SPM 1 - Coastal BW

c?
^m^t^
U.S. Army
Coastal Engineering
Research Centei
DATA LIBRARY
Woods Hole Oceanographic Institution
SHORE PROTECTION
MANUAL
DEPARTMENT OF THE ARMY

1/. 1 CORPS OF ENGINEERS
1975
Reprint or republication of any of this material shall give appropriate
credit to the U.S. Army Coastal Engineering Research Center.
U.S. Army Coastal Engineering Research Center

Kingman Building
Fort Belvoir, Virginia 22060
SONOMA COAST, CALIFORNIA (GOAT ROCK) - 10 December 1958

SHORE PROTECTION
MANUAL
DATA LIBRARY
Institution
Woods Hole Oceanographic
VOLUME I
I
m
i cr ( Chapters 1 Through 4
; J
s a
: -0
°
:
= o
i 1^
: m
S°a ;
U.S. ARMY COASTAL ENGINEERING RESEARCH CENTER
1975
Second Edition
For sale by the Superintendent of Documents, U.S. Government Printing Office

Washintrton, D.C. 20402 - Price $15.05 per 3-part set. (sold in sets only)
Stock Number 008-022-00077-1 Catalog Number D 103.42/6:SH7/V.l-3
PREFACE
The U.S. Army Coastal Engineering Research Center, (CERC) formerly the Beach Erosion Board, has,
since 1930, conducted studies on shore processes and methods of shore protection. CERC continues an
extensive research and development program to improve both shore protection and offsliore engineering
techniques. The scientific and engineering aspects of coastal processes and offshore structures are in the
developmental stage and the requirement for improved techniques for use in design and engineering of
coastal structures is evident. This need was met in 1954, to the extent of available knowledge, by
publication of Teclmical Report Number 4, "Shore Protection, Planning and Design" (TR 4); revised
editions thereof appeared in 1957, 1961, and 1966.

Significant advances in knowledge and capability have been made since the 1966 revision. This Shore
Protection Manual (SPM) incorporates new material with appropriate information extracted from TR 4,
and expands coverage within the shore protection field. This SPM is a replacement volume covering
guidelines and techniques for functional and structural design for shore protection works. Accordingly,
further editions of TR 4 are not planned.
The Shore Protection Manual is in three volumes. Volume 1 describes the physical environment in the
coastal zone starting with an introduction of coastal engineering, continuing with discussions of mechanics
of wave motion, wave and water level predicitons, and finally littoral processes.
Volume II translates the interaction of the physical environment and coastal structures into design
parameters for use in the solution of coastal engineering problems. It discusses planning, analysis, structural
features, and structural design as related to physical factors, and shows an example of a coastal engineering
problem which utilizes the technical content of material presented in all three volumes.
Volume III contains four appendixes including a glossary of coastal engineering terms, a list of symbols,
tables and plates, and a subject index.
R. A. Jachowski, Chief, Design Branch, Engineering Development Division, was the project engineer
responsible for preparation and assemblage of the text, under the general supervision of G. M. Watts, Chief,
Engineering Development Division. At the time of approval for publication by the Coastal Engineering
Research Board, LTC Don S. McCoy was Commander and Director and Thorndike Saville, Jr. was Technical
Director. Members of the Coastal Engineering Research Board were: MG John W. Morris (President),
MG Daniel A. Raymond, MG Ernest Graves, Jr., BG George B. Fink, Dean Morrough P. O'Brien, Dr. Arthur
T. Ippen, and Prof. Robert G. Dean. The board members were intimately involved in both the planning and
review of this manual.
Preparation of this manual includes the contribution, review and suggestions of numerous engineers,
scientists, technical and support personnel. Present members of the CERC staff who made significant
technical contributions to this manual are: R. H. Allen, B. R. Bodine, M. T. Czemiak, A. E. DeWall, D. B.

Duane, C. J. Galvin, R. J. Hallermeier, D. L. Harris, R. A. Jachowski, W. R. James, 0. M. Madsen, P. C.
Pritchett, A. C. Rayner, R. L. Rector, R. P. Savage, T. Saville, Jr., P. N. Stoa, P. G. Teleki, G. M. Watts, J.
R. Weggel and D. W. Woodard. Technical editor for this manuscript was R. H. Allen. Typing and composing
were done by M. L. Vrooman and C. M. Lowe; and drafting by H. J. Bruder and J. S. Rivas. LCDR K. E.
Fusch was responsible for the completion of the final manuscript.

The manual format and binding were selected to optimize its use by scientists and engineers as a learning
text as well as a field and office engineering reference. Chapters include a bibliography. The binding
facilitates text and chart removal for separate use or rebinding in loose leaf form.
Comments or suggestions on material in this manual are invited.
This report is published under authority of Public Law 166, 79'^ Congress, approved July 31, 1945, as
supplemented by Public Law 172, 88''' Congress, approved November 7, 1963.
Ill
TABLE OF CONTENTS
VOLUME n
CHAPTER PAGE
5 PLANNING ANALYSIS
5.1 GENERAL 5-1
5.2 SEAWALLS. BULKHEADS, AND REVETMETS 5-3
5.3 PROTECTIVE BEACHES . ^ 5-7
5.4 SAND DUNES 5-21
5.5 SAND BYPASSING 5-24
5.6 GROINS 5-31
5.7 JETTIES 5-46
5.8 BREAKWATERS-SHORE-CONNECTED 5-49
5.9 BREAKWATERS-OFFSHORE 5-50
5.10 ENVIRONMENTAL CONSIDERATIONS 5-57
REFERENCES AND SELECTED BIBLIOGRAPHY 5-58
6 STRUCTURAL FEATURES
6.1 INTRODUCTION 6-1
6.2 SEAWALLS, BULKHEADS, AND REVETMENTS 6-1
6.3 PROTECTIVE BEACHES 6-16
6.4 SAND DUNES 6-36
6.5 SAND BYPASSING 6-54
6.6 GROINS 6-76
6.7 JETTIES 6-84
6.8 BREAKWATERS-SHORE-CONNECTED 6-88
6.9 BREAKWATERS-OFFSHORE 6-96
6.10 CONSTRUCTION MATERIALS 6-96
6.11 MISCELLANEOUS DESIGN PRACTICES 6-98
7 STRUCTURAL DESIGN-PHYSICAL FACTORS

7.1 WAVE CHARACTERISTICS 7-1
7.2 WAVE RUNUP, OVERTOPPING, AND TRANSMISSION 7-15
7.3 WAVE FORCES 7-63
7.4 VELOCITY FORCES-STABILITY OF CHANNEL REVETMENTS 7-203
7.5 IMPACT FORCES 7-204
7.6 ICEFORCES 7-206
7.7 EARTH FORCES 7-208
8 ENGINEERING ANALYSIS-CASE STUDY

8.2 DESIGN PROBLEM CLACULATIONS-ARTIFICIAL OFFSHORE ISLAND . . 8-2
REFERENCES 8-132
VOLUME III
APPENDIX
A GLOSSARY OF TERMS A-1
B LIST OF SYMBOLS B-1
C MISCELLANEOUS TABLES AND PLATES C-1
D SUBJECT INDEX D-1
IV
TABLE OF CONTENTS
VOLUME I
CHAPTER 1 - INTRODUCTION TO COASTAL ENGINEERING
SECTION PAGE
1.1 INTRODUCTION TO THE SHORE PROTECTION MANUAL 1-1
1.2 THE SHORE ZONE 1-2

1.21 NATURAL BEACH PROTECTION 1-2
1.22 NATURAL PROTECTIVE DUNES 1-2
1.23 BARRIER BEACHES, LAGOONS AND INLETS 1-5
1.24 STORM ATTACK 1-5
1.25 ORIGIN AND MOVEMENT OF BEACH SANDS 1-5
1.3 THE SEA IN MOTION 1-5

1.31 TIDES AND WINDS 1-5
1.32 WAVES 1-5
1.33 CURRENTS AND SURGES 1-7
1.34 TIDAL CURRENTS 1-9
1.4 THE BEHAVIOR OF BEACHES 1-9

1.41 BEACH COMPOSITION 1-9
1.42 BEACH CHARACTERISTICS 1-9
1.43 BREAKERS 1-10
1.44 EFFECTS OF WIND WAVES 1-10
1.45 LITTORAL TRANSPORT 1-10
1.46 EFFECT OF INLETS ON BARRIER BEACHES 1-12
1.47 IMPACT OF STORMS 1-12
1.48 BEACH STABILITY 1-13
1.5 EFFECTS OF MAN ON THE SHORE 1-13

1.51 ENCROACHMENT ON THE SEA 1-13
1.52 NATURAL PROTECTION 1-14
1.53 SHORE PROTECTION METHODS 1-14
1.54 BULKHEADS, SEAWALLS AND REVETMENTS 1-16
1.55 BREAKWATERS 1-16
1.56 GROINS 1-17
1.57 JETTIES 1-17
1.58 BEACH RESTORATION AND NOURISHMENT 1-18
1.6 CONSERVATION OF SAND 1-21
CHAPTER 2 - MECHANICS OF WAVE MOTION
2.2 WAVE MECHANICS 2-1

2,21 GENERAL 2-1
SECTION PAGE
2.22 WAVE FUNDAMENTALS AND CLASSIFICATION OF WAVES 2-3

2.23 ELEMENTARY PROGRESSIVE WAVE THEORY
(Small-Amplitude Wave Theory) 2-7
2.231 Wave Celerity, Length and Period 2-7
2.232 The Sinusoidal Wave Profile 2-10
2.233 Some Useful Functions 2-10
2.234 Local Fluid Velocities and Accelerations 2-12
2.235 Water Particle Displacements 2-15
2.236 Subsurface Pressure 2-22
2.237 Velocity of a Wave Group 2-24
2.238 Wave Energy and Power 2-27
2.239 Summary - Linear Wave Theory 2-33
2.24 HIGHER ORDER WAVE THEORIES 2-33
2.25 STOKES' PROGRESSIVE, SECOND-ORDER WAVE THEORY 2-36
2.251 Wave Celerity, Length and Surface Profile 2-37
2.252 Water Particle Velocities and Displacements 2-37
2.253 Mass Transport Velocity 2-38
2.254 Subsurface Pressure 2-39
2.255 Maximum Steepness of Progressive Waves 2-39
2.256 Comparison of the First- and Second-Order Theories . . 2-39
2.26 CNOIDAL WAVES 2-47
2.27 SOLITARY WAVE THEORY 2-59
2.28 STREAM FUNCTION WAVE THEORY 2-62
2.3 WAVE REFRACTION 2-62

2.32 GENERAL - REFRACTION BY BATHYMETRY 2-65
2.321 Procedures in Refraction Diagram Construction -
Orthogonal Method 2-69
2.322 Procedure when a is Less than 80 Degrees 2-71
2.323 Procedure when a is Greater than 80 Degrees - The
R/J Method 2-73
2.324 Refraction Fan Diagrams 2-73
2.325 Other Graphical Methods of Refraction Analysis .... 2-75
2.326 Computer Methods for Refraction Analysis 2-75
2.327 Interpretation of Results and Diagram Limitations. . . 2-75
2.328 Refraction of Ocean Waves 2-78
2.4 WAVE DIFFRACTION 2-79

2.42 DIFFRACTION CALCULATIONS 2-81
2.421 Waves Passing a Single Breakwater 2-81
2.422 Waves Passing a Gap of Width Less than Five
Wavelengths at Normal Incidence 2-98
2.423 Waves Passing a Gap of Width Greater than Five
Wavelengths at Normal Incidence 2-98
2.424 Diffraction at a Gap-Oblique Incidence 2-98
2.43 REFRACTION AND DIFFRACTION COMBINED 2-98
VI
SECTION PAGE
2.5 WAVE REFLECTION 2-110

2.51 GENERAL 2-110
2.52 REFLECTION FROM IMPERMEABLE, VERTICAL WALLS
(Linear Theory) 2-113
2.53 REFLECTIONS IN AN ENCLOSED BASIN 2-115
2.54 WAVE REFLECTION FROM BEACHES 2-117
2.6 BREAKING WAVES 2-120

2.61 DEEP WATER 2-120
2.62 SHOALING WATER 2-121
CHAPTER 3 - WAVE AND WATER LEVEL PREDICTIONS
3.2 CHARACTERISTICS OF OCEAN WAVES 3-2

3.21 SIGNIFICANT WAVE HEIGHT AND PERIOD 3-2
3.22 WAVE HEIGHT VARIABILITY 3-5
3.23 ENERGY SPECTRA OF WAVES 3-11
3.24 DIRECTIONAL SPECTRA OF WAVES 3-13
3.3 WAVE FIELD 3-15

3.31 DEVELOPMENT OF A WAVE FIELD 3-15
3.32 VERIFICATION OF WAVE HINDCASTING 3-17
3.33 DECAY OF A WAVE FIELD 3-17
3.4 WIND INFORMATION NEEDED FOR WAVE PREDICTION 3-20

3.41 ESTIMATING THE WIND CHARACTERISTICS 3-22
3.42 DELINEATING A FETCH 3-27
3.43 FORECASTS FOR LAKES, BAYS, AND ESTUARIES 3-29
3.431 Wind Data 3-29
3.432 Effective Fetch 3-29
3.5 SIMPLIFIED WAVE-PREDICTION MODELS 3-33

3.51 SMB METHOD FOR PREDICTING WAVES IN DEEP WATER 3-35
3.52 EFFECTS OF MOVING STORMS AND A VARIABLE WIND SPEED
AND DIRECTION 3-40
3.53 VERIFICATION OF SIMPLIFIED WAVE HINDCAST PROCEDURES. . . 3-40
3.54 ESTIMATING WAVE DECAY IN DEEP WATER 3-42
3.6 WAVE FORECASTING FOR SHALLOW WATER 3-42

3.61 FORECASTING CURVES 3-42
3.62 DECAY IN LAKES, BAYS, AND ESTUARIES 3-52
3.7 HURRICANE WAVES 3-52

3.71 DESCRIPTION OF HURRICANE WAVES 3-52
3.72 MODEL WIND AND PRESSURE FIELDS FOR HURRICANES 3-54
3.73 PREDICTION TECHNIQUE 3-57
vJi
SECTION PAGE
3.8 WATER LEVEL FLUCTUATIONS 3-69

3.81 ASTRONOMICAL TIDES 3-71
3.82 TSUNAMIS 3-71
3.83 LAKE LEVELS 3-76
3.84 SEICHES 3-78
3.85 WAVE SETUP 3-80
3.86 STORM SURGE AND WIND SETUP 3-82
3.861 General 3-82
3.862 Storms 3-83
3.863 Factors of Storm Surge Generation 3-84
3.864 Initial Water Level 3-85
3.865 Storm Surge Prediction 3-85
CHAPTER 4 - LITTORAL PROCESSES

4.11 DEFINITIONS 4-1
4.111 Beach Profile 4-1
4.112 Areal View 4-1
4.12 ENVIRONMENTAL FACTORS 4-1
4.121 Waves 4-1
4.122 Currents 4-4
4.123 Tides and Surges 4-5
4.124 Winds 4-5
4.125 Geologic Factors 4-6
4.126 Other Factors 4-6
4.13 CHANGES IN THE LITTORAL ZONE 4-6
4.2 LITTORAL MATERIALS 4-11

4.21 CLASSIFICATION 4-11
4.211 Size and Size Parameters 4-11
4.212 Composition 4-18
4.213 Other Characterisitcs 4-18
4.22 SAND AND GRAVEL 4-18
4.23 COHESIVE MATERIALS 4-20
4.24 CONSOLIDATED MATERIAL 4-21
4.241 Rock 4-21
4.242 Beach Rock 4-21
4.243 Organic Reefs 4-21
4.25 OCCURRENCE OF LITTORAL MATERIALS ON U.S. COASTS 4-22
4.251 Atlantic Coast 4-22
4.252 Gulf Coast 4-22
4.253 Pacific Coast 4-24
4.254 Alaska 4-24
4.255 Hawaii 4-24
4.256 Great Lakes 4-24
4.26 SAMPLING LITTORAL MATERIALS 4-24
VIII
SECTION PAGE
SIZE ANALYSES 4-26

Sieve Analysis 4-26
Settling Tube 4-26
LITTORAL WAVE CONDITIONS 4-27

EFFECT OF WAVE CONDITIONS ON SEDIMENT TRANSPORT 4-27
FACTORS DETERMINING LITTORAL WAVE CLIMATE 4-28
Offshore Wave Climate 4-28
Effect of Bottom Topography 4-28
Winds and Storms 4-29
INSHORE WA'/E CLIMATE 4-29
Mean Value Data on U.S. Littoral Wave Climates .... 4-29
Mean vs. Extreme Conditions 4-30
OFFICE STUDY OF WAVE CLIMATE 4-35
EFFECT OF EXTREME EVENTS 4-37
NEARSHORE CURRENTS 4-39

WAVE- INDUCED WATER MOTION ,
4-39
FLUID MOTION IN BREAKING WAVES 4-42
ONSHORE-OFFSHORE CURRENTS 4-43
Onshore-Offshore Exchange 4-43
Diffuse Return Flow 4-45
Rip Currents 4-45
LONGSHORE CURRENTS 4-45
Velocity and Flow Rate 4-45
Velocity Prediction 4-48
SUMMARY 4-50
LITTORAL TRANSPORT 4-50

INTRODUCTION 4-50
Importance of Littoral Transport 4-50
Zones of Transport 4-53
Profiles 4-54
Profile Accuracy 4-57
ONSHORE- OFFSHORE TRANSPORT 4-60
Sediment Effects 4-60
Initiation of Sediment Motion 4-61
Seaward Limit of Significant Transport 4-65
Beach Erosion and Recovery 4-70
Bar-Berm Prediction 4-78
Slope of the Foreshore 4-85
LONGSHORE TRANSPORT RATE 4-88
Definitions and Methods 4-88
Energy Flux Method 4-89
Energy Flux Example (Method 3) 4-102
Empirical Prediction of Gross Longshore Transport
Rate (Method 4) 4-107
4.535 Method 4 Example 4-108
ix
SECTION PAGE
4.6 ROLE OF FOREDUNES IN SHORE PROCESSES 4-111

4.61 BACKGROUND 4-111
4.62 ROLE OF FOREDUNES 4-113
4.621 Prevention of Overtopping 4-113
4.622 Reservoir of Beach Sand 4-115
4.623 Long-Term Effects 4-115
4.7 SEDIMENT BUDGET 4-116

4.71 DEFINITIONS 4-116
4.711 Sediment Budget 4-116
4.712 Elements of Sediment Budget 4-117
4.713 Sediment Budget Boundaries 4-119
4.72 SOURCE OF LITTORAL MATERIAL 4-119
4.721 Rivers 4-119
4.722 Erosion of Shores and Cliffs 4-121
4.723 Transport from Offshore Slope 4-121
4.724 Windblown Sediment Sources 4-123
4.725 Carbonate Production 4-123
4.726 Beach Replenishment 4-123
4.73 SINKS FOR LITTORAL MATERIALS 4-124
4.731 Inlets and Lagoons 4-124
4.732 Overwash 4-124
4.733 Backshore and Dune Storage 4-124
4.734 Offshore Slopes 4-127
4.735 Submarine Canyons 4-127
4.736 Deflation 4-129
4.737 Carbonate Loss 4-129
4.738 Mining and Dredging 4-129
4.74 CONVECTION OF LITTORAL MATERIALS 4-131
4.75 RELATIVE CHANGE IN SEA LEVEL 4-131
4.76 SUMMARY OF SEDIMENT BUDGET 4-131
4.8 ENGINEERING STUDY OF LITTORAL PROCESSES 4-139

4.81 OFFICE STUDY 4-139
4.811 Sources of Data 4-139
4.812 Interpretation of Shoreline Position 4-140
4.82 FIELD STUDY 4-147
4.821 Wave Data Collection 4-147
4.822 Sediment Sampling 4-147
4.823 Surveys 4-147
4.824 Tracers 4-150
4.83 SEDIMENT TRANSPORT CALCULATIONS 4-152
4.831 Longshore Transport Rate 4-152
4.832 Onshore-Offshore Motion 4-153
4.833 Sediment Budget 4-154

LIST OF FIGURES
FIGURE PAGE
1-1 Beach Profile - Related Terms 1-3

1-2 Sand Dunes Along the South Shore of Lake Michigan 1-4
1-3 Sand Dunes, Honeyman State Park, Oregon 1-4
1-4 Barrier Beach Island Developed as Recreation Park - Jones
Beach State Park, Long Island, New York 1-6
},-5 Large Waves Breaking Over a Breakwater 1-8
1-6 Wave Characteristics 1-8
1-7 Schematic Diagram of Storm Wave Attack on Beach and Dune . 1-11
1-8 Backshore Damage at Sea Isle City, New Jersey 1-15
1-9 Weir Jetty at Masonboro Inlet, North Carolina 1-19
1-10 Wrightsville Beach, North Carolina, after Completion of
Beach Restoration and Hurricane Protection Project . , . 1-20
2-1 Approximate Distribution of Ocean Surface Wave Energy

Illustrating the Classification of Surface Waves by
Wave Band, Primary Disturbing Force and Primary
Restoring Force 2-5
2-2 Definition of Terms - Elementary, Sinusoidal, Progressive
Wave 2-8
2-3 Local Fluid Velocities and Accelerations 2-14
2-4 Water Particle Displacements from Mean Position for
Shallow- Water and Deepwater Waves 2-17
2-5 Formation of Wave Groups by the Addition of Two Sinusoids
Having Different Periods 2-26
2-6 Summary - Linear (Airy) Wave Theory - Wave
Characteristics 2-34
2-7 Regions of Validity for Various Wave Theories 2-35
2-8 Comparison of Second-Order Stokes' Profile with Linear
Profile 2-42
2-9 Cnoidal Wave Surface Profiles as a Function of k^ 2-50
2-10 Cnoidal Wave Surface Profiles as a Func tion of k^ 2-51
2-11 Relationship Between k^, H/d and T/g/d 2-52
2-12 Relationship Between k^ and L^H/d^ 2-53
2-13 Relationships Between k^ and L^H/d^ and Between (y^-d)/H,
(y^ -d)/H)+ 1 and L^H/d^ 2-54
2-14 Relationship Between T/g/d y^/d, H/y^ and L^H/d^ 2-55
2-15 Relationship Between C/>/gy^, H/y^ and L^H/d^ 2-56
2-16 Functions M and N in Solitary Wave Theory 2-61
2-17 Wave Refraction at West Hampton Beach, Long Island,
New York 2-64
2-18 Refraction Template 2-68
2-19 Changes in Wave Direction and Height Due to Refraction on
Slopes with Straight, Parallel Depth Contours 2-70
2-20 Use of the Refraction Template 2-72
2-21 Refraction Diagram Using R/J Method 2-74
2-22 Use of Fan-Type Refraction Diagram 2-76
2-23 Refraction Along a Straight Beach with Parallel Bottom
Contours 2-77
XI
FIGURE PAGE
2-24 Refraction by a Submarine Ridge and Submarine Canyon . . . 2-77

2-25 Refraction Along an Irregular Shoreline 2-77
2-26 Wave Incident on a Breakwater 2-80
2-27 Wave Diffraction at Channel Islands Harbor Breakwater,
California 2-82
2-28 Wave Diffraction Diagram - 15° Wave Angle 2-83
2-40 Diffraction for a Single Breakwater Normal Incidence . . . 2-95
2-41 Schematic Representation of Wave Diffraction Overlay . . . 2-97
2-42 Generalized Diffraction Diagram for a Breakwater Gap
Width of Two Wave Lengths (B/L = 2) 2-99
2-43 Contours of Equal Diffraction Coefficient; Gap Width =
0.5 Wave Length (B/L =0.5) 2-100
1 Wave Length (B/L = 1) 2-100
1.41 Wave Lengths (B/L = 1.41) 2-101
1.64 Wave Lengths (B/L =1.64) 2-101
1.78 Wave Lengths (B/L =1.78) 2-102
2 Wave Lengths (B/L =2) 2-102
2.50 Wave Lenghts (B/L =2.50) 2-103
2.95 Wave Lengths (B/L = 2,95) 2-103
3.82 Wave Lengths (B/L = 3.82) 2-104
5 Wave Lengths (B/L =5) 2-104
2-53 Diffraction for a Breakwater Gap of Width > 5L (B/L > 5) . 2-105
2-54 Wave Incidence Oblique to Breakwater Gap 2-105
2-55 Diffraction for a Breakwater Gap of One Wave Length Width
((J)
= and 15°) 2-106
((fi = 30 and 45°) 2-107
= 60 and 75°)
((|) 2-108
Xtl
FIGURE PAGE
2-58 Diffraction Diagram for a Gap of Two Wave Lengths and a

45° Approach Compared with that for a Gap Width /2 Wave
Lengths with a 90° Approach 2-109
2-59 Single Breakwater - Refraction - Diffraction Combined. . . 2-111
2-60 Wave Reflection at Hamlin Beach, New York 2-112
2-61 Standing Wave (Clapotis) System - Perfect Reflection from
a Vertical Barrier - Linear Theory 2-114
2-62 (H^/L^) max vs Beach Slope 2-119
2-63 X2 vs Beach Slope for Various Values H^/L^ 2-119
2-64 Wave of Limiting Steepness in Deep Water 2-121
2-65 Breaker Height Index vs Deep Water Wave Steepness 2-122
2-66 Dimensionless Depth at Breaking vs Breaker Steepness . . . 2-123
2-67 Spilling Breaking Wave 2-125
2-68 Plunging Breaking Wave 2-125
2-69 Surging Breaking Wave 2-126
2-70 Collapsing Breaking Wave 2-126
3-1 Sample Wave Records 3-3

3-2 Waves in a Coastal Region 3-4
3-3 Theoretical and Observed Wave-Height Distributions .... 3-7
3-4 Theoretical and Observed Wave-Height Distributions .... 3-8
3-5 Theoretical Wave-Height Distributions 3-9
3-6 Typical Wave Spectra from the Atlantic Coast 3-14
3-7 Observed and Hindcasted Significant Wave Heights vs Time . 3-18
3-8 Map of North Atlantic Grid Points, Ocean Weather Ship
(OWS) Stations and Argus Island 3-19
3-9 Surface Synoptic Chart for 0030Z, 27 October 1950 3-23
3-10 Sample Plotted Report 3-25
3-11 Geostrophic Wind Scale 3-26
3-12 Possible Fetch Limitations 3-28
3-13 Relation of Effective Fetch to Width-Length Ratio for
Rectangular Fetches 3-31
3-14 Computation of Effective Fetch for Irregular Shoreline . . 3-32
3-15 Deepwater Wave Forecasting Curves (for Fetches of 1 to
1000 miles) 3-36
3-16 Deepwater Wave Forecasting Curves (for Fetches of 100
to More than 1000 miles) 3-37
3-17 Location of Wave Hindcasting Stations and Summary of
Synoptic Meterological Observations (SSMO) Areas .... 3-41
3-18 A Comparison of Shipboard Observations and Hindcasts . . . 3-43
3-19 Decay Curves 3-44
3-20 Travel Time of Swell Based on t^, = D/C^ 3-45
3-21 Forecasting Curves for Shallow-Water Waves; Constant
Depth = 5 Feet 3-47
Depth = 10 Feet 3-47
XIII
FIGURE PAGE
3-24 Forecasting Curves for Shallow- Water Waves; Constant

3-31 Typical Hurricane Wave Spectra 3-53
3-32 Composite Wave Charts 3-55
3-33 Pressure and Wind Distribution in Model Hurricane 3-57
3-34 Isolines of Relative Significant Wave Height for Slow
Moving Hurricane 3-59
3-35 Relationship for Friction Loss Over a Bottom of
Constant Depth 3-64
3-36 Typical Tide Curves Along Atlantic and Gulf Coasts .... 3-72
3-37 Typical Tide Curves Along Pacific Coasts of the United
States 3-73
3-38 Sample Tsunami Records from Tide Gages 3-75
3-39 Typical Water Level Variations in Lake Erie 3-77
3-40 Long-Wave Surface Profiles 3-79
3-41 Storm Surge and Observed Tide Chart 3-86
3-42 High Water Mark Chart for Texas, Hurricane Carla,
7-12 September 1961 3-88
3-43 Notation and Reference Frame 3-93
3-44 Storm Surge Chart 3-98
3-45 Schematic of Forces and Responses for Bathystrophic
Approximation 3-102
3-46 Various Setup Components Over the Continental Shelf. . . . 3-107
3-47 Track for Hurricane Camille, August 1969 3-110
3-48 Foot Surface Isovels (knots), Hurricane Camille,
August 1969 3-111
3-49 Seabed Profile Used for Hurricane Camille, August 1969 . . 3-114
3-50 Open Coast Surge Hydrograph, Hurricane Camille,
August 1969 3-118
3-51 Preliminary Estimate of Peak Surge 3-119
3-52 Shoaling Factors on Gulf Coast 3-121
3-53 Shoaling Factors on East Coast 3-122
3-54 Correction Factor for Storm Motion 3-123
3-55 Comparison of Observed and Computed Peak Surges (for 43
Storms with a Landfall South of New England from
1893 - 1957) 3-124
3-56 Surge Profile Along Coast Hurricane Camille,
August 1969 3-127
XIV
FIGURE PAGE
3-57 Lake Surface Contours on Lake Okeechobee, Florida

