maximum flood flow velocity reaches up to 1.40
m/s and the maximum ebb flow velocity is up to 1.32 m/s [1].

2.2. Wave climate

The Qinzhou Bay is influenced by subtropical monsoon and the waves within the Bay are
mainly wind-driven with some surge waves traveled from the open sea.

The waves in winter season (October~Apirl) prevail in N-NE direction while they prevail in
S-SW direction in summer season (May-September) and the stronger waves propagate in
SSW, SSE directions; the mean wave height is 0.52m and mean wave period is 3.1s [2].

3. Sediment characteristics

3.1. Suspended sediment

Suspended sediment concentration within Qinzhou Bay water is generally low. In summer
2009, the mean full tidal concentration is 0.022 kg/m3, among which, it is 0.035 kg/m3 in
spring tide, 0.020 kg/m3 in middle tide and 0.013 kg/m3 in neap tide; the maximum concen‐
tration is 0.081 kg/m3, occurred in spring ebb tide, the maximum middle tidal concentration
332 Sediment Transport Processes and Their Modelling Applications

is 0.034 kg/m3, occurred also in ebb tidal period, the maximum neap tidal concentration is
0.025 kg/m3, occurred also in ebb tidal period [1].

The medium diameters of suspended sediment vary within 0.0067~0.0152mm with a mean
value of 0.0101mm. The suspended sediment is mainly clayey silt with 30.8% clay particles,
53.4% silt particles and 15.8% sand particles (Table 1). The sorting index is 1.90 [1].

Grading sand silt clay Sorting index

percentage% 15.8 53.4 30.8 1.90

Table 1. Grading of suspended sediment.

3.2. Bottom sediment

The bottom sediments in Qinzhou Bay mainly consist of gravel, coarse sand, medium sand,
fine sand, silty clay and clayey silt, etc.

Figure 2. Bottom sediment grain size distribution in Qinzhou Bay.

 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 333
[Link]

The gain sizes vary remarkably in 0.0027~1.099mm (Figure 2). The spatial mean median di‐
ameter (D50) in the inner bay is 0.334mm; it is 0.356mm in the deep channel but becomes
0.0041mm in the shallow parts in the Yingling inlet; in the outer bay, the spatial mean D50 is
about 0.298mm with an overall deposition pattern: coarser in waterways but finer in shoals,
and coarser in the western part than in the eastern part of the bay [1].

3.3. Sediment source

The Qinzhou Bay has been a drowned rocky valley by the last sea level transgression since
7,000-8,000 year before present [3]. Therefore, the huge amount of sand deposits in the outer
bay has come from the deposits by paleo-Maolingjiang river and paleo-Qinjiang river, they
have been reformed into the contemporary channel-shoal geomorphology by tidal dynamics.
At the present day, sediments delivered by these two rivers are deposited within the inner bay
with limited amount of fine particles transported into the outer bay and open sea; meanwhile,
limited amount of sediment eroded from the adjacent slopes by storm rains also enter the outer
bay. Generally speaking, sediment from the open sea into the Qinzhou Bay is very limited.

4. Numerical model

A 3D unstructured grid, finite-volume coastal ocean model (called FVCOM) has been devel‐
oped in the Marine Ecosystem Dynamics Modeling Laboratory led by Dr. C. Chen at the
University of Massachusetts–Dartmouth (UMASS-D) in collaboration with Dr. R. Beardsley
at the Woods Hole Oceanographic Institute. FVCOM is a three-dimensional (3D) primitive
equation ocean model, consisting of momentum, continuity, sediment, temperature, salinity,
and density equations and is closed physically and mathematically using the Mellor and Ya‐
mada level-2.5 turbulent closure submodel; the irregular bottom slope is represented using a
σ-coordinate transformation, and the horizontal grids comprise unstructured triangular
cells; the finite-volume method used in the model combines the advantages of a finite-ele‐
ment method for geometric flexibility and a finite-difference method for simple discrete
computation; current, sediment, temperature, and salinity in the model are computed in the
integral form of the equations, which provides a better representation of the conservative
laws for mass, momentum, and heat in the coastal region with complex geometry [4].

4.1. The primitive equations

The governing equations consist of the following momentum, continuity, temperature, sal‐
inity, and density equations:

(1) continuity equation

¶u ¶v ¶w
+ + =0 (1)
¶x ¶y ¶z
334 Sediment Transport Processes and Their Modelling Applications

(2) x -direction momentum equation

¶u ¶u ¶u ¶u 1 ¶P ¶ ¶u
+u + v + w - fv = - + ( K m ) + Fu (2)
¶t ¶x ¶y ¶z r0 ¶x ¶z ¶z

(3) y -direction momentum equation

¶v ¶v ¶v ¶v 1 ¶P ¶ ¶v
+ u + v + w + fu = - + (K ) + Fv (3)
¶t ¶x ¶y ¶z r0 ¶y ¶z m ¶z

(4) temperature equation

¶T ¶T ¶T ¶T ¶ ¶T
+u +v +w = (Kh ) + FT (4)
¶t ¶x ¶y ¶z ¶z ¶z

(5) salinity equation

¶s ¶s ¶s ¶s ¶ ¶s
+ u + v + w = ( K h ) + Fs (5)
¶t ¶x ¶y ¶z ¶z ¶z

(6) pressure equation

¶P
= -r g (6)
¶t

r = r (T , s ) (7)

where x, y, and z are the east, north, and vertical axis of the Cartesian coordinate; u, v, and w
are the x, y, z velocity components; T is the potential temperature; s is the salinity; P is the
pressure; f is the Coriolis parameter; g is the gravitational acceleration; Km is the vertical ed‐
dy viscosity coefficient; and Kh is the thermal vertical eddy diffusion coefficient. Here Fu, Fy,
FT, and Fs represent the horizontal momentum, thermal, and salt diffusion terms.

4.2. Numerical solutions

The momentum and continuity equations are solved using a ‘model splitting’ method [4],
that is, the current is divided into external and internal modes that can be computed using
two distinct time steps. The external mode is used to solve the 2D vertically integrated mo‐
mentum and continuity equations while the internal mode is computed for the 3D equa‐
 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 335
[Link]

tions, the latter is solved numerically using a simple combined explicit and implicit scheme,
in which the local change of the current is integrated using the first-order accuracy upwind
scheme; the advection terms are computed explicitly by a second-order accuracy Runge–
Kutta time-stepping scheme [4].

4.3. Sediment computation

FVCOM adopts the Community Numerical Modeling System to simulate erosion, trans‐
port, deposition and the fate of sediments in the coastal ocean developed by experts from
USGS [5]. The sediment-transport algorithms are implemented for an unlimited number
of user-defined noncohesive/cohesive sediment classes. Each class has attributes of grain
diameter, density, settling velocity, critical stress threshold for erosion, and erodibility
constant. These properties are used to determine bulk properties of each bed layer. Sus‐
pended-sediment transport in the water column is computed with the same advection-dif‐
fusion algorithm used for all passive tracers and an additional algorithm for vertical
settling that is not limited by the CFL criterion [5].

4.3.1. Suspended sediment

Suspended sediment transport equation is:

¶Ci ¶uCi ¶vCi ¶ (w - wi )Ci ¶ ¶C ¶ ¶C ¶ ¶C

+ + + = ( As ,h i ) + ( As ,h i ) + ( K s ,h i ) (8)
¶t ¶x ¶y ¶z ¶x ¶x ¶y ¶y ¶z ¶z

here Ci is the concentration of the i th sediment class, As,h is the horizontal sediment diffusivi‐
ty, K s,h is the vertical sediment diffusivity, ωi is the settling velocity of the i th
sediment class
given by the user.

At the top boundary, the vertical diffusive flux is set to be zero:

¶Ci
K s ,h = 0, z = z (9)
¶t

here ζ is surface elevation. At the bottom boundary, the vertical sediment flux is specified by:

¶Ci
K s ,h = Ei - Di , z = - D (10)
¶t

here D is water depth. While the erosion flux of the i th sediment class is computed as:

tb
Ei = DtE0i (1 - Pb ) Fbi ( - 1) (11)
t ci
336 Sediment Transport Processes and Their Modelling Applications

where E0i is an empirical constant set by the user as the erosion rate of the i th sediment class,
Pbis the porosity of the bed,F bi is the fraction of the i th
sediment class, τbis the bottom flow
shear stress, τci is the critical stress for the incipient motion of the i th sediment class.

When the bottom shear stress passes the critical stress erosion occurs on the bed. The sedi‐
ment concentration profile in the water body is determined by horizontal convection, diffu‐
sion, vertical diffusion, settling and bottom erosion flux [5].

4.3.2. Bedload transport computation

Bedload transport rate is computed using established empirical formula, i.e. the Meyer-Peter
and Muller formula or using formulas that the modeler considers appropriate, for example,
the theoretical-based formula [6].

4.3.3. Sediment bed

The sediment bed consists of a constant number layers, and each layer is initialized with a
thickness, sediment-class distribution, porosity, and age, the mass of each sediment class can
be determined from these values and the grain density; the bed evolving properties include
bulk properties of the surface layer (active-layer thickness, mean grain diameter, mean den‐
sity, mean settling velocity, mean critical stress for erosion) [5].

Figure 3. Distribution of vertical layers in bed model (from [5]).

 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 337
[Link]

The bed layers are modified at each time step to account for erosion and deposition (Figure
3) and track stratigraphy; at the beginning of each time step, an active-layer thickness z a is
calculated based on the relation of Harris and Wiberg [7]:

za = max éë k1 (t sf - t ce ) r0 , 0 ùû + k2 D50 (12)

where τsf is bottom skin-friction stress; τ¯

ce is the critical stress for erosion; and the overbar
indicates this is averaged over all sediment classes; D50 is the median grain diameter of sur‐
face sediment; and k 1 and k 2 are empirical constants, 0.007 and 6.0, respectively.

Each sediment class can be transported by suspended-load and/or bedload; suspended-load

mass is exchanged vertically between the water column and the top bed layer; mass of each
sediment class available for transport is limited to the mass available in the active layer; bed‐
load mass is exchanged horizontally between the top layers of the bed; mass of each sedi‐
ment class available for transport is limited to the mass available in the top layer [5].

If continuous deposition results in a top layer thicker than a user-defined threshold, a new lay‐
er is provided to begin accumulation of depositing mass; the bottom two layers are then com‐
bined to conserve the number of layers; after erosion and deposition have been calculated, the
active-layer thickness is recalculated and bed layers are readjusted to accommodate it [5].

5. Model setup

5.1. Model domain

Figure 4 shows the computation domain consisting of unstructured triangular grids.

Figure 4. Unstructured grids of Qinzhou Bay.

It has 16466 nodes and 29722 elements in each horizontal layer and 7 sigma-levels in the ver‐
tical. The horizontal grid resolution varies from 2,000 m at the open boundary to 350m in the
338 Sediment Transport Processes and Their Modelling Applications

channel-shoal region to 150 m in the Yingling inlet, especially, down-to-30-m elements are
interpolated around islands and in the estuarine channels.

5.2. Boundary conditions

The open boundary uses observed water level as input condition, for this purpose half-
month water levels at each open boundary grid are interpolated from two tidal station, i.e.
the Beihai Station and the Bailongwei Station (Figure 4). River boundaries use annually-
mean discharges as input conditions.

Suspended-sediment concentrations at the open boundary grids are interpolated from ad‐
jacent observation sites, while those at river input grids are annually-mean suspended-
sediment discharges.

The bathymetry in the computation domain consists of a local bathymetric survey in the
Qinzhou Bay in 2008 and sea maps surveys in 2004 and 1997 to supplement other parts.

The time step for both of external mode and internal mode is 1 s.

6. Calibrations

TCZC [1] measured half month water level at three temporary tidal gauges, namely, Guo‐
zishan, Shabadun and Wulei, also measured 26-hour spring, middle and neap tidal flow ve‐
locity & direction and suspended-sediment concentrations at 10 sites (Figure 5).

Figure 5. Tidal gauges (red square) and flow observations sites (green triangular) in Qinzhou Bay.
 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 339
[Link]

The present study firstly performs the calibrations of water level, flow velocity& direction
and suspended-sediment concentrations.

6.1. Tidal water level

The Calibrations for water levels at three tidal gauges are shown in Figure 6.

Figure 6. Tidal water level calibrations (the ordinate unit of is meter).

6.2. Flow calibrations

The flow calibrations are shown in Figure 7. For limited space only calibrations for spring
tide are show here.
340 Sediment Transport Processes and Their Modelling Applications

Figure 7. Flow calibrations.

6.3. Sediment calibration

Sediment calibrations for selected sites are shown on Figure 8. It needs to explain that the
overall sediment calibrations are satisfactory, but results for some sites are not satisfactory
enough due to observation and computation errors, ship activities and dredging during
the observation period, etc.
 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 341
[Link]

Figure 8. Suspended-sediment calibrations at selected sites for spring tide (the ordinate unit is kg/m3).
342 Sediment Transport Processes and Their Modelling Applications

6.4. Morphological calibration

Based upon a local bathymetric survey conducted in 2008 summer and a sea map sur‐
veyed in 2004, statistics shows that the total siltation amount at this area is about 1.52
million m3with a spatial average value of 0.592m (the deposition volume is divided by the
deposition area, the same hereinafter), the total erosion amount is nearly 3.35 million m3
with a spatial average value of 1.104m (the erosion volume is divided by the erosion area,
the same hereinafter), the net eroded sediment amounts to 1.83 million m3 (Figure 9).

Figure 9. Erosion & deposition distribution map in a part of Qinzhou Bay.

The present study computes the morphological evolution using the 2004 bathymetry as the
initial bathymetry, the computed four-year accumulative erosion & siltation distribution is
shown in Figure 10. The computed total siltation amount is 1.244 million m3 with a spatial
average value of 0.461m, amounting 81.89% and 77.87% of surveyed quantities, respectively;
the computed total erosion amount is 2.97 million m3 with a spatial average value of 1.073m,
amounting 88.53% and 97.19% of surveyed quantities, respectively; the computed net erod‐
ed sediment amounts to 1.72 million m3, amounting 94.03% of surveyed quantities.

In view of the discrepancies of computed results vs. surveyed quantities the present mor‐
phological calibration is quite satisfactory. This lays down very good basis for further mor‐
phodynamic study.
 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 343
[Link]

Figure 10. The computed 2004~2008 erosion & deposition distribution map.

7. Computation results

7.1. Tidal flow field

As for neap tide and middle tide, tidal water floods into the Qinzhou Bay in the northeast‐
ern direction while it floods into the Bay nearly in the northern direction during spring tide.
As constrained by the shoreline and channel-shoal geomorphology tidal water propagates
in the northwestern direction into the Yingling inlet and further into the inner bay, where it
flows anticlockwise till stack water; then tidal water rushes into Yingling inlet and diverges
among east, middle and west waterways; finally it leaves the Qinzhou Bay in the southwest‐
ern direction to the South China Sea (Figure 11).

Generally speaking, the tidal flow field in the Qinzhou Bay is characteristics of reciprocating
flow in parallel to the major waterways, large scale eddies occur during flow reversal periods.

The computation results show that the ebb-mean velocities of spring tide and middle tide
are all larger than the flood-mean velocities; while flood-mean velocity becomes larger than
ebb-mean velocity during neap tide in the overall flow field of Qinzhou Bay.
344 Sediment Transport Processes and Their Modelling Applications

Figure 11. a) Flood peak flow of spring tide in Qinzhou Bay, b) Ebb peak flow of spring tide in Qinzhou Bay.
 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 345
[Link]

Figure 12. a) Upper-layer flood peak flow of spring tide in Qinzhou Bay, b) Middle-layer flood peak flow of spring tide
in Qinzhou Bay, c) Lower-layer flood peak flow of spring tide in Qinzhou Bay.
346 Sediment Transport Processes and Their Modelling Applications

Among three major tidal channels in the outer bay, the flood-mean and ebb-mean velocities
of spring, middle, neap tides in the middle channel are all larger than those in the east and
west channels; though flood-mean velocity in spring tide in the west channel is somewhat
smaller than that in the east channel the flood-mean velocities in middle & neap tides in the
west channel are all larger than those in the east channel; as for ebb-mean velocity, they are
all larger in the west channel than those in the east channel. These data demonstrates that
the west channel is the dominant channel for tidal water flowing into and out the Qinzhou
Bay, the middle channel comes second and the east channel is third; ebb tide dominates in
the west channel and middle channel but flood tide dominates in the east channel.

Generally speaking, flow velocity at the upper water layer is the largest and decreases from top
to bottom (Figure 12). The depth-averaged residual flow is shown in Figure 13. Various-sized
residual eddies occur in the Qinzhou Bay. Mean residual flow velocity in the Yingling inlet is
around 0.15m/s with largest velocity of 0.489m/s, it is generally below 0.05m/s in other parts.

Figure 13. Depth-averaged residual flow in Qinzhou Bay.

7.2. Sediment transport

The suspended-sediment sources include those delivered by Maolingjiang river and Qinjing
river and limited amount transported from the open sea, but the majority is eroded and re‐
 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 347
[Link]

suspended in situ in the Bay. As a result, the spatial & temporal variations of suspended-
sediment concentrations are in accordance with the processes of tidal flows.

Generally speaking, suspended-sediment concentrations are larger in major channels than

those on shoals and intertidal zones (Figure 14), decrease from bottom to top. The majority
of sediment delivered by Maolingjiang and Qinjiang rivers is deposited within the inner bay
with limited amount of finer particles transported into the outer bay and deep water.

Figure 14. a) Suspended-sediment concentration field at flood peak of spring tide, b) Suspended-sediment concentra‐
tion field at ebb peak of middle tide.

The computation results show that suspended-sediment concentration at spring tidal flood
peak is 0.037 kg/m3, 0.021 kg/m3 at spring tidal flood stack, 0.034 kg/m3 at spring tidal ebb
peak and 0.023 kg/m3 at spring tidal ebb stack, respectively, the mean spring-tidal concen‐
tration is 0.029 kg/m3; sediment concentration is 0.031 kg/m3 at middle tidal flood peak,
0.023 kg/m3 at middle tidal flood stack, 0.031 kg/m 3 at middle tidal ebb peak and 0.018 kg/m3
at middle tidal ebb stack, respectively, the mean middle-tidal concentration is 0.026 kg/m3;
the sediment concentration is 0.013 kg/m3 at neap tidal flood peak, 0.013 kg/m3 at neap tidal
flood stack, 0.013kg/m3 at neap tidal ebb peak and 0.012 kg/m3 at neap tidal ebb stack, re‐
spectively, The mean neap-tidal concentration is 0.013 kg/m3. Generally speaking, suspend‐
ed-sediment concentration is relatively low in the Qinzhou Bay.

The west channel has been the major channel to transport sediment from outer bay thought
Yingling inlet to inner bay and vice versa; the middle channel comes second and the east
348 Sediment Transport Processes and Their Modelling Applications

channel contributes the least. The sediment discharges at ebb tide are all larger than those at
flood tide in the three channels, demonstrating net sediment transport into the open deep
water. The sediment discharges at spring tide in three channels are all larger than those at
middle & neap tide, and the dividing ratio of west channel, middle channel and east channel
is nearly [Link] for spring flood tide and [Link] for spring ebb tide, [Link] for middle flood tide
and [Link] for middle ebb tide, respectively.
The sediment transport pattern is normally accordance with tidal flow asymmetry in
three channels, that is, ebb flow strength and discharge are superior to flood flow
strength and discharge.

7.3. Morphological evolution

7.3.1. 2009-year erosion and deposition

Due to lack of data on the deposit thickness distribution in Qinzhou Bay and considering
that rock is exposed locally within the deep channel in Yingling inlet by strong tidal flows
[8], the present study assumes the initial deposit thickness is 0.3m within the deep channel
in Yingling inlet and 20m in other parts of the Qinzhou Bay. The morphological computa‐
tion starts from year 2008.
The computed 2009 annual erosion & deposition distribution map is shown in Figure 15.
It can be observed that erosions mainly occur within channels including three major chan‐
nels in the outer bay, deep-water channel in the Yingling inlet and those in the inner bay
while depositions occur at shoal & ridge area and at the end of channels. This asserts that
tide flow is really the dominant force for maintaining and reforming the channel-shoal
morphology in the Qinzhou Bay.
Generally speaking, eroded sediments exceed deposited sediment for the whole Qinzhou
Bay with net erosion nearly up to 10.288 million m3. Except for the inner bay where net dep‐
osition occurs with a quantity of 3.190 million m3 net erosions all occur in the Yingling inlet
and the outer bay, they are 2.999 million m3 and 3.503 million m3, respectively. Due to lack
sediment supply, the offshore slope outside the Qinzhou Bay is also subjected to net erosion
of 6.976 million m3, where erosion mainly occurs at the middle and southeastern part while
deposition occurs at the southwestern part.
The total deposition in the west channel (bounded by -5m bathymetric contour, the same for
middle channel and east channel) in the outer bay is roughly 490,887.486 m3, the total ero‐
sion is roughly 2,257,125.612 m3, and the net erosion is about 1,766,238.1 m3; the spatial mean
deposition is 0.097m, the spatial mean erosion is -0.232m, the maximum deposition is
0.355m and the maximum erosion is -1.244m.
The total deposition in the middle channel in the outer bay is roughly 9,830.569 m3, the total
erosion is roughly 551,595.451 m3, and the net erosion is about 453,285.88 m3; the spatial
mean deposition is 0.073m, the spatial mean erosion is -0.394m, the maximum deposition is
0.157m and the maximum erosion is -1.785m. The erosion mainly occurs at the channel
mouth connecting with the Yingling inlet.
 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 349
[Link]

Figure 15. 2009-year annual erosion & deposition distribution map.

The total deposition in the east channel in the outer bay is roughly 499,259.028 m3, the total
erosion is roughly 1,488,779.676 m3, and the net erosion is about 989,520.6 m3; the spatial
mean deposition is 0.110m, the spatial mean erosion is -0.213m, the maximum deposition is
0.623m and the maximum erosion is -1.285m.