Hurricane, 26-27 August 1949 3-129
3-58 Grid System 3-133
3-59 Lake Erie 3-135
3-60 Cross-Section Area and Surface Width - Lake Erie 3-136
3-61 Mean Bottom Profile of Lake Erie 3-137
3-62 Wind Speed and Direction for Lake Erie - Storm, March 1955. 3-158
3-63 Wind Setup Hydrograph for Buffalo and Toledo - Storm,
March 1955 3-141
3-64 Wind Setup Profile for Lake Erie - Storm, March 1955. . . . 3-142
4-1 Typical Profile Changes with Time, Westhampton Beach, N.Y.. 4-2
4-2 Three Types of Shoreline 4-3
4-3 Shoreline Erosion near Shipbottom, N.J 4-7
4-4 Shoreline Accretion and Erosion Near Beach Haven, N.J.. . . 4-8
4-5 Stable Shoreline Near Peahala, N.J 4-9
4-6 Fluctuations in Location of Mean Sea Level Shoreline on
Seven East Coast Beaches 4-12
4-7 Grain Size Scales 4-14
4-8 Example Size Distribution 4-17
4-9 Sand Size Distribution Along the U.S. Atlantic Coast. . . . 4-23
4-10 Mean Monthly Nearshore Wave Heights for Five Coastal
Segments 4-31
4-11 Mean Monthly Nearshore Wave Period for Five Coastal
Segments 4-32
4-12 Distribution of Significant Wave Heights from Coastal
Wave Gages for 1-year Records 4-34
4-13 Nearshore Current System Near LaJolla Canyon, California. . 4-44
4-14 Typical Rip Currents, Ludlam Island, N.J 4-46
4-15 Distribution of Longshore Current Velocities 4-47
4-16 Measured versus Predicted Longshore Current Speed 4-49
4-17 Coasts in the Vicinity of New York Bight 4-51
4-18 Three Scales of Profiles, Westhampton, Long Island 4-55
4-19 Unit Volume Change versus Time Between Surveys for Profiles
on South Shore of Long Island 4-59
4-20 Maximum Wave Induced Bottom Velocity as a Function of
Relative Depth 4-62
4-21 Maximum Bottom Velocity from Small Amplitude Theory .... 4-63
4-22 Initiation of Ripple Motion 4-66
4-23 Wave Conditions Producing Maximum Bottom Velocity of
0.5 ft/sec 4-67
4-24 Nearshore Bathymetry with Shore-Parallel Contours off
Panama City, Florida 4-68
4-25 Nearshore Bathymetry with Shore-Parallel Contours and
Linear Bars off Manasquan, N.J 4-69
4-26 Slow Accretion of Ridge-and-Runnel at Crane Beach, Mass.. . 4-76
4-27 Rapid Accretion of Ridge-and-Runnel - Lake Michigan .... 4-77
4-28 Typical Berm and Bar Profiles from Prototype Size
Laboratory Wave Tank 4-79
XV
FIGURE PAGE
4-29 Berra -Bar Criterion Based on Dimensionless Fall Time and

Deep Water Steepness 4-82
4-30 Berm - Bar Criterion Based on Dimensionless Fall Time and
Height to Grain Size Ratio 4-83
4-31 Fall Velocity of Quartz Spheres in Water as a Function of
Diameter and Temperature 4-84
4-32 Data Trends - Median Grain Size versus Foreshore Slope. . . 4-86
4-33 Data - Median Grain Size versus Foreshore Slope 4-87
4-34 Longshore Component of Wave Energy Flux in Dimensionless
Form as a Function of Breaker Conditions 4-92
4-35 Longshore Component of Wave Energy Flux as a Function of
Deepwater Wave Conditions 4-94
4-36 Transport Rate versus Energy Flux Factor for Field and
Laboratory Conditions 4-99
4-37 Relationship Between Wave Energy and Longshore Transport. . 4-100
4-38 Longshore Transport Rate as a Function of Breaker Height
and Breaker Angle 4-103
4-39 Longshore Transport Rate as a Function of Deepwater Wave
Height and Deepwater Angle 4-104
4-40 Upper Limit on Longshore Transport Rates 4-109
4-41 Typical Barrier Island Profile Shape 4-112
4-42 Event Frequency per 100 Years the Stated Level is Equalled
or Exceeded on the Open Coast, South Padre Island, Texas. 4-114
4-43 Basic Example of Sediment Budget 4-120
4-44 Erosion Within Littoral Zone During Uniform Retreat of an
Idealized Profile 4-122
4-45 Sediment Trapped Inside Old Drum Inlet, N.C 4-125
4-46 Overwash on Portsmouth Island, N.C 4-126
4-47 Growth of a Spit in to Deep Water, Sandy Hook, N.J 4-128
4-48 Dunes Migrating Inland Near Laguna Point, California. . , . 4-130
4-49 Materials Budget for Littoral Zone 4-132
4-50 Summary of Example Problem Conditions and Results 4-135
4-51 Variation of y with Distance Along Spit 4-136
4-52 Growth of Sandy Hook, N.J., 1835-1932 4-141
4-53 Transport Directions at New Buffalo Harbor Jetty on
Lake Michigan 4-142
4-54 Sand Accumulation at Point Mugu, California 4-143
4-55 Tombolo and Pocket Beach at Greyhound Rock, California. . . 4-144
4-56 A Nodal Zone of Divergence Illustrated by Sand Accumulation
at Groins, South Shore Staten Island, N.Y 4-145
4-57 South Shore of Long Island, N.Y., Showing Closed,
Partially Closed and Open Inlets 4-146
4-58 Four Types of Barrier Islands Offset 4-148
4-59 Fire Island Inlet, New York - Overlapping Offset 4-149
4-60 Old Drum Inlet, North Carolina - Negligible Offset 4-149
XVI
LIST OF TABLES
TABLE PAGE
1-1 Shoreline Characteristics 1-3
2-1 Distribution of Wave Heights in a Short Train of Waves. . . 2-32

2-2 Example Computations of Values of C,/C„ for Refraction
Analysis 2-71
3-1 Correction for Sea-Air Temperature 3-27

3-2 Wind- Speed Adjustment, Nearshore 3-29
3-3 Values of Kg or (H/H^) 3-63
3-4 Computations for Wind Waves Over the Continental Shelf. . . 3-66
3-5 Tidal Ranges 3-74
3-6 Fluctuations in Water Levels - Great Lakes System
(1860 through 1970) 3-76
3-7 Short-Period Fluctuations in Lake Levels at Selected
Gage Sites 3-78
3-8 Highest and Lowest Water Levels 3-89
3-9 Systems of Units for Storm Surge Computations 3-108
3-10 Manual Surge Computations 3-116
4-1 Seasonal Profile Changes on Southern California Beaches . . 4-10

4-2 Density of Littoral Materials 4-19
4-3 Minerals Occurring in Beach Sand 4-20
4-4 Mean Wave Height at Coastal Localities of Conterminous
United States 4-33
4-5 Storm- Induced Beach Changes 4-72
4-6 Longshore Transport Rates from U.S. Coasts 4-90
4-7 Longshore Energy Flux ?i, for a Single Periodic Wave in
Any Specified Depth 4-97
4-8 Approximate Formulas for Computing Longshore Energy Flux
Factor, Pj,g, Entering the Surf Zone 4-97
4-9 Assumptions for P^g Formulas in Table 4-8 4-98
4-10 Deepwater Wave Heights, in Percent by Direction, off East
Facing Coast of Inland Sea 4-105
4-11 Computed Longshore Transport for East - Facing Coast of
Inland Sea 4-106
4-12 Estimate of Gross Longshore Transport Rate for Shore of
Inland Sea 4-110
4-13 Classification of Elements in the Littoral Zone Sediment
Budget 4-118
4-14 Sand Budget of the Littoral Zone 4-133
XVII
CHAPTER 1
INTRODUCTION
TO
COASTAL ENGINEERING
ÂAÂt<c~
HALEIWA BEACH, OAHU, HAWAII - 23 August 1970
CHAPTER 1
INTRODUCTION TO COASTAL ENGINEERING
1.1 INTRODUCTION TO THE SHORE PROTECTION MANUAL
This Shore Protection Manual has been prepared to assemble in a

single three-volume publication coastal-engineering practices for shore
protection. "Coastal Engineering" is defined as the application of the
physical and engineering sciences to the planning, design, and construc-
tion of works to modify or control the interaction of the air, sea, and
land in the coastal zone for the benefit of man and for the enhancement
of natural shoreline resources. "Shore protection," as used in this
Manual, applies to works designed to stabilize the shores of large bodies
of water where wave action is the principal cause of erosion. Much of
the material is applicable to the protection of navigation channels and
harbors.
The nature and degree of required shore-protection measures vary

widely at different localities. Proper solution of any specific problem
requires systematic and thorough study. The first requisite for such
study is a clear definition of the problem and the objectives sought.
The first factor to be determined is the cause of the problem. Ordinarily
there will be more than one method of obtaining the immediate objective.
Therefore, the long-term effects of each method should be studied. The
immediate and long-term effects of each method should be evaluated not
only within the problem area, but also in adjacent shore areas. All
physical and environmental effects, advantageous and detrimental, should
be considered in comparing annual costs and benefits to determine the
justification of protection methods.
Detailed summaries of applicable methods, techniques, and useful data

pertinent to the solution of shore protection problems have been included
in this Manual.
By replacing Shore Proteatioriy Planning and Design with the Shore

Protection Manual^ CERC is providing coastal engineers with an improved
tool for solving shore-protection problems. The Manual is designed as
an advanced text, but contains sufficient introductory material to allow
a person with an engineering background to obtain an understanding of
coastal phenomena and to solve related engineering problems.
Chapter 1 presents a basic introduction to the subject. Chapter 2,

"Mechanics of Wave Motion," treats wave theories, wave refraction and
diffraction, wave reflection, and breaking waves. Chapter 3, "Wave and
Water Level Predictions," discusses wave forecasting, hurricane waves,
storm surge, and water level fluctuations. Chapter 4, "Littoral Proc-
esses, y treats the characteristics and sources of littoral material
nearshore currents, littoral transport, and sand budget techniques.
Chapter 5, "Planning Analyses," treats the functional planning of shore-
protection measures. Chapter 6, "Structural Features," illustrates the
l-l
functional design of various structures. Chapter 7, "Structural
—
Design Physical Factors," treats tke effects of environmental forces
on the design of protective works. Chapter 8, "Engineering .Analysis
Case Study," presents a series of calculations for the preliminary
design of an offshore island facility in the mouth of the Delaware
Bay.
Each chapter contains its own bibliography. This Manual concludes

with four appendixes. Because the meanings of coastal engineering terms
differ from place to place, the reader is urged to use Appendix A,
Glossary of Terms, that defines the terms used in this Manual. Appen-
dix B lists the symbols used. Appendix C is a collection of miscella-
neous tables and plates that supplement the material in the chapters.
Appendix D is the subject index.
1,2 THE SHORE ZONE
Table 1-1 summarizes regional shoreline characteristics. The infor-

mation obtained from the "Report on the National Shoreline Study," by the
Department of the Army, Corps of Engineers (1971), indicates that of the
total 84,240 miles of U.S. shoreline, there are 34,520 miles (41 percent)
of exposed shoreline and 49,720 miles (59 percent) of sheltered shoreline
(i.e., in bays, estuaries and lagoons). About 20,500 miles of the shore-
line (or 24 percent of the total) are eroding. Of the total length of
shoreline, exclusive of Alaska (36,940 miles), about 12,150 miles (33
percent) have beaches; the remaining 24,790 miles have no beach.
1.21 NATURAL BEACH PROTECTION
Where the land meets the ocean at a sandy beach, the shore has
natural defenses against attack by waves, currents and storms. First
of these defenses is the sloping nearshore bottom that causes waves to
break offshore, dissipating their energy over the surf zone. The pro-
cess of breaking often creates an offshore bar in front of the beach
that helps to trip following waves. The broken waves re-form to break
again, and may do this several times before finally rushing up the beach
foreshore. At the top of wave uprush a ridge of sand is formed. Beyond
this ridge, or crest of the berm, lies the flat beach berm that is
reached only by higher storm waves. A beach profile and its related
terminology are shown in Figure 1-1.
1.22 NATURAL PROTECTIVE DUNES
Winds blowing inland over the foreshore and berm move sand behind the
beach to form dunes. (See Figures 1-2 and 1-3.) Grass, and sometimes
bushes and trees, grow on the dunes, and the dunes become a natural levee
against sea attack. Dunes are the final natural protection line against
wave attack, and are also a reservoir for storage of sand against storm
waves
1-2
Figure 1-2. Sand Dunes along the South Shore of Lake Michigan
1.23 BARRIER BEACHES, LAGOONS AND INLETS
In some areas, an additional natural protection for the mainland is

provided in the form of barrier beaches. (See Figure 1-4.) Nearly all
of the U.S. east coast from Long Island to Mexico is comprised of bar-
rier beaches. These are long narrow islands or spits lying parallel to
the shoreline. Barrier beaches generally enclose shallow lagoons that
separate the mainland from the ocean. During severe storms these barrier
beaches absorb the brunt of the wave attack. When barrier-beach dunes
are breached, the result may be the cutting of an inlet. The inlet per-
mits sand to enter the lagoon and settle to the bottom, removing sand
from the beach.
1.24 STORM ATTACK
Diiring storms, strong winds generate high waves. Storm surge and
waves may raise the water level near the shore. If storm surge does
occur, large waves can then pass over the offshore bar formation without
breaking. If the storm occurs at high tide, storm surge super-elevates
the water, and some waves may break on the beach or even at the base of
the dunes. After a storm or storm season, natural defenses may again be
re-formed by normal wave and wind action.
lo25 ORIGIN AND MOVEMENT OF BEACH SANDS
Most of the sands of the beaches and nearshore slopes are normally
small, resistant rock particles that have traveled many miles from in-
land mountains. When the sand reaches the shore, it is moved alongshore
by waves and littoral currents. This alongshore transport is a constant
process, and great volumes may be transported. In most coastal segments
the direction of movement changes as direction of wave attack changes.
1.3 THE SEA IN MOTION
1.31 TIDES AND WINDS
The motions of the sea originate in the gravitational effects of the

sun, the moon, and earth; and from air movements or winds caused by dif-
ferential heating of the earth.
The moon, and to a lesser extent the sun, creates ocean tides by
gravitational forces. These forces of attraction, and the fact that the
sun, moon, and earth are always in motion with relation to each other,
cause waters of ocean basins to be set in motion. These tidal motions of
water masses are a form of very long period wave motion, resulting in a
rise and fall of the water surface at a point. There are normally two
tides per day, but some localities have only one per day.
1.32 WAVES
The familiar waves of the ocean 'are wind waves generated by winds
blowing over water. They may vary in size from ripples on a pond to
1-5
I
-6
large ocean waves as high as 100 feet. (See Figure 1-5.) Wind waves
cause most of the damage to the ocean coasts. Another type of wave,
the tsunami, is created by earthquakes or other tectonic disturbances
on the ocean bottom. Tsunamis have caused spectacular damage at times,
but fortunately, major tsunamis do not occur frequently.
Wind waves are of the type known as oscillatory waves, and are
usually defined by their height, length, and period. (See Figure 1-6.)
Wave height is the vertical distance from the top of the crest to the
bottom of the trough. Wavelength is the horizontal distance between
successive crests„ Wave period is the time between successive crests
passing a given point.
As waves propagate in water, only the form and part of the energy
of the waves move forward; the water particles remain.
The height, length, and period of wind waves are determined by the
fetch (the distance the wind blows over the sea in generating the waves),
the wind speed, the length of time the wind blows, and the deoay distance
(the distance the wave travels after leaving the generating area). Gener-
ally, the longer the fetch, the stronger the wind; and the longer the
time the wind blows, the larger the waves. The water depth, if shallow
enough, will also affect the size of wave generated. The wind simul-
taneously generates waves of many heights, lengths, and periods as it
blows over the sea.
If winds of a local storm blow toward the shore, the generated

waves will reach the beach in nearly the form in which they are gener-
ated. Under these conditions, the waves are steep; that is, the wave-
length is 10 to 20 times the wave height. Such waves are called seas.
If waves are generated by a distant storm, they may travel through
hundreds or even thousands of miles of calm areas before reaching the
shore. Under these conditions, waves decay - short, steep waves are
eliminated, and only relatively long, low waves reach the shore. Such
waves have lengths from 30 to more than 500 times the wave height, and
are called swell.
1.33 CURRENTS AND SURGES
Currents are created in oceans and adjacent bays and lagoons when
water in one area becomes higher than water in another area. Water in
the higher area flows toward the lower area, creating a current. Some
causes of differences in the elevation of the water surface in the oceans
are tides, wind, waves breaking on a beach, and streams. Changes in water
temperature or salinity cause changes in water density that may also pro-
duce currents.
Wind creates currents because, as it blows over the water surface,

it creates a stress on surface water particles, and starts these parti-
cles moving in the direction in which the wind is blowing. Thus, a sur-
face current is created. When such a current reaches a barrier, such as
coastline, water tends to pile up against the land. In this way, wind
1-7
li^Mirc l-fi. Large Wuve:; lii'caking ovci- u lireakwater
Oifi'clion of Wave Irovel
Wove Crest -^
Crest Lcngtti
Region
Stillwoler level
TrouQti Lcnqtii
Region d^ Oeptti
Oceon Bottom-
Figure 1-0. W.ivc Cliar.Kt c-ri sties
1-8
setup or storm surges are created by strong winds. The height of stomi
surge depends on wind velocity and dii'ection, fetch, water depth, ajid
ncarshore slope. In violent storms, storm surge may raise sea level at
the shore as much as 20 feet. In the United States, larger surges occur
on the Gulf coast because of the shallower and broader shelf off that
coast compared to the shelves off the Atlantic and Pacific coasts. Storm
surges may also be increased by a funneling effect in converging estuaries.
When waves approach the beach at an angle, they create a current in

shallow water parallel to the shore, known as the lotuji^hovc c-'uvveyit. This
current, under certain conditions, may turn and run out to sea in what is
known as a rip aurvent.
1.34 TIDAL CURRENTS
If water level rises :uid falls at an area, then water must flow into
and out of the area. Significant currents generated by tides occur at in-
lets to lagoons and bays or at entrances to harbors. At such constricted
places, tidal currents generally flow in when the tide is rising (flood
tide) and flow out as the tide falls (ebb tide). Exceptions can occur
at times of high river discharge or strong winds, or when density cur-
rents are an important part of the current system.
In addition to creating currents, tides const juitly chaiige the level

at which waves attack the bcnclu
1.4 THE BEHAVIOR OF BEACHES
1.41 BEACH COMPOSITION
The size and character of sediments on a beacli ai"e related to forces

to which the beach is exposed and the type of material available at the
shore. Most beaches are composed of fine or coarse sand and. in some
areas, of small stones called shingle or gravel. This material is sup-
plied to the beach zone by streams, by erosion of the shores caused by
waves and currents and, in some cases, by onshore movement of material
from deeper water. Clay and silt do not usually remain on ocean beaches
because the waves create such turbulence in the water along the shore
that these fine materials ;ire kept in suspension. It is only after nnw-
ing away from the beaches into quieter or deeper water that these fine
particles settle out and deposit on the bottom.
1.42 BEACH CHARACTERISTICS
Characteristics of a beach are usually described in terms of average

size of the sand particles that make up the beach, range and distribution
of sizes of those particles, sand composition, elevation and width of berm,
slope or steepness of the foreshore, the existence (or lack) of a bar, and
the general slope of the inshore zone fronting the beach. Generally, the
larger the sand particles tlic steeper the beach slope. Beaches witli
gently sloping foreshores and inshore zones usually have a prepondor.ince
1-9
of the finer sizes of sand. Daytona Beach, Florida, is a good example
of a gently sloping beach composed of fine sand.
lo43 BREAKERS
As a wave moves toward shore, it reaches a depth of water so shallow

that the wave collapses or breaks. This depth is equal to about 1.3 times
the wave height. Thus a wave 3 feet high will break in a depth of about
4 feet. Breaking can occur in several different ways (plunging, spilling,
surging, or collapsing). Breaking results in a dissipation of the energy
of the wave and is manifested by turbulence in the water. This turbulence
stirs up the bottom materials. For most waves, the water travels forward
after breaking as a foaming, turbulent mass, expending most of its remain-
ing energy in a rush up the beach slope.
1.44 EFFECTS OF WIND WAVES
Wind waves affect beaches in two major ways. Short steep waves,
which usually occur during a storm near the coast, tend to tear the
beach down. (See Figure 1-7.) Long swells, which originate from dis-
tant storms, tend to rebuild the beaches. On most beaches, there is a
constant change caused by the tearing away of the beach by local storms
followed by gradual rebuilding by swells. A series of violent local
storms in a short time can result in severe erosion of the shore if
there is not enough time between storms for swells to rebuild the
beaches. Alternate erosion and accretion of beaches may be seasonal
on some beaches; the winter storms tear the beach away, and the summer
swells rebuild it. Beaches may also follow long-term cyclic patterns.
They may erode for several years, and then accrete for several years.
1.45 LITTORAL TRANSPORT
Littovat transport is defined as the movement of sediments in the

nearshore zone by waves and currents and is divided into two general
classes; transport parallel to the shore (longshore transport) and
transport perpendicular to the shore (on shore- offshore transport). This
transport is distinguished from the material moved, which is called
littoral drift.
Onshore -offshore transport is determined primarily by wave steepness,

sediment size, and beach slope. In general, high steep waves move material
offshore, and low waves of long period (low steepness waves) move material
onshore. This onshore-offshore process associated with storm waves is
illustrated in Figure 1-7.
Longshore transport results from the stirring up of sediment by the

breaking wave, and the movement of this sediment by the component of the
wave in an alongshore direction, and by the longshore current generated
by the breaking wave. The direction of longshore transport is directly
related to the direction of wave approach, and the angle of the wave to
the shore. Thus, due to the variability of wave approach, longshore
l-IO
•Dune Crest
Crest :.:. I I
- Storm wove ottock

Lowering Profile C
of foredune
••••••••
iô^
•"••::•'\•:'K•V•:^;?>:3J^^^ MLW
Crest
ACCRETION
Profile A
M.L.W.
Profile D - After storm wove ottock,
normol wove oction
ACCRETION
Profile A
Figure 1-7, Schematic Diagram of Storm Wave Attack on Beach and Dune
l-ll
transport direction can vary from season to season, day to day or hour
to hour. These reversals of transport direction are quite common for
most United States shores. Direction may vary at random, but in most
areas the net effect is seasonal.
The rate of longshore transport is dependent on both angle of wave

approach, and wave energy. Thus, high storm waves will generally move
more material per unit time than low waves. However, if low waves
exist for a much longer time than do high waves, the low waves may be
more significant in moving sand than the high waves.
Because reversals in transport direction occur, and because different

types of waves transport material at different rates, two components of
the longshore transport rate become important. The first is the net rate,
the net amount of material passing a particular point in the predominant
direction during an average year. The second component is the gross rate,
the total of all material moving past a given point in a year regardless
of direction. Most shores consistently have a net annual longshore trans-
port in one direction. Determining the direction and average net and
gross annual amount of longshore transport is important in developing
shore protection plans.
In landlocked water of limited extent, such as the Great Lakes, a

longshore transport rate in one direction can normally be expected to be
no more than about 150,000 cubic yards per year. For open ocean coasts,
the net rate of transport may vary from 100,000 to more than 2 million
cubic yards per year. The rate depends on the local shore conditions
and shore alignment as well as the energy and direction of wave action.
1.46 EFFECT OF INLETS ON BARRIER BEACHES
Inlets may have significant effects on adjacent shores by interrupt-

ing the longshore transport and trapping onshore- offshore moving sand.
On ebb current, sand moved to the inlet by waves is carried a short dis-
tance out to sea and deposited on an outer bar. When this bar becomes
large enough, the waves begin to break on it, and sand again begins to
move over the bar back toward the beach. On the flood tide, when water
flows through the inlet into the lagoon, sand in the inlet is carried a
short distance into the lagoon and deposited. This process creates shoals
in the landward end of the inlet known as middlegroimd shoals or inner
bars. Later, ebb flows may bring some of the material in these shoals
back to the ocean, but some is always lost from the stream of littoral
drift and thus from the downdrift beaches. In this way, tidal inlets
may store sand and reduce the supply of sand to adjacent shorelines.
1.47 IMPACT OF STOM/IS
Hurricanes or severe storms moving over the ocean near the shore may
greatly change beaches. Strong winds of a storm often create a storm surge.
This surge raises the water level and exposes to wave attack higher parts
of the beach not ordinarily vulnerable to waves. Such storms also generate
1-12
large, steep waves. These waves carry large quantities of sand from the
beach to the nearshore bottom. Land structures, inadequately protected
and located too close to the water, are then subjected to the forces of
waves and may be damaged or destroyed. Low- lying areas next to the ocean,
lagoons, and bays are often flooded by storm surge. Storm surges are
especially damaging if they occur concurrently with astronomical high tide.
Beach berms are built naturally by waves to about the highest eleva-
tion reached by normal storm waves. Berms tend to absorb the wave energy;
however, overtopping permits waves to reach the dunes or bluffs in back of
the beach and damage unprotected upland features.
When storm waves erode the berm and carry the sand offshore, the pro-
tective value of the berm is reduced and large waves can overtop the beach.
The width of the berm at the time of a storm is thus an important factor
in the amount of upland damage a storm can inflict.
Notwithstanding changes in the beach that result from storm-wave

attack, a gently sloping beach of adequate width and height is the most
effective method known for dissipating wave energy.
1.48 BEACH STABILITY
Although a beach may be temporarily eroded by storm waves and later

partly or wholly restored by swells, and erosion and accretion patterns
may occur seasonally, the long-range condition of the beach - whether
eroding, stable or accreting - depends on the rates of supply and loss
of littoral material. The shore accretes or progrades when the rate of
supply exceeds the rate of loss. The shore is considered stable (even
though subject to storm and seasonal changes) when the long-term rates
of supply and loss are equal.
1.5 EFFECTS OF MAN ON THE SHORE
1.51 ENCROACHMENT ON THE SEA
During the early days of the United States, natural beach processes
continued to mold the shore as in ages past. As the country developed,
activity in the shore area was confined principally to harbor areas.
Between harbor areas, development along the shore progressed slowly as
small, isolated, fishing villages. As the national economy grew, im-
provements in transportation brought more people to the beaches. Gradu-
ally, extensive housing, commercial, recreational and resort developments
replaced fishing villages as the predominant coastal manmade features.
Examples of this development are Atlantic City and Miami Beach.
Numerous factors control the growth of development at beach areas,

but undoubtedly the beach environment is the development's basic asset.
The desire of visitors, residents, and industries to find accommodations
as close to the ocean as possible has resulted in man's encroachment on
the sea.
1-13
There are places where the beach has been gradually widened, as well
as narrowed, by natural processes over the years. This is evidenced by
lighthouses and other structures that once stood on the beach, but now
stand hundreds of feet inland.
In their eagerness to be as close as possible to the water, developers

and property owners often forget that land comes and goes, and that land
which nature provides at one time may later be reclaimed by the sea. Yet
once the seaward limit of a development is established, this line must be
held if large investments are to be preserved. This type of encroachment
has resulted in great monetary losses due to storm damage, and in ever-
increasing costs of shore protection.
lo52 NATURAL PROTECTION
While the sloping beach and beach berm are the outer line of defense
to absorb most of the wave energy, dunes are the last zone of defense in
absorbing the energy of storm waves that overtop the berm. Although dunes
erode during severe storms, they are often substantial enough to afford
complete protection to the land behind them. Even when breached by waves
of a severe storm, dunes may gradually rebuild naturally to provide pro-
tection during future storms. Continuing encroachment on the sea with
manmade development has often taken place without proper regard for the
protection provided by dunes. Large dune areas have been leveled to make
way for real estate developments, or have been lowered to permit easy
access to the beach. Where there is inadequate dune or similar protection
against storm waves, the storm waters may wash over low- lying land, mov-
ing or destroying everything in their path, as illustrated by Figure 1-8.
1.53 SHORE PROTECTION METHODS
Where beaches and dunes protect shore developments, additional pro-

tective works may not be required. However, when natural forces do
create erosion, storm waves may overtop the beach and damage backshore
structures. Manmade structures -must then provide protection. In gene-
ral, measures designed to stabilize the shore fall into two classes:
structures to prevent waves from reaching erodible material (seawalls,
bulkheads, revetments); and an artificial supply of beach sand to make
up for a deficiency in sand supply through natural processes. Other
manmade structures, such as groins and jetties, are used to retard the
longshore transport of littoral drift. These may be used in conjunction
with seawalls or beachfills or both.
Separate protection for short reaches of eroding shores (as an indi-

vidual lot frontage) within a larger zone of eroding shore, is difficult
and costly. Such protection often fails at its flanks as the adjacent
unprotected shores continue to recede. Partial or inadequate protective
measures may even accelerate erosion of adjacent shores. Coordinated
action under a comprehensive plan that considers erosion processes over
the full length of the regional shore compartment is much more effective
and economical.
1-14
2
M
•H
0)
I—
(A
to
E
nj
Q
u
CO
I -15
1.54 BULKHEADS, SEAWALLS AND REVETMENTS
Protection on the upper part of the beach which fronts backshore

development is required as a partial substitute for the natural pro-
tection that is lost when the dunes are destroyed. Shorefront owners
have resorted to shore armoring by wave- resist ant walls of various
types. A vertical wall in this location is known as a bulkhead, and
serves as a secondary line of defense in major storms. Bulkheads are
constructed of steel, timber, or concrete piling. For ocean-exposed
locations, bulkheads do not provide a long-term solution, because a
more substantial wall is required as the beach continues to recede and
larger waves reach the structure. Unless combined with other types of
protection, the bulkhead must be enlarged into a massive seawall capable
of withstanding the direct onslaught of the waves. Seawalls may have
vertical, curved or stepped faces. While seawalls may protect the up-
land, they can create a local problem. Downward forces of water created
by waves striking the wall, can rapidly remove sand from in front of the
wall. A stone apron is often necessary to prevent excessive scouring
and undermining.
A revetment armors the slope face of a dune or bluff. It is usually

composed of one or more layers of stone or is of concrete construction.
This sloping protection dissipates wave energy with less damaging effect
on the beach than waves striking vertical walls.
1.55 BREAKWATERS
Beaches and bluffs or dunes can be protected by an offshore break-

water that reduces the wave energy reaching the shore. However, offshore
breakwaters are usually more costly than onshore structures, and are sel-
dom built solely for shore protection. Offshore breakwaters are construc-
ted mainly for navigation purposes. A breakwater protecting a harbor
area provides shelter for boats. Breakwaters have both beneficial and
detrimental effects on the shore. All breakwaters reduce or eliminate
wave action and thus protect the shore immediately behind them. Whether
offshore or shore-connected, the elimination of wave action reduces long-
shore transport, obstructing the movement of sand along the shore and
starving the downdrift beaches.
At a harbor breakwater, the sand stream generally can be restored by

pumping sand through a pipeline from the side where sand accumulates to
the eroded downdrift side. This type of operation has been in use for
many years at Santa Barbara, California.
Even without a shore arm, an offshore breakwater reduces wave action

and creates quiet water between it and the shore. In the absence of wave
action to move sand, it is deposited and builds the shore seaward toward
the breakwater. The buildup serves as a barrier which also blocks the
movement of littoral materials. If the offshore breakwater is placed
immediately updrift from a navigation opening, the structure impounds
sand, prevents it from entering the navigation channel, and affords
1-16
shelter for a floating dredge to pump the impounded material across the
navigation opening back onto the downdrift beach. This method is used
at Channel Island Harbor near Port Hueneme, California.
1.56 GROINS
The groin is a barrier-type structure that extends from the backshore

into the littoral zone. The basic purposes of a groin are to interrupt
longshore sand movement, to accumulate sand on the shore, or to retard
sand losses. Trapping of sand by a groin is done at the expense of the
adjacent downdrift shore unless the groin or groin system is artificially
filled with sand to its entrapment capacity from other sources. To reduce
the potential for damage to property downdrift of a groin, some limitation
must be imposed on the amount of sand permitted to be impounded on the
updrift side. Since more and more shores are being protected, and less
and less sand is available as natural supply, it is now desirable, and
frequently necessary, to place sand artificially to fill the area between
the groins, thereby ensuring a more or less uninterrupted passage of the
sand to the downdrift shores.
Groins have been constructed in various configurations using timber,

steel, concrete or rock. Groins can be classified as high or low, long
or short, permeable or impermeable, and fixed or adjustable.
A high groin, extending through the breaking zone for ordinary or

moderate storm waves, initially entraps nearly all of the longshore
moving sand within that intercepted area until the areal pattern or sur-
face profile of the accumulated sand mass allows sand to pass around the
seaward end of the groin to downdrift shores. Low groins (top profile no
higher than that of desired beach dimensions) function like high groins,
except that sand also passes over the top of the structure. Permeable
groins permit some of the wave energy and moving sand to pass through the
structure.
1.57 JETTIES
Jetties are generally employed at inlets in connection with naviga-

tion improvements. When sand being transported along the coast by waves
and currents arrives at an inlet, it flows inward on the flood tide to
form an inner bar, and outward on the ebb tide to form an outer bar.
Both formations are harmful to navigation through the inlet, and must be
controlled to maintain an adequate navigation channel. The jetty is sim-
ilar to the groin in that it traps sand moving along the beach. Jetties
are usually constructed of steel, concrete, or rock. The jetty type
depends on foundation conditions, wave climate, and economic considera-
tions. Jetties are much larger than groins, since jetties sometimes ex-
tend from the shoreline seaward to a depth equivalent to the channel
depth desired for navigation purposes. To be efficient in maintaining
the channel, the jetty must be high enough to completely obstruct sand
movement
1-17
Jetties aid navigation by reducing movement of sand into the channel,
by stabilizing the location of the channel, and by shielding vessels from
waves. Sand is impounded at the updrift jetty, and the supply of sand to
the shore downdrift from the inlet is reduced, thus causing erosion of
that shore. Before the installation of a jetty, nature supplied sand by
transporting it across the inlet intermittently along the outer bar to the
downdrift shore.
To eliminate undesirable downdrift erosion, some projects provide for

dredging the sand impounded by the updrift jetty and pumping it through a
pipeline (bypassing the inlet) to the eroding beach. This provides an
intermittent flow of sand to nourish the downdrift beach, and also prevents
shoaling of the entrance channel.
A more recent development for sand bypassing provides a low section

or weir in the updrift jetty over which sand moves into a sheltered pre-
dredged, deposition basin. By dredging the basin periodically, deposition
in the channel is reduced or eliminated. The dredged material is normally
pumped across the inlet to provide nourishment for the downdrift shore.
A weir jetty of this type at Masonboro Inlet, North Carolina, is shown in
Figure 1-9.
1.58 BEACH RESTORATION AND NOURISHMENT
As previously stated, beaches are very effective in dissipating wave

energy. When maintained to adequate dimensions, they afford protection
for the adjoining backshore. Therefore, a protective beach is classed as
a shore-protection structure. When studying an erosion problem, it is gen-
erally advisable to investigate the feasibility of mechanically or hydrau-
lically placing borrow material on the shore to form and maintain an ade-
quate protective beach. The method of placing beach fill to ensure sand
supply at the required replenishment rate is important. Where stabiliza-
tion of an eroding beach is the problem, suitable beach material may be
stockpiled at the updrift sector of the problem area. The establishment
and periodic replenishment of such a stockpile is termed artificial
beaoh nourishment. To restore an eroded beach and stabilize it at the
restored position, fill is placed directly along the eroded sector, and
then the beach is artificially nourished by the stockpiling method.
When conditions are suitable for artificial nourishment, long reaches

of shore may be protected by this method at a relatively low cost per
linear foot of protected shore. An equally important advantage is that
artificial nourishment directly remedies the basic cause of most erosion
problems - a deficiency in natural sand supply - and benefits rather than
damages the adjacent shore. An added consideration is that the widened
beach has value as a recreation feature. A project for beach restoration
with an artificial dune for protection against hurricane wave action, com-
pleted in 1965 at Wrightsville Beach, North Carolina, is shown in Figure
l-lOo
1-18
Figure 1-9. Weir Jetty at Masonboro Inlet, North Carolina
1-19
I
-20
Sometimes structures must be provided to protect dunes, to maintain
a specific beach dimension, or to reduce nourishment requirements. In
each case, the cost of such structures must be weighed against the bene-
fits they would provide. Thus, measures to provide and keep a wider pro-
tective and recreational beach for a short section of an eroding shore
would require excessive nourishment without supplemental structures such
as groins to reduce the rate of loss of material from the widened beach.
A long, high terminal groin or jetty is frequently justified at the down-
drift end of a beach restoration project to reduce losses of fill into an
inlet and to stabilize the lip of the inlet.
1.6 CONSERVATION OF SAND
Experience and study have demonstrated that sand from dunes, beaches,
and nearshore areas is the best material available naturally in suitable
form to protect shores. Where sand is available in abundant quantities,
protective measures are greatly simplified and reduced in cost. When
dunes and broad, gently sloping beaches can no longer be provided, it is
necessary to resort to alternative structures, and the recreational attrac-
tion of the seashore is lost or greatly diminished.
Sand is a diminishing natural resource. Sand was once available to

our shores in adequate supply from streams, rivers and glaciers, and by
coastal erosion. Now cultural development in the watershed areas and along
previously eroding shores has progressed to a stage where large areas of
our coast now receive little or no sand through natural geological pro-
cesses. Continued cultural development in both inland and shore areas
tends to further reduce coastal erosion with resulting reduction in sand
supply to the shore. It thus becomes apparent that sand must be conserved.
This does not mean local hoarding of beach sand at the expense of adjoin-
ing areas, but rather the elimination of wasteful practices and the pre-
vention of losses from the shore zone whenever feasible.
Fortunately, nature has provided extensive stores of beach sand in

bays, lagoons, estuaries and offshore areas that can be used as a source
of beach and dune replenishment where the ecological balance will not be
disrupted. Massive dune deposits are also available at some locations,
though these must be used with caution to avoid exposing the area to flood
hazard. These sources are not always located in the proper places for
economic utilization, nor will they last forever. When they are gone, we
must face increasing costs for the preservation of our shores. Offshore
sand deposits will probably become the most important source in the future.
Mechanical bypassing of sand at coastal inlets is one means of conser-

vation that will come into increasing practice. Mining of beach sand for
commercial purposes, formerly a common procedure, is rapidly being reduced
as coastal communities learn the need for regulating this practice. Modem
hopper dredges, used for channel maintenance in coastal inlets, are being
equipped with a pump-out capability so their loads can be discharged near
the shore instead of being dumped at sea. On the California coast, where
large volumes of sand are lost into deep submarine canyons near the shore.
1-21
facilities are being considered that will trap the sand before it reaches
the canyon and transport it mechanically to a point where it can resume
normal longshore transport. Dune planting with appropriate grasses and
shmbs reduces landward windbome losses and aids in dune preservation.
Sand conservation is an important factor in the preservation of our

seacoasts, and must be included in long-range planning. Protection of
our seacoasts is not a simple problem; neither is it insurmountable. It
is a task and a responsibility that has increased tremendously in impor-
tance in the past 50 years, and is destined to become a necessity in future
years. While the cost will mount as time passes, it will be possible
through careful planning, adequate management, and sound engineering to
do the job properly and economically.
1-22
CHAPTER 2
MECHANICS
OF
WAVE MOTION
LEO CARRILLO STATE BEACH, CALIFORNIA - July 1968
CHAPTER 2
MECHANICS OF WAVE MOTION
2.1 INTRODUCTION
The effects of water waves are of paramount importance in the field

of coastal engineering. Waves are the major factor in determining the
geometry and composition of beaches, and significantly influence planning
and design of harbors, waterways, shore protection measures, coastal
structures, and other coastal works. Surface waves generally derive their
energy from the winds. A significant amount of this wave energy is finally
dissipated in the nearshore region and on the beaches.
Waves provide an important energy source for forming beaches; assort-

ing bottom sediments on the shoreface; transporting bottom materials on-
shore, offshore, and alongshore; and for causing many of the forces to
which coastal structures are subjected. An adequate understanding of the
fundamental physical processes in surface wave generation and propagation
must precede any attempt to understand complex water motion in the near-
shore areas of large bodies of water. Consequently, an understanding of
the mechanics of wave motion is essential in the planning and design of
coastal works.
This chapter presents an introduction to surface wave theories.

Surface and water particle motion, wave energy, and theories used in
describing wave transformation due to interaction with the bottom and
with structures are described. The purpose is to provide an elementary
physical and mathematical understanding of wave motion, and to indicate
limitations of selected theories. A number of wave theories have been
omitted. References are cited to provide information on theories not
discussed and to supplement the theories presented.
The reader is cautioned that man's ability to describe wave phenomena

is limited, especially when the region under consideration is the coastal
zone. Thus, the results obtained from the wave theories presented should
be carefully interpreted for application to actual design of coastal struc-
tures or description of the coastal environment.
2.2 WAVE MECHANICS
2.21 GENERAL
Waves in the ocean often appear as a confused and constantly changing

sea of crests and troughs on the water surface because of the irregularity
of wave shape and the variability in the direction of propagation. This
is particularly true while the waves are under the influence of the wind.
The direction of wave propagation can be assessed as an average of the
directions of individual waves. A description of the sea surface is
difficult because of the interaction between individual waves. Faster
waves overtake and pass through slower ones from various directions.
Waves sometimes reinforce or cancel each other by this interaction, and
often collide with each other and are transformed into turbulence, and
2-1
spray. When waves move out of the area where they are directly affected
by the wind, they assume a more ordered state with the appearance of
definite crests and troughs and with a more rhythmic rise and fall. These
waves may travel hundreds or thousands of miles after leaving the area in
which they were generated. Wave energy is dissipated internally within
the fluid by interaction with the air above, by turbulence on breaking,
and at the bottom in shallow depths.
Waves which reach coastal regions expend a large part of their energy
in the nearshore region. As the wave nears the shore, wave energy may be
dissipated as heat through turbulent fluid motion induced by breaking and
through bottom friction and percolation. While the heat is of little
concern to the coastal engineer, breaking is important since it affects
both beaches and manmade shore structures. Thus, shore protection measures
and coastal structure designs are dependent on the ability to predict wave
forms and fluid motion beneath waves, and on the reliability of such
predictions. Prediction methods generally have been based on simple waves
where elementary mathematical functions can be used to describe wave motion,
For some situations, simple mathematical formulas predict wave conditions
well, but for other situations predictions may be unsatisfactory for
engineering applications. Many theoretical concepts have evolved in the
past two centuries for describing complex sea waves; however, complete
agreement between theory and observation is not always found.
In general, actual water-wave phenomena are complex and difficult

to describe mathematically because of nonlinearities, three-dimensional
characteristics and apparent random behavior. However, there are two
classical theories, one developed by Airy (1845) and the other by Stokes
(1880), that describe simple waves. The Airy and Stokes theories gener-
ally predict wave behavior better where water depth relative to wavelength
is not too small. For shallow water, a cnoidal wave theory often provides
an acceptable approximation of simple waves. For very shallow water near
the breaker zone, solitary wave theory satisfactorily predicts certain
features of the wave behavior. These theories will be described according
to their fundamental characteristics together with the mathematical equa-
tions which describe wave behavior. Many other wave theories have been
presented in the literature which, for some specific situations, may pre-
dict wave behavior more satisfactorily than the theories presented here.
These other theories are not included, since it is beyond the scope of
this Manual to cover all theories.
The most elementary wave theory, referred to as small-amplitude or

linear wave theory, was developed by Airy (1845). It is of fundamental
importance since it not only is easy to apply, but is reliable over a
large segment of the whole wave regime. Mathematically, the Airy theory
can be considered a first approximation of a complete theoretical descrip-
tion of wave behavior. A more complete theoretical description of waves
may be obtained as the sum of an infinite number of successive approxima-
tions, where each additional term in the series is a correction to preced-
ing terms. For some situations, waves are better described by these
higher order theories which are usually referred to as finite amplitude
2-2
theories. The first finite amplitude theory, known as the trochoidal
theory, was developed by Gerstner (1802). It is so called because the
free surface or wave profile is a trochoid. This theory is mentioned
only because of its classical interest. It is not recommended for appli-
cation, since the water particle motion predicted is not that observed in
nature. The trochoidal theory does, however, predict wave profiles quite
accurately. Stokes (1880) developed a finite-amplitude theory which is
more satisfactory than the trochodial theory. Only the second-order Stokes'
equations will be presented, but the use of higher order approximations is
sometimes justified for the solution of practical problems.
For shallow-water regions, cnoidal wave theory, originally developed

by Korteweg and De Vries (1895) ,
predicts rather well the waveform and
associated motions for some conditions. However, cnoidal wave theory has
received little attention with respect to actual application in the solu-
tion of engineering problems. This may be due to the difficulties in
making computations. Recently, the work involved in using cnoidal wave
theory has been substantially reduced by introduction of graphical and
tabular forms of functions. (Wiegel, 1960), (Masch and Wiegel, 1961.)
Application of the theory is still quite involved. At the limit of cnoidal
wave theory, certain aspects of wave behavior may be described satisfacto-
rily by solitary wave theory. Unlike cnoidal wave theory, the solitary
wave theory is easy to use since it reduces to functions which may be
evaluated without recourse to special tables.
Development of individual wave theories is omitted, and only the

results are presented since the purpose is to present only that infor-
mation which may be useful for the solution of practical engineering
problems. Many publications are available such as Wiegel (1964), Kinsman
(1965) and Ippen (1966a), which cover in detail the development of some of
,
the theories mentioned above as well as others. The mathematics used here
generally will be restricted to elementary arithmetic and algebraic opera-
tions. Emphasis is placed on selection of an appropriate theory in accord-
ance with its application and limitations.
Numerous example problems are provided to illustrate the theory

involved and to provide some practice in using the appropriate equations
or graphical and tabular functions. Some of the sample computations give
more significant digits than are warranted for practical applications.
For instance, a wave height could be determined to be 10.243 feet for
certain conditions based on purely theoretical considerations. This
accuracy is unwarranted because of the uncertainty in the basic data used
and the assumption that the theory is representative of real waves. A
practical estimate of the wave height given above would be 10 feet. When
calculating real waves, the final answer should be rounded off.
2.22 WAVE FUNDAMENTALS AND CLASSIFICATION OF WAVES
Any adequate physical description of a water wave involves both its

surface form and the fluid motion beneath the wave. A wave which can be
described in simple mathematical terms is called a simple wave. Waves
2-3
which are difficult to describe in form or motion, and which may be com-
prised of several components are termed complex waves. Sinusoidal or
simple harmonic waves are examples of simple waves since their surface
profile can be described by a single sine or cosine function. A wave is
periodic if its motion and surface profile recur in equal intervals of
time. A wave form which moves relative to a fluid is called a progressive
wave; the direction in which it moves is termed the direction of wave
propagation. If a wave form merely moves up and down at a fixed position,
it is called a complete standing wave or a clapotis. A progressive wave
is said to be a wave of permanent form if it is propagated without experi-
encing any changes in free surface configuration.
Water waves are considered osoiltatory or nearly oscillatory if the

water particle motion is described by orbits, which are closed or nearly-
closed for each wave period. Linear, or Airy, theory describes pure
oscillatory waves. Most finite amplitude wave theories describe nearly
oscillatory waves since the fluid is moved a small amount in the direction
of wave advance by each successive wave. This motion is termed mass
transport of the waves. When water particles advance with the wave, and
do not return to their original position, the wave is called a wave of
translation. A solitary wave is an example of a wave of translation.
It is important to distinguish between various types of water waves

that may be generated and propagated. One way to classify waves is by
wave period T (the time for a wave to travel a distance of one wave
length), or by the reciprocal of T, the wave frequency f. One illustra-
tion of classification by period or frequency is given by Kinsman (1965)
and shown in Figure 2-1. The figure shows the relative amount of energy
contained in ocean waves having a particular frequency. Of primary concern
are those waves referred to as gravity waves in Figure 2-1, having periods
from 1 to 30 seconds. A narrower range of wave periods, from 5 to 15
seconds, is usually more important in coastal engineering problems. Waves
in this range are referred to as gravity waves since gravity is the
principal restoring force; that is, the force due to gravity attempts to
bring the fluid back to its equilibrium position. Figure 2-1 also shows
that a large amount of the total wave energy is associated with waves
classified as gravity waves; hence gravity waves are extremely important
in dealing with the design of coastal and offshore structures.
Gravity waves can be further separated into two states:
(a) seas, when the waves are under the influence of wind in a
generating area, and
(b) swell, when the waves move out of the generating area and
are no longer subjected to significant wind action.
Seas are usually made up of steeper waves with shorter periods and
lengths, and the surface appears much more confused than for swell. Swell
behaves much like a free wave, i.e., free from the disturbing force that
caused it, while seas consist to some extent of forced waves, i.e., waves
on which the disturbing force is applied continuously.
2-4
Ocean waves are complex. Many aspects of fluid mechanics necessary
to a complete discussion have only a minor influence on solving most
coastal engineering problems. Thus, a simplified theory which omits most
of the complicating factors is useful. The assumptions made in developing
the simple theory should be understood, because not all of the assumptions
are justified in all problems. When an assumption is not valid in a
particular problem, a more complete theory should be employed.
The most restrictive of common assumptions is that waves are small

perturbations on the surface of a fluid which is otherwise at rest. This
leads to a wave theory which is variously called, small-amplitude theory,
linear theory, or Airy theory. The small-amplitude theory provides in-
sight for all periodic wave behavior and a description of the periodic flow
adequate for most practical problems. This theory is unable to account for
mass transport due to waves (Section 2.253 Mass Transport Velocity), or the
fact that wave crests depart further from the mean water level than do the
troughs. A more general theory, usually called the finite amplitude, or
nonlinear wave theory is required to account for these phenomena as well as
most interactions between waves and other flows. The nonlinear wave theory
also permits a more accurate evaluation of some wave properties than can be
obtained with linear theory.
Several assumptions, commonly made in developing a simple wave theory

are listed below.
(a) The fluid is homogeneous and incompressible; therefore, the

density p is a constant.
(b) Surface tension can be neglected.
(c) Coriolis effect can be neglected.
(d) Pressure at the free surface is uniform and constant.
(e) The fluid is ideal or inviscid (lacks viscosity)
(f) The particular wave being considered does not interact with
any other water motions.
(g) The bed is a horizontal, fixed, impermeable boundary which

implies that the vertical velocity at the bed is zero.
(h) The wave amplitude is small and wave form is invariant in

time and space.
(i) Waves are plane or long crested (two-dimensional)
The first three are acceptable for virtually all coastal engineering
problems. It will be necessary to relax assumptions (d) , (e), and (f)
for some specialized problems not considered in this Manual. Relaxing
the three final assumptions is essential in many problems, and is con-
sidered later in this chapter.
2-6
In applying assumption (g) to waves in water of varying depth encoun-
tered when waves approach a beach the local depth is usually used. This
can be rigorously justified, but not without difficulty, for most practical
cases in which the bottom slope is flatter than about 1 on 10. A progres-
sive wave moving into shallow water will change its shape significantly.
Effects due to viscosity and vertical velocity on a permeable bottom may
be measurable in some situations, but these effects can be neglected in
most engineering problems.
2.23 ELEMENTARY PROGRESSIVE WAVE THEORY (Small -Amplitude Wave Theory)
The most fundamental description of a simple sinusoidal oscillatory

wave is by its length L (the horizontal distance between corresponding
points on two successive waves) height H (the vertical distance to its
;
crest from the preceding trough) ; period T (the time for two successive
crests to pass a given point) ; and depth d (the distance from the bed
to the Stillwater level). (See Appendix B for a list of common symbols.)
Figure 2-2 shows a two-dimensional simple progressive wave propagating

in the positive x-direction. The symbols used here are presented in the
figure. The symbol n denotes the displacement of the water surface
relative to the Stillwater level (SWL) and is a function of x and time.
At the wave crest, t\is equal to the amplitude of the wave a, or one-
half of the wave height.
Small -amplitude wave theory and some finite-amplitude wave theories

can be developed by introduction of a velocity potential <t)(x, z, t) . Hori-
zontal and vertical components of the water particle velocities are defined
at a point (x, z) in the fluid as u = 8<(i/9x and w = 3(|)/3z. The velocity
potential, Laplace's equation, and Bernoulli's dynamic equation together
with the appropriate boundary conditions provide the necessary information
needed in deriving the small-amplitude wave formulas. Such a development
has been shown by Lamb (1932), Eagleson and Dean (See Ippen 1966b), and
others
2.231 Wave Celerity, Length and Period The speed at which a wave form
.
propagates is termed the phase velocity or wave celerity, C. Since the

distance traveled by a wave during one wave period is equal to one wave-
length, the wave celerity can be related to the wave period and length by
C = (2-1)
^
An expression relating the wave celerity to the wavelength and water depth
is given by
C = (2-2)
2-7
o
CO
o :
Q. :
>
o

> •H
o „ o
> S in
X \ o
5 o M
o c= r- *- O
o U
o OJ X
o.
o ^^
c
(V
ro X I
II
.^ I"
II
o •H
o O
I
m
II II
I-
O l_
o C
•H
o
a>
+->
c
0)
£
0)
1—1
II
IM J)
B
u
E H
0)
a> o
a> 4h
o O
C
o
Hc
<4-l
a>
Q
u
3
W)
2-8
From Equation 2-1, it is seen that 2-2 can be written as
C = —
gT
2iT
,
tanh
/27rd
L
(2-3)
\ >
The values 2Tr/L and 2Tr/T are called the wave number k and wave angular
frequency w, respectively. From Equations 2-1 and 2-3 an expression
for wavelength as a function of depth and wave period may be obtained.
L = tanh (2-4)
2-n
Use of Equation 2-4 involves some difficulty since the unknown L, appears
on both sides of the equation. Tabulated values in Appendix C may be used
to simplify the solution of Equation 2-4.
Gravity waves may also be classified by the depth of water in which

they travel. Classification is made according to the magnitude of d/L
and the resulting limiting values taken by the function tanh(27rd/L)
Classifications are:
Classification
subscript is omitted. (Ippen, 1966b, pp 21-240 If units of feet and
seconds are specified, the constant g/2Tr is equal to 5.12 ft/sec^ and
C
"
= ^
2-n
= 5.12 T (ft/sec) , (2-7)
and
L^ = ^— = 5.12 T^ (ft). (2-8)
If Equation 2-7 is used to compute wave celerity when the relative depth
is d/L = 0.25, the resulting error will be about 9 percent. It is evi-
dent that a relative depth of 0.5 is a satisfactory boundary separating
deepwater waves from waves in water of transitional depth. If a wave is
traveling in transitional depths. Equations 2-2 and 2-3 must be used with-
out simplification. Care should be exercised to use Equations 2-2 and 2-3
when necessary, that is, when the relative depth is between 1/2 and 1/25.
When the relative water depth becomes shallow, i.e., 2TTd/L < 1/4 or
d/L < 1/25, Equation 2-2 can be simplified to
C = yf^. (2-9)
This relation, attributed to Lagrange, is of importance when dealing with

long-period waves, often referred to as long waves. Thus, when a wave
travels in shallow water, wave celerity depends only on water depth.
2.232 The Sinusoidal Wave Profile The equation describing the free
.
surface as a function of time t, and horizontal distance x, for a

simple sinusoidal wave can be shown to be
17 = a cos I
— 1 = — cos I
— -— ) , (2-10)
where n is the elevation of the water surface relative to Stillwater

level, and H/2 is one-half the wave height equal to the wave amplitude
a. This expression represents a periodic, sinusoidal, progressive wave
traveling in the positive x-direction. For a wave moving in the negative
x-direction, one need only replace the minus sign before 2TTt/T with a
plus sign. When (2Trx/L - 2Trt/T) equals 0, Tr/2, tt, 37v/2, the corresponding
values of n are H/2, 0, - H/2, and 0, respectively.
2.233 Some Useful Functions It can be shown by dividing Equation 2-3 by