7.3.2. 2012-year erosion and deposition

The computed 2012-year annual erosion & deposition distribution map is shown in Figure
16. Eroded sediments still exceed deposited sediment for the whole Qinzhou Bay with net
erosion nearly up to 10.469 million m3. The inner bay continues to accommodate net deposi‐
tion of 2.832 million m3, net erosions still occur in the Yingling inlet and the outer bay, they
are 0.809 million m3 and 4.161 million m3, respectively; the offshore slope outside the Qinz‐
hou Bay is still subjected to net erosion of 8.331 million m3.

The total deposition in the west channel in the outer bay is roughly 240,510.501 m3, the total
erosion is roughly 2,031,819.599 m3, and the net erosion is about 1,791,309.0 m3; the spatial
350 Sediment Transport Processes and Their Modelling Applications

mean deposition is 0.063m, the spatial mean erosion is -0.185m, the maximum deposition is
0.232m and the maximum erosion is -0.889m.

The total deposition in the middle channel in the outer bay is roughly 32,702.536 m3, the to‐
tal erosion is roughly 435,482.067 m3, and the net erosion is about 402,779.53 m3; the spatial
mean deposition is 0.044m, the spatial mean erosion is -0.217m, the maximum deposition is
0.095m and the maximum erosion is -0.772m.

Figure 16. 2012-year annual erosion & deposition distribution map.

The total deposition in the east channel in the outer bay is roughly 356,241.632 m3, the total
erosion is roughly 1,160,837.006 m3, and the net erosion is about 804,595.37 m3; the spatial
mean deposition is 0.084m, the spatial mean erosion is -0.159m, the maximum deposition is
0.527m and the maximum erosion is -0.705m.

These data in the three channels reflect that the erosion and deposition in the three major
channels in the outer bay has steadily decreased. In particular, the erosion length in the west
 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 351
[Link]

channel has increased substantially, leading almost to whole-channel erosion, and tidal
channels have further developed in the inner bay.

7.3.3. 2020-year erosion and deposition

The computed 2020-year annual erosion & deposition distribution map is shown in Figure
17. Eroded sediments still exceed deposited sediment for the whole Qinzhou Bay with net
erosion nearly up to 11.136 million m3. The inner bay continues to accommodate net deposi‐
tion of 2.601 million m3, net erosions still occur in the Yingling inlet and the outer bay, they
are 0.677 million m3 and 3.095 million m3, respectively; the offshore slope outside the Qinz‐
hou Bay is still subjected to net erosion of 9.965 million m3.

The overall morphological evolution trend is that total erosion & deposition amount have
dropped steadily in all parts of Qinzhou Bay though net deposition might moderately in‐
crease or decrease in different parts of the Bay.

Figure 17. 2020-year annual erosion & deposition distribution map.

352 Sediment Transport Processes and Their Modelling Applications

The total deposition in the west channel in the outer bay is roughly 207,835.846 m3, the
total erosion is roughly 1,685,565.419 m3, and the net erosion is about 1,477,729.5 m3; the
spatial mean deposition is 0.074m, the spatial mean erosion is -0.141m, the maximum
deposition is 0.311m and the maximum erosion is -0.475m.

The total deposition in the middle channel in the outer bay is roughly 17,775.557 m3, the
total erosion is roughly 296,870.467 m3, and the net erosion is about 279,094.91 m3; the
spatial mean deposition is 0.025m, the spatial mean erosion is -0.145m, the maximum
deposition is 0.049m and the maximum erosion is -0.405m.

The total deposition in the east channel in the outer bay is roughly 251,859.055 m3, the
total erosion is roughly 929,903.211 m3, and the net erosion is about 678,044.15 m3; the
spatial mean deposition is 0.069m, the spatial mean erosion is -0.118m, the maximum
deposition is 0.431m and the maximum erosion is -0.412m.

7.3.4. 2040-year erosion and deposition

The computed 2040-year annual erosion & deposition distribution map is shown in Figure 18.

Figure 18. 2040-year annual erosion & deposition distribution map.

 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 353
[Link]

Eroded sediments still exceed deposited sediment for the whole Qinzhou Bay with net ero‐
sion nearly up to 13.915 million m3. The inner bay continues to accommodate net deposition
of 1.063 million m3, but net deposition has occurred in the Yingling inlet with a value of
0.125 million m3, net erosion in the outer bay is 1.592 million m3, the offshore slope outside
the Qinzhou Bay has been subjected to increased net erosion of 13.512 million m3.

The total deposition in the west channel in the outer bay is roughly 387,349.439 m3, the total
erosion is roughly 1,114,549.259 m3, and the net erosion is about 727,199.81 m3; the spatial
mean deposition is 0.098m, the spatial mean erosion is -0.103m, the maximum deposition is
0.335m and the maximum erosion is -0.429m.

The total deposition in the middle channel in the outer bay is roughly 37,682.474 m3, the to‐
tal erosion is roughly 202,893.855m3, and the net erosion is about 165,211.38 m3; the spatial
mean deposition is 0.054m, the spatial mean erosion is -0.099m, the maximum deposition is
0.310m and the maximum erosion is -0.379m.

The total deposition in the east channel in the outer bay is roughly 170,181.197 m3, the total
erosion is roughly 420,256.804 m3, and the net erosion is about 250,075.6 m3; the spatial mean
deposition is 0.051m, the spatial mean erosion is -0.051m, the maximum deposition is
0.271m and the maximum erosion is -0.259m.

The overall morphological evolution trend is that total erosion & deposition amount have
continuously decreased, sub-channels haves occurred at the shoal between west channel
and middle channel and new channel-shoal morphology has been developed within the
inner bay (Figure 19).

Figure 19. Bathymetry maps of 2008 vs. 2040.

354 Sediment Transport Processes and Their Modelling Applications

8. Long-term morphological evolution

The Qinzhou Harbor has used the east channel as its major navigation channel to deliver
cargos. Two large-scale dredgings were performed within this channel, i.e. 2009/9-2002/12
with a total dredged sediment of 8.234 million m3, 2004/2-2008/12 with a total dredged sedi‐
ment of 45.307 million m3 [9]. These two dredging activities have exerted profound influen‐
ces upon the morphological evolution of the channel-shoal system in the Qinzhou bay. The
present computation has just reflected this evolution process and trend. During the adjust‐
ing process, the deeper channels have experienced further erosion while the shallower
shoals (ridges) have accreted further higher, and the overall stability of the channel-shoal
system has been maintained without horizontal migration or sign of merging or perishing;
the inner bay has not only accepted sediments delivered by Maolingjiang river and Qinjiang
river, but also sediments transported by flood tidal flows from the outer bay; the remaining
part of the net eroded sediments from the outer bay has been transported into the offshore
slope and deeper water by ebb tidal flows.

The magnitude of morphological adjustment by the above-mentioned channel dredging has

been initially large but decreasing steadily with time. It could be estimated the morphologi‐
cal adjustment process at the outer bay would finished one hundred years later while other
parts of the Qinzhou Bay might experience even longer adjustments. It should be clarified
that such an estimation has assumed that no new engineering projects to be constructed and
the computation conditions such as spatial bed thickness, horizontal & vertical sediment
grading, shoreline configuration, tidal force and river discharges are unchanged.

9. Concluding remarks

Qinzhou Bay is characteristics of a unique dumbbell in shape, consisting of an inner en‐

closed bay, a trumpet-shaped outer bay and irregular rocky tidal inlet connecting them.
Within the outer bay a complex channel-shoal (ridge) system has been present with the ma‐
jor channels serving as the navigation course for the Qinzhou Harbor.

FVCOM is a 3D unstructured grid, finite-volume coastal ocean model for the study of coast‐
al oceanic and estuarine circulation, sediment transport and morphodynamics.. Having per‐
formed good calibrations of observed tidal water level, flow velocity and direction,
suspended-sediment concentration at hydrographic sites [1] and morphological variation in
the period of 2004 through 2008, the present study further investigates the diurnal tidal cir‐
culation including tidal asymmetry, residual eddy, and accompanying sediment transport
processes in order to ascertain the water & sediment exchanges between the inner basin and
the outer bay, especially, among the branching channels.

It is found that the inner basin has been acting as a sediment storage basin to accept sedi‐
ments delivered by river flows and those by asymmetric tidal flows from the outer bay;
Coriolis force together with rocks at the inlet mouth has controlled the dividing ratios of wa‐
 Numerical Modeling Tidal Circulation and Morphodynamics in a Dumbbell-Shaped Coastal Embayment 355
[Link]

ter and sediment among the branching channels, the west channel has been the dominant
course for tidal flow and sediment to pass the outer bay, the middle channel comes secondly
important and the east channel contribute the least.

Two large-scale dredging activities conducted in 2000/9-2002/12 and 2004/2-2008/12 in the

east channel for deep navigation course development have exerted profound influence upon
the morphodynamic evolution of the channel-shoal system. The erosion & deposition pat‐
tern, i.e. erosion in channels and deposition in shoals (ridges) has clearly demonstrated that
tidal flow is the predominant force for maintaining and reforming the channel-shoal mor‐
phology; the dredging in the east channel has caused lasting erosions in the major channels
in the outer bay, Yingling inlet and the inner bay as well as the offshore slope, meanwhile,
depositions accumulate on shoals (ridges) and at the end parts of the channels.

Generally speaking, the overall channel-shoal system has been stable with channels becoming
deeper and shoals becoming higher, and such a morphological adjustment process will proba‐
bly finished over one hundred years later, if no new coastal engineering activity intervenes.

Acknowledgements

This study is supported by a grant from National Natural Science Foundation (No.
51179211) and a Young Researcher Fund of IWHR (No. NJ1009).

Author details

Yu-Hai Wang*

Address all correspondence to: wangyuhai-2166@[Link]

Department of Sediment Research, China Institute of Water Resources and Hydropower Re‐
search, Beijing, China

References

[1] Tianjing Coastal Zone Engineering Company (TCZC). (2009). Fengchenggang Nucle‐
ar Power Plant Summer Full Tide Hydrological Survey and Analysis. Technical Re‐
port, in Chinese.

[2] Yan, X.-X., & Liu, G.-T. (2006). Study on deposition characteristics and channel silta‐
tion in offshore zone of Qinzhou Bay. Journal of Waterway and Harbor, 27(2), 79-83, in
Chinese.
356 Sediment Transport Processes and Their Modelling Applications

[3] Li, G.-Z., Liang, W., & Liu, J.-H. (2002). Discussion on the source and transport ten‐
dency of silt in the Qinzhou Bay in terms of the dynamic partition zones of heavy
minerals in the sediments. Marine Science Bulletin, 21(5), 61-68, in Chinese.

[4] Chen, C., Liu, H., & Beardsley, R. (2003). An unstructured grid, finite-volume, three-
dimensional, primitive equations ocean model: Application to coastal ocean and es‐
tuaries. Journal of Atmospheric and Oceanic Technology, 20(1), 159-186.

[5] Warner, J. C., Sherwood, C. R., Signell, R. P., Harris, C., & Arango, H. G. (2008). De‐
velopment of a three-dimensional, regional, coupled wave, current, and sediment-
transport model. Computers and Geosciences, 34, 1284-1306.