.
Equation 2-6, and by dividing Equation 2-4 by Equation 2-8 that
(2-11)
2-10
If both sides of Equation 2-11 are multiplied by d/L, it becomes:
—
d d
= - tanh
, /27rd\
(2-12)
The term d/L^ has been tabulated as a function of d/L by Wiegel (1954),
and is presented in Appendix C on Table C-1. Table C-2 includes d/L as
a function of d/L^ iri addition to other useful functions such as 2ird/L
and tanh(2Trd/L) These functions simplify the solution of wave problems
.
described by the linear theory.
An example problem illustrating the use of linear wave theory and

the tables in Appendix C follows:
********* *** * EXAMPLE PROBLEM **************

GIVEN A wave with a period of T = 10 seconds is propagated shoreward
:
over a uniformly sloping shelf from a depth of d = 600 feet to a depth

of d = 10 feet.
FIND The wave celerities C and lengths

: L corresponding to depths of
d = 600 feet and d = 10 feet.
SOLUTION:
Using Equation 2-8,
L^ = 5.12 T^ = 5.12 (10)^ = 512 feet.
For d = 600 feet
d 600
1.1719.
Lo 512
From Table C-1 it is seen that for values of
d
->1.0
K L'
therefore
L - L^ = 512 (deepwater wave, since

feet fa. — > —
2-1
By Equation 2-1
L 512
T T
C = —
512
= 51.2 ft/sec.
For d = 10 feet
0.0195.
Z; ~ 512
Entering Table C-1 with d/L^ it is found that.
- = 0.05692.
L
hence
10
= 176. r I . . ,
reet [transitional depth, since
, ,
—<
1 d
— l\
<—),
0.05692 \ Ad i-f A j
176
c = = 17.6 ft/sec.
10
2.234 Local Fluid Velocities and Accelerations In wave force studies, .
it is often desirable to know the local fluid velocities and accelerations

for various values of z and t during the passage of a wave. The
horizontal component u, and the vertical component w, of the local
fluid velocity are given by;
H gT cosh[27r(z + d)/L] 27rx

u = cos
2 L cosh (27rd/L)
H gT sinh[27r(z + d)/L]
w = — sin
2 L cosh (27rd/L)
occurs when 9 = 0, 2tt, etc., while the maximum horizontal velocity in the
negative direction occurs when 9 = tt. Sit, etc. On the other hand the
maximum positive vertical velocity occurs when 9 = tt/2, 5tt/2, etc., and
the maximum vertical velocity in the negative direction occurs when
e = 377/2, 777/2, etc. (See Figure 2-3.)
The local fluid particle accelerations are obtained from Equations

2-13 and 2-14 by differentiating each equation with respect to t. Thus,
gTTH cosh[27r(z + d)/L] . , _..^

277X „„. , ._ ^_.
"- = ""
IT cosh (277d/L) "" '-^ - ^^- ^'"''^
I/)
c
o
i-i
<U
•—<
<u
o
u
<
T5
C
If)
0)
u
o
o
o
to
I
CM
0)
2-14
hence
50
L = = 270 feet
0.1854
Evaluation of the constant terms in Equations 2-13 through 2-16 gives
HgT 18 (32.2) (8)

4.88,
2L cosh (27rd/L) 2(270) (1.758)
gTTH 1 18 (32.2) (tt)

= 3.84.
cosh (27rd/L) (270) (1.758)
Substitution into Equation 2-13 gives
27:(50- 15)
u = 4.88 cosh [cos 60°] = 4.88 [cosh (0.8145)] (0.500).
270
From Table C-1 find
lird
= 0.8145 ,
and by interpolation
cosh (0.8145) = 1.3503,
and
sinh (0.8145) = 0.9074.
Therefore
u = 4.88 (1.3503) (0.500) = 3.29 ft/sec ,
w= 4.88 (0.9074) (0.866) = 3.83 ft/sec ,
"x = 3.84 (1.3503) (0.866) - 4.49 ft/sec^
az = -3.84 (0.9074) (0.500) = - 1.74 ft/sec^,
Figure 2-3, a sketch of the local fluid motion, indicates that the
fluid under the crest moves in the direction of wave propagation and
returns during passage of the trough. Linear theory does not predict any
mass transport; hence the sketch shows only an oscillatory fluid motion.
*************************************
2.235 Water Particle Displacements Another important aspect of linear
.
wave mechanics deals with the displacements of individual water particles

within the wave. Water particles generally move in elliptical paths in
shallow or transitional water and in circular paths in deep water. If the
2-15
mean particle position is considered to be at the center of the ellipse or
circle, then vertical particle displacement with respect to the mean
position cannot exceed one-half the wave height. Thus, since the wave
height is assigned to be small, the displacement of any fluid particle from
its mean position is small. Integration of Equations 2-13 and 2-14 gives
the horizontal and vertical particle displacement from the mean position,
respectively. (See Figure 2-4.)
Thus,
HgT^ cosh[27r(z + d)/L]
I = - sin (2-17)
47rL cosh (27rd/L)
HgT^ smh[27:(z + d)/L] /27rx l-nt^

f
^
= + —2 ; ; cos (2-18)
477 L cosh (277 d/L) \ L T ;
The above equations can be simplified by using the relationship
—^
2ire
tanh
, 277d
.
Thus,
H
—
cosh [277(z + d)/L] .
/27rx 277t
— — (2-19)
; sin I
2 sinh(277d/L) \ L T
H
—
sinh [277(z + d)/L] /277X
—- \
277t
; cos I I. (2-20)
2 sinh (277d/L) \ L T /
Writing Equations 2-19 and 2-20 in the following forms:
277X 277t i, sinh (277d/L)

sin
a cosh [277(z+d)/L.
27rt j sinh (27rd/L)

cos
T a sinh[277(z+d)/L]_
and adding, gives:
— + ^r = 1, (2-21)
in which
H cosh[277(z+d)/L]
A = (2-22)
2 sinh (277d/L)
H sinh[277(z+d)/L]
^ " (2-23)
2 sinh(277d/L) '
2-16
(D
>
CO
<D
CI,
(D
Q
TD
C
cfl
c
o
o
a.
c
0)
S
o
u
U)
c
0)
E
(D
O
c«
i-H
cx
Q
(U
I—
o
ni
3
•H
2-17
Equation 2-21 is the equation of an ellipse with a major (horizontal)
semiaxis equal to A, and a minor (vertical) semiaxis equal to B. The
lengths of A and B are measures of the horizontal and vertical dis-
placements of the water particles. Thus, the water particles are predicted
to move in closed orbits by linear wave theory: i.e., each particle returns
to its initial position after each wave cycle. Morison and Crooke (1953),
compared laboratory measurements of particle orbits with wave theory and
found, as had others, that particle orbits were not completely closed. This
difference between linear theory and observations is due to the mass trans-
port phenomenon which is discussed in a subsequent section.
Examination of Equations 2-22 and 2-23 shows that for deepwater

conditions A and B are equal and particle paths are circular. The
equations become
A=B = -e'^^/L
2
for->-.
L 2
(2-24)
For shallow-water conditions, the equations become
B
SOLUTION :
(a) Equation 2-3,
2ir \ L ,
Equation 2-1,
T
Therefore, equating 2-1 and 2-3,
T 2n \ L r
and multiplying both sides by (2tt)2/lt
(2i:y L _ (2nr gT , /27:d^

LT T = IF ^ '^"MX;
Hence,
i2ny 27rg ^ Ayrd^
(bj Equation 2-13 may be written
gTH cosh[27r(z+d)/L] /2nx

^ ^ _ 27Tt\
2L cosh (27rd/L) """M L ~ T J
^ 1 gH cosh[27r(z + d)/L] hnx 27rt^
C 2 ~
cosh(27rd/L) ^""llT F;
since
T _ 1
L ~ C
Since
C = C
2n
.anh M
\ L J
^ ^ ^T 1_ cosh[27r(z + d)/L] /27rx _ 2n^

tanh(27rd/L) cosh (27rd/L) ^°' I L ~T"y
2-19
and since,
/27rd\_ sinh (27rd/L)
L ~ cosh (27rd/L)
y y '
therefore,
nH cosh-—[27r(z+d)/L] (inx 27rt\

U = — ;
COS I
— I .
T sinh (27rd/L) \ L T /
*************************************
************** EXAMPLE PROBLEM **************
GIVEN A wave in a depth of d = 40 feet, height of H = 10 feet, and a
:
period of T = 10 seconds. The corresponding deepwater wave height is

Hq = 10.45 feet.
FIND :
(a) The horizontal and vertical displacement of a water particle from

its mean position when z = 0, and when z = - d.
(b) The maximum water particle displacement at a depth d = 25 feet when

the wave is in infinitely deep water.
(c) For the deepwater conditions of (b) above, show that the particle
displacements are small relative to the wave height when z = - Lo/2.
SOLUTION :
(a) L^ = 5.12 T^ = 5.12(10)^ = 512 feet ,
d 40
0.0781 .
Lc 512
From Appendix C, Table C-1
sinh (
— 1= 0.8394 ,
1
/27rd\
tanh I 1= 0.6430 .
When z = 0, Equation 2-22 reduces to
H 1
A =
2 tanh(27rd/L) '
2-20
and Equation 2-23 reduces to
H
B = -
2
Thus,
10 1
A = = 7.78 feet ,
2 (0.6430)
H 10
B = -
2
= —
2
= 5.0 feet
When z = - d,
A = = = 5 96 feet
2sinh(27rd/L) 2(0.8394)
and, B = 0.
(b) With Uo = 10-45 feet, and z = - 25, evaluate the exponent of e

for use in Equation 2-24, noting that L = L^,
/27rz\
= —
27r(- 25)
^ = - 0.307
\L
,
I 512
thus
e-0.307 = 0.736 .
Therefore,
H
A=B= — 10.45
27:2/
e ^^ = (0.736) = 3.85 feet .
2 2
The maximum displacement or diameter of the orbit circle would

be 2(3.85) = 7.70 feet.
ô "512
r
Cc)
^
z = = = - 256 feet ,
2 2
27rz 2-n(- 256)

-— = = - 3.142 .
L 512
Using Table C-4 of Appendix C,

g-3.142 ^ 0.043.
H^
A = B ~ 2
e
2mL
f^
10.45
= -^^— (0.043) = 0.225
2
feet .
2-21
Thus the maximum displacement of the particle is 0.45 feet which is small
when compared with the deepwater height, H^ = 10.45 feet.
*************************************
2.236 Subsurface Pressure Subsurface pressure under a wave is the
.
summation of two contributing components, dynamic and static pressures,

and is given by
coshl27r(z + d)/L] H /27rx 27rt\

P '^" ^ ^^
= '^ cosh (2.d/L) 2
^°^
\y~^l ' '
^'-'^^
where p' is the total or absolute pressure, p^^ is the atmospheric

pressure and p = w/g is the mass density of water (for saltwater,
p = 2.0 lbs sec2/ft'+= 2.0 slugs/ft^; for fresh water, p = 1.94 slugs/ft^).
The first term of Equation 2-26 represents a dynamic component due to
acceleration, while the second term is the static component of pressure.
For convenience, the pressure is usually taken as the gage pressure
defined as
cosh[27r(z+d)/L] H /27rx 27rt\

^'"''^
^=^'-â = P% eosh(2.d/L) 2
'^"^
IÎT
"^
T"]" '^' '
Equation 2-27 can be written as
cosh[27r(z+d)/L]
p = Pgr? Pgz ,
(2-28)
cosh (27Td/L)
since
H /27rx 27rtl
7? = — cos I —
2 \ L T ;
The ratio
cosh[27r(z+d)/L]
K^ = -^-^ . (2-29)
cosh(27rd/L)
is termed the pressure response factor. Hence, Equation 2-28 can be

written as
p = pg(rjK -z) . (2-30)
2-22
The pressure response factor K, for the pressure at the bottom when
z = - d,
K, =
2
K = — 1
1
.^
cosh (ZTrd/L)
,,.-> ,
'
(2-31)
is tabulated as a function of d/L^ and d/L in Tables C-1 and C-2 of

Appendix C.
It is often necessary to determine the height of surface waves based

on subsurface measurements of pressure. For this purpose it is convenient
to rewrite Equation 2-30 as,
_ N(p+pgz)
r? - ,
(2-32)
where z is the depth below the SWL of the pressure gage, and N is a
correction factor equal to unity if the linear theory applies. Several
empirical studies have found N to be a function of period, depth, wave
amplitude and other factors. In general, N decreases with decreasing
period, being greater than 1.0 for long-period waves and less than 1.0 for
short-period waves.
A complete discussion of the interpretation of pressure gage wave

records is beyond the scope of this Manual. For a more detailed discussion
of the variation of N with wave parameters, the reader is referred to
Draper (1957), Grace (1970), and Esteva and Harris (1971).
************** EXAMPLE PROBLEM **************

GIVEN An average maximum pressure of p = 2590 lbs/ft^ is measured by a
:
subsurface pressure gage located 2 feet above the bed in water at d = 40

feet. The average frequency f = 0.0666 cycles per second.
FIND :The height of the wave H assuming that linear theory applies and
the average frequency corresponds to the average wave amplitude.
SOLUTION:
1 1
T = — = ^ _^^^ ~ 15 seconds
f (0.0666)
L^ = 5.12 T^ = 5.12(15)^ = 1152 feet
40
0.0347
K 1152
2-23
From Table C-1 o£ Appendix C, entering with d/L^,
d
0.07712
L
hence
40
L = ; = 519 feet ,
(0.07712)
and
cosh I 1= 1.1197
Therefore, from Equation 2-29
cosh[27r(z + d)/L] cosh [27r(- 38 + 40)/519] 1.0003

K
z
= . .„
cosh(27rd/L)
.,^. =
1.1197
=
1.1197
= 0.8934 .
Since n = a = H/2 when the pressure is maximiom (under the wave crest) ,
and N = 1.0 since linear theory is assumed valid.
H N(p+pgz) 1.0 [2590 + (64.4) (-38)]

- = ,,
= = 2.47 feet
2 pgK (64.4) (0.8934)
Therefore,
H = 2 (2.47) ^ 5 feet
Note that the tabulated value of K in Appendix C, Table C-1, could not
be used since the pressure was not measured at the bottom.
2.237 Velocity of a Wave Group The speed with which a group of waves
.
or a wave train travels is generally not identical to the speed with which
individual waves within the group travel. The group speed is termed the
group velocity, C_; the individual wave speed is the phase velocity or
wave celerity given by Equations 2-2 or 2-3. For waves propagating in
deep or transitional water with gravity as the primary restoring force,
the group velocity will be less than the phase velocity. (For those waves
propagated primarily under the influence of surface tension, i.e., capil-
lary waves, the group velocity may exceed the velocity of an individual
wave.
The concept of group velocity can be described by considering the

interaction of two sinusoidal wave trains moving in the same direction
2-24
with slightly different wavelengths and periods, The equation of the
water surface is given by:
H /27rx 27rt\
+ n, = — cos I
' 2 \ L.
v^'n^^Vz ênvelope
-0.2 -0.1 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 I.I 1.2 1.3
X
~2
/J-2J-1
\~r r
\
)~
t
~2 \'~j J
——
/Jj_~Xl \
J
( offer Kinsman, 1965)
Figure 2-5. Formation of Wave Groups by the Addition of Two Sinusoids

Having Different Periods
2-26
hence the group and phase velocities are equal. Thus in shallow water,
because wave celerity is fully determined by the depth, all component waves
in a wave train will travel at the s^nie speed precluding the alternate
reinforcing and cancelling of components. In deep and transitional water,
wave celerity depends on the wavelength; hence slightly longer waves travel
slightly faster, and produce the small phase differences resulting in wave
groups. These waves are said to be dispersive or propagating in a
dispersive mediioriy i.e. in a medium where their celerity is dependent on
wavelength
Outside of shallow water, the phase velocity of gravity waves is

greater than the group velocity, and an observer moving along with a group
of waves at the group velocity will see waves that originate at the rear
of the group move forward through the group traveling at the phase velocity,
and disappear at the front of the wave group.
Group velocity is important because it is with this velocity that wave

energy is propagated.
Although mathematically, the group velocity can be shown rigorously

from the interference of two or more waves (Lamb, 1932), the physical
significance is not as obvious as it is in the method based on the con-
sideration of wave energy. Therefore an additional explanation of group
velocity is provided on wave energy and energy transmission.
2.238 Wave Energy and Power . The total energy of a wave system is the
sum of its kinetic energy and its potential energy. The kinetic energy is
that part of the total energy due to water particle velocities associated
with wave motion. Potential energy is that part of the energy resulting
from part of the fluid mass being above the trough - the wave crest.
According to the Airy theory, if the potential energy is determined
relative to mean water level, and all waves are propagated in the same
direction, potential and kinetic energy components are equal, and the
total wave energy in one wavelength per unit crest width is given by
Subscripts k and p refer to kinetic and potential energies. Total

average wave energy per unit surface area, termed the specific energy or
energy density, is given by
E
E = - = ^
pgH*
. (2-39)
Wave energy flux is the rate at which energy is transmitted in the

direction of wave propagation across a vertical plane perpendicular to the
direction of wave advance and extending down the entire depth. The average
2-27
energy flux per unit wave crest width transmitted across a plane perpen-
dicular to wave advance is
P = InC = EC .
(2-40)
g
Energy flux F is frequently called wave power and
47rd/L
1 +
sink (47rd/L)
If a plane is taken other than perpendicular to the direction of wave

advance F = F C sin (J), where is the angle between the plane across
(j)
which the energy is being transmitted and the direction of wave advance.
For deep and shallow water. Equation 2-40 becomes
P^ = - E^C (deep water). (2-41)

o 2 " " *^
P = EC^ = EC (shallow water) .

(2-42)
An energy balance for a region through which waves are passing will
reveal, that for steady state, the amount of energy entering the region
will equal the amount leaving the region provided no energy is added or
removed from the system. Therefore, when the waves are moving so that
their crests are parallel to the bottom contours.
Eo
************** EXAMPLE PROBLEM **************
GIVEN A deepwater oscillatory wave with a wavelength of L^ = 512 feet,
:
a height of H^ = 5 feet and a celerity of C^^ = 51.2 ft/sec, moving

shoreward with its crest parallel to the depth contours. Any effects
due to reflection from the beach are negligible.
FIND :
(a) Derive a relationship between the wave height in any depth of water
and the wave height in deep water, assuming that wave energy per
unit crest width is conserved as a wave moves from deep water into
shoaling water.
(b) Calculate the wave height for the given wave when the depth is 10
feet.
(c) Determine the rate at which energy per unit crest width is trans-
ported toward the shoreline and the total energy per unit width
delivered to the shore in 1 hour by the given waves
SOLUTION :
(a) Since the wave crests are parallel to the bottom contours, refraction
does not occur, therefore H^ = H^. (See Section 2.3. WAVE
REFRACTION.)
From Equation 2-43,

Therefore,
1-1 2 n C
and since from Equations 2-3 and 2-6
—
C
= tanh
,
/27Td^
and from Equation 2-35 where
47:d/L
1 +
sink (477 d/L)
H 1 1
K. (2-44)
tanh (27rd/L) (47rd/L)
1 +
sinh (4rrd/L)
where Kg or H/H^ is termed the shoaling coefficient. Values

of H/Hq as a function of d/Lo and d/L have been tabulated
in Tables C-1 and C-2 of Appendix C.
(b) For the given wave, d/L^ = 10/512 = 0.01953. Either from Table
C-1 or from an evaluation of Equation 2-44 above.
H
7 = 1.233
H
Therefore
H = 1.233(5) = 6.165 ft.
(c) The rate at which energy is being transported toward shore is the
wave energy flux.
P = - E^C
o o
= nEC .
Since it is easier to evaluate the energy flux in deep water,

the left side of the above equation will be used.
- 1 - 1 PgCH^)' 51.2 1 64(5)^

P = - E^C = = 51.2 ,
2 " " 2 8 2 8
ft-lb
P = 5120 per ft . of wave crest ,
sec
5120
P = 9.31 horsepower per ft. of wave crest .
550
2-30
This represents the expenditure of
5120
ft-lb
sec
X 3600 —
sec
hr
= 18.55 X
, .
10* ft-lbs
,,
,
of energy each hour on each foot of beach.
*************************************
The mean rate of energy transmission associated with waves propagating
into an area of calm water provides a better physical description of the
concept of group velocity. An excellent treatment of this subject is given
by Sverdrup and Munk (1947) and is repeated here.
Quoting from the Beach Erosion Board Technical

Report No. 2, (1942) "As the first wave in the group
:
advances one wave length, its form induces correspond-

ing velocities in the previously undisturbed water and
the kinetic energy corresponding to those velocities
must be drawn from the energy flowing ahead with the
form. If there is equipartition of energy in the wave,
half of the potential energy which advanced with the
wave must be given over to the kinetic form and the
wave loses height. Advancing another wave length
another half of the potential energy is used to supply
kinetic energy to the undisturbed liquid. The process
continues until the first wave is too small to identify.
The second, third, and subsequent waves move into water
already disturbed and the rate at which they lose height
is less than for the first wave. At the rear of the
group, the potential energy might be imagined as moving
ahead, leaving a flat surface and half of the total
energy behind as kinetic energy. But the velocity
pattern is such that flow converges toward one section
thus developing a crest and diverges from another
section forming a trough. Thus the kinetic energy is
converted into potential and a wave develops in the
rear of the group."
This concept can be interpreted in a quantitative

manner, by taking the following example from R. Gatewood
(Gaillard 1904, p. 50). Suppose that in a very long
trough containing water originally at rest, a plunger
at one end is suddenly set into harmonic motion and
starts generating waves by periodically imparting an
energy E/2 to the water. After a time interval of n
periods there are n waves present. Let m be the posi-
tion of a particular wave in this group such that m=l
refers to the wave which has just been generated by
the plunger, m={n+\)/2 to the center wave, and m=n to
the wave furthest advanced. Let the waves travel with
constant velocity C, and neglect friction.
2-31
After the first complete stroke one wave will be
present and its energy is E/2. One period later this
wave has advanced one wave length but has left one-
half of its energy or E/4 behind. It now occupies a
previously undisturbed area to which it has brought
energy E/4. In the meantime, a second wave has been
generated, occupying the position next to the plunger
where E/4 was left behind by the first wave. The
energy of this second wave equals E/4 + E/2 = 3E/4.
Repeated applications of this reasoning lead to the
results shown in Table 2-1.
The series number n gives the total number of

waves present and equals the time in periods since
the first wave entered the area of calm; the wave
number m gives the position of the wave measured from
the plunger and equals the distance from the plunger
expressed in wave lengths. In any series, n, the
deviation of the energy from the value E/2 is
symmetrical about the center wave. Relative to the
center wave all waves nearer the plunger show an
excess of energy and all waves beyond the center wave
show a deficit. For any two waves at equal distances
from the center wave the excess equals the deficiency.
In every series, n, the energy first decreases slowly
with increasing distance from the plunger, but in the
vicinity of the center wave it decreases rapidly.
Thus, there develops an "energy front" which advances
with the speed of the central part of the wave system,
that is, with half the wave velocity.
According to the last line in Table 2-1 a definite

pattern develops after a few strokes: the wave closest
to the plunger has an energy E(2"-l)/2" which approaches
the full amount E, the center wave has an energy E/2,
and the wave which has traveled the greatest distance
has very little energy (E/2^)
Table 2-1. Distribution of Wave Heights in a Short Train of Waves
Series
number
n
With a large number of waves (a large n), energy decreases with
increasing m, and the leading wave will eventually lose its identity.
At the group center, energy increases and decreases rapidly - to nearly
maximum and to nearly zero. Consequently, an energy front is located
at the center wave group for deepwater conditions. If waves had been
examined for shallow rather than deep water, the energy front would
have been found at the leading edge of the group. For any depth, the
ratio of group to phase velocity (C„/C) generally defines the energy
front. Also, wave energy is transported in the direction of phase propa-
gation, but moves with the group velocity rather than phase velocity.
2,239 Summary - Linear Wave Theory . Equations describing water surface

profile particle velocities, particle accelerations, and particle displace-
ments for linear (Airy) theory are summarized in Figure 2-6.
2.24 HIGHER ORDER WAVE THEORIES
Solution of the hydroynamic equations for gravity-wave phenomena can

be improved. Each extension of the theories usually produces better
agreement between theoretical and observed wave behavior. The extended
theories can explain phenomena such as mass transport that cannot be
explained by linear theory. If amplitude and period are known precisely,
the extended theories can provide more accurate estimates of such derived
quantities as the velocity and pressure fields due to waves than can
linear theory. In shallow water, the maximum wave height is determined
by depth, and can be estimated without wave records.
When concern is primarily with the oscillating character of waves,

estimates of amplitude and period must be determined from empirical data.
In such problems, the uncertainty about the accurate wave height and
period leads to a greater uncertainty about the ultimate answer than does
neglecting the effect of nonlinear processes. Thus it is unlikely that
the extra work involved in using nonlinear theories is justified.
The engineer must define regions where various wave theories are
valid. Since investigators differ on the limiting conditions for the
several theories, some overlap must be permitted in defining the regions.
Le Mehaute (1969) presented Figure 2-7 to illustrate approximate limits
of validity for several wave theories. Theories discussed here are indi-
cated as are Stokes' third- and fourth-order theories. Dean (1973),
after considering three analytic theories, presents a slightly different
analysis. Dean (1973) and Le Mehaute (1969) agree in recommending cnoidal
theory for shallow-water waves of low steepness, and Stokes' higher order
theories for steep waves in deep water, but differ in regions assigned
to Airy theory. Dean indicates that tabulated stream function theory is
most internally consistent over most of the domain considered. For the
limit of low steepness waves in transitional and deep water, the differ-
ence between stream function theory and Airy theory is small. Additional
wave theories not presented in Figure 2-7 may also be useful in studying
2-33
^'^
0.001
itiil:]ll[ii#i#ll''fî'l-|!in']-]1îti^|tiMr!^1-linî^^
II I
0.01 0.02 0.04 0.06 0.1 0.2 0.4 0.6 0.8 1.0 2.0 4.0 6.0 8.0 100
d
(ft/sec2) (after Le Mehaute,l969)
Figure 2-7. Regions of Validity for Various Wove Theories
2-35
wave phenomena. For given values of H, d and T, Figure 2-7 may be used
as a guide in selecting an appropriate theory. The magnitude of the
Ursell parameter Ud shown in the figure may be used to establish the
boiondaries of regions where a particular wave theory should be used.
The Ursell parameter is defined by
T 2 14
"« = y^ .
(2-45)
For linear theory to predict accurately the wave characteristics, both

wave steepness, H/gT^, and the Ursell parameter must be small as shown in
Figure 2-7.
2.25 STOKES' PROGRESSIVE, SECOND-ORDER WAVE THEORY
Wave formulas presented in the preceding sections on linear wave

theory are based on the assumption that the motions are so small that the
free surface can be described to the first order of approximation by
Equation 2-10:
H /277X
—
27rt\ H
v = — cos 1
= — cos or a cos d .
2 \L T
I
I
2
More specifically, it is assumed that wave amplitude is small, and the

contribution made to the solution by higher order terms is negligible. A
more general expression would be:
T? = acos(0) + a^Bj (L, d) cos (20)

(2-46)
+ a'Bj (L, d) cos(3fl) + • • • a"B„ (L, d) cos(n0) ,
where a = H/2, for first- and second-orders, but a < H/2 for orders
higher than the second, and Bg, B3 etc. are specified functions of the
wavelength L, and depth d.
Linear theory considers only the first term on the right side of
Equation 2-46. To consider additional terms represents a higher order of
approximation of the free surface profile. The order of the approxima-
tion is determined by the highest order term of the series considered.
Thus, the ordinate of the free surface to the third order is defined by
the first three terms in Equation 2-46.
When the use of a higher order theory is warranted, wave tables, such
as those prepared by Skjelbreia (1959), and Skjelbreia and Hendrickson
(1962), should be used to reduce the possibility of numerical errors made
in using the equations. Although Stokes (1880) first developed equations
for finite amplitude waves, the equations presented here are those of
Miche (1944).
2-36
2.251 Wave Celerity, Length, and Surface Profile It can be shovm that, .
for second-order theories, expressions for wave celerity and wavelength

are identical to those obtained by linear theory. Therefore,
^ = 2- gT , /27rd'
C tanh (2-3)
2tt \ L
and
L = ^^ tanh
'-^ (2-4)
2n \ L
The above equations, corrected to the third order, are given by:
+ +
C = —
gT
271
,
tanh
/27rd^
\ L
1 +
T
5 2 cosh (47rd/L)
8 sinh" (27rd/L)
2 cosh^ (47rd/L)
,(2-47)
and
'nH\ 5 + 2 cosh (47rd/L) + 2 cosh^ (47rd/L)

L = ^— tanh 1 + (2-48)
27r I L 8 sinh" (2;rd/L)
The equation of the free surface for second-order theory is
H
—
/27rx 27rt
'
cos I
2 \ L
(2-49)
+
cosh (27rd/L)
2 + cosh(47rd/L) cos
u.
(8L sinh^ (27rd/L) t)
For deep water, (d/L > 1/2) Equation 2-49 becomes,
H„ --. /4jrx
T? =
o
COS
J2-nx
I-..- —
2-ni ^ '^"o
+ cos — —
4fft\
(2-50)
2
I
\L, TJ 4L^ \l^ t/
2.252 Water Particle Velocities and Displacements The periodic x and z .
components of the water particle velocities to the second order are given
by
HgT cosh [27r(z + d)/L] /27rx lux^