[6] Wang, Y. H. (2007). Formula for predicting bedload transport rate in oscillatory sheet
flows. Coastal Engineering, 54(8), 594-601.

[7] Harris, C. K., & Wiberg, P. L. (1997). Approaches to quantifying long-term continen‐
tal shelf sediment transport with an example from the northern California STRESS
mid-shelf site. Continental Shelf Research, 17, 1389-1418.

[8] Li, G.-Z., Liang, W., & Liu, J.-H. (2001). Features of underway dynamic geomorphol‐
ogy of the Qinzhou Bay. Geography and Territorial Research, 17(4), 70-75, in Chinese.

[9] Wang, Y. H., Wang, C. H., Liu, D. B., & Lin, Y.-X. (2010). Preliminary study on chan‐
nel stability in Qinzhou Bay. Port & Waterway Engineering, 8, 76-80, in Chinese.
 Chapter 14

Numerical Modeling of Flow and Sediment Transport

in Lake Pontchartrain due to Flood Release from
Bonnet Carré Spillway

Xiaobo Chao, Yafei Jia and A. K. M. Azad Hossain

Additional information is available at the end of the chapter

[Link]

1. Introduction

Lake Pontchartrain is a brackish estuary located in southeastern Louisiana, United States. It

is the second-largest saltwater lake in U.S. The lake covers an area of 1630 square km with a
mean depth of 4.0 meters. It is an oval-shaped quasi-enclosed water body with the main
east-west axis spanning 66 km, while the shorter north–south axis is about 40 km. It is con‐
nected to the Gulf of Mexico via Rigolets strait, to Lake Borgne via Chef Menteur Pass, and
to Lake Maurepas via Pass Manchac. These lakes form one of the largest estuaries in the
Gulf Coast region. It receives fresh water from a few rivers located on the north and north‐
west of the lake. The estuary drains the Pontchartrain Basin, an area of over 12,000 km2 situ‐
ated on the eastern side of the Mississippi River delta plain.
Lake Pontchartrain has served the surrounding communities for more than two centuries.
The coastal zone of the Lake and its basin has offered opportunities for fishing, swimming,
boating, crabbing and other recreational activities. The Lake Basin is Louisiana's premier ur‐
ban estuary and nearly one-third of the state population live within this area. Over the past
decades, rapid growth and development within the basin have resulted in significant envi‐
ronmental degradation and loss of critical habitat in and around the Lake. Human activities
associated with pollutant discharge and surface drainage have greatly affected the lake wa‐
ter quality (Penland et al., 2002).
In order to protect the city of New Orleans from the Mississippi River flooding, the Bonnet
Carré Spillway (BCS) was constructed from 1929 to 1936 to divert flood water from the river
into Lake Pontchartrain and then into the Gulf of Mexico. However, a BCS opening event
may cause many environmental problems in the lake. To evaluate the environmental im‐
358 Sediment Transport Processes and Their Modelling Applications

pacts of the flood water on lake ecosystems, it is important to understand the hydrodynam‐
ics as well as sediment transport in the lake.

Lake Pontchartrain is a large shallow lake and the water column is well mixed. In general,
the water movements within the lake are affected by wind and tide. During the BCS open‐
ing for flood release, the flow discharge over the spillway produces significant effects on the
lake hydrodynamics.

Numerical models generally are cost-effective tools for predicting the flow circulation and
pollutant transport in a lake environment. In recent years, numerical models have been ap‐
plied to simulate the flow and pollutant distribution in Lake Pontchartrain. Hamilton et al.
(1982) developed a 2D-vertical integrated model to simulate the flow circulation in Lake
Pontchartrain. The model is an explicit numerical difference scheme based on leap-frog inte‐
gration algorithm, and a coarse uniform mesh with a spacing of 1 km was selected for nu‐
merical simulation. Signell (1997) applied the coastal and ocean model (ECOM), developed
by Hydroqual to simulate the tide and wind driven circulation processes in the lake. It was
found that water levels in Mississippi Sound influence the circulation patterns in the eastern
part of the lake, while the wind force dominates the flow pattern of the western part.
McCorquodale et al. (2005) applied a 3D coastal ocean model, ECOMSED, to simulate the
flow fields and mass transport in Lake Pontchartrain. Dortch et al. (2008) applied the CH3D-
WES hydrodynamic model, developed by the US Army Corps of Engineers, to simulate the
lake flow field and post-Hurricane Katrina water quality due to the large amount of conta‐
minated floodwater being pumped into the Lake. McCorquodale et al. (2009) developed a
1D model for simulating the long term tidal flow, salinity and nutrient distributions in the
Lake. Most of those researches focus on hydrodynamics and pollutant transport in the Lake.

In this study, the flow fields and sediment transport in Lake Pontchartrain during a flood
release from BCS was simulated using the computational model CCHE2D developed at the
National Center for Computational Hydroscience and Engineering (NCCHE), the University
of Mississippi (Jia and Wang 1999, Jia et al. 2002). This model can be used to simulate free
surface flows and sediment transport, and the capabilities were later extended to simulate
the water quality, pollutant transport and contaminated sediment (Chao et al. 2006, Zhu et
al. 2008). CCHE2D is an integrated numerical package for 2D-depth averaged simulation
and analysis of flows, non-uniform sediment transport, morphologic processes, water quali‐
ty and pollutant transport. There are several turbulence closure schemes available within
the model for different purposes, including the parabolic eddy viscosity, mixing length, k–ε
and nonlinear k–ε models. A friendly Graphic User Interface (GUI) is available to help users
to setup parameters, run the simulation and visualize the computational results. In addition
to general data format, CCHE2D has capabilities to produce the simulation results in ArcGIS
and Google Earth data formats (Hossain et al., 2011). Those capabilities greatly improve the
model’s applications.

The simulated flow and sediment distribution during the BCS opening were compared with
satellite imagery and field measured data provided by the United States Geological Survey
(USGS) and the United States Army Corp of Engineers (USACE). Good agreements were ob‐
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 359
[Link]

tained from the numerical model. This model provides a useful tool for lake water quality
management.
(USGS) and the United States Army Corp of Engineers (USACE). Good agreements were
obtained from the numerical model. This model provides a useful tool for lake water quality
2. Bonnet Carré Spillway (BCS) opening for flood release
management.

In response to theCarré
2. Bonnet highSpillway
flood stage
(BCS) ofOpening
the Mississippi
for FloodRiver and to protect the city of New
Release
Orleans, the Bonnet Carré Spillway (BCS) was built to divert
In response to the high flood stage of the Mississippi River and toMississippi River
protect the city flood waters
of New
Orleans, the Bonnet Carré Spillway (BCS) was built to divert Mississippi River flood waters
to the Gulf of Mexico via Lake Pontchartrain (Fig. 1).
to the Gulf of Mexico via Lake Pontchartrain (Fig. 1).

North

Lake Pontchartrain

New Orleans
Bonnet Carré Spillway

8 km
Mississippi River

Figure 1. The location of Bonnet Carré

Fig. Spillway
1. The (BCS)
location of Bonnet Carré Spillway (BCS)

The construction of the

The construction spillway
of the spillwaywas
was completed in 1931.
completed in 1931. It is located
It is located in [Link],
in St. Charles Charles Parish,
Louisiana - about 19 km west of New Orleans. The spillway consists of two basic
Louisiana - about 19 km west of New Orleans. The spillway consists of two basic compo‐
components: a 2.4 km long control structure along the east bank of the Mississippi River and
nents: aa2.4
9.7 km long control
km floodway structure
that transfers along the
the diverted floodeast bank
waters of lake.
to the the Mississippi River and a 9.7
The design capacity
km floodway that transfers
of the spillway is 7080 mthe diverted flood waters to the lake. The design capacity of the
3/s and will be opened when the Mississippi river levels in New

Orleans approached the flood stage of 5.2 m. It was first operated in 1937 and nine times
spillwaythereafter
is 7080 (1945,
m3/s 1950,
and will be opened when the Mississippi river levels in New Orleans
1973, 1975, 1979, 1983, 1997, 2008 and 2011). The maximum flow
approached the flood
discharges stage
and days of 5.2 m.
of opening for It was
each firstareoperated
event in 1937
listed in Table and nine
1 (USACE 2011;times
GEC thereafter
(1945, 1950,
1998).1973, 1975, 1979, 1983, 1997, 2008 and 2011). The maximum flow discharges and
days of During
opening for each event are listed in Table 1 (USACE 2011; GEC 1998).
the BCS opening, a large amount of fresh water and sediment discharged from the
During Mississippi River into Lake Pontchartrain and then into the Gulf of Mexico. The flow
the BCS opening, a large amount of fresh water and sediment discharged from the
discharge over the spillway produces significant effects on the lake hydrodynamics. It also
Mississippi River into Lake Pontchartrain and then into the Gulf of Mexico. The flow dis‐
charge over the spillway produces significant effects on the lake hydrodynamics. It also
changes the distributions of salinity, nutrients and suspended sediment (SS) in the lake dra‐
matically. During a flood releasing event, the fresh water dominated the whole lake and the
lake salinity reduced significantly. A lot of sediment deposited into the lake or was trans‐
ported into the Gulf of Mexico. The contaminated sediment from Mississippi River could
bring a lot of pollutants, such as nutrients, Al, Cu, Cr, Hg, Pb, Zn, etc., to the lake, and
360 Sediment Transport Processes and Their Modelling Applications

caused a lot of environmental problems. The algal bloom occurred in a large area of the lake
after a flood release event. The blooms produced high levels of heptatoxins and caused de‐
creases of dissolved oxygen in the lake (Dortch et al., 1998; Penland et al., 2002).

Max. discharge
Year Date opened Date Closed Days opened
m3/s

1937 Jan28 Mar 16 48 5975

1945 Mar 23 May 18 57 9005

1950 Feb 10 Mar 19 38 6315

1973 Apr 8 Jun 21 75 5522

1975 Apr 14 Apr 26 13 3115

1979 Apr 17 May 31 45 5409

1983 May 20 Jun 23 35 7589

1997 Mar 17 Apr 18 31 6881

2008 Apr 11 May 8 28 4531

2011 May 9 June 20 42 8892

Table 1. Information of Bonnet Carré Spillway opening for flood release

Due to a large amount of sediment discharged /deposited into the lake, the bed form of the
lake changed. The BCS opening event produced significant changes in flow pattern, salinity
and water temperature, which greatly affected the lake fish habitat, and caused negative im‐
pacts to oyster beds and fishery nursery grounds in the lake. In response to the dynamic
changes in the salinity, temperature, water surface elevation, and bed form of the lake, it
was observed that some species, particularly brown shrimp, shifted and moved. It may take
a long time for the fisheries resources to recover from the flood release event.

To understand the impact of the BCS flood release event on the ecosystem of Lake
Pontchartrain, the flow circulation and sediment transport are most important key tasks
to be studied.

3. Model descriptions

To simulate the flow field and sediment transport in Lake Pontchartrain, a two-dimensional
depth-averaged model, CCHE2D, was applied. CCHE2D is a 2D hydrodynamic and sedi‐
ment transport model that can be used to simulate unsteady turbulent flows with irregular
boundaries and free surfaces (Jia and Wang 1999, Jia et al. 2002). It is a finite element model
utilizing a special method based on the collocation approach called the “efficient element
method”. This model is based on the 2D Reynolds-averaged Navier-Stokes equations. By
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 361
[Link]

applying the Boussinesq approximation, the turbulent stress can be simulated by the turbu‐
lent viscosity and time-averaged velocity. There are several turbulence closure schemes
available within CCHE2D, including the parabolic eddy viscosity, mixing length, k–ε and
nonlinear k–ε models. In this model, an upwinding scheme is adopted to eliminate oscilla‐
tions due to advection, and a convective interpolation function is used for this purpose due
to its simplicity for the implicit time marching scheme which was adopted in this model to
solve the unsteady equations. The numerical scheme of this approach is the second order.
The velocity correction method is applied to solve the pressure and enforce mass conserva‐
tion. Provisional velocities are solved first without the pressure term, and the final solution
of the velocity is obtained by correcting the provisional velocities with the pressure solution.
The system of the algebraic equations is solved using the Strongly Implicit Procedure (SIP)
method (Stone 1968).

3.1. Governing equations

The free surface elevation of the flow is calculated by the continuity equation:

¶h ¶uh ¶vh
+ + =0 (1)
¶t ¶x ¶y

The momentum equations for the depth-integrated two-dimensional model in the Cartesian
coordinate system are:

¶u ¶u ¶u ¶h 1 æ ¶ht xx ¶ht xy ö t sx - t bx
+u +v = -g + ç + ÷+ + fCor v (2)
¶t ¶x ¶y ¶x h çè ¶x ¶y ÷
ø rh

¶v ¶v ¶v ¶h 1 æ ¶ht yx ¶ht yy ö t sy - t by
+u +v = -g + ç + ÷+ - fCor u (3)
¶t ¶x ¶y ¶x h çè ¶x ¶y ÷
ø rh

where u and v are the depth-integrated velocity components in x and y directions, respec‐
tively; t is the time; g is the gravitational acceleration; η is the water surface elevation; ρ is
the density of water; h is the local water depth; fCor is the Coriolis parameter; τxx, τxy, τyx and
τyy are depth integrated Reynolds stresses; and τsx and τsy are surface share stresses in x and
y directions, respectively; and τbx and τby are shear stresses on the interface of flow and bed
in x and y directions, respectively.

The turbulence Reynolds stresses in equations (2) and (3) are approximated according to the
Bousinesq’s assumption that they are related to the main rate of the strains of the depth-
averaged flow field and an eddy viscosity coefficient νt which is computed using the Sma‐
gorinsky scheme (Smagorinsky 1993):
362 Sediment Transport Processes and Their Modelling Applications

1/2
éæ ¶u ö 2 1 æ ¶v ¶u ö 2 æ ¶v ö 2 ù
n t = aDxDy êç ÷ + ç + ÷ + ç ÷ ú (4)
êè ¶x ø 2 è ¶x ¶y ø è ¶y ø ú
ë û

The parameter α ranges from 0.01 to 0.5. In this study, it was taken as 0.1.

In CCHE2D model, three approaches are adopted to simulate non-uniform sediment trans‐
port. One is the bed load transport, which is to simulate the bed load only without consider‐
ing the diffusion of suspended load. The second approach is the suspended load transport,
which simulates suspended load and treats bed -material load as suspended load. The third
approach is to simulate bed load and suspended load separately (Jia and Wang 1999, Jia et
al. 2002, Wu 2008).

In this study, CCHE2D was used to simulate sediment transport in Lake Pontchartrain dur‐
ing the BCS opening for flood release. In this period, sediment transport in the lake is pri‐
marily dominated by suspended sediment. So the second sediment transport approach,
suspended load, was used for this study, and the non-uniform suspended sediment (SS)
transport equation can be written as:

¶c k ¶c ¶c ¶ æ ¶c ö ¶ æ ¶c ö
+ u k + v k = ç Dcx k ÷ + ç Dcy k ÷ + Sck (5)
¶t ¶x ¶y ¶x è ¶x ø ¶y è ¶y ø

Where ck is the depth-averaged concentration of the kth size class of SS; Dcx and Dcy are the
mixing coefficients of SS in x and y directions, respectively; Sck is the source term and can be
calculated by:

atwsk
Sck = - ( c k - ct * k ) (6)
h

Where ct*k is the equilibrium sediment concentration of the kth size class of suspended load;
ωsk is the settling velocity of the kth size class; αt is the adaptation coefficient of suspended
load, and it can be estimated using the formula proposed by Wu (2008).

Settling velocity is calculated using Zhang's formula (Zhang and Xie 1993):

æ n ö ægs ö g
wsk = çç 13.95 ÷÷ + 1.09 ç - 1 ÷ gdk - 13.95 (7)
è d k ø è g ø d k

where ν is the kinematic viscosity; dk is the diameter of the kth size class of sediment; γs and
γ are the densities of water and sediment; g is the gravity acceleration.
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 363
[Link]

The equilibrium sediment concentration ct*k can be calculated based on sediment transport
capacities of fractional suspended load and bed load. Based on field and laboratory data,
Wu et al (2000) proposed a formula to calculate the fractional suspended load transport ca‐
pacity φsk :

1.74
éæ t ö U ù
j sk = 0.0000262 êçç - 1 ÷÷ ú (8)
ëêè t ck ø wsk ûú

where U is the depth-averaged velocity; τ is the shear stress; τck is the critical shear stress
and can be calculated by

0.6
æp ö
t ck = 0.03(g s - g )dk çç hk ÷÷ (9)
è pek ø

in which phk and pek are the hiding and exposure probabilities for the k-th size class of sedi‐
ment, they can be defined as:

N dj
phk = å pbj (10)
j =1 dk + d j

N dk
pek = å pbj (11)
j =1 dk + d j

where N is the total number of particle size classes in the non-uniform sediment mixture; pbj
is the probability of particles dj staying in front of particles dk. A relationship between phk and
pek is known as:

phk + pek = 1 (12)

In Eq. (8) φsk can also be expressed by:

qs * k
j sk =
æg ö (13)
pbk ç s - 1 ÷ gdk3
èg ø
364 Sediment Transport Processes and Their Modelling Applications

in which pbk is the bed material gradation; and qs*k is the equilibrium transport rate of the k-
th size class of suspended load per unit width. Based on Eqs. (8) and (13), the following
equation can be obtained to calculate qs*k :

1.74
æg ö éæ t ö U ù æg ö
qs* k = j sk pbk ç s - 1 ÷ gdk3 = 0.0000262 êçç - 1 ÷÷ ú pbk ç s - 1 ÷ gdk3 (14)
è g ø t
ëêè ck ø wsk úû è g ø

Wu et al. (2000) also proposed a formula to calculate the fractional bed load transport ca‐
pacity φbk:

2.2
éæ ' ö1.5 ù
n t
j bk = 0.0053 êç ÷ - 1ú (15)
êçè n ÷ø t ck ú
ë û

where n is the Manning's roughness coefficient; n' is the Manning's coefficient correspond‐
ing to the grain roughness, n ' = d50
1/6
/ 20; the transport capacity φbk can also be expressed as:

qb* k
j bk =
æg ö (16)
pbk ç s - 1 ÷ gdk3
è g ø

in which qb*k is the equilibrium transport rate of the k-th size class of bed load per unit
width. Based on Eqs. (15) and (16), the following equation can be obtained to calculate qb*k:

2.2
éæ ' ö1.5 ù
ægs ö 3 êç n ÷ t - 1ú æg ö
qb* k = j bk pbk ç - 1 gd
÷ k = 0.0053 pbk ç s - 1 ÷ gdk3 (17)
èg ø êçè n ÷ø t ck ú è g ø
ë û

Based on Eqs.(14) and (17), the equilibrium sediment concentration ct*k in Eq. (6) can be cal‐
culated by:

æg ö
pbk ç s - 1 ÷ gdk3 ì 1.74 éæ ' ö1.5 ù ü
2.2
q + qb* k èg ø ï éæ t ö U ù n t ï (18)
ct * k = s* k = í0.0000262 êçç - 1 ÷÷ ú + 0.0053 êç ÷ - 1ú ý
Uh Uh ê t w ú ê çn÷ t ú
ï ëè ck ø sk û ëè ø ck
û ïþ
î

The wind shear stresses (τsx and τsy) at the free surface are expressed by
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 365
[Link]

2 2
t sx = r aCdUwind Uwind + Vwind (19)

2 2
t sy = r aCd Vwind Uwind + Vwind (20)

where ρa is the air density; Uwind and Vwind are the wind velocity components at 10 m eleva‐
tion in x and y directions, respectively. Although the drag coefficient Cd may vary with wind
speed (Koutitas and O’Connor 1980; Jin et al. 2000), for simplicity, many researchers as‐
sumed the drag coefficient was a constant on the order of 10-3 (Huang and Spaulding 1995;
Rueda and Schladow 2003; Chao et al 2004; Kocyigit and Kocyigit 2004). In this study, Cd
was taken as 1.5 × 10−3.
In this study, the decoupled approach was used to simulate sediment transport. At one time
step, the flow fields, including water elevation, velocity components, and eddy viscosity pa‐
rameters were first obtained using the hydrodynamic model, and then the suspended sedi‐
ment concentration was solved numerically using Eq. (5).

4. Model verification

4.1. Tide-induced flow

A comparison of model simulation to an analytical solution was performed for a tidally

forced flow in a square basin with constant water depth and no bottom friction. It was as‐
sumed that right side of the basin is an open boundary, and the other three sides are the
closed solid wall. The tidal flow was simulated by driving the free surface elevation at the
open boundary of the basin. A standing cosine wave with the maximum amplitude of Am at
the open boundary was introduced. This case has been used by other researchers to verify
their models (Huang and Spaulding 1995; Zhang and Gin 2000). The analytical solution for
surface elevation (ξ) and velocity (U) were given by Ippen (1966):

x = Am cos(wt ) (21)

Amw x
U( x , t ) = sin(wt ) (22)
H

in which Am is the tide amplitude; ω is the angular frequency, ω = 2π / T ; T is the tidal

period; t is the time; x is the distance from the left closed basin boundary; U is the veloc‐
ity in x direction; and H is the base water depth. In the numerical simulation, the follow‐
ing values were adopted: basin length is 12km, width is 12km, H=10m, T=12h, Am=0.5m.
Figs.2 and 3 show the comparison of the analytical solution and the numerical simula‐
366 Sediment Transport Processes and Their Modelling Applications

tion results for water surface elevation and velocity. The numerical results are in good
agreement with analytical solutions.