U = ;
cos I
—
2L cosh(27rd/L) \ L T y
(2-51)
3 /ttHV cosh [47r(z + d)/L] ^ttx 4jrtl

+ — C . . ^ ,. TTTT cos
sinh* (27rd/L)
2-37
w =
nU
—L C
sinh [27r(z
r~r-r:
sinh(27rd/L)
—+7r~^ —
d)/L]
sin 1
\
2itx
L
(2-52)
+ 1 i'^\ r
si"h [4n(z + d)/L] . Mttx
4 \l/ sinh'»(27rd/L) ''" \l
Second-order equations for water-particle displacements from their
mean position for a finite amplitude wave are:
HgT^ cosh [27r(z + d)/L] Ilrrx _ 2-nt 7:H^

= -
.
?
+
471 cosh (27rd/L) ^"^ \ L T 8L sinh^ (2;rd/L)
(2-53)
3 cosh [47r(z + d)/L] Mtt: /ttHV Ct cosh [47r(z + d)/L]

- sin +
1
2 sinh^ (27:d/L) It) T sinhM27rd/L)
and
HgT^ sinh [27r(z + d)/L] 2nx l-nt
t = —2 cos —
^
47rL cosh (27rd/L) \ L T
(2-54)
3 ttH^ sinh [47r(z + d)/L] Mttx
+ ~ n. ; cos
16 L sinh''(27rd/L) \L
2.253 Mass Transport Velocity The last term in Equation 2-53 is of

.
particular interest; it is not periodic, but is the product of time and

a constant depending on the given wave period and depth. The term predicts
a continuously increasing net particle displacement in the direction of
wave propagation. The distance a particle is displaced during one wave
period when divided by the wave period gives a mean drift velocity, U(z)
called the mass transport velocity. Thus,
/ttHV C cosh [47r(z + d)/L]

U(z) = (2-55)
T) 2 sinhM27rd/L)
Equation 2-53 indicates that there is a net transport of fluid by

waves in the direction of wave propagation. If the mass transport, indi-
cated by Equation 2-55 leads to an accumulation of mass in any region, the
free surface must rise, thus generating a pressure gradient. A current,
formed in response to this pressure gradient, will reestablish the distri-
bution of mass. Studies of mass transport, theoretical and experimental,
have been conducted by Longuet-Higgins (1953, 1960), Mitchim (1940), Miche
(1944), Ursell (1953), and Russell and Osorio (1958). Their findings
indicate that the vertical distribution of the mass transport velocity is
modified so that the net transport of water across a vertical plane is
zero.
2-38
2.254 Subsurface Pressure . The pressure at any distance below the fluid
surface is given by
H cosh [27r(z + d)/L]

^°'
^§ 2 cosh (27rd/L)
+
3
8
ttH^
L
tanh(27rd/L)
sinhM27rd/L)
cosh [47r(z
sinh^ (27rd/L)
d)/L]
cos r^ - ^ (2-56)
ttH^ tanh (27rd/L) 4;r(z + d)

cosh
s'^T sinh^ (27rd/L)
2.255 Maximum Steepness of Progressive Waves A progressive gravity wave .
is physically limited in height by depth and wavelength. The upper limit,

or breaking wave height in deep water is a function of the wavelength, and
in shallow and transitional water is a function of both depth and wavelength.
Stokes (1880) predicted theoretically that a wave would remain stable

only if the water particle velocity at the crest was less than the wave
celerity or phase velocity. If the wave height were to become so large
that the water particle velocity at the crest exceeded the wave celerity,
the wave would become unstable and break. Stokes found that a wave having
a crest angle less than 120 degrees would break (angle between two lines
tangent to the surface profile at the wave crest). The possibility of
existence of a wave having a crest angle equal to 120 degrees was shown by
Wilton (1914) Michell (1893) found that in deep water the theoretical
.
limit for wave steepness was
'H
— 1 = 0.142 « - (2-57)
"olmax '
Havelock (1918) confirmed Michell 's findings.
Miche (1944) gives the limiting steepness for waves traveling in

depths less than L^/2 without a change in form as
27rd\ , /27rd\
= 0.142 tanh (2-58)
Laboratory measurements by Danel (1952) indicate that the above

equation is in close agreement with an envelope curve to laboratory
observations. Additional discussion of breaking waves in deep and
shoaling water is presented in Section 2.6, BREAKING WAVES.
2.256 Comparison of the First- and Second-Order Theories A comparison .
of first- and second-order theories is useful to obtain insight about the
2-39
choice of a theory for a particular problem. It should be kept in mind
that linear (or first-order) theory applies to a wave which is symmetrical
about Stillwater level and has water particles that move in closed orbits.
On the other hand, Stokes' second-order theory predicts a wave form that
is unsymmetrical about the Stillwater level but still symmetrical about a
vertical line through the crest, and has water particle orbits which are
open.
************** EXAMPLE PROBLEM **************

GIVEN : A wave traveling in water depth of d = 20 feet, with a wavelength
of L = 200 feet and a height of H = 4 feet.
FIND :
(a) Compare the wave profiles given by the first- and second-order
theories
(b) What is the difference between the first- and second- order
horizontal velocities at the surface under both the crest and
trough?
(c) How far in the direction of wave propagation will a water particle
move from its initial position during one wave period when z = 0?
(d) What is pressure at the bottom under the wave crest as predicted
by both the first- and second-order theories?
(e) What is the wave energy per unit width of crest predicted by the
first-order theory?
SOLUTION :
(a) The first-order profile Equation 2-10 is:
H
7? = — cos 6 ,
where
(2iTX 27rt
L T
and the second-order profile Equation 2-49 is:
H ttH^ cosh(27rd/L)
V = — cos 6 + — ; 2 + cosh I cos 26
2 8L sinh^(27rd/L)
for
20
-d = = 0.1 ,
L 200
2-40
and from Table C-2
/27rd\
cosh t
1 = 1.2040 ,
sinh I 1 = 0.6705 ,
/47rd\
cosh = 1.8991 ,
ttH^ cosh(2rrd/L) 47rd\

2 + cosh = 0.48
8L sinh' (27rd/L)
Therefore
7?
= 2 COS0 + 0.48 cos 2d ,
Tj, = 2.48 ft.,

2
r?^ = ~ 1-52 ft.

2
at the crest
where n^ 2 ^"'^ ^t 2 ^^^ ^^^ values of n
(i.e. cos 9=1, cos 26 = 1) and trough (i.e. cos 6 = - 1,
cos 26 = 1) according to second-order theory.
Figure 2-8 shows the surface profile n as a function of 6. The

second-order profile is more peaked at the crest and flatter at the
trough than the first-order profile. The height of the crest above
SWL is greater than one-half the wave height; consequently the
distance below the SWL of the trough is less than one-half the
height. Moreover, for linear theory, the elevation of the water
surface above the SWL is equal to the elevation below the SWL;
however, for second-order theory there is more height above SWL
than below.
(b) For convenience, let
u^ 7 = value of u at crest according to first-order theory,
^t 1 ~ v^l"® °^ " at a trough according to first-order theory,
u„ 2 ~ value of u at a crest according to second-order theory,
u+ 9 = value of u at a trough according to second-order theory.
2-41
CVJ
in
>- -T. <»
LJ
According to first-order theory, a crest occurs at z = H/2,
cos 6=1 and a trough at z = - H/2, cos = - 1. Equation 2-13
therefore implies
with
Entering Table C-2 with d/L = 0.10, find tanh (2TTd/L) = 0.5569.
From Equation 2-3 which is true for both first- and second-order
theories.
C^ = —
27r
tanh
(t)
27rd\
=
(32.2) (200) (0.5569)
27r
= 571 (ft/sec)'
or
C = 23.9 ft/sec.
As a consequence.
T 1
= 0.0418 sec/ft
L C
Referring again to Table C-2, it is found that when
H
^=2 '
27r(z + d)
cosh = cosh[27r(0.11)] = 1.249
and when
H
z = — — '
2
27r(z + d)
cosh = cosh [27r(0.09)] = 1.164
Thus, the value of u at a crest and trough respectively according

to first-order theory is
\,1 = \ (32.2) (0.0418) ^^ = 2.8 ft/sec ,
4 1.164
u, , = - (32.2) (0.0418) = — 2.6 ft/sec
*'^
2 1.2040
2-44
Entering Table C-2 again, it is found that when
z = 1?^ 2 = 2.48 feet
27r(z + d)
cosh = cosh [27r(0.1124)] = 1.260.
47r(z + d)
cosh = cosh [47r(0.1 124)] = 2.175
When
z = T?^
2 = ~ 1-52 feet
27r(z + d)
cosh cosh [27r(0.0924)] = 1.174
47r(z + d)
cosh cosh [47r(0.0924)] = 1.753
Thus, the value of u at a crest and trough respectively according

to second-order theory is
u^2 =
'''''
4
:;
2
(32.2) (0.0418) ——1.260
1.2040
- +
3
- -—
/
4 \200)
47r\^
(23.9)
2.175
-^^
0.202
= 3.6 ft/sec ,
''2
u,y =
4
(32.2) (0.0418) —1.174
1. 2040
+ -—
3 /
4 \200/
47r\2
(23 9)
1.753
0.202
2.0 ft/ sec
(c) To find the horizontal distance that a particle moves during one
wave period at z = 0, Equation 2-55 can be written as:
AX(z) _ /ttH ^ C cosh [47r(z + d)/L]

U(z) =
T ~ \ L 2 sinh^ (27rd/L)
where AX(z) is the net horizontal distance traveled by a water

particle, z feet below the surface, during one wave period.
2-45
For the example problem, when
z = ,
cosh (47rd/L) C
AX(0) >= ^ (t)" sinh2(27rd/L) '
1.
(ttH)^ cosh(47Td/L)_ (;r4)^ (1.899)

= 1.67 ft.
2L sinh^(27rd/L) 2(200) (0.6705)'
(d) The first-order approximation for pressure under a wave is;
pgH cosh [27r(z + d)/L]

P
= cos 6 — pgz
cosh (27rd/L)
when
6 = (i.e. the wave crest), cos = 1,

and when
27r(z + d)
z = — d, cosh = cosh(O) =1.0
Therefore
p
= (2) (32.2) (4)
——
1
- (2) (32.2) (- 20) = 107 + 1288 = 1395 lbs/ ft^
M J. (Û^
at a depth of 20 feet below the SWL. The second-order terms

according to Equation 2-56 are
3 -nW tanh(27rd/L) cosh [47r(z + d)/L] 1

cos 26
8 ^^ IT sinh' (27rd/L) sinh' (27rd/L) 3
1 ttH^ tanh (27rd/L) 47r(z + d)

cosh
sinh' (27rd/L)
Substituting in the equation:
3 '^(4)' (0.5569) 1
- ,o^ .oo ..
(2) (32.2) (1)
8 200 (0.6705)' (0.6705)'
-i (2) (32.2) ^_(i)-

(0-5569)
, ,
8 200 (0.6705)'
2-46
Thus, second-order theory predicts a pressure,
p = 1395 + 14 = 1409 lbs/ft^ .
(e) Using Equation 2-38, the energy in one wavelength per unit width
of crest given by the first-order theory is:
^^„^^ft-lbs
E = ^
pgtfL
=
(2) (32.2) (4)^200)
^^^-^^ i^jL_:^ .'
= 25,800 —r- •
8 8 ft
Evaluation of the hydrostatic pressure component (1288 lbs/ft^)

indicates that Airy theory gives a dynamic component of 107 lbs/ft^
while Stokes theory gives 121 lbs/ft^. Stokes theory shows a
dynamic pressure component about 13 percent greater than Airy
theory.
2.26 CNOIDAL WAVES
Long, finite-amplitude waves of permanent form propagating in shallow

water are frequently best described by anoidal wave theory. The existence
in shallow water of such long waves of permanent form may have first been
recognized by Boussinesq (1877) However, the theory was originally
,
.
developed by Korteweg and DeVries (1895) The term anoidal is used since
.
the wave profile is given by the Jacobian elliptical cosine function

usually designated en .
In recent years, cnoidal waves have been studied by many investigators.

Wiegel (1960) summarized much of the existing work on cnoidal waves, and
presented the principal results of Korteweg and DeVries (1895) and Keulegan
and Patterson (1940) in a more usable form. Masch and Wiegel (1961)
presented such wave characteristics as- length, celerity and period in
tabular and graphical form, to facilitate application of cnoidal theory.
The approximate range of validity for the cnoidal wave theory as

determined by Laitone (1963) and others is d/L < 1/8, and the Ursell
parameter, L^H/d^ > 26. (See Figure 2-7.) As wavelength becomes long,
and approaches infinity, cnoidal wave theory reduces to the solitary wave
theory which is described in the next section. Also, as the ratio of wave
height to water depth becomes small (infinitesimal wave height), the wave
profile approaches the sinusoidal profile predicted by the linear theory.
Description of local particle velocities, local particle accelerations,

wave energy, and wave power for cnoidal waves is difficult; hence their
description is not included here, but can be obtained in graphical form
from Wiegel (1960, 1964) and Masch (1964).
Wave characteristics are described in parametric form in terms of the

modulus k of the elliptic integrals. While k itself has no physical
2-4-
significance, it is used to express the relationships between the various
wave parameters. Tabular presentations of the elliptic integrals and
other important functions can be obtained from the above references. The
ordinate of the water surface Xg measured above the bottom is given by
Jt
+ Hcn^ 2K(k)^-| U (2-59a)
where y^ is the distance from the bottom to the wave trough, en is the
elliptic cosine function, K(k) is the complete elliptic integral of the
first kind, and k is the modulus of the elliptic integrals. The argument
of cn^ is frequently denoted simply by ( ), thus. Equation 2-59a above
can be written as
y^ + H cnM )
(2 -59b)
The elliptic cosine is a periodic function where cn2[2K(k) ((x/L) - (t/T))]

has a maximum amplitude equal to unity. The modulus k is defined over
the range between and 1. When k = 0, the wave profile becomes a
sinusoid as in the linear theory, and when k = 1, the wave profile
becomes that of a solitary wave.
The distance from the bottom to the wave trough. yt' as used in
Equations 2-59a and b, is given by
yj
d d
H
d
—
16d'
3V
— ,
K(k) [K(k)
,
- E(k)] + 1-7Hd (2-60)
where is the distance from the bottom to the crest and E(k) is the
complete elliptic integral of the second kind. Wavelength is given by
'I6d^
L = kK(k) ,
(2-61)
3H
and wave period by
kK(k)
(2-62)
H E(k)\
1 +
y? K(k)
Cnoidal waves are periodic and of permanent form thus L = CT.
Pressure under a cnoidal wave at any elevation y, above the bottom

depends on the local fluid velocity, and is therefore complex. However,
it may be approximated in a hydrostatic form as
p = Pg (y^
- y)
(2-63)
2-48
that is, the pressure distribution may be assumed to vary linearly from
pgyg at the bed to zero at the surface.
Figures 2-9 and 2-10 show the dimensionless cnoidal wave surface
profiles for various values of the square of the modulus of the elliptic
integrals k^, while Figures 2-11 through 2-15 present dimensionless
plots of the parameters which characterize cnoidal waves. The ordinates
of Figures 2-11 and 2-12 should be read with care, since values of k^
are extremely close to 1.0 (k^ = l-lO"^ = 1-0. 1 = 0.99) It is the
.
exponent a of k^ = 1-10"" that varies along the vertical axis of

Figures 2-11 and 2-12.
Ideally, shoaling computations might best be performed using cnoidal

wave theory since this theory best describes wave motion in relatively
shallow (or shoaling) water. Simple, completely satisfactory procedures
for applying cnoidal wave theory are not available. Although linear wave
theory is often used, cnoidal theory may be applied by using figures such
as 2-9 through 2-15.
The following problem will illustrate the use of these figures.
************** EXAMPLE PROBLEM **************

GIVEN A wave traveling in water depth of d = 10 feet, with a period of
:
T = 15 seconds, and a height of H = 2.5 feet.
FIND :
(a) Using cnoidal wave theory, find the wavelength L and compare this
length with the length determined using Airy theory.
(b) Determine the celerity C. Compare this celerity with the celerity
determined using Airy theory.
(c) Determine the distance above the bottom of the wave crest (y^)
and wave trough (y^)
(d) Determine the wave profile.
SOLUTION:
(a) Calculate
H
T
d
= —
2.5
10
= 0.25
and
''
^ "' Jw - ^'''
2-49
c
o
o
3
in
o
o
\—
CL
a>
o
o
>*-
cn
>
o
•o
o
oc
a>
I
CM
a>
k.
3
o
c
o
o
cr
3
o
CO
o
a.
a>
o
o
3
CO
a>
>
o
oo
o
c
2-51
-10" 100
-10 10
-10"
-0.1
l-IO
40 60 100 200 400 600 1,000 2,000 4,000
J /g (oft«r WitstI, I9«0)
V d
Figure 2-11. Relationship Between k*^, H /d and T vg/d
2-52
to
^«.
X
^-^
c
o
T3
I
c
Q>

aJ
m
(/>
Q.
c
o
_o
cm
ro
CM
a>
i_
3
<»
340 1
——
1
320
300
280
260
240
I
220
CM
200
180
160
140
120
» -o
100
80
601-
40
20
to
o
X
'^
Dc
o
c
Qi
a>
00
to
c
o
13
M6A
V (»l)3 I / I H ^'/ =
2-56
From Figure 2-11, entering with H/d and T/g/d', determine the square
of the modulus of the complete elliptical integrals, k^.
F = 1 - 10-^-9-y
Entering Figure 2-12 with the value of k^ gives
or
(c) The percentage of the wave height above the SWL may be determined
from Figure 2-13. Entering the figure with L^H/d^ = 190, the
value of (y^ - d)/H is found to be 0.833 or 83.3 percent.
Therefore
Yc
= 0.833 H + d.
y^ = 0.833(2.5) + 10 = 2.083 + 10 = 12.083 feet .
Also from Figure 2-13,
(y, - d)
+ 1 = 0.833
H
thus
y^
= (0.833- 1) (2.5) + 10 = 9.583 ft.
(d) The dimensionless wave profile is given on Figure 2-9 and is

approximately the one drawn for k^ = i - 10"^. The results
obtained in (c) above can also be checked by using Figure 2-9.
For the wave profile obtained with k^ = 1 - 10"'*, it is seen that
the SWL is approximately 0.17 H above the wave trough or 0.83 H
below the wave crest.
The results for the wave celerity determined under (b) above
can now be checked with the aid of Figure 2-15. Calculate
0.261 .
The difference between this number and the 18.38 ft/sec calculated
under (b) above is the result of small errors in reading the
curves
2.27 SOLITARY WAVE THEORY
Waves considered in the previous sections were oscillatory or nearly

oscillatory waves. The water particles move backward and forward with the
passage of each wave, and a distinct wave crest and wave trough are evident,
A solitary wave is neither oscillatory nor does it exhibit a trough. In
the pure sense, the solitary wave form lies entirely above the Stillwater
level. The solitary wave is a wave of translation relative to the water
mass.
Russell (1838, 1845) first recognized the existence of a solitary

wave. The original theoretical developments were made by Boussinesq
(1872), Lord Rayleigh (1876), and McCowan (1891), and more recently by
Keulegan and Patterson (1940), Keulegan (1948), and Iwasa (1955).
In nature it is difficult to form a truly solitary wave, because at

the trailing edge of the wave there are usually small dispersive waves.
However, long waves such as tsunamis and waves resulting from large dis-
placements of water caused by such phenomena as landslides, and earthquakes
sometimes behave approximately like solitary waves. When an oscillatory
wave moves into shallow water, it may often be approximated by a solitary
wave, (Munk, 1949). As an oscillatory wave moves into shoaling water, the
wave amplitude becomes progressively higher; the crests become shorter and
more pointed, and the trough becomes longer and flatter.
The solitary wave is a limiting case of the cnoidal wave. When k^ =

1, K(k) = K(l) =
oo, and the elliptic cosine reduces to the hyperbolic
secant function, y- = d, and Equation 2-59 reduces to
H
d + H sech^ (x - Ct)
or
3 H_
1? = H sech^ (x - Ct) (2-64)
4 d^
where the origin of x is at the wave crest. The volume of water within
the wave above the still water level per unit crest width is
16 %
V = d' H (2-65)
2-59
An equal amount of water per unit crest length is transported
forward past a vertical plane that is perpendicular to the direction of
wave advance. Several relations have been presented to determine the
celerity of a solitary wave; these equations differ depending on the
degree of approximation. Laboratory measurements by Daily and Stephan
(1953) indicate that the simple expression
C = Vg (H + d)" ,
(2-66)
gives a reasonably accurate approximation to the celerity.
The water particle velocities for a solitary wave, as found by

McCowan (1891) and given by Munk (1949), are
1 + cos(My/d) cosh(Mx/d) ^^ ^^^

u = CN —'- (2-67)
[cos(My/d) + cosh(Mx/d)]' '
sin(My/d) sinh (Mx/d)

w = CN (2-68)
[cos(My/d) + cosh(Mx/d)]^
where M and N are the functions of H/d shown on Figure 2-16, and y
is measured from the bottom. The expression for horizontal velocity u,
is often used to predict wave forces on marine structures sited in shallow
water. The maximum velocity u^nox' occurs when x and t are both equal
to zero; hence,
CN
~^^'
"'"'« " 1 + cos(My/d)
Total energy in a solitary wave is about evenly divided between

kinetic and potential energy. Total wave energy per unit crest width is,
E = ^^pgH3/2d3/2 , (2-70)
and the pressure beneath a solitary wave depends upon the local fluid
velocity as does the pressure under a cnoidal wave; however, it may be
approximated by
P = Pg(y,-y). (2-71)
Equation 2-71 is identical to that used to approximate the pressure

beneath a cnoidal wave.
As a solitary wave moves into shoaling water it eventually becomes

unstable and breaks. McCowan (1891) assumed that a solitary wave breaks
2-60
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
LI
Relative wave height -r-

(afterMunk,l949)
Figure 2-16. Functions M and N in Solitary Wave Theory
2-61
when the water particle velocity at the wave crest becomes equal to the
wave celerity. This occurs when
(2-72)
Laboratory investigations have shown that the value of (H/d)^^^^ = 0-78

agrees better with observations for oscillatory waves than for solitary
waves. Ippen and Kulin (1954) and Galvin (1969) have shown that the near-
shore slope has a substantial effect on this ratio. Other factors such as
bottom roughness may also be involved. For slopes of 0.0, 0.05, 0.10, and
0.20, Galvin found that H/d ratios were approximately equal to 0.83,
1.05, 1.19, and 1.32, respectively. Thus, it must be concluded that for
some conditions. Equation 2-72 is unsatisfactory for predicting breaking
depth. Further discussion of the breaking of waves with experimental
results is in Section 2.6 - BREAKING WAVES.
2.28 STREAM FUNCTION WAVE THEORY
In recent years, numerical approximations to solutions of hydrodynamic

equations describing wave motion have been proposed and developed by Dean
(1965a, 1965b, 1967) and Monkmeyer (1970). The approach by Dean, termed
a symmetric, stream function theory, is a nonlinear wave theory which is
similar to higher order Stokes' theories. Both are constructed of suras of
sine or cosine functions that satisfy the original differential equation
(Laplace equation). The theory, however, determines the coefficient of
each higher order term so that a best fit, in the least-squares sense, is
obtained to the theoretically posed, dynamic, free-surface boundary con-
dition. Assumptions made in the theory are identical to those made in the
development of the higher-order Stokes' solutions. Consequently, some of
the same limitations are inherent in the stream function theory; however,
it represents a better solution to the equations used to approximate the
wave phenomena. More important is that the stream function representation
appears to better predict some of the wave phenomena observed in laboratory
wave studies (Dean and L^Mehaut^, 1970), and may possibly describe naturally
occurring wave phenomena better than other theories
The long tedious computations involved in evaluating the terms of the

series expansions that make up the higher-order stream function solutions,
make it desirable to use tabular or graphical presentations of the
solutions. These tables, their use and range of validity have been
developed by Dean (1973).
2.3 WAVE REFRACTION
2.31 INTRODUCTION
Equation 2-2 shows that wave celerity depends on the depth of water
in which the wave propagates. If the wave celerity decreases with depth,
wavelength must decrease proportionally. Variation in wave velocity occurs
2-62
along the crest of a wave moving at an angle to underwater contours because
that part of the wave in deeper water is moving faster than the part in
shallower water. This variation causes the wave crest to bend toward align-
ment with the contours, (See Figure 2-17.) This bending effect, called
refraction, depends on the relation of water depth to wavelength. It is
analogous to refraction for other types of waves such as, light and sound.
In practice, refraction is important for several reasons such as:
(1) Refraction, coupled with shoaling, determines the wave height

in any particular water depth for a given set of incident deepwater wave
conditions, that is wave height, period, and direction of propagation in
deep water. Refraction therefore has significant influence on the wave
height and distribution of wave energy along a coast.
(2) The change of wave direction of different parts of the wave

results in convergence or divergence of wave energy, and materially affects
the forces exerted by waves on structures
(3) Refraction contributes to the alteration of bottom topography

by its effects on the erosion and deposition of beach sediments. Munk and
Traylor (1947) confirmed earlier work by many indicating the possible inter-
relationships between refraction, wave energy distribution along a shore,
and the erosion and deposition of beach materials.
(4) A general description of the nearshore bathymetry of an area

can sometimes be obtained by analyzing aerial photography of wave refraction
patterns. While the techniques for performing such analyses are not well
developed, an experienced observer can obtain a general picture of simple
bottom topography.
In addition to refraction caused by variations in bathymetry, waves

may be refracted by currents, or any other phenomenon which causes one part
of a wave to travel slower or faster than another part. At a coastal inlet,
refraction may be caused by a gradient in the current. Refraction by a
current occurs when waves intersect the current at an angle. The extent
to which the current will refract incident waves depends on the initial
angle between the wave crests and the direction of current flow, the
characteristics of the incident waves, and the strength of the current.
In at least two situations, wave refraction by currents may be of practical
importance. At tidal entrances, ebb currents run counter to incident waves
and consequently increase wave height and steepness. Also, major ocean
currents such as the Gulf Stream may have some effect on the height, length
and direction of approach of waves reaching the coasts. Quantitative
evaluation of the effects of refraction by currents is difficult. Addi-
tional research is needed in this area. No detailed discussion of this
problem will be presented here, but an introduction is presented by
Johnson (1947).
The decrease in wave celerity with decreasing water depth can be con-
sidered an analog to the decrease in the speed of light with an increase in
2-63
Figure 2-17. Wave Refraction at Westhampton Beach, Long Island, New York
2-64
the refractive index of the transmitting medium. Using this analogy,
O'Brien (1942) suggested the use of Snell's law of geometrical optics for
solving the problem of water-wave refraction by changes in depth. The
validity of this approach has been verified experimentally by Chien (1954),
Ralls (1956), and Wiegel and Arnold (1957). Chao (1970) showed analyti-
cally that Format 's principle and hence Snell's law followed from the
governing hydrodynamic equations, and was a valid approximation when applied
to the refraction problem. Generally, two basic techniques of refraction
analysis are available - graphical and numerical. Several graphical pro-
cedures are available, but fundamentally all methods of refraction analyses
are based on Snell's law.
The assumptions usually made are:
(1) Wave energy between wave rays or orthogonals remains constant.

(Orthogonals are lines drawn perpendicular to the wave crests, and extend
in the direction of wave advance.) (See Figure 2-17.)
(2) Direction of wave advance is perpendicular to the wave crest,

that is, in the direction of the orthogonals.
(3) Speed of a wave of given period at a particular location

depends only on the depth at that location.
(4) Changes in bottom topography are gradual.
(5) Waves are long-crested, constant-period, small -amplitude, and

monochromatic.
(6) Effects of currents, winds, and reflections from beaches, and

underwater topographic variations, are considered negligible.
2.32 GENERAL - REFRACTION BY BATHmETRY
In water deeper than one-half the wavelength, the hyperbolic tangent

function in the formula
is nearly equal to unity, and Equation 2-2 reduces to
" 2-n
In this equation, the velocity C^, does not depend on depth; therefore
in those regions deeper than one-half the wavelength (deep water) , refrac-
tion by bathymetry will not be significant. Where the water depth is
between 1/2 and 1/25 the wavelength (transitional water) , and in the region
where the water depth is less than 1/25 the wavelength (shallow water),
refraction effects may be significant. In transitional water, wave velocity
2-65
must be computed from Equation 2-2; in shallow water, tanh(2Trd/L) becomes
nearly equal to 2TTd/L and Equation 2-2 reduces to Equation 2-9.
^2 _ gd or C = (gd)*^ . (2-9)
Both Equations 2-2 and 2-9 show the dependence of wave velocity on depth.
To a first approximation, the total energy in a wave per unit crest width
may be written as
(2-38)
8
It has been noted that not all of the wave energy E is transmitted
forward with the wave; only one-half is transmitted forward in deep water.
The amount of energy transmitted forward for a given wave remains nearly
constant as the wave moves from deep water to the breaker line if energy
dissipation due to bottom friction (K^ = 1.0), percolation and reflected
wave energy is negligible.
In refraction analyses, it is assumed that for a wave advancing toward

shore, no energy flows laterally along a wave crest; that is the transmitted
energy remains constant between orthogonals. In deep water the wave energy
transmitted forward across a plane between two adjacent orthogonals (the
average energy flux) is
ô = \KKô' (2-73)
where h^ is the distance between the selected orthogonals in deep water.