Figure 2. Comparison of analytical solution and numerical simulation for water surface elevation

Figure 3. Comparison of analytical solution and numerical simulation for depth-averaged velocity
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 367
[Link]

4.2. Verification of mass transport

To verify the transport simulation model, the numerical results were tested against an ana‐
lytical solution for predicting salinity intrusion in a one-dimensional river flow with con‐
stant depth. It was assumed that the downstream end of the river is connected with the
ocean with salt water. At the end of the river reach (x=0), there is a point source with a con‐
stant salinity, S0, from the ocean, and the salt water may intrude into the river due to disper‐
sion (Fig.4). Under the steady-state condition, the salinity in the river can be expressed as:

¶S ¶ 2S
U = Dx 2 (23)
¶x ¶x

where U is the velocity (no tidal effect); S is the salinity in river; Dx is the dispersion coeffi‐
cient; and x is the displacement from downstream seaward boundary (point O). An analyti‐
cal solution given by Thomann and Mueller (1988) is:

Ux
S( x) = S0 exp( ) (24)
Dx

in which S0 is the salinity at downstream seaward boundary. In this test case, it was as‐
sumed that the water depth = 10 m, U=0.03m/s, Dx=30 m2/s, and S0 = 30 ppt. Fig. 5
shows the salinity concentration distributions obtained by the numerical model and ana‐
lytical solution. The maximum error between the numerical result and analytical solution
is less than 2%.

River Ocean
x O
11400 m

U=0.03 m/s
U=0.03 m/s
S0 =30ppt

Figure 4. Test river for model verification

Fig. 4. Test river for model verification
368 Sediment Transport Processes and Their Modelling Applications

Figure 5. Salinity distribution along the river

5. Model application to lake pontchartrain

5.1. Study area

Fig. 6 shows the bathymetry and locations of field measurement stations of the study
site — Lake Pontchartrain. The circulation in Lake Pontchartrain is an extremely compli‐
cated system. It is affected by tide, wind, fresh water input, etc. The lake has a diurnal
tide with a mean range of 11 cm. Higher salinity waters from the Gulf of Mexico can en‐
ter the lake through three narrow tidal passes: the Rigolets, Chef Menteur, and a man-
made Inner Harbor Navigation Canal (IHNC). Freshwater can discharge into the lake
through the Tchefuncte and Tangipahoa Rivers, the adjacent Lake Maurepas, and from
other watersheds surrounding the lake. The Bonnet Carré Spillway (BCS) is located at
the southwest of the lake.
Based on the bathymetric data, the computational domain was descritized into an irregular
structured mesh with 224×141 nodes using the NCCHE Mesh Generator (Zhang and Jia,
2009).

5.2. Boundary conditions

As shown in Fig.6, there are two inlet boundaries located at the northwest of the lake, and
three tidal boundaries located at the south and east of the lake. The flow discharges at Tche‐
functe and Tangipahoa Rivers obtained from USGS were set as two inlet boundary condi‐
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 369
[Link]

tions. The hourly water surface elevation data at the Rigolets Pass obtained from USGS was
set as a tidal boundary. Due to the lack of measured surface elevation data at Chef Menteur
Pass, the Rigolets data was used (McCorquodale et al., 2005). After the BCS was opened for
Based onthe
flood release, theflow
bathymetric data, the
discharge computational
at BCS was set asdomain
inletwas descritized
boundary into an irregular
conditions.
structured mesh with 224×141 nodes using the NCCHE Mesh Generator (Zhang and Jia,
2009).

Tchefuncte
Tangipahoa N

Mandeville

Rigolets
Lake
Pontchartrain

Norco Gage
South Lake Westend Chef Menteur

IHNC 10 km
BCS spillway

Fig. 6. The
Figure 6. The bathymetry bathymetry
and and fieldstations
field measurement measurement
in Lakestations in Lake Pontchartrain
Pontchartrain

The other
5.2 tidal pass,
Boundary IHNC, is a man-made canal which connects the Lake Pontchartrain and
Conditions
Mississippi River with a lock structure. It is also connected with both the Gulf Intracoastal
As shown in Fig.6, there are two inlet boundaries located at the northwest of the lake, and
Waterwaythreeand
tidal the Mississippi
boundaries locatedRiver
at the Gulf
south Outlet
and east(MRGO).
of the [Link]
The measured
flow dischargesdaily
at water sur‐
face elevation
Tchefuncte data
and is the onlyRivers
Tangipahoa available data
obtained at USGS
from IHNC. In set
were general,
as two theinlet daily water surface
boundary
conditions. The hourly water surface elevation data at the Rigolets Pass obtained from
elevation data can not represent the variations of tidal boundary. It would cause problems if
USGS was set as a tidal boundary. Due to the lack of measured surface elevation data at
the measured dailyPass,
Chef Menteur datathewas directly
Rigolets set used
data was as tidal boundaryetconditions
(McCorquodale at IHNC.
al., 2005). After the BCS To resolve
was opened for flood release, the flow discharge at BCS was set as inlet boundary
this problem, the relationship between measured daily water surface elevations at Rigolets
conditions.
and IHNC tidal passes were established and adopted to convert the hourly data at Rigolest
The otherdata
to the hourly tidal at
pass, IHNC,Since
IHNC. is a man-made canal which
both IHNC connects the
and Rigolets Lakepasses
tidal Pontchartrain and
are connected with
Mississippi River with a lock structure. It is also connected with both the Gulf Intracoastal
the GulfWaterway
of Mexico, and the
the tide effectsRiver
Mississippi at these
Gulftwo places
Outlet are assumed
(MRGO). The measured to bedaily
similar.
waterBy compar‐
ing thesurface
measured daily
elevation datawater surface
is the only elevations
available at Rigolets
data at IHNC. In general,and IHNC,
the daily waterno obvious phase
surface
elevation data can not represent the variations of tidal boundary. It would cause problems if
differences were observed (McCorquodale et al., 2005). Fig.7 shows the
the measured daily data was directly set as tidal boundary conditions at IHNC. To resolve comparison of meas‐
ured daily water surface elevations at Rigolets and IHNC tidal passes.
this problem, the relationship between measured daily water surface elevations at Rigolets

The measured results show that the surface elevations at the two locations have a close line‐
ar relation with the correlation coefficient r2 being 0.92:

hi = 1.0484hr - 0.0055 (25)

370 Sediment Transport Processes and Their Modelling Applications

Where ηi and ηr are the daily surface elevations at IHNC and Regolets, respectively. It was
assumed that the hourly water surface elevations at IHNC and Regolets have the similar re‐
lationships, and Eq. (25) was adopted to calculate the hourly water surface elevations at
IHNC from the measured hourly surface elevation at Rigolets. The calculated hourly data
was plotted together with the measured daily data at IHNC (Fig.8), a similarity distribution
was observed. So the calculated hourly surface elevation was used as tidal boundary condi‐
tion at IHNC.

Measured daily elevations at IHNC (m)

m easurem ents
0.8 fitted line

0.6

0.4 2
r =0.92

0.2

0
-0.4 -0.2 0 0.2 0.4 0.6 0.8
Measured daily elevations at Rigolets (m )
-0.2

-0.4

Figure 7. Comparison of measured daily water surface elevations at Regolets and IHNC

5.3. Model calibration

After obtaining the inlet boundaries, outlet boundaries, and wind speeds and directions,
the developed model was applied to simulate the flow circulation and sediment trans‐
port in Lake Pontchartrain. Some field measured data sets were used for model calibra‐
tion and validation.

A period from March 1 to 31, 1998, was selected for model calibration. For calibration runs,
several parameters, such as drag coefficient Cd, Manning’s roughness coefficient, and the pa‐
rameter α in Smagorinsky scheme (Eq.4), were adjusted to obtain a reasonable reproduction
of the field data. In this study, Cd = 0.0015, Manning’s roughness coefficient = 0.025, and α =
0.1. Simulated water surface elevations and depth-averaged velocities were compared with
the field measured data. Fig. 9 shows the simulated and measured water surface elevations
at the Mandeville. Fig. 10 and Fig.11 show the simulated and measured depth-averaged ve‐
locities in x and y directions at the South Lake Site, respectively.
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 371
[Link]

0.8 measurements (daily)

calculations (hourly)
Surface elevations (m)

0.6

0.4

0.2

-0.2

-0.4
0 5 10 15 20 25 30
Day (March 1-31, 1998)

Figure 8. The calculated hourly water surface elevation at IHNC using Eq. (25)

0.8 measurements
model
Surface elevations (m)

0.6

0.4

0.2

-0.2

-0.4
0 5 10 15 20 25 30
Day (March 1-31, 1998)

Figure 9. Simulated and measured water surface elevations at the Mandeville Station

A set of statistics error analysis, including root mean square error (RMSE), relative RMSE
(RMSE/range of observed data) and correlation coefficient (r2), were used to assess the per‐
formance of the model for the calibration case (Table 2). The RMSE between simulated and
observed water surface elevations at Mandeville Station was 0.037m and the relative RMSE
of water surface elevations at this station was 3.4%. The r2 of simulated and observed water
surface elevations at this station was 0.98. The measured velocity data set at South Lake sta‐
tion was used for model comparison. It can be observed that the RMSE of u-velocity and v-
velocity were 0.044 m/s and 0.019 m/s. The relative RMSE for u-velocity and v-velocity were
15% and 12%; and r2 of simulated and observed u-velocity and v-velocity were 0.52 and 0.41,
respectively. In general, the flow fields produced by the numerical model are in agreement
with field measurements.
372 Sediment Transport Processes and Their Modelling Applications

0.25

0.2
measurements
0.15 model
Velocity (m/s)

0.1

0.05

-0.05

-0.1

-0.15

-0.2
0 5 10 15 20 25 30
Day (March 1-31, 1998)

Figure 10. Simulated and measured depth-averaged velocities in west-east direction at the South Lake Site

0.08

0.06

0.04

0.02
Velocity (m/s)

-0.02

-0.04

-0.06

-0.08 measurements
model
-0.1

-0.12
0 5 10 15 20 25 30
Day (March 1-31, 1998)

Figure 11. Simulated and measured depth-averaged velocities in south-north direction at the South Lake Site

Station Variable RMSE Observed range Relative RMSE r2

Mandeville Water level 0.037 m 1.11 m 3.40% 0.98

South Lake u-velocity 0.044 m/s 0.29 m/s 15% 0.52

South Lake v-velocity 0.019 m/s 0.16 m/s 12% 0.41

Table 2. Statistics error analysis of the model calibration case

5.4. Modeling the suspended sediment during the BCS opening for 1997 flood release

After the Bonnet Carré Spillway (BCS) was built to divert Mississippi River flood waters
to the Gulf of Mexico via Lake Pontchartrain, there were 10 times opening events occur‐
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 373
[Link]

red from 1937 to 2011. In this study, the 1997 flood release event was selected for model
simulation.

In 1997, the BCS was opened for flood release from 3/17 to 4/18. The maximum flow dis‐
charge was about 6881 m3/s, and over 31 days of flood release. The average discharge was
about 4358 m3/s. Fig. 12 shows the flow hydrograph at the spillway (Department of Natural
Resources, 1997; McCorquodale et al. 2007). The total volume of sediment-laden water enter‐
ing Lake Pontchartrain was approximately 1.18×1010 m3, or twice the volume of the lake
(Turner et al., 1999). The total amount of sediment entering the lake was about 9.1 million
tons, more than 10 times as much as the normal yearly sediment loads of the lake. The sus‐
pended sediment (SS) concentration at the spillway gate was about 240 mg/l (Manheim and
Hayes, 2002).