The subscript o always refers to deepwater conditions. This power may
be equated to the energy transmitted forward between the same two orthogonals
in shallow water
P = nbEC, (2-74)
where b is the spacing between the orthogonals in the shallower water.

Therefore, (1/2) b^ E^C^ = nb EC, or
\/b„\/C„\
(2-75)
From Equation 2-39,
(2-76)
2-66
and combining Equations 2-75 and 2-76,
H_
(2-77)
Ho
The term /(1/2) (1/n) (C^/C)' is known as the shoaling coefficient Kg

or H/H^. This shoaling coefficient is a function of wavelength and water
depth. Kg and various other functions of d/L, such as 2T7d/L, 47rd/L,
tanh(27rd/L) and sinh(4TTd/L) are tabulated in Appendix C, (Table C-1 for
,
even increments of d/L^, and Table C-2 for even increments of d/L).
Equation 2-77 enables determination of wave heights in transitional

or shallow water, knowing the deepwater wave height when the relative
spacing between orthogonals can be determined. The square root of this
relative spacing, /b^/b', is the refraction coefficient K^.
Various methods may be used for constructing refraction diagrams.

The earliest approaches required the drawing of successive wave crests.
Later approaches permitted the immediate construction of orthogonals,
and also permitted moving from the shore to deep water (Johnson, O'Brien
and Isaacs, 1948), (Arthur, et al , 1952), (Kaplan, 1952) and (Saville
.
and Kaplan, 1952) .
The change of direction of an orthogonal as it passes over relatively

simple hydrography may be approximated by
sina, = M- sin a (Snell'slaw) (2-78)
where
a^ is the angle a wave crest (the perpendicular to an orthogonal)

makes with the bottom contour over which the wave is passing,
a^ is a similar angle measured as the wave crest (or orthogonal)

passes over the next bottom contour,
C, is the wave velocity (Equation 2-2) at the depth of the first

contour, and
C2 is the wave velocity at the depth of the second contour.
From this equation, a template may be constructed which will show the
angular change in a that occurs as an orthogonal passes over a
particular contour interval, and construct changed-direction orthogonal.
Such a template is shown in Figure 2-18. In application to wave refrac-
tion problems, it is simplest to construct this template on a transparent
material.
2-67
<J>a)^-^r)^f^vK>cJ — o
o o oo o
Refraction may be treated analytically at a straight shoreline with
parallel offshore contours, by using Snell's law directly:
... -ig ...

where a. is the angle between the wave crest and the shoreline, and ot^^
is the angle between the deepwater wave crest and the shoreline.
For example, if cxq = 30° and the period and depth of the wave
are such that C/C^ = 0.5, then
a = sin-' [0.5 (0.5)] = 14.5 degrees
cos a = 0.968
and
cosa^ = 0.866
cosaA'^ / 0.866 \^
= 0.945
cos a / \ 0.968
Figure 2-19 shows the relationships between a, a^, period, depth, and
K^ in graphical form.
2.321 Procedures in Refraction Diagram Construction - Orthogonal Method .
Charts showing the bottom topography of the study area are obtained. Two
or more charts of differing scale may be required, but the procedures are
identical for charts of any scale. Underwater contours are drawn on the
chart, or on a tracing paper overlay, for various depth intervals. The
depth intervals chosen depend on the degree of accuracy desired. If
overlays are used, the shoreline should be traced for reference. In
tracing contours, small irregularities must be smoothed out, since bottom
features that are comparatively small in respect to the wavelength do not
affect the wave appreciably.
The range of wave periods and wave directions to be investigated is

determined by a hindcasting study of historical weather charts or from
other historical records relating to wave period and direction. For each
wave period and direction selected, a separate diagram must be prepared.
C1/C2 values for each contour interval may then be marked between contours,
The method of computing C1/C2 is illustrated by Table 2-2; a tabulation
of C1/C2 for various contour intervals and wave periods is given in
Table C-4 of Appendix C.
To construct orthogonals from deep to shallow water, the deepwater

direction of wave approach is first determined. A deepwater wave front
(crest) is drawn as a straight line perpendicular to this wave direction.
2-69
(O
and suitably spaced orthogonals are drawn perpendicular to this wave front
and parallel to the chosen direction of wave approach. Closely spaced
orthogonals give more detailed results than widely spaced orthogonals.
These lines are extended to the first depth contour shallower than L^/2
where L^ (in feet) = 5.12 T^.
TABLE 2-2 EXAMPLE COMPUTATIONS OF VALUES OF

Cj/C2 FOR REFRACTION ANALYSIS
T = 10 seconds
Templote Orthogonal Line
Contour
"-X
20'"^
Tangent to Mid- ^ 'Ci

Contour
'"'^Wc.-r-l- /~
bO \^
Incoming Orttiogonol
Turning Point-
Template Orttiogonal' Line

Contour Turned Orthogonal
20 ^^
II- 1071
Tangent to Mid-
Contour -^
ô \<^ 1.0461
Incoming Orthogonal
NOTE: The template has been turned obout R until the volue '/r ; 1.045
intersects the tangent to the mid-contour. The templote "orthogonal line
lies in the direction of the turned orthogonol. This direction is to be laid
off ot some point " B"on the incoming orthogonol which is equidistant
from the two contours along the incoming and outgoing orthogonals.
Figure 2-20. Use of the Refraction Template
2-72
(c) Rotate the template about the turning point until the C2/C2 value
corresponding to the contour interval being crossed intersects the tangent
to the midcontour. The orthogonal line on the chart now lies in the direc-
tion of the turned orthogonal on the template (Figure 2-20 bottom)
(d) Place a triangle along the base of the template and construct
a perpendicular to it so that the intersection of the perpendicular with
the incoming orthogonal is midway between the two contours when the dis-
tances are measured along the incoming orthogonal and the perpendicular
(See Point B in Figure 2-20 bottom) . Note that this point is not neces-
sarily on the midcontour line. This line represents the turned orthogonal;
(e) Repeat the above steps for successive contour intervals
If the orthogonal is being constructed from shallow to deep water, the

same procedure may be used, except that C2/C]^ values are used instead of
A template suitable for attachment to a drafting machine can be made.

Palmer (1957) ,and may make the procedure simpler if many diagrams are to
be used.
2.323 Procedure when a is Greater than 80 Degrees - The R/J Method . In

any depth, when a becomes greater than 80 degrees, the above procedure
cannot be used. The orthogonal no longer appears to cross the contours,
but tends to run almost parallel to them. In this case, the contour
interval must be crossed in a series of steps. The entire interval is
divided into a series of smaller intervals. At the midpoint of the indi-
vidual subintervals, orthogonal-angle turnings are made.
Referring to Figure 2-21, the interval to be crossed is divided into

segments or boxes by transverse lines. The spacing R, of the transverse
lines is arbitrarily set as a ratio of the distance J, between the con-
tours. For the complete interval to be crossed, C2/C1 is computed or
found from Table C-4 of Appendix C. (C2/C^, not 0.-^/^2-^
On the template (Figure 2-18), a graph showing orthogonal angle

turnings Aa, is plotted as a function of the C2/C1 value for various
values of the ratio R/J. The Aa value is the angle turned by the in-
coming orthogonal in the center of the subinterval
The orthogonal is extended to the middle of the box, Aa is read

from the graph, and the orthogonal turned by that angle. The procedure
is repeated for each box in sequence, until a at a plotted or interpo-
lated contour becomes smaller than 80 degrees. At this point, this method
of orthogonal construction must be stopped, and the preceding technique
for a smaller than 80 degrees used, otherwise errors will result.
2.324 Refraction Fan Diagrams . It is often convenient, especially where

sheltering land forms shield a stretch of shore from waves approaching in
certain directions, to construct refraction diagrams from shallow water
2-73
J = Distance between contours at turning points,*
R = Distance along orttfogonal
T = 12 seconds
Lo= 737 ft.
For contour interval from 40fm to 30 fm C /C = 1.045, 0/0=0.957

12 2 1
Acc = 20-25
Figure 2-21. Refraction Diagram Using R/J Method
2-74
toward deep water. In such cases, a sheaf or fan of orthogonals may be
projected seaward in directions some 5 or 10 degrees apart. See Figure
2-22a. With the deepwater directions thus determined by the individual
orthogonals, companion orthogonals may be projected shoreward on either
side of the seaward projected ones to determine the refraction coefficient
for the various directions of wave approach. (See Figure 2 -22b.)
2.325 Other Graphical Methods of Refraction Analysis . Another graphical

method for the construction of refraction diagrams is the wave-front
method (Johnson, et al., 1948). This method is particularly applicable
to very long waves where the crest alignment is also desired. The method
is not presented here, where many diagrams are required, because, where
many diagrams are required, it is overbalanced by the advantages of the
orthogonal method. The orthogonal method permits the direct construction
of orthogonals and determination of the refraction coefficient without the
intermediate step of first constructing successive wave crests. Thus,
when the wave crests are not required, significant time is saved by using
the orthogonal method.
2.326 Computer Methods for Refraction Analysis . Harrison and Wilson (1964)
developed a method for the numerical calculation of wave refraction by use
of an electronic computer. Wilson (1966) extended the method so that, in
addition to the numerical calculation, the actual plotting of refraction
diagrams is accomplished automatically by use of a computer. Numerical
methods are a practical means of developing wave refraction diagrams when
an extensive refraction study of an area is required, and when they can
be relied upon to give accurate results. However, the interpretation of
computer output requires care, and the limitations of the particular scheme
used should be considered in the evaluation of the results. For a dis-
cussion of some of these limitations, see Coudert and Raichlen (1970).
For additional references, the reader is referred to the works of Keller
(1958), Mehr (1962), Griswold (1963), Wilson (1966), Lewis, et al., (1967),
Dobson (1967), Hardy (1968), Chao (1970), and Keulegan and Harrison (1970),
in which a number of available computer programs for calculation of refrac-
tion diagrams are presented. Most of these programs are based on an algo-
rithm derived by Munk and Arthur (1951) and, as such, are fundamentally
based on the geometrical optics approximation. (Fermat's Principle.)
2.327 Interpretation of Results and Diagram Limitations . Some general

observations of refraction phenomena are illustrated in Figures 2-23, 24,
and 25. These figures show the effects of several common bottom features
on passing waves. Figure 2-23 shows the effect of a straight beach with
parallel evenly spaced bottom contours on waves approaching from an angle.
Wave crests turn toward alignment with the bottom contours as the waves
approach shore. The refraction effects on waves normally incident on a
beach fronted by a submarine ridge or submarine depression are illustrated
in Figure 2-24a and 2-24b. The ridge tends to focus wave action toward
the section of beach where the ridge line meets the s horeli ne. The ortho-
gonals in this region are more closely spaced; hence-v/b^/o is greater
than 1.0 and the waves are higher than they would be if no refraction
occurred. Conversely, a submarine depression will cause orthogonals to
2-75
u
C
o
H
P
o
ni
^(
a>
ai
a>
P-
X
I
C
uu
O
O
t/)
(N
I
(Nl
2-76
Beach Line
'Breakers
»*JL-'X«'L^xxx,xxxr;x)i»^*»* _x1
Figure 2-23. Refraction Along a Straight Beach with Parallel

Bottom Contours
Beach Line Beach Line
Contours Contours
Orthogonals Orthogonals
(a) (b)
Figure 2-24. Refraction by a Submarine Ridge (a) and Submarine
Canyon (b)
Figure 2-25. Refraction Along an Irregular Shoreline
2-77
diverge, resulting in low heights at the shore. (bî/b less than 1.0.)
Similarly, heights will be greater at a headland than in a bay. Since
the wave energy contained between two orthogonals is constant, a larger
part of the total energy expended at the shore is focused on projections
from the shoreline; consequently, refraction straightens an irregular
coast. Bottom topography can be inferred from refraction patterns on
aerial photography. The pattern in Figure 2-17 indicates the presence
of a submarine ridge.
Refraction diagrams can provide a measure of changes in waves

approaching a shore. However, the accuracy of refraction diagrams is
limited by the validity of the theory of construction and the accuracy
of depth data. The orthogonal direction change (Equation 2-78) is
derived for straight parallel contours. It is difficult to carry an
orthogonal accurately into shore over complex bottom features (Munk and
Arthur, 1951). Moreover, the equation is derived for small waves moving
over mild slopes.
Dean (1973) considers the combined effects of refraction and shoal-

ing including nonlinearities applied to a slope with depth contours
parallel to the beach but not necessarily of constant slope. He finds
that non-linear effects can significantly increase (in coirparison with
linear theory) both amplification and angular turning of waves of low
steepness in deep water.
Strict accuracy for height changes cannot be expected for slopes

steeper than 1:10, although model tests have shown that direction
changes nearly as predicted even over a vertical discontinuity (Wiegel
and Arnold, 1957). Accuracy where orthogonals bend sharply or exhibit
extreme divergence or convergence is questionable because of energy
transfer along the crest. The phenomenon has been studied by Beitinjani
and Brater (1965), Battjes (1968) and Whalin (1971). Where two ortho-
gonals meet, a caustic develops. A caustic is an envelope of ortho-
gonal crossings, caused by convergence of wave energy at the caustic
point. An analysis of wave behavior near a caustic is not available;
however, qualitative analytical results show that wave amplitude decays
exponentially away from a caustic in the shadow zone, and there is a
phase shift of it/2 across the caustic (Whalin 1971). Wave behavior
near a caustic has also been studied by Pierson (1950), Chao (1970) and
others. Little quantitative information is available for the area
beyond a caustic.
2.328 Refraction of Ocean Waves . Unlike Monochromatic waves, actual

ocean waves are more complicated. Their crest lengths are short; their
form does not remain permanent; and their speed, period, and direction
of propagation vary from wave to wave.
2-78
Pierson (1951), Longuet-Higgins (1957), and Kinsman (1965), have sug-
gested a solution to the ocean-wave refraction problem. The sea surface
waves in deep water become a number of component monochromatic waves, each
with a distinct frequency and direction of propagation. The energy spec-
trum for each component may then be found and the conventional refraction
analysis techniques applied. Near the shore, the wave energy propagated
in a particular direction is approximated as the linear sum of the spectra
of wave components of all frequencies refracted in the given direction from
all of the deepwater directional components.
The work required for this analysis, even for a small number of indi-
vidual components, is laborious and time consuming. More recent research
by Borgman (1969) and Fan and Borgman (1970), has used the idea of direc-
tional spectra which may provide a technique for solving complex refraction
problems more rapidly.
2.4 WAVE DIFFRACTION
2.41 INTRODUCTION
Diffraction of water waves is a phenomenon in which energy is trans-

ferred laterally along a wave crest. It is most noticeable where an other-
wise regular train of waves is interrupted by a barrier such as a breakwater
or an islet. If the lateral transfer of wave energy along a wave crest and
across orthogonals did not occur, straight, long-crested waves passing the
tip of a structure would leave a region of perfect calm in the lee of the
barrier, while beyond the edge of the structure the waves would pass un-
changed in form and height. The line separating two regions would be a
discontinuity. A portion of the area in front of the barrier would, how-
ever, be disturbed by both the incident waves and by those waves reflected
by the barrier. The three regions are shown in Figure 2-26a for the hypo-
thetical case if diffraction did not occur, and in Figure 2-26b for the
actual phenomenon as observed. The direction of the lateral energy trans-
fer is also shown in Figure 2-26a. Energy flow across the discontinuity
is from Region II into Region I. In Region III, the superposition of
incident and reflected waves results in the appearance- of short-crested
waves if the incident waves approach the breakwater obliquely. A partial
standing wave will occur in Region III if the waves approach perpendicular
to the breakwater.
This process is also similar to that for other types of waves, such
as light or sound waves.
Calculation of diffraction effects is important for several reasons.

Wave height distribution in a harbor or sheltered bay is determined to
some degree by the diffraction characteristics of both the natural and
manmade structures affording protection from incident waves. Therefore,
a knowledge of the diffraction process is essential in planning such
facilities. Proper design and location of harbor entrances to reduce
such problems as silting and harbor resonance also require a knowledge
of the effects of wave diffraction. The prediction of wave heights near
2-79
Figure 2-26a. Wave Incident on a Breakwater
No Diffraction
Figure 2- 26b. Wave Incident on a Breakwater

Diffraction Effects
2-80
the shore is affected by diffraction caused by naturally occurring changes
in hydrography. An aerial photograph illustrating the diffraction of
waves by a breakwater is shown in Figure 2-27.
Putnam and Arthur (1948) presented experimental data verifying a

method of solution proposed by Penny and Price (1944) for wave behavior
after passing a single breakwater. Wiegel (1962) used a theoretical
approach to study wave diffraction around a single breakwater. Blue and
Johnson (1949) dealt with the problem of the behavior of waves after
passing through a gap, as between two breakwater arms.
The assumptions usually made in the development of diffraction

theories are:
(1) Water is an ideal fluid, i.e., inviscid and incompressible.
(2)Waves are of small -amplitude and can be described by linear

wave theory.
(3)Flow is irrotational and conforms to a potential function which

satisfies the Laplace equation.
(4) Depth shoreward of the breakwater is constant.
2.42 DIFFRACTION CALCULATIONS
2.421 Waves Passing a Single Breakwater From a presentation by Wiegel

.
(1962), diffraction diagrams have been prepared which, for a uniform depth
adjacent to an impervious structure, show lines of equal wave height re-
duction. These diagrams are shown in Figures 2-28 through 2-39; the graph
coordinates are in units of wavelength. Wave height reduction is given in
terms of a diffraction coefficient K' which is defined as the ratio of a
wave height H, in the area affected by diffraction to the incident wave
height H^, in the area unaffected by diffraction. Thus, H and H^ are
determined by H = K'H^.
The diffraction diagrams shown in Figures 2-28 through 2-39 are con-
structed in polar coordinate form with arcs and rays centered at the struc-
ture's tip. The arcs are spaced one radius -wavelength unit apart and rays
15 apart In application, a given diagram must be scaled up or down so
.
that the particular wavelength corresponds to the scale of the hydrographic

chart being used. Rays and arcs on the refraction diagrams provide a
coordinate system that makes it relatively easy to transfer lines of
constant K' on the scaled diagrams.
When applying the diffraction diagrams to actual problems, the wave-

length must first be determined based on the water depth at the tip of the
structure. The wavelength L, in water depth dg, may be found by com-
puting dg/L^ = dg/5.12T2 and using Appendix C, Table C-1 to find the
corresponding value of dg/L. Dividing dg by dg/L will give the shallow
water wave length L. It is then useful to construct a scaled diffraction
2-81
Figure 2-21 . Wave Diffraction at Channel Islands Harbor Breakwater,
California
2-82
r——
C
<
>
o
o
00
o06 O
<4-i
•H
Q
u
3
•H
2-84
2-85
2-86
C
<
0)
>
C
o
o
U
4-1
0)
>
to
I
CM
0)
3
bO
2-87
o
iT)
O
00
<
ID
>
03
o
O
6
U
c
o
•H
o06 p
o
(D
o
o
o
o.
Q.
 U
o
•H
o
c
o
«3 O
2-88
t—<
c
<
>
o
o
00
c
o
4-1
•H
Q
>
to
I
<N
D
U
3
•H
2-89
o
o
(N
C
o
• H
Q
CM
V
3
00
•H
2-90
C
<
>
o
LO
to
OS
o
o06 H
+J
O
nj
U
1-1
4-1
•H
Q
>
to
3
00
2-91
C
<
>
B
U
o
• H
+J
O
nJ
u
4-1
4^
>
00
^0
3
H00
2-93
o
diagram overlay template to correspond to the hydrographic chart being
used. In constructing this overlay, first determine how long each of its
radius-wavelength units must be. As noted previously, one radius -wave length
unit on the overlay must be identical to one wavelength on the hydrographic
chart. The next step is to construct and sketch all overlay rays and arcs
on clear plastic or translucent paper. This allows penciling in of the
scaled lines of equal K for each angle of wave approach that may be
considered pertinent to the problem. Thus, after studying the wave field
for one angle of wave approach, K lines may be erased for a subsequent
analysis of a different angle of wave approach.
The diffraction diagrams in Figures 2-28 through 2-39 show the break-
water extending to the right as seen looking toward the area of wave dif-
fraction; however, for some problems the structure may extend to the left.
All diffraction diagrams presented may be reversed by simply turning the
transparency over to the opposite side.
Figure 2-40 illustrates the use of a template overlay. Also indicated

is the angle of wave approach which is measured counterclockwise from the
breakwater. This angle would be measured clockwise from the breakwater if
the diagram were turned over. Figure 2-40 also shows a rectangular coordi-
nate system with distance expressed in units of wavelength. Positive
x direction is measured from the structure's tip along the breakwater and
positive y direction is measured into the diffracted area.
Template Overlay
Angle of Wave Approach
Breakwater
O.I. I. '././. 1. 1. 1. '.I. '.I. J. '7^
I.
f'/'/'/'r /=z= -•-X
Wave Crests
Figure 2-40. Diffraction for a Single Breakwater Normal Incidence
The following problem illustrates determination of a single wave

height in the diffraction area.
************** EXAMPLE PROBLEM **************
GIVEN Waves with a period of T = 8 seconds and height of H = 10 feet
:
impinge upon a breakwater at an angle of 135 degrees. The water depth

at the tip of the breakwater toe is dg = 15 feet. Assume that one inch
on the hydrographic chart being used is equivalent to 133 feet.
2-95
FIND: The wave height at a point P having coordinates in units of wave-
length of X = 3 and y = 4. (Polar coordinates of x and y are r=5 at 53°.)
SOLUTION :
Since dg = 15 feet, T = 8 seconds,
d.
Using Table C-1 with

OVERLAY
(Figure 2-36)
X and y are measured in units

of wavelength.
(These units vory depending
on the wavelength and the
chart scale.)
180'
Wave Crests
Direction of Wave Approach
Figure 2-41. Schematic Representation of Wave Diffraction Overlay
2-97
2.422 Waves Passing a Gap of Width Less than Five Wavelengths at Normal
Incidence The solution of this problem is more complex than that for a
.
single breakwater, and it is not possible to construct a single diagram

for all conditions. A separate diagram must be drawn for each ratio of
gap width to wavelength B/L. The diagram for a B/L-ratio of 2 is shown
in Figure 2-42 which also illustrates its use. Figures 2-43 through 2-52
(Johnson, 1953) show lines of equal diffraction coefficient for B/L-ratios
of 0.50, 1.00, 1.41, 1.64, 1.78, 2.00, 2.50, 2.95, 3.82 and 5.00. A
sufficient number of diagrams have been included to represent most gap
widths encountered in practice. In all but Figure 2-48 (B/L = 2.00), the
wave icrest lines have been omitted. Wave crest lines are usually of use
only for illustrative purposes. They are, however, required for an
accurate estimate of the combined effects of refraction and diffraction.
In such cases, wave crests may be approximated with sufficient accuracy
by circular arcs. For a single breakwater, the arcs will be centered on
the breakwater tip. That part of the wave crest extending into unprotected
water beyond the K' =0.5 line may be approximated by a straight line.
For a breakwater gap, crests that are more than eight wavelengths behind
the breakwater may be approximated by an arc centered at the middle of
the gap; crests to about six wavelengths may be approximated by two arcs,
centered on the two ends of the breakwater and may be connected by a
smooth curve (approximated by a circular arc centered at the middle of
the gap) . Only one-half of the diffraction diagram is presented on the
figures since the diagrams are symmetrical about the line x/L = 0.
2.423 Waves Passing a Gap of Width Greater Than Five Wavelengths at

Normal Incidence . Where the breakwater gap width is greater than five
wavelengths, the diffraction effects of each wing are nearly independent,
and the diagram (Figure 2-33) for a single breakwater with a 90° wave
approach angle may be used to define the diffraction characteristic in
the lee of both wings (See Figure 2-53.)
2.424 Diffraction at a Gap-Oblique Incidence . When waves approach at an

angle to the axis of a breakwater, the diffracted wave characteristics
differ from those resulting when waves approach normal to the axis. An
approximate determination of diffracted wave characteristics may be
obtained by considering the gap to be as wide as its projection in the
direction of incident wave travel as shown in Figure 2-54. Calculated
diffraction diagrams for wave approach angles of 0°, 15°, 30°, 45°, 60°
and 75° are shown in Figures 2-55, 56 and 57. Use of these diagrams will
give more accurate results than the approximate method. A comparison of
a 45° incident wave using the approximate method and the more exact diagram
method is shown in Figure 2-58.
2.43 REFRACTION AND DIFFRACTION COMBINED
Usually the bottom seaward and shoreward of a breakwater is not

flat; therefore, refraction occurs in addition to diffraction. Although
a general unified theory of the two has not yet been developed, some in-
sight into the problem is presented by Battjes (1968). An approximate
picture of wave changes may be obtained by: (a) constructing a refraction
2-98
u
o
*->
cd
s
u
—
/
CIS <N
U II
B OQ
^-^
nj
OO c/l
•r-l 4->
Q 00
c
t-> >
4-1 O
4-1
T3 O
N
•H
^
4->
(L>
(U nJ
L3 O
u
3
2-99
y/L
..0 2 4- 6 8 10 (2 14 16 18 20
GAP
B'UIL K =1.145 K:|.0 K'=0.8 K'=0.6 K'eO.4
" DIRECTION OF ( Johnson, 1952 )
INCIDENT WAVE
Figure 2-45. Contours of Equal Diffraction Coefficient
Gap Width =1.41 Wave Lengths (B/L = 1.41)
, Diffrocted Wove Height

DIRECTION OF Incident Wove Height
INCIDENT WAVE
K'«0.4
( Johnson, 1952)

2-101
PI DIRECTION OF , ( Johnson, 1952)
', INCIDENT WOVF

,_ Diffrocted Wove Height

Incident Wave Height
(Johnson, 1952)

Gap Width = 2 Wave Lengths (B/L = 2)
2-102
GAP
B=2.50L K =0.6
K'=l.2 K'= 1.0 K'=0.8
I
I DIRECTION OF ,
* Johnson, 1952)
ra INCIDENT WAVE (

,_ Diffrocted Wove Height

DIRECTION OF
Incident Wave Height
INCIDENT WAV^
( Johnson, 1952)

2-103
(Johnson, 1952)

ri
'BREAKWATER
i^ , Diffrocted Wove Height

'^
Incident Wove Height
DIRECTION OF
^
INCIDENT WAVE
GAP K'sl.O K'sl.O K'« 1.282 K't|.2
B:SL
( Johnson, 1952)
Gap Width = 5 Wave Lengths (B/L = 5)
2-104
Template Overlays
Wave Crests
Figure 2-53. Diffraction for a Breakwater Gap of Width > 5L (B/L > 5)
YZZZZZZZZZZZZZ.
Breakwater
maginary Equivalent Gap
Wave Crests
( Johnson, 1952)
Figure 2-54. Wave Incidence Oblique to Breakwater Gap
2-105
= O»o
B = IL
V,
10
0=30
B=IL
10
12 4 6 8 10 12 14 16 le 20
y/L
-.
2
07 75"
I
B=IL '
10
12
( Johnson, 1952)
Figure 2-58. Diffraction Diagram for a Gap of Two Wave

Lengths and a 45° Approach Compared with
That for a Gap Width ^2^ Wave lengths with
a 90° Approach
2-109
diagram shoreward to the breakwater; (b) at this point, constructing a
diffraction diagram carrying successive crests three or four wavelengths
shoreward, if possible; and (c) with the wave crest and wave direction
indicated by the last shoreward wave crest determined from the diffraction
diagram, constructing a new refraction diagram to the breaker line. The
work of Mobarek (1962) on the effect of bottom slope on wave diffraction
indicates that the method presented here is suitable for medium-period
waves. For long-period waves the effect of shoaling (Section 2.32) should
be considered. For the condition when the bottom contours are parallel to
the wave crests, the sloping bottom probably has little effect on diffrac-
tion. A typical refraction-diffraction diagram and the method for deter-
mining combined refraction-diffraction coefficients are shown in Figure
2-59. When a wave crest is not of uniform height, as when a wave is under-
going refraction, a lateral flow of energy - wave diffraction - will occur
along the wave crest. Therefore diffraction can occur without the wave
moving past a structure although the diffraction effects are visually more
dramatic at the structure.
2.5 WAVE REFLECTION
2.51 GENERAL
Water waves may be either partially or totally reflected from both

natural and manmade barriers. (See Figure 2-60.) Wave reflection may
often be as important a consideration as refraction and diffraction in the
design of coastal structures, particularly for structures associated with
development of harbors. Reflection of waves implies a reflection of wave
energy as opposed to energy dissipation. Consequently, multiple reflections
and absence of sufficient energy dissipation within a harbor complex can
result in a buildup of energy which appears as wave agitation and surging
in the harbor. These surface fluctuations may cause excessive motion of
moored ships and other floating facilities, and result in the development
of great strains on mooring lines. Therefore seawalls, bulkheads and
revetments inside of harbors should dissipate rather than reflect incident
wave energy whenever possible. Natural beaches in a harbor are excellent
wave energy dissipaters and proposed harbor modifications which would
decrease beach areas should be carefully evaluated prior to construction.
Hydraulic model studies are often necessary to evaluate such proposed
changes. The importance of wave reflection and its effect on harbor
development are discussed by Bretschneider (1966), Lee (1964), and
LeMehaute (1965) ; harbor resonance is discussed by Raichlen (1965)
A measure of how much a barrier reflects waves is given by the ratio

of the reflected wave height H^, to the incident wave height H^ which
is termed the reflection coefficient xJ hence x = H^j/H^. The magnitude
of X varies from 1.0 for total reflection to for no reflection; how-
ever, a small value of x does not necessarily imply that wave energy is
dissipated by a structure since energy may be transmitted through such
structures as permeable, rubble-mound breakwaters. A transmission co-
efficient may be defined as the ratio of transmitted wave height H^, to
incident wave height H^. In general, both the reflection coefficient
2-110
b| Lines of Equal Diffraction Coefficient (k')
Breakwater
X\\\\\\\\\\V^^^\\\\S\^V\^^^^^^^
Wave Crests
Over-oil retroction- diffraction coefficient is given
by (Kr) (k') A/b^TTj

Where Kp^Refroction coefficient to breakwater.
K' ^Diffraction coefficient ot point on
diffracted wave crest from which
orthogonolis drawn.
bÔrthogonal spocing at diffracted wove
crest
b =Orthogonol spacing nearer shore.
Figure 2-59. Single Breakwater - Refraction - Diffraction Combined
2-III
December 1952
Figure 2-60. Wave Reflection at Hamlin Beach, New York
2-JI2
and the transmission coefficient will depend upon the geometry and compo-
sition of a structure and the incident wave characteristics such as wave
steepness and relative depth d/L, at the structure site.
2.52 REFLECTION FROM IMPERMEABLE, VERTICAL WALLS (LINEAR THEORY)
Impermeable vertical walls will reflect almost all incident wave

energy unless they are fronted by rubble toe protection or are extremely
rough. The reflection coefficient x is therefore equal to approximately
1.0 and the height of a reflected wave will be equal to the height of the
incident wave. Although some experiments with smooth, vertical, impermeable
walls appear to show a significant decrease of x with increasing wave
steepness, Domzig (1955), Coda and Abe (1968) have shown that this paradox
probably results from the experimental technique, based on linear wave
theory, used to determine x- The use of a higher order theory to describe
the water motion in front of the wall gives a reflection coefficient of
1.0 and satisfies the conservation of energy principle.
Wave motion in front of a perfectly reflecting vertical wall subjected

to monochromatic waves moving in a direction perpendicular to the barrier
can be determined by superposing two waves with identical wave niombers,
periods and amplitudes but traveling in opposite directions. The water
surface of the incident wave is given to a first order (linear) approxi-
mation by Equation 2-10,
(2-10)
and the reflected wave by.

H.
T? = cos +
' 2 L T
Consequently, the water surface is given by the sum of n^ and n^ °^-

since H^ = H^,
H, /27rx 2jrt\ /27rx 27rt^

T? = 7?,- + T?^ cos — + cos +
\L T/ I L T ,
which reduces to
27rx 2-nt
n = H. cos cos -— (2-79)
'
L T
Equation 2-79 represents the water surface for a standing wave or olapotis
which is periodic in time and in x having a maximum height of 2H^ when
both cos(2ttx/L) and cos(2TTt/T) equal 1. The water surface profile as a
function of 2ttx/L for several values of 27rt/T are shown in Figure 2-61.
There are some points (nodes) on the profile where the water surface
2-113
2-114
remains at the SWL for all values of t and other points (antinodes) where
the water particle excursion at the surface, is 2 H^ or twice the incident
wave height. The equations describing the water particle motion show that
the velocity is always horizontal under the nodes and always vertical under
the antinodes. At intermediate points, the water particles move along
diagonal lines as shown in Figure 2-61. Since water motion at the anti-
nodes is purely vertical, the presence of a vertical wall at any antinode
will not change the flow pattern described since there is no flow across
the vertical barrier and equivalently, there is no flow across a vertical
line passing through an antinode. (For the linear theory discussion here,
the water contained between any two antinodes will remain between those
two antinodes.) Consequently, the flow described here is valid for a
barrier at 2ttx/L = (x = 0) since there is an antinode at that location.
2.53 REFLECTIONS IN AN ENCLOSED BASIN
Some insight can be obtained about the phenomenon of the resonant

behavior of harbors and other enclosed bodies of water by examining the
standing wave system previously described. The possible resonant oscillat-
tions between two vertical walls can be described by locating the two
barriers so that they are both at antinodes; for example, barriers at
X = and or x =
tt and 2tt, etc. represent possible modes of oscillation.
If the barriers are taken at x = and x = it, there is one-half of a wave
in the basin or, if Iq is the basin length, Iq - L/2. Since the wave-
length is given by Equation 2-4
gT^
L = ^- tanh
2jr
,
—
/27rd\
\ L
/
,
the period of this fundamental mode of oscillation is,
47^4
T = (2-80)
gtanh (jrd/^)
The next possible resonant mode occurs when there is one complete wave in
the basin (barriers at x = and x = 2ir) and the next mode when there are
3/2 waves in the basin (barriers at x = and x = 37t/2, etc. In general,
Jlg = jL/2, where j = 1, 2, In reality, the length of a natural or
manmade basin 2.^, is fixed and the wavelength of the resonant wave con-
tained in the basin will be the variable; hence,
L = —
in
j
= 1,2,----, (2-81)
may be thought of as defining the wavelengths capable of causing resonance

in a basin of length l^. The general form of Equation 2-80 is found by
2-115
substituting Equation 2-81 into the expression for the wavelength; there-
fore,
47.^3 V2
= , j = 1,2,---- (2-82)
^J jg tanh (7rjd/4)
For an enclosed harbor, of approximately rectangular planform with length,

%, waves entering through a breakwater gap having a predominant period
close to one of those given by Equation 2-82 for small values of j, may
cause significant agitation unless some effective energy dissipation
mechanism is present. The addition of energy to the basin at the resonant
(or excitation) frequency (f^- = 1/Ty) is said to excite the basin.
Equation 2-82 was developed by assxaming the end boundaries to be

vertical; however, it is still approximately valid so long as the end
boundaries remain highly reflective to wave motion. Sloping boundaries,
such as beaches, while usually effective energy dissipaters, may be signifi-
cantly reflective if the incident waves are extremely long. The effect of
sloping boundaries and their reflectivity to waves of differing character-
istics is given in Section 2.54, Wave Reflection from Beaches.
Long-period resonant oscillations in large lakes and other large

enclosed bodies of water are termed seiches. The periods of seiches may
range from a few minutes up to several hours, depending upon the geometry
of the particular basin. In general, these basins are shallow with respect
to their length; hence, tanh (irjd/JJ.g) in Equation 2-82 becomes approximately
equal to 7rjd/Jlg and
24
T.
]
= -r
j
— 1
r-w
(gd)'^
j = 1,2, (small values) . (2-83)
Equation 2-83 is termed Merian's equation. In natural basins, complex

geometry and variable depth will make the direct application of Equation
2-83 difficult; however, it may serve as a useful first approximation for
enclosed basins. For basins open at one end, different modes of oscilla-
tion exist since resonance will occur when a node is at the open end of
the basin and the fundamental oscillation occurs when there is one quarter
of a wave in the basin; hence, ilg' = L/4 for the fundamental mode and
T = 4£g'//gT; In general ^g' = (2- - l)L/4, and
44' 1
'î
=
O^ (^^ j
= 1, 2, • • • • (small values) .
(2-84)
Note that higher modes occur when there are 3, 5, ...., 2^ - 1, etc.,
quarters of a wave within the basin.
2-116
************** EXAMPLE PROBLEM **************
GIVEN: Lake Erie has a mean depth of d = 61 feet and its length % is
220" miles or 116,160 feet.
FIND : The fundamental period of oscillation T-, if j = 1.
SOLUT ION : From Equation 2-83 for an enclosed basin,
4
211,
1
T 1
2(116,160)
T. =
» 1 [32.2(61)]^^
Tj = 52,420 sec. or 14.56 hrs.
Considering the variability of the actual lake cross-section, this

result is surprisingly close to the actual observed period of 14.38
hours (Platzman and Rao, 1963). Such close agreement may not always
result.
*************************************
Note: Additional discussion of seiching is presented in Section 3.84.
2.54 WAVE REFLECTION FROM BEACHES
The amount of wave energy reflected from a beach depends upon the
roughness, permeability and slope of the beach in addition to the steep-
ness and angle of approach of incident waves. Miche (1951) assumed that
the reflection coefficient for a beach x> could be described as the
product of two factors by the expression,
X = X,X2 , (2-85)
where x-, depends on the roughness and permeability of the beach and is
independent of the slope, while Xz depends on the beach slope and the
wave steepness.
Based on measurements made by Schoemaker and Thijsse (1949), Miche

found that x, *» 0-8 for smooth impervious beaches. A value of x * 0.3
2-117
to 0.6 has been recommended for rough slopes and step-faced structures
The second factor X2 is given by
01 o (2-86a)
01 o
(2-86b)
'\L'o \ o /max
I
where (n^/Lo^rnax is constant for a particular beach slope and is given by
H,
(2-87)
in which 3 is the angle the beach makes with the horizontal (tan B =
beach slope) and Hq/^q is the incident wave steepness in deep water.
,
^ô/ô^max '-^^ ^® considered a cut-off steepness; waves steeper than

(H^/L^^max will be only partially reflected while waves with a steepness
less than (Mo/ô^max "iH ê almost totally reflected. Equation 2-87 is
given in graphical form in Figure 2-62, and Equation 2-86 is shown in
Figure 2-63. These equations and figures are valid for impervious beaches
with waves approaching at a right angle to the beach.
************** EXAMPLE PROBLEM ***************

GIVEN A wave with a period of T = 10 second, and a deepwater height of
:
Hq = 5 feet impinges on an impermeable revetment with a slope of tan 3

= 0.20.
FIND :
(a) Determine the reflection coefficient x.
(h) What will be the steepest incident wave almost totally reflected
from the given revetment?
SOLUTION. Calculate:
tanj3 =
(ofttr Micht, 1951)
cotangent ^ -
tangent )S
Figure 2-62. (Ho/Lo)max Versus Beoch Slope
max 0.5 -
X,-
2
/.«»/,n«onf
3
-
4
I
56789(ofUr Micht,
10
1951)
cotangent p
fl - -tt
tongent p
Figure 2-63. Xg Versus Beach Slope for Various Values of Hq/Lq
2-119
Entering Figure 2-63 with cot B = 5, and using the curve for Ho/ho -
0.01, a value o£ "
X2 ~ 0-41 is found. Assuming that since the beach
is impermeable, Xi = 0.8 and
X = X, X2 = 0.8(0.41) = 0.33 .
The steepest incident wave which will be nearly perfectly reflected from
the given revetment is, from Figure 2-62,
0.005 .
It is interesting to note the effectiveness of flat beaches in dissipat-

ing wave energy by considering the above wave on a beach having a slope
of 0.02 (1:50). From Equation 2-87 (noting that 3 « sin B « tan B = 0.02),
I— = 0.000014.
\ ol max
Hence
X = X, Xj = 0.8(0.0014) = 0.0011,
or the height of the reflected wave is about 0.1 percent of the incident
wave height
As indicated by the dependence of reflection coefficient on incident

wave steepness, a beach will selectively dissipate wave energy, dissipat-
ing the energy of relatively short steep waves and reflecting the energy
of the longer, flatter waves.
*************************************
2.6 BREAKING WAVES
2.61 DEEP WATER
The maximum height of a wave travelling in deep water is limited by

a maximum wave steepness for which the wave form can remain stable. Waves
reaching the limiting steepness will begin to break and in so doing, will
dissipate a part of their energy. Based on theoretical considerations,
Michell (1893) found the limiting steepness to be given by,
Ho
— = 0.142 =*
1
- . (2-88)
which occurs when the crest angle as shown in Figure 2-64 is 120 This .
limiting steepness occurs when the water particle velocity at the wave
crest just equals the wave celerity; a further increase in steepness would
2-120
result in particle velocities at the wave crest greater than the wave
celerity and, consequently, instability.
Limiting steepness -p- = 0.142
Figure 2-64. Wave of Limiting Steepness in Deep Water
2.62 SHOALING WATER
When a wave moves into shoaling water, the limiting steepness which
it can attain decreases, being a function of both the relative depth d/L,
and the beach slope m, perpendicular to the direction of wave advance.
A wave of given deepwater characteristics will move toward a shore until
the water becomes shallow enough to initiate breaking, this depth is
usually denoted as d]-, and termed the breaking depth. Munk (1949) derived
several relationships from a modified solitary wave theory relating the
breaker height H^, the breaking depth d^,, the unrefracted deepwater
wave height H^, and the deepwater wavelength L^. His expressions are
given by
1
(2-89)
3.3(HyL^)H
and
= 1.28, (2-90)
H.
The ratio H^/H^ is frequently termed the breaker height index. Subse-
quent observations and investigations by Iversen (1952, 1953), Galvin
(1969), and Coda (1970) among others, have established that H2,/H^ and
dy/U-jj depend on beach slope and on incident wave steepness. Figure 2-65
shows Coda's empirically derived relationships between H^j/H^ and H^/L^
for several beach slopes. Curves shown on the figure are fitted to widely
scattered data; however they illustrate a dependence of H^j/H^^ on the
beach slope. Empirical relationships between d^/H^, and H^^/gT^ for
various beach slopes are presented in Figure 2-66. It is recommended
that Figures 2-65 and 2-66 be used, rather than Equations 2-89 and 2-90,
for making estimates of the depth at breaking or the maximum breaker
2-121
2-122
o
(VJ
height in a given depth since the figures take into consideration the
observed dependence of d^^/H^, and H^j/H^ on beach slope. The curves
in Figure 2-66 are given by
^
_± ^ '
(2-91)
Hfc b-(aHj,/gT^)
where a and b are functions of the beach slope m, and may be approxi-
mated by
a = 1.36g(l-e-i5'") C2-92)
1-56 (2-93)
b =
(l + e-19.5m)
Breaking waves have been classified as spilling, plunging or surging

depending on the way in which they break (Patrick and Wiegel, 1955), and
(Wiegel, 1964). Spilling breakers break gradually and are characterized
by white water at the crest. (See Figure '2-67.) Plunging breakers curl
over at the crest with a plunging forward of the mass of water at the
crest. (See Figure 2-68.) Surging breakers build up as if to form a
plunging breaker but the base of the wave surges up the beach before the
crest can plunge forward. (See Figure 2-69.) Further subdivision of
breaker types has also been proposed. The term collapsing breaker is
sometimes used (Galvin, 1968) to describe breakers in the transition from
plunging to surging. (See Figure 2-70.) In actuality, the transition
from one breaker type to another is gradual without distinct dividing
lines; however, Patrick and Wiegel (1955) presented ranges of H^/L^ for
several beach slopes for which each type of breaker can be expected to
occur. This information is also presented in Figure 2-65 in the form of
three regions on the Uj^/ô vs. H^/L^ plane. An example illustrating the
estimation of breaker parameters follows.
************** EXAMPLE PROBLEM *****

GIVEN A beach having a 1:20 slope; a wave with deepwater height of
:
H^ = 5 feet and a period of T = 10 seconds. Assume that a refraction

analysis gives a refraction coefficient, Kff = (bo/b)^/^ = 1.05 at the
point where breaking is expected to occur.
FIND : The breaker height Hh and the depth db at which breaking occurs,
SOLUTION : The unrefracted deepwater height H^ can be found from
h' b \''''
— = Kp =
/
T- (See Section 2.32),
2-124
Figure l-dl . Spilling Breaking Wave
Figure 2-68. Plunging Breaking Wave
2-125
Figure 2-69. Surging Breaking Wave
Figure 2-70. Collapsing Breaking Wave
2-126
hence,
H^ = 1.05(5) = 5.25 feet,
and since, L^ = 5,12T^ (linear wave theory).
% 5.25
= 0.010
Lo 5.12(10)^
From Figure 2-65, entering with H^/L^ = 0.010 and intersecting the curve
for a slope of 1:20 (m = 0.05) results in H^/H^ = 1.65. Therefore
H^ = 1.65(5.25) = 8.66 feet .
To determine the depth at breaking calculate:
^b 8.66
0.00269 ,
gT^ 32.2(10)2
and enter Figure 2-66 for m = 0.050.
^b
= 0.90
Thus djy = 0.90 (8.66) =7.80 feet, and therefore the wave will break
when it is approximately 7.80/(0.05) = 156 feet from the shoreline,
assuming a constant nearshore slope. The initial value selected for
the refraction coefficient should now be checked to determine if it is
correct for the actual breaker location as found in the. solution. If
necessary, a corrected value for the refraction coefficient should be
used and the breaker height recomputed. The example wave will result
in a plunging breaker. (See Figure 2-65.)
2-127
REFERENCES AND SELECTED BIBLIOGRAPHY
AIRY, G.B., "On Tides and Waves," Enoyolopaedia Metropolitanaj 1845.

(Date of original writing unknown.)
ARTHUR, R.S., MUNK, W.H., and ISSACS, J.D., "The direct Construction of
Wave Rays," Transaotions of the American Geophysical Union^ Vol. 33,
No, 6, 1952, pp. 855-865.
BATTJES, J. A., "Refraction of Water Waves," Journal of the Waterways and

Harbors Division, ASCE, Vol. 94, WW4, No. 6206, Nov. 1968.
BEITINJANI, K.I., and BRATER, E.F., "Study on Refraction of Waves in

Prismatic Channels," Journal of the Waterways and Harbors Division,
ASCE, Vol. 91, WW3, No. 4434, 1965, pp. 37-64.
BLUE, F.L., JR. and JOHNSON, J.W., "Diffraction of Water Waves Passing
through a Breakwater Gap," Transaotions of the American Geophysical
Union, Vol. 30, No. 5, Oct. 1949, pp. 705-18.
BORGMAN, L.E., "Directional Spectra Models for Design Use," HEL-1-12,

University of California, Berkeley, 1969.
BOUSSINESQ, J., "Theory of Waves and Swells Propagated in a Long Horizontal

Rectangular Canal, and Imparting to the Liquid Contained in this Canal
Approximately Equal Velocities from the Surface to the Bottom," Journal
de Mathematiques Pares et Appliquees, Vol. 17, Series 2, 1872.
BOUSSINESQ, J., "Essai sur la Theorie des Eaux Courantes," Mem. divers
Savants a L'Academie des Science, 32:56, 1877.
BRETSCHNEIDER, C.L., "Wave Refraction, Diffraction and Reflection,"

Estuaj'y and Coastline Hydrodynamics, A.T. Ippen, ed., McGraw-Hill,
New York, 1966, pp. 257-280.
CALDWELL, J.M., "Reflection of Solitary Waves," TM-11, U.S. Army, Corps

of Engineers, Beach Erosion Board, Washington, D.C., Nov. 1949.
CARR, J.H. and STELZRIEDE, M.E., "Diffraction of Water Waves by Breakwaters,"

Gravity Waves, Circ. No. 521, National Bureau of Standards, Washington,
D.C., Nov. 1952, pp. 109-125.
CHAO, Y.Y., "The Theory of Wave Refraction in Shoaling Water," TR-70-7,

Department of Meteorology and Oceanography, New York University, New
York, 1970.
CHIEN, NING, "Ripple Tank Studies of Wave Refraction," Transactions of the

American Geophysical Union, Vol. 35, No. 6, 1954, pp. 897-904.
COUDERT, J.F., and RAICHLEN, F., "Wave Refraction Near San Pedro Bay,
California," Journal of the Waterways, Harbors and Coastal Engineering
Division, ASCE, WW3, 1970, pp. 737-747.
2-129
DAILY, J.W. and STEPHAN, S.C.,JR., "Characteristics of a Solitary Wave,"
Transactions of the Amerioan Society of Civil Engineers ^ Vol. 118, 1953,
pp. 575-587.
DANEL, P., "On the Limiting Clapotis," Gravity Waves^ Circ. No. 521,
National Bureau of Standards, Washington, D.C. ,1952, pp. 35-38.
DEAN, R.G., "Stream Function Representation of Nonlinear Ocean Waves,"

Journal of the Geophysical Research ^ Vol. 70, No. 18, Sep. 1965a.
DEAN, R.G., "Stream Function Wave Theory; Validity and Application,"

Coastal Engineering^ Santa Barbara Specialty Conference j Ch. 12,
Oct. 1965b.
DEAN, R.G., "Relative Validities of Water Wave Theories," Proceedings of

the Conference on Civil Engineering in the Oceans J, San Francisco,
Sep. 1967.
DEAN, R.G., "Evaluation and Development of Water Wave Theories for

Engineering Application," Engineering and Industrial Experiment Station
Report, College of Engineering, University of Florida, Gainesville, 1970,
(Draft copy.)
DEAN, R.G., "Evaluation and Development of Water Wave Theories for

Engineering Application, U.S. Army, Corps of Engineers, Coastal
Engineering Research Center. (To be published in 1973).
DEAN, R.G. and LE MEHAUTE, B., "Experimental Validity of Water Wave

Theories," National Structural Engineering Meeting , ASCE, Portland,
Oregon, Apr. 1970.
— Reprints on Stream Function Wave Theory and Application to Wave Force

Calculations, Engineering and Industrial Experiment Station, College of
Engineering, University of Florida, Gainesville, 1970.
DOBSON, R.S., "Some Applications of Digital Computers to Hydraulic

Engineering Problems," TR-80, Ch. 2, Department of Civil Engineering,
Stanford University, Calif., Jun. 1967.
DOMZTG, H., "Wellendruck und druckerzeugender Seegang," Mitteilungen der

Hannoverschen Versuchsanstalt fur Grundhau und Wasserbau^ Hannover,
Germany, Heft 8, 1955.
DRAPER, L., "Attenuation of Sea Waves with Depth," La Houille Blanchej

Vol. 12, No. 6, 1957, pp. 926-931.
EAGLESON, P.S. and DEAN, R.G., "Small Amplitude Wave Theory," Estuary and
Coastline Hydrodynamics , A.T. Ippen, ed., McGraw-Hill, New York, 1966,
pp. 1-92.
ESTEVA, D. and HARRIS, D.L., "Comparison of Pressure and Staff Wave Gages
Records," Proceedings of the l2th Conference on Coastal Engineering
Washington, D.C, 1971.
2-130
FAN, S.S., "Diffraction of Wind Waves," HEL-1-10, Hydraulics Laboratory,
University of California, Berkeley, Aug. 1968.
FAN, S.S. and BORGMAN, L.E., "Computer Modeling of Diffraction of Wind

Waves," Proceedings of the 12th Conference on Coastal Engineering
Washington, D.C., Sep. 1970.
GALVIN, C.J., JR., "Breaker Type Classification on Three Laboratory Beaches,"

Journal of Geophysical Research^ Vol. 73, No. 12, 1968.
GALVIN, C.J., JR., "Breaker Travel and Choice of Design Wave Height,"
Journal of the Waterways and Harobrs Division^ ASCE, WW2, No. 6569,
1969, pp. 175-200.
GATEWOOD, R., in GAILLARD, D.D., "Wave Action in Relation to Engineering

Structures," Professional Paper No. 21j U.S. Army, Corps of Engineers,
1904.
GERSTNER, F., "Theorie die Wellen," Abhandlungen der Konigiohen Bohmisohen

Gesellsahaft der Wis sens chaften^ Prague, 1802.
GODA, Y, "A Synthesis of Breaker Indices," Transactions of the Japanese

Society of Civil Engineers ^ Vol. 2, Part 2, 1970.
GODA, Y. and ABE, Y., "Apparent Coefficient of Partial Reflection of

Finite Amplitude Waves," Rpt. No. 3, Vol. 7, Port and Harbor Research
Institute, Nagase, Yokosuka, Japan, Sep. 1968.
GRACE, R.A., "How to Measure Waves," Ocean Industry, Vol. 5, No. 2, 1970,
pp. 65-69.
GRESLAU, L. and MAHE Y., "Study of the Reflection Coefficient of a Wave

on an Inclined Plane," Proceedings of the Sth Conference on Coastal
Engineering, Grenoble, France (French), Sep. 1954.
GRISWOLD, G.M., "Numerical Calculation of Wave Refraction," Journal of

Geophysical Research, Vol. 68, No. 6, 1963.
HARDY, J.R., "Some Grid and Projection Problems in the Numerical

Calculation of Wave Refraction," Journal of Geophysical Research,
Vol. 73, No. 22, Nov. 1958.
HARRISON, W. and WILSON, W.S., "Development of a Method for Numerical

Calculation of Wave Refraction," TM-6, U.S. Army, Corps of Engineers,
Coastal Engineering Research Center, Washington, D.C., 1964.
HAVELOCK, T.H., "Periodic Irrotational Waves of Finite Height," Proceedings

of the Royal Society of London, Vol. 95, Series A, No. A665, 1918,
pp. 37-51.
HEALY, J.J., "Wave Damping Effect of Beaches," Conference of International

Hydraulics, lAHR-ASCE, Minneapolis, Sep. 1953.
2-131
HOM-MA, M., HORIKAWA, K. , and KOMORI, S., "Response Characteristics of
Underwater Wave Gauge," Proceedings of the lOth Conference on Coastal
Engineering J Vol. 1, Tokyo, 1966, pp. 99-114.
HUNT, "Winds, Wind Setup and Seiches on Lake Erie," U.S. Army
I. A.,
Engineer District, Lake Survey, 1959.
IPPEN, A.T., "Waves and Tides in Coastal Processes," Journal of the Boston
Society of Civil Engineers^ Vol. 53, No. 2, 1966a., pp. 158-181.
IPPEN, A.T., ed.. Estuary and Coastline Hydrodynamics j McGraw-Hill,

New York, 1966b.
IPPEN, A. and KULIN, G., "The Shoaling and Breaking of the Solitary Wave,"
Proceedings of the 6th Conference on Coastal Engineering ^ Grenoble,
France, 1954, pp. 27-49.
IVERSEN, H.W., "Laboratory Study of Breakers," Gravity WaveSj Circ. No. 521,
National Bureau of Standards, Washington, D.C., 1952.
IVERSEN, H.W., "Waves and Breakers in Shoaling Water," Proceedings of the

3rd Conference on Coastal Engineering ^ Cambridge, Mass., 1953.
IWASA, Y., "Analytical Consideration on Cnoidal and Solitary Waves,"

Memoirs of the Faculty of Engineering ^ Reprint, Kyoto University,
Japan, 1955.
JOHNSON, J.W., "The Refraction of Surface Waves by Currents," Transactions

of the American Geophysical Union^ Vol. 28, No. 6, Dec. 1947, pp. 867-874.
JOHNSON, J.W., "Generalized Wave Diffraction Diagrams," Proceedings of the

2nd Conference on Coastal Engineering ^ Houston, Texas, 1952.
JOHNSON, J.W., "Engineering Aspects of Diffraction and Refraction,"

Transactions of the American Society of Civil Engineers, Vol. 118,
No. 2556, 1953, pp. 617-652.
JOHNSON, J.W., O'BRIEN, M.P., and ISAACS, J.D,, "Graphical Construction of

Wave Refraction Diagrams," HO No. 605, TR-2, U.S. Naval Oceanographic
Office, Washington, D.C., Jan. 1948.
KAPLAN, K., "Effective Height of Seawalls," Vol. 6, Bui. No. 2, U.S. Army,
Corps of Engineers, Beach Erosion Board, Washington, D.C., 1952.
KELLER, J.B., "Surface Waves on Water of Non-uniform Depth," Journal of

Fluid Mechanics, Vol. 4, 1958, pp. 607-614.
KEULEGAN, G.H., "Gradual Damping of Solitary Waves," Journal of Research

of the National Bureau of Standards, Reprint, Vol. 40, 1948, pp. 487-498.
2-132
KEULEGAN, G.H. and HARRISON, J., "Tsunami Refraction Diagrams by Digital
Computer," Journal of the Waterways, Harbors and Coastal Engineering
Division, ASCE, Vol. 96, WW2, No. 7261, May 1970, pp. 219-233.
KEULEGAN, G.H. and PATTERSON, G.W., "Mathematical Theory of Irrotational