8000

6000
Discharge (m3/s)

4000

2000

0
0 5 10 15 20 25 30 35
Day (Mar.17-April 18, 1997)

Figure 12. The flow hydrograph at the Bonnet Carré Spillway during the 1997 event

The calibrated CCHE2D model was applied to simulate the lake flow fields and sediment
transport during the BCS opening in 1997. In this period, the flow discharge was very
strong, and the “suspended load approach” was adopted for simulating sediment transport
in the lake. The observed flow discharge was set as inlet boundary condition at BCS. The
water surface elevations at Rigolets and Chef Menteur were set as tidal boundaries. The
wind speeds and directions at the New Orleans International Airport were used for model
simulation. The observed SS concentration was set as inlet sediment boundary condition at
BCS. In general, the sediment in Lake Pontchartrain is cohesive sediment. However, during
the BCS opening, sediment concentration in Lake Pontchartrain is dominated by the sedi‐
ment coming from the Mississippi River. It was assumed that the effect of sediment cohe‐
sion on suspended sediment transport is not significant. Due to the lack of measured
sediment data, the classes of non-uniform sediment size at BCS were estimated based on the
observed sediment data in the lower Mississippi River (Thorne et al. 2008). Four size classes,
including 0.005mm, 0.01mm, 0.02mm and 0.04mm were assumed to represent the non-uni‐
form sizes of suspended sediment discharged into the lake from BCS. The fall velocity of
374 Sediment Transport Processes and Their Modelling Applications

each size class of sediment was calculated using the Eq. (7) proposed by Zhang and Xie
(1993). During this period, the flow discharge over the spillway dominated the lake hydro‐
dynamics and suspended sediment transport. The bottom shear stress due to water flow as
well as wind driven flow were obtained using the hydrodynamic model. The critical shear
stress was calculated using Eq. (9) proposed by Wu (2008). The equilibrium sediment con‐
centration ct*k was calculated using Eq. (18).
water surface elevations at Rigolets and Chef Menteur were set as tidal boundaries. The
Fig. 13 shows the computed
wind speeds flow circulations
and directions at the NewinOrleans
Lake Pontchartrain
Internationalduring
Airportthe BCSused
were open‐
for model
ing. Duesimulation.
to the flood release, the entire lake water were moved eastward through
The observed SS concentration was set as inlet sediment boundary condition at Rigolets
and Chef [Link] into the
In general, the Gulf of Mexico,
sediment in Lakewhich was completely
Pontchartrain different
is cohesive fromHowever,
sediment. the flow during
patternsthe induced by tide and
BCS opening, wind. Fig.
sediment 14 shows theincomparisons
concentration of SS concentration
Lake Pontchartrain is dominated ob‐ by the
tained from the numerical
sediment coming fromsimulation and remote
the Mississippi sensing
River. imageries
It was assumed(AVHRR
that the data) provid‐
effect of sediment
cohesionThe
ed by NOAA. on suspended
simulated SS sediment transportare
concentrations is not significant.
generally Due to
in good the lack of
agreement measured
with
satellite sediment
[Link],
The the classesprocesses
transport of non-uniform
of SS insediment size atreproduced
the lake were BCS were estimated based on the
by the numeri‐
cal [Link] sediment
The simulated data in
results andthesatellite
lower Mississippi River (Thorne
imageries revealed et al. amount
that a large 2008). Four size classes,
of sedi‐
including 0.005mm, 0.01mm, 0.02mm and 0.04mm were assumed to represent the non-
ment discharged into the lake, moved eastward along the south shore and gradually
uniform sizes of suspended sediment discharged into the lake from BCS. The fall velocity of
expanded northward, eventually affecting the entire Lake after one month of diversion.
each size class of sediment was calculated using the Eq. (7) proposed by Zhang and Xie
(1993). During this period, the flow discharge over the spillway dominated the lake
hydrodynamics and suspended sediment transport. The bottom shear stress due to water
flow as well as wind driven flow were obtained using the hydrodynamic model. The critical
shear stress was calculated using Eq. (9) proposed by Wu (2008). The equilibrium sediment
concentration ct*k was calculated using Eq. (18).

Fig. 13 shows the computed flow circulations in Lake Pontchartrain during the BCS
opening. Due to the flood release, the entire lake water were moved eastward through
Rigolets and Chef Menteur into the Gulf of Mexico, which was completely different from
the flow patterns induced by tide and wind. Fig. 14 shows the comparisons of SS
concentration obtained from the numerical simulation and remote sensing imageries
Figure 13. Flow circulations in Lake Pontchartrain during BCS opening in 1997
(AVHRR data) provided by NOAA. The simulated SS concentrations are generally in good
agreement with satellite imageries. The transport processes of SS in the lake were
reproduced by the numerical model. The simulated results and satellite imageries revealed
that a large amount of sediment discharged into the lake, moved eastward along the south
shore and gradually expanded northward, eventually affecting the entire Lake after one
month of diversion.
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 375
[Link]

SS concentration ( kg/m3)

0.10
0.09
0.02 0.08
0.02 0.09 0.06
0.07
0.09 0.06 0.06
0.05
0.04
0.08 0.08
0.10 0.03
0.10
0.02
0.01
0.00

Remote sensing imagery Simulation results

Fig. 14. Comparisons of simulated depth-averaged SS concentration and remote sensing

Figure 14. Comparisons of simulated depth-averaged SS concentration and remote sensing imageries (4/7/1997)
imageries (4/7/1997)

6. Discussion

6. Discussion
In general, wind and tide are the major driving mechanisms of circulation in Lake
Pontchartrain. When the tidal level changes, most of the water that enters or leaves the lake
must come through the three narrow tidal passes at the east and south end of the lake. Since
In general, wind
the tidal andaretide
passes very are thethemajor
narrow, driving
tidal force mechanisms
may affect of near
the flow fields circulation
the tidal in Lake
Pontchartrain.
passes. When
When thethe windtidal
blows level changes,
over the most
lake, it may offlow
affect thecirculations
water that enters
of the whole or leaves the
lake.
Fig. 15 shows the general flow pattern of the lake induced by tide and wind. It was
lake mustcompletely
come through the three
different from the one narrow
when the tidal
BCS was passes
openedatfor
the east
flood andand
release south end of the
caused
lake. Since
the the
entiretidal passes
lake water aremoved
to be veryeastward
narrow, intothe
the tidal
Gulf offorce
Mexicomay
(Fig. affect
13). Duethe flow fields
to the
near the effects of tide and
tidal passes. wind, the
When the stronger
wind blowscurrentsover
occurthealong the shoreline
lake, where the
it may affect flowwater
circulations
depth is shallow and near the narrow tidal passes, and weaker currents are in the center of
of the whole lake.
the lake. [Link]
These 15 shows thetogeneral
are similar flow pattern
results obtained by other of the lake(Signell
researchers induced by tide and
and List
wind. It 1997,
was McCorquodale
completely et different
al., 2005). from the one when the BCS was opened for flood re‐
lease and caused the entire lake water to be moved eastward into the Gulf of Mexico
Under the normal condition, sediment in Lake Pontchartrain is dominated by cohesive
(Fig. 13).sediments
Due toexceptthe for
effects
a smallof tide ofand
number areaswind,
near thethe
riverstronger
mouths and currents occur
tidal passes (Flocksalong the
shorelineetwhere theIn water
al. 2009). responsedepth
to wind is and
shallow and near
tidal induced flow the narrow
shown in Fig. tidal passes,
15, sediment and weak‐
may
transport/resuspend near shoreline and tidal passes, and deposit in the lake
er currents are in the center of the lake. These results are similar to results obtained by center.

other researchers
During the(Signell andfor
BCS opening List 1997,
flood McCorquodale
release et al., of
in 1997, large amount 2005).
sediment discharged
from the Mississippi River into Lake Pontchartrain. Sediment transport in Lake
Under thePontchartrain
normal condition, sediment
is dominated in Lake sediment
by the suspended Pontchartrain
from theis dominated
Mississippi byand
River, cohesive
the sedi‐
ments except
effect for a smallcohesion
of sediment numbercan of be
areas nearInthe
ignored. thisriver mouths
period, the flowand tidal
fields werepasses
majorly(Flocks et
al. 2009). determined
In response on the flow discharge
to wind and tidalat the BCS. They
induced were
flow also affected
shown by 15,
in Fig. the wind and tide
sediment may trans‐
induced flows. The simulated results and satellite imageries indicated that the suspended
port/resuspend near shoreline and tidal passes, and deposit in the lake center.
sediment moved eastward along the south shore first, gradually expanded northward, and
was distributed throughout almost the entire lake about three weeks after the BCS opening.
During the BCS
It was opening
reported due tofor
theflood releasea significant
flood release, in 1997, amount
large amount
of nutrientsofwere
sediment discharged
discharged
from the Mississippi River into Lake Pontchartrain. Sediment transport in Lake Pontchar‐
train is dominated by the suspended sediment from the Mississippi River, and the effect of
sediment cohesion can be ignored. In this period, the flow fields were majorly determined
by the flow discharge at the BCS. They were also affected by the wind and tide induced
flows. The simulated results and satellite imageries indicated that the suspended sediment
moved eastward along the south shore first, gradually expanded northward, and was dis‐
tributed throughout almost the entire lake about three weeks after the BCS opening. It was
reported due to the flood release, a significant amount of nutrients were discharged into
 376 Sediment Transport Processes and Their Modelling Applications

into Lake
LakePontchartrain,
Pontchartrain, which
which maymayresult
resultinina amassive
massive algal
algal bloom
bloom in lake.
in the the lake. However,
However, the
the sediment
sediment and nutrients were simultaneously discharged into the lake, and the algaethe
and nutrients were simultaneously discharged into the lake, and algae
growth
growthrate
ratewas
was restricted
restricted as aasresult
a result of extremely
of extremely high high suspended
suspended sediment sediment concentration
concentration in the
in the lake (Chaoetetalal2007).
lake (Chao 2007).SoSothere
there was
was no no algal
algal bloombloom observed
observed in theinlake
the during
lake during
the BCS the
BCS opening.
[Link] the BCS
After the BCSwas wasclosed,
closed,
thethe sediment
sediment derivedderived
from thefrom the Mississippi
Mississippi River
River gradu‐
gradually
ally dispersed
dispersed ininthe
thewater
water column
column andand deposited
deposited to thetolake
thebed.
lakeInbed. In response
response to the lowtoSSthe
low SSconcentration,
concentration,
highhigh nutrients,
nutrients, temperature
temperature and lightand light the
intension, intension,
algal bloomtheoccurred
algal bloom
in a
occurred in area
large a large arealake,
of the of the
andlake, and the
the peak of thepeak of bloom
algal the algal bloominobserved
observed mid-June,inabout
mid-June,
two
months
about two after the
months spillway
after closure closure
the spillway (Dortch et al., 1998).
(Dortch et In
al.,general,
1998). Inafter the BCSafter
general, was the
closed,
BCS
it took about two to three months for the SS concentration in the lake recovered
was closed, it took about two to three months for the SS concentration in the lake recovered to the sea‐
sonal average
to the seasonal [Link].
average

0.5m/s

Fig. 15.
Figure [Link]
Generalflow flow circulations
circulations in Lakedue
in Lake Pontchartrain Pontchartrain
to tide and wind due to tide and wind

7. Conclusions
7. Conclusions
A numerical model was applied to simulate the flow circulations and suspended sediment
transport in Lake Pontchartrain in Louisiana, under tide, wind and flood release. It is one of
A numerical model was applied to simulate the flow circulations and suspended sediment
the most significant
transport in Lakereal life problems
Pontchartrain we canunder
in Louisiana, find tide,
withwind
reasonable
and floodfield measurements
release. It is one of
obtained from USGS and USACE. Additional satellite imageries were obtained from NOAA
the most significant real life problems we can find with reasonable field measurements ob‐
for model
tained validation.
from USGSThe andresults
[Link]
these comparisons are in good
satellite imageries agreement
were obtained fromwithin
NOAA the
accuracy limitations of both the approximate numerical model solutions and
for model validation. The results of these comparisons are in good agreement within the ac‐the field
measurements under theofdifficult
curacy limitations both theconditions.
approximate numerical model solutions and the field meas‐
urements under the difficult conditions.
In the BCS flood release event, a vast amount of fresh water, sediment and nutrients were
In the BCS flood release event, a vast amount of fresh water, sediment and nutrients were
discharged into Lake Pontchartrain. The dispersion and transport processes of the
discharged into Lake Pontchartrain. The dispersion and transport processes of the suspend‐
suspended sediment in the lake were simulated successfully using the numerical model. .
The simulated SS concentrations are generally in good agreement with satellite imageries
provided by NOAA. The differences of flow circulation and sediment transport in the lake
under normal condition and BCS opening event were discussed. The simulated results and
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 377
[Link]

ed sediment in the lake were simulated successfully using the numerical model. The simu‐
lated SS concentrations are generally in good agreement with satellite imageries provided
by NOAA. The differences of flow circulation and sediment transport in the lake under nor‐
mal condition and BCS opening event were discussed. The simulated results and satellite
imageries show that after the BCS opening, a large amount of sediment discharged into the
lake, moved eastward and gradually expanded northward, eventually affecting the entire
lake. After the BCS closure, the sediment derived from the Mississippi River gradually de‐
posited to the lake bed and it took two to three months for the SS concentration in the lake
recovered to the seasonal average level.
This research effort has positively demonstrated that the numerical model is capable of pre‐
dicting free surface flow and sediment transport in Lake Pontchartrain under extreme natu‐
ral conditions. It is a useful tool for providing information on hydrodynamics and sediment
transport in such a big lake where the field measurements may not be sufficient. All the in‐
formation obtained from the numerical model is very important for lake restoration and wa‐
ter quality management.

Acknowledgements

This research was funded by the US Department of Homeland Security and was sponsored
by the Southeast Region Research Initiative (SERRI) at the Department of Energy's Oak
Ridge National Laboratory. The authors would like to thank Rich Signell and David Walters
of the USGS, and George Brown of the USACE for providing field measured data in Lake
Pontchartrain. The technical assistance from Yaoxin Zhang, Weiming Wu and suggestions
and comments provided by Yan Ding, Tingting Zhu and Kathy McCombs of the University
of Mississippi are highly appreciated.

Author details

Xiaobo Chao, Yafei Jia and A. K. M. Azad Hossain

The University of Mississippi, USA

References

[1] Chao, X, Jia, Y, & Shields, D. (2004). Three dimensional numerical simulation of flow
and mass transport in a shallow oxbow lake. Proc., World Water & Environmental
Resources Congress 2004, ASCE, Resyon, Va. (CD-Rom).
[2] Chao, X, Jia, Y, & Zhu, T. (2006). CCHE_WQ Water Quality Module. Technical Re‐
port : NCCHE-TR-The University of Mississippi., 2006-01.
378 Sediment Transport Processes and Their Modelling Applications

[3] Chao, X, Jia, Y, Shields, D, & Wang, S. S. Y. (2007). Numerical Modeling of Water
Quality and Sediment Related Processes. Ecological Modelling, 201, 385-397.

[4] Department of Natural Resources(1997). Bonnet Carre Spillway Opening. Hydro‐

graphic Data Report, Final Report.

[5] Dortch, M. S, Zakikhani, M, Kim, S. C, & Steevens, J. A. (2008). Modeling water and
sediment contamination of Lake Pontchartrain following pump-out of Hurricane Ka‐
trina floodwater. Journal of Environmental ManagementS: , 429-442.

[6] Dortch, Q, Peterson, J, & Turner, R. (1998). Algal bloom resulting from the opening of
the Bonnet Carre spillway in 1997. Fourth Bi-annual Basics of the Basin Symposium.
United States Geological Survey, , 28-29.

[7] Flocks, J, Kindinger, J, Marot, M, & Holmes, C. (2009). Sediment characterization and
dynamics in Lake Pontchartrain, Louisiana. Journal of Coastal ResearchSpecial ,
54(54), 113-126.

[8] Gulf Engineers & Consultants (GEC) ((1998). Biological and Recreational Monitoring
of the Impacts of 1997 Opening of the Bonnet Carré Spillway Southeastern Louisiana.
Final Report to U.S Army Corps of Engineers, New Orleans, Louisiana.

[9] Hamilton, G. D, Soileau, C. W, & Stroud, A. D. (1982). Numerical modeling study of

Lake Pontchartrain. Journal of the Waterways, Port, Coastal and Ocean Division,
ASCE. , 108, 49-64.

[10] Hossain, A, Jia, Y, Ying, X, Zhang, Y, & Zhu, T. T. (2011). Visualization of Urban Area
Flood Simulation in Realistic 3D Environment, Proceedings of 2011 World Environ‐
mental & Water Resources Congress, May Palm Springs, California., 22-26.

[11] Huang, W, & Spaulding, M. (1995). 3D model of estuarine circulation and water
quality induced by surface discharges. Journal of Hydraulic Engineering, 121(4),
300-311.

[12] Ippen, A. T. (1966). Estuary and Coastline Hydrodynamics. McGraw Hill, New York.

[13] Kocyigit, M. B, & Kocyigit, O. (2004). Numerical study of wind-induced currents in

enclosed homogeneous water bodies. Turkish J. Engineering & Environmental Sci‐
ence, , 28, 207-221.

[14] Koutitas, C, & Connor, O. B., ((1980). Modeling three-dimensional wind-induced

flows. ASCE, Journal of Hydraulic Division, 106 (11), 1843-1865.

[15] Jia, Y, & Wang, S. S. Y. (1999). Numerical model for channel flow and morphological
change studies. Journal of Hydraulic Engineering, 125(9), 924-933.

[16] Jia, Y, Wang, S. Y. Y, & Xu, Y. (2002). Validation and application of a 2D model to
channels with complex geometry. International Journal of Computational Engineering
Science, 3(1), 57-71.
 Numerical Modeling of Flow and Sediment Transport in Lake Pontchartrain 379
[Link]

[17] Jin, K. R, Hamrick, J. H, & Tisdale, T. (2000). Application of three dimensional hydro‐
dynamic model for Lake Okeechobee. J. Hydraulic Engineering, 126(10), , 758
EOF-772 EOF.

[18] Manheim, F. T, & Hayes, L. (2002). Sediment database and geochemical assessment
of Lake Pontchartrain Basin, in Manheim FT, and Hayes L (eds.), Lake Pontchartrain
Basin: Bottom Sediments and Related Environmental Resources: U.S. Geological Sur‐
vey Professional Paper 1634.

[19] Mccorquodale, J. A, Georgiou, I, Chilmakuri, C, Martinez, M, & Englande, A. J.

(2005). Lake Hydrodynamics and Recreational Activities in the South Shore of Lake
Pontchartrain, Louisiana. Technical Report for NOAA, University of New Orleans.

[20] Mccorquodale, J. A, Georgiou, I, Retana, A. G, Barbe, D, & Guillot, M. J. (2007). Hy‐

drodynamic modeling of the tidal prism in the Pontchartrain Basin, Technical report,
The University of New Orleans.

[21] Mccorquodale, J. A, Roblin, R. J, Georgiou, I, & Haralampides, K. A. (2009). Salinity,

nutrient, and sediment dynamics in the Pontchartrain Estuary. Journal of Coastal Re‐
searchSpecial , 54(54), 71-87.

[22] Penland, S., Beall, A. and Kindinger, J. (2002). Environmental Atlas of the Lake
Pontchartrain Basin. USGS Open File Report 02-206.

[23] Rueda, F.J. and Schladow, S.G. (2003). Dynamics of large polymictic lake. II: Numeri‐
cal Simulations. Journal of Hydraulic Engineering. 129(2): 92-101.

[24] Signell, R.P. and List, J.H. (1997). Modeling Waves and Circulation in Lake Pontchar‐
train. Gulf Coast Association of Geological Societies Transactions. 47: 529-532.

[25] Smagorinsky, J. (1993). Large eddy simulation of complex engineering and geophysi‐
cal flows, in Evolution of Physical Oceanography, edited by B. Galperin, and S. A.
Orszag, pp. 3-36. Cambridge University Press.

[26] Stone, H.L. (1968). Iterative solution of implicit approximation of multidimensional

partial differential equations. SIAM (Society for Industrial and Applied Mathematics)
Journal on Numerical Analysis. 5: 530–558.

[27] Thomann, R.V. and Mueller, J.A. (1988). Principles of Surface Water Quality Model‐
ing and Control. Harper&Row Publication, New York.

[28] Thorne, C., Harmar, O., Watson, C., Clifford, N., Biedenham, D. and Measures, R.
(2008). Current and Historical Sediment Loads in the Lower Mississippi River. Final
Report, Department of Geology, Nottingham University.

[29] Turner, R.F., Dortch, Q. and Rabalais, N.N. (1999). Effects of the 1997 Bonnet Carré
Opening on Nutrients and Phytoplankton in Lake Pontchartrain. Final report sub‐
mitted to the Lake Pontchartrain Basin Foundation, 117 p.
380 Sediment Transport Processes and Their Modelling Applications

[30] United States Army Corps of Engineers (USACE) (2011). Bonnet Carre Spillway
overview, Spillway pace ([Link]
[Link]).

[31] Wu, W. (2008). Computational River Dynamics, Taylor & Francis Group, London,
UK.

[32] Wu, W., Wang, S.S.Y. and Jia, Y. (2000). Non-uniform sediment transport in alluvial
river, Journal of Hydraulic Research, IAHR 38 (6) 427-434.

[33] Zhang, Q.Y. and Gin, K.Y.H. (2000). Three-dimensional numerical simulation for ti‐
dal motion in Singapore’s coastal waters. Coastal Engineering. 39: 71–92.

[34] Zhang, R.J. and Xie, J.H. (1993). Sedimentation Research in China, Systematic Selec‐
tions. Water and Power Press. Beijing, China.

[35] Zhang, Y, & Jia, Y. (2009). CCHE Mesh Generator and User’s Manual, Technical Re‐
port No. NCCHE-TR-University of Mississippi., 2009-1.

[36] Zhu, T, Jia, Y, & Wang, S. S. Y. (2008). CCHE2D water quality and chemical model
capabilities. World Environmental and Water Resources Congress. Hawaii (CD-
ROM).
SEDIMENT TRANSPORT
PROCESSES AND THEIR
MODELLING APPLICATIONS

Edited by Andrew J. Manning