Translation Waves," RP No. 1272, National Bureau of Standards,
Washington, D.C., 1940, pp. 47-101,
KINSMAN, B., Wind Waves, their Generation and Propagation on the Ocean
Surface, Prentice Hall, New Jersey, 1965.
KORTEWEG, D.J. and DE VRIES, G., "On the Change of Form of Long Waves
Advancing in a Rectangular Canal, and on a New Type of Long Stationary
Waves," Philosophical Magazine, 5th Series, 1895, pp. 422-443.
LACOMB, H., "The Diffraction of a Swell - A Practical Approximate Solution

and Its Justification," Gravity Waves, Circ. No. 521, National Bureau
of Standards, Washington, D.C., 1952, pp. 129-140.
LAITONE, E.V., "Higher Approximation to Non-linear Water Waves and the

Limiting Heights of Cnoidal, Solitary and Stokes Waves," 1^-133,
U.S. Army, Corps of Engineers, Beach Erosion Board, Washington, D.C.,
1963.
LAMB, H., Hydrodynamics, 6th ed. Cambridge University Press, London -
New York, 1932.
LEE, C, "On the Design of Small Craft Harbors," Proceedings of the 9th
Conference on Coastal Engineering, Lisbon, Portugal, Jun 1964.
LE MEHAUTE, B., "Wave Absorbers in Harbors," Contract Report No. 2-122,

National Engineering Science Company for the U.S. Army, Waterways
Experiment Station, Vicksburg, Miss., Jun. 1965.
LE MEHAUTE, B., "An Introduction to Hydrodynamics and Water Waves,"

Water Wave Theories, Vol. II, TR ERL 118-POL-3-2, U.S. Department of
Commerce, ESSA, Washington, D.C., 1969.
LEWIS, R.M., BLEISTEIN, N., and LUDWIG, D., "Uniform Assymptolic Theory of
Creeping Waves," Communications on Pure and Applied Mathematics, Vol. 20,
No. 2, 1967.
LONGUET-HIGGINS, M.S., "Mass Transport in Water Waves," Philosophical

Transactions of the Boyal Society of London, Series A, Vol. 245,
No. 903, 1953.
LONGUET-HIGGINS, M.S., "On the Transformation of a Continuous Spectrum by

Refraction," Proceedings of the Cambridge Philosophical Society, Vol. 53,
Part I, 1957, pp. 226-229.
2-133
LONGUET-HIGGINS, M.S., "Mass Transport in the Boundary Layer at a Free
Oscillating Surface," Vol. 8, National Institute of Oceanography,
Wormley, Surrey, England, 1960.
LOWELL, S.C, "The Propagation of Waves in Shallow Water," Communications

on Pure and Applied Mathematios , Vol. 12, Nos.2 and 3, 1949.
McCOWAN, J., "On the Solitary Wave," Philosophical Magazine^ 5th Series,
Vol. 32, No. 194, 1891, pp. 45-58.
MASCH, F.D., "Cnoidal Waves in Shallow Water," Proceedings of the 9th

Confevence of Coastal Engineering ^ Ch. 1, Lisbon, 1964.
MASCH, F.D. and WIEGEL, R.L., "Cnoidal Waves. Tables of Functions,"

Council on Wave Research j The Engineering Foundation ^ Richmond,
Calif., 1961.
MATTSSON, A., "Reflection of Gravity Waves," Congress on the International

Association of Hydraulic Research^ London, 1963.
MEHR, E., "Surf Refraction - Computer Refraction Program," Statistical

Laboratory, New York University, New York, Sep. 1962.
MICHE, M., "Mouvements Ondulatoires de la Mer en Profondeur Constante ou

Decroissante," Annales des Ponts et Chaussees, 1944, pp. 25-78, 131-164,
270-292, and 369-406.
MICHE, M., "The Reflecting Power of Maritime Works Exposed to Action of

the Waves," Annals des Ponts et Chaussees Jun. 1951. (Partial trans-
,
lation by the Beach Erosion Board, Vol. 7, Bui. No. 2, Apr. 1953.)
MICHELL, J.H., "On the Highest Waves in Water," Philosophical Magazine^

5th Series, Vol. 36, 1893, pp. 430-437.
MITCHIM, C.F., "Oscillatory Waves in Deep Water," The Military Engineer^

March-April 1940, pp. 107-109.
MOBAREK, ISMAIL, "Effect of Bottom Slope on Wave Diffraction," HEL-1-1,

University of California, Berkeley, Nov. 1962.
MONKMEYER, P.L., "Higher Order Theory for Symmetrical Gravity Waves,"

Ch. 33, Proceedings of the 12th Conference on Coastal Engineering
Washington, D.C., 1970.
MORISON, J.R. and CROOKE, R.C., "The Mechanics of Deep Water, Shallow
Water, and Breaking Waves," TM-40, U.S. Army, Corps of Engineers,
Beach Erosion Board, Washington, D.C., 1953.
MORSE, P.M. and RUBINSTEIN, P.J., "The Diffraction of Waves by Ribbons

and Slits," Physical Reviews, Vol. 54, Dec. 1938, pp. 895-898.
2-134
MUNK, W.H., "The Solitary Wave Theory and Its Application to Surf Problems,"
Annals of the New York Aoademy of Soienoes, Vol. 51, 1949, pp. 376-462.
MUNK, W.H. and ARTHUR, R.S., "Wave Intensity Along a Refracted Ray,"
Symposium on Gravity WaveSy Circ. 521, National Bureau of Standards,
MUNK, W.H. and TRAYLOR, M.A., "Refraction of Ocean Waves," Journal of

Geology, Vol. LV, No. 1, 1947.
NIELSEN, A.H., "Diffraction of Periodic Waves Along a Vertical Breakwater

of Small Angles of Incidence," HEL-1-2, Institute of Engineering
Research, University of California, Berkeley, Dec. 1962.
O'BRIEN, M.P., "A Summary of the Theory of Oscillatory Waves," TR-2,

1942.
ORR, T.E. and HERBICH, J.B., "Numerical Calculation of Wave Refraction by

Digital Computer," COE Rpt No. 114, Sea Grant Publication, No. 209,
.
Department of Coastal and Ocean Engineering, Texas A§M University,

Dec. 1969.
PALMER, R.Q., "Wave Refraction Plotter," Vol. 11, Bui. No. 1, U.S. Army,
Corps of Engineers, Beach Erosion Board, Washington, D.C., 1957.
PATRICK, D.A. and WIEGEL, R.L., "Amphibian Tractors in the Surf," Frooeed-
ings of the First Conference on Ships and Waves , Council on Wave
Research and Society of Naval Architects and Marine Engineers, 1955.
PENNEY, W.G. and PRICE, A.T., "Diffraction of Sea Waves by a Breakwater,"

Artificial Harbors , Technical History No. 26, Sec. 3-D, Directorate
of Miscellaneous Weapons Development, 1944.
PIERSON, W.J., JR., "The Interpretation of Crossed Orthogonals in Wave

Refraction Phenomena," TM-21, U.S. Army, Corps of Engineers, Beach
Erosion Board, Washington, D.C., 1950.
PIERSON, W.J., JR., "The Accuracy of Present Wave Forecasting Methods with
Reference to Problems in Beach Erosion on the New Jersey and Long
Island Coasts," TM-24, U.S. Army, Corps of Engineers, Beach Erosion
Board, Washington, D.C., 1951.
PIERSON, W.J., JR., "Observing and Forecasting Ocean Waves," H.O. No. 603,
U.S. Naval Oceanographic Office, Washington, D.C., 1965.
PLATZMAN, G.W., and RAO, D.B., "The Free Oscillations of Lake Erie," TR-8,
Department of Geophysical Sciences, University of Chicago, for the
U.S. Weather Bureau, Sep. 1963.
2-135
POCINKI, L.S., "The Application of Conformal Transformations to Ocean Wave
Refraction Problems," Department of Meteorology and Oceanography, New
York University, New York, 1950.
PUTNAM, J. A. and ARTHUR, R.S., "Diffraction of Water Waves by Breakwaters,"

Transactions of the Amevioan Geophysical Union, Vol. 29, No. 4, Aug.
1948, pp. 317-374.
RAICHLEN, F., "Long Period Oscillations in Basins of Arbitrary Shapes,"

Coastal Engineering Santa Barbara Specialty Conference , Ch. 7, Oct. 1965.
pp. 115-148, American Society of Coastal Engineering, New York, 1966.
RALLS, G.C., "A Ripple Tank Study of Wave Refraction," Journal of the
Waterways and Harbors Division^ ASCE, Vol. 82, WWl, No. 911, Mar. 1956.
RAYLEIGH, LORD, "On Waves," Philosophical Magazine and Journal of Soienoej

Vol. 1, No. 4, 1876, pp. 257-279.
RUSSELL, J.S., "Report of the Committee on Waves," 7th Meeting of the

British Association for the Advancement of Soienoey 1838, p. 417.
RUSSELL, J.S., "Report on Waves," I'ith Meeting of the British Association

for the Advancement of Science, 1845, p. 311.
RUSSELL, R.C.H. and OSORIO, J.D.C., "An Experimental Investigation of

Drift Profiles in a Closed Channel," Proceedings of the 6th Conference
on Coastal Engineering, Miami, Council on Wave Research, University of
California, Berkeley, 1958, pp. 171-183.
SAVILLE, T., JR., and KAPLAN, K., "A New Method for the Graphical
Construction of Wave Refraction Diagrams," Vol. 6, Bui. No. 3, U.S.
Army, Corps of Engineers, Beach Erosion Board, Washington, D.C., 1952,
pp. 23-24.
SHOEMAKER, H.J. and THIJSSE, J.TH. "Investigations of the Reflection of

Waves," 3rd. Meeting of the International Association for Hydraulic
Structures Research, Grenoble, France, Sep. 1949.
SILVESTER, R., "Design Wave for Littoral Drift Model," Journal of the
Waterways and Harbors Division, ASCE, Vol. 89, WW3, No. 360, Aug. 1963.
SKJELBREIA, L., Gravity Waves. Stokes' Third Order Approximation. Tables

of Functions, University of California, Council on Wave Research, The
Engineering Foundation, Berkeley, 1959.
SKJELBREIA, L. and HENDRICKSON, J. A., Fifth Order Gravity Wave Theory and
Tables of Functions, National Engineering Science Company, 1962.
SOMMERFELD, A., "Matheraatische Theorie der Diffraction," Mathematische

Annals, Vol. 47, 1896, pp. 317-374.
2-136
STOKES, G.C., "On the Theory of Oscillatory Waves," Mathematical and
Physical Papers^ Vol. 1, Cambridge University Press, Cambridge, 1880.
SVERDRUP. H.U. and MUNK, W.H., "Wind, Sea, and Swell: Theory of Relations
for Forecasting," TR-1, H.O. No. 601, U.S. Naval Hydrographic Office,
Washington, D.C., 1947, p. 7.
U.S. ARMY, CORPS OF ENGINEERS, "A Summary of the Theory of Oscillatory

Waves," TR-2, Beach Erosion Board, Washington, D.C., 1942.
URSELL, F., "Mass Transport in Gravity Waves," Proceedings of the Cambridge

Philosophical Society^ Vol. 49, Pt. 1, Jan. 1953, pp. 145-150.
WEGGEL, J.R., "Maximum Breaker Height," Journal of the Wateruays^ Harbors

and Coastal Engineering division^ ASCE, Vol. 98, WW4, Nov. 1972.
WHALIN, R.W., "The Limit of Applicability of Linear Wave Refraction Theory

in a Convergence Zone," H-71-3, U.S. Army, Corps of Engineers,
Waterways Experiment Station, Washington, D.C., 1971, pp. 156.
WIEGEL, R.L., Gravity Wooes. Tables of Functions, University of California,

Council on Wave Research, The Engineering Foundation, Berkeley, 1954.
WIEGEL, R.L., "A Presentation of Cnoidal Wave Theory for Practical

Application," Journal of Fluid Mechanics, Vol. 7, Pt 2, Cambridge
.
University Press, 1960, pp. 273-286.
WIEGEL, R.L., "Diffraction of Waves by a Semi-infinite Breakwater," Journal

of the Hydraulics Division, ASCE, Vol. 88, HYl, Jan. 1962, pp. 27-44.
WIEGEL, R.L., Oceanographical Engineering, Fluid Mechanics Series, Prentice

Hall, New Jersey, 1964.
WIEGEL, R.L. and ARNOLD, A.L., "Model Studies of Wave Refraction," TM-103,
U.S. Army, Corps of Engineers, Beach Erosion Board, Washington, D.C.
Dec. 1957.
WILLIAMS, L.C., "CERC Wave Gages," Tll-30, U.S. Army, Corps of Engineers,
Coastal Engineering Research Center, Washington, D.C, 1969.
WILSON, W.S., "A Method for Calculating and Plotting Surface Wave Rays,"
TM-17, U.S. Army, Corps of Engineers, Coastal Engineering Research
Center, Washington, D.C, Feb. 1966.
WILTON, J.R., "On Deep Water Waves," Philosophical Magazine, 6th Series,
1914, pp. 385-394.
WORTHINGTON, H.W. and HERBICH, J.B., "A Computer Program to Estimate the
Combined Effects of Refraction and Diffraction of Water Waves,"
Sea Grant Publication, No. 219, Texas A§M University, Aug. 1970.
2-137
CHAPTER 3
WAVE AND
WATER LEVEL
PREDICTIONS
OCEAN CITY, NEW JERSEY - 9 March 1962
CHAPTER 3
WAVE AND WATER-LEVEL PREDICTIONS
3.1 INTRODUCTION
Chapter 2, treated phenomena associated with surface waves as though

each phenomenon could be considered separately without regard to other
phenomena. Surface waves were discussed from the standpoint of motions
and transformations without regard to wave generation. Furthermore, the
water level, stiltwater level (SWL) , on which the waves propagated was
assumed known.
In this chapter, wave observations are presented to show characteris-
tics of surface waves in nature. The characteristics of real waves are
much less regular than those implied by theory. Also presented are pro-
cedures for representing the complexity of the real sea by a small number
of parameters. Deviations between the actual waves and the parameter
values are discussed.
Theory for wave generation is reviewed to show progress in explaining
and predicting the actual complexity of the sea. Wave prediction is
called hindaasting when based on past meteorological conditions, and fore-
casting when based on predicted conditions. The same procedures are used
for hindcasting and forecasting; the only difference is the source of
meteorological data. The most advanced prediction techniques currently
available can be used only in a few laboratories, because of the need for
electronic computers, the sophistication of the models, and the need for
correct weather data. However, simplified wave prediction techniques,
suitable for use by field offices or a design group are presented.
While simplified prediction systems will not solve all problems, they
can be used to indicate probable wave conditions for most design studies.
Simplified wave prediction can also be used to obtain statistical wave
data over several years.
Review of prediction theories is presented to give the reader more
perspective for the simplified prediction methods that are presently
available. This will justify confidence in some applications of the
simplified procedures, will aid in recognizing unexpected difficulties
when they occur, and will indicate some conditions in which they are not
adequate
The graphs in Sections 3.5, Simplified Wave Prediction Models, and

3.6, Wave Forecasting for Shallow Water Areas, may be read with the pre-
cision justified by the underlying theory. The equations were derived
originally from graphs, and do not provide any physical understanding.
Calculations with the graphs should be carried out to tenths or hundredths
where this is feasible, and then rounded off in the final result.
Predictions are compared with available observations wherever possible

to indicate their accuracy. Calibration of techniques applied to a spe-
cific geographic area by comparison with available observations is always
desirable.
3-
The problem of obtaining wind information for wave hindcasting is
discussed, and specific instructions for estimating wind parameters are
given.
Water levels continuously change. Changes due to astronomical tides

are predictable, and are well documented for many areas. Fluctuations due
to meteorological conditions are not as predictable, and are less well
documented.
Many factors govern water levels at a shore during a storm. The

principle factor is the effect of wind blowing over water. Some of the
largest increases in water level are due to severe storms, such as hurri-
canes, which can cause storm surges higher than 25 feet at some locations
on the open coast and even higher water levels in bays and estuaries.
Estimating water levels caused by meteorological conditions is complex,
even for the simplest cases; and unfortunately, the best approaches avail-
able for predicting these water levels are elaborate computational tech-
niques which require the use of large digital computers.
3.2 CHARACTERISTICS OF OCEAN WAVES
The earlier discussion of waves was concerned with idealized, mono-

chromatic waves. The pattern of waves on any body of water exposed to
atmospheric winds generally contains waves of many periods. Typical
records from a recording gage during periods of steep waves (Fig. 3-1)
indicate that heights and periods of real waves are not as constant as is
assumed in theory. Wavelengths and directions of propagation are also
variable. (See Figure 3-2.) The prototype is so complex that some
idealization is required.
3.21 SIGNIFICANT WAVE HEIGHT AND PERIOD
An early idealized description of ocean waves postulated a significant

height and significant period, that would represent the characteristics of
the real sea in the form of monochromatic waves.
The representation of a wave field by significant height and period

has the advantage of retaining much of the insight gained from the theo-
retical studies. Its value has been demonstrated in the solution of
engineering problems. For some problems this representation appears
adequate; for others it is useful, but not entirely satisfactory.
To apply the significant wave concept it is necessary to define the

height and period parameters from wave observations. Munk (1944) defined
significant wave height, as the average height of the one-third highest
waves, and stated that it was about equal to the average height of the
waves as estimated by an experienced observer. This definition, while
useful, has some drawbacks in wave-record analysis. It is not always
clear which irregularities in the wave record should be counted to deter-
mine the total number of waves on which to compute the average height of
the one-third highest. The significant wave height is written as H1/3 or
simply Hg.
3-2
The significant wave period obtained by visual observations of waves
is likely to be the average period of 10 to 15 successive prominent waves.
When determined from gage records, the significant period is apt to be the
average period of the subjectively estimated most prominent waves, or the
average period of all waves whose troughs are below and whose crests are
above the mean water level, (zero up crossing method).
3.22 WAVE HEIGHT VARIABILITY
When the heights of individual waves on a wave record are ranked from
the highest to lowest, the frequency of occurrence of waves above any given
value is given to a close approximation by the cumulative form of the
Rayleigh distribution. This fact can be used to estimate the average
height of the one-third highest waves from measurements of a few of the
highest waves, or to estimate the height of a wave of any arbitrary
frequency from a knowledge of the significant wave height. According to
the Rayleigh distribution function, the probability that the wave height
H is more than some arbitrary value of H referred to as H is given
by
'""'
P(H > H) = e ^ (3-1)
where Hmis is a parameterôf the distribution, and P(H > H) is the number
n of waves larger than H divided by the total number N of waves in
the record. Thus P has the form n/N. The value Hrms is called the
root-mean- square height and is defined by
^rm. = V. .2. H • (3-2)
It was shown in Section 2.238, Wave Energy and Power, that the total energy
per unit surface area is given by
E = '-fil
The average energy per unit surface area for a number of waves is given by
Pg 1 N
where Hj is the height of successive individual waves, and (¥)^ is the

average energy per unit surface area of all waves considered. Thus H^ms
is a measure of average wave energy. Calculation of Hj^g by Equation 3-2
is somewhat less subjective than direct evaluation of the Hg because more
emphasis is placed on the larger, better defined waves. The calculation
can be made more objective by substitution of n/N for P(H > iT) in
Equation 3-1 and taking natural logarithms of both sides to obtain
Ln(n) = Ln(N) - (H;^pH^ .

(3_4)
3-5
By making the substitutions
y(n) = Ln(n), a = Ln(N), b = - H;^^, x(n) = H'(n) .
Equation 3-4 may be written as
y(n) = a + bx(n) (3-5)
The constants a and b can be found graphically or by fitting a least-

square regression line to the observations. The parameters N and Harris
may be computed from a and b. The value of N found in this way, is
the value that provides the best fit between the observed distribution of
identified waves and the Rayleigh distribution function. It is generally
a little larger than the number of waves actually identified in the record.
This seems reasonable because some very small waves are generally neglected
in interpreting the record. When the observed wave heights are scaled by
^rms) that is, made dimensionless by dividing each observed height by
^rms> then data from all observations may be combined into a single plot.
Points from scaled 15-minute samples are superimposed on Figure 3-3 to
show the scatter to be expected from analyzing individual observations in
this manner.
Data from 72 scaled 15-minute samples representing 11,678 observed

waves have been combined in this manner to produce Figure 3-4. The theo-
retical height appears to be about 5 percent greater than the observed
height for a probability of 0.01 and 15 percent at a probability of 0.0001.
It is possible that the difference between the actual and theoretical
heights of highest waves is due to breaking of the very highest waves
before they reach the coastal wave gages.
Equation 3-1 can be established rigorously for restrictive conditions,

and empirically for a much wider range of conditions. If Equation 3-1 is
accepted as an exact law, the probability density function can be obtained
in the form
^^'""'
f[(H-AH) < H < (fi+ AH)] = l^—\ He . (3-6)
The height of the wave with any given probability n/N of being exceeded
may be determined approximately from curve a in Figure 3-5 or from the
equation.
(3-7)
3-6
0.000
0.0005
0.001
0.005
~ 0.0
O
o
a>
>
0.05
From fifteen 15-minute records
3
E containing a total of 2,342 waves
S (3,007 waves calculated)
(See discussion below Equation 3-8)

_l I I I I
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2
1.42
Scaled Heigtit (H/Hrms)
Figure 3-3. Theoretical and Observed Wave-Height Distributions.
Observed distributions for 15 individual IS-minute
observations from several Atlantic coast wave gages
are superimposed on the Rayleigh distribution curve
3-7
0.000
0.0005
0.001
Q. 0.005
- 0.0
o
.a
o
a>
>
0.05
3 From seventy-two 15-minute samples
E
containing a total of 11,678 waves
(15,364 calculated waves)
0.5
Hg Theory
J__L
1.0 1.2 1.4 I 2.0 2.2 2.4 2.6 2.8 3.0 3.2
A
1.42
Scaled Heigtit (H/Hrms)
Figure 3-4. Theoretical and Observed Wave-Height Distributions.

Observed waves from 72 individual 15-minute
observations from several Atlantic coast wave gages
are superimposed on the Rayleigh distribution curve
3-8
The average height of all waves with heights greater than H (H) can be
obtained from the equation.
"^'"^
/ h' e
'
' d H
H
H(H) = (3-8)
/ H eVHrmJ dH
H
or from curve b in Figure 3-5. By setting H = 0, all waves are con-

sidered, and it is found that the average wave height is given by
H = 0.886 H rms '

(3-9)
and the significant wave height is given by
H = 1.416H „, « vTh (3-10)
In the analysis system used by CERC from 1960 to 1970/ and whenever
digital recordings cannot be used, the average period of a few of the best
formed waves is selected as the significant wave period. An estimated
number of equivalent waves in the record is obtained by dividing the
duration of the record by this significant period. The highest waves
are then ranked in order with the highest wave ranked 1. The height of
the wave ranked nearest 0.135 times the total number of waves is taken as
the significant wave height. The derivation of this technique is based on
the assumption that the Rayleigh distribution law is exact. Harris (1970)
showed that this procedure agrees closely with values obtained by more
rigorous procedures which require the use of a computer. These procedures
are described in Section 3.23, Energy Spectra of Waves.
The following problem illustrates the use of the theoretical wave

height distribution curves given in Figure 3-5.
************** EXAMPLE PROBLEM **************

GIVEN Based on an analysis of wave records at a coastal location, the
:
significant wave height Hg was estimated to be 10 feet.
FIND :
(a) Hjo (Average of the highest 10 percent of all waves)
(b) H-^ (Average of the highest 1 percent of all waves)
3-10
SOLUTION : 1? = Hg = 10 feet
Using Equation 3-10
Hs = 1.416 H^,
rms
or
H''"^ .-1!^ = -15-. 7.06 ft

1.416 1.416
(a) From Figure 3-5, curve b, it is seen that for P = 0.1 (10 percent)
^^ = 1.80(7.06) =
« 1.80; Hjo = 1.80 H 12.7 feet, say 13 feet.
rms
(b) Similarly, for P = 0.01 (1 percent)
"/
2.36; H^ = 2.36 H^^^ = 2.36 (7.06) = 16.7 ft., say 17 ft.
rtns
Note that;
^10 12.7
or H,„
^<^
= 1.27 H.
H$ 10
and
^ _
~
1^ or H^ = 1.67 H^
H, 10
*************************************
Goodknight and Russell (1963) analyzed wave-gage observations taken
on an oil platform in the Gulf of Mexico during four hurricanes. They
found agreement adequate for engineering_application between such important
parameters as Hg, Hio, ihiax, ^rms^ and H, although they did not find
consistently good agreement between measured wave-height distributions
and the entire Rayleigh distribution. Borgman (1972) substantiates this
conclusion from wave observations from other hurricanes. These findings
are consistent with Figures 3-3, and 3-4, based on wave records recently
obtained by CERC from shore-based gages. The CERC data include waves from
both extratropical storms and hurricanes.
3.23 ENERGY SPECTRA OF WAVES
The significant wave analysis, although simple in concept, is diffi-

cult to do objectively, and does not provide all of the information needed
in engineering design.
appears from Figure 3-1 that the wave field might be better de-
It
scribed by a sum of sinusoidal terms. That is, the curves in Figure 3-1
might be better represented by expressions of the type
N
77(t)= S a. cos(w.-0.) , (3-11)
J
j= 1 J '
where n(t) is the departure of the water surface from its average posi-
tion as a function of time, aj is the amplitude, toj is the frequency,
and (jjj is the phase of the j*^ wave at the time t = 0. The values
of oj are arbitrary, and cj may be assigned any value within suitable
limits. In analyzing records, however, it is convenient to set
- ZiTJ/D,
(iijj where j is an integer and D is the duration of the obser-
vation. The dij will be large only for those w^- that are prominent in
the record. When analyzed in this manner, the significant period may be
defined as D/j where j
, is the value of j corresponding to the
largest a^-.
It was shown by Kinsman (1965) , that the average energy of the wave
train is proportional to the average value of [n(t)]^. This is identical
to a^ where o is the standard deviation of the wave record. It can
also be shown that
1 ^
o' = - S a?. (3-12)
Experimental results and calculations based on the Rayleigh distri-

bution function show that the significant wave height is approximately
equal to 4a. Thus, recalling that
". '^ v/2 H^^^
and
H^ « 4a
then
a == 0.25 \/2^H^^^, (3-13)
or
^rms * 2 V2' o. (3-14)
The a? may be regarded as approximations to the energy spectrum

function E(oj) where
a?
E(cj) Aw = -^ . (3-15)
3-12
Thus
E(co)da; . (3-16)
The spectrum E((jj) permits one to assign specific portions of the

total wave energy to specific frequency intervals, to recognize that two
or more periods may be important in describing the wave field, and to give
an indication of their relative importance. This also permits a first
approximation to the calculation of velocities and accelerations from a
record of the wave height in a complex wave field. Several energy spectra,
computed from coastal zone wave records obtained by CERC, are shown in
Figure 3-6.
The international standard unit for frequency measure is the hertz^

defined as one cycle per second. The units aycZes pev second and radians
per second are also widely used. One hertz = 2it radians per second.
3.24 DIRECTIONAL SPECTRA OF WAVES
A more complete description of the wave field is required to recognize

that not all waves are traveling in the same direction. This may be
written as
Tj (x, y, t) = 2 a. cos [co-t - 0- - k (x cos fly + y sin 6-^ , (3- 17)
where k = Ztt/L, and 9^- is the angle between the x axis and the
direction of wave propagation, and â is the phase of the j^^ wave
at The energy density E(e,u)
t = 0. represents the concentration of
energy at a particular wave direction 6 and frequency w
, therefore ,
the total energy is obtained by integrating E(e,{jo) over all directions

and frequencies, thus
27r oo
=
// E(0,a))dwd9 . (3-18)
The concept of directional wave spectra is essential for advanced

wave-prediction models, but technology has not yet reached the point where
directional spectra can be routinely recorded or used in engineering design
studies. Therefore, directional wave spectra are not discussed extensively
here.
3-13
Period (Seconds)
2.5 10 2.5
+ 0.4
iiJ
e
> + 0.2
o
—
o
o
0.0
OjO 0.2 0.4 0-0 2 0.4
Frequency (Hertz)
Savannah Coast Guard Light Tower
Period (Seconds)
10 2.5 10 2.5
0.4
T = 10.8 Sec
H, 6.9 Ft
liJ
> 0.2
o
o
I-
0.0 ^-«
0.0 0.2 4
Frequency (Hertz)
Nags Head, N.C.
Figure 3-6. Typical Wave Spectra from the Atlantic Coast.
The ordinate scale is the fraction of total
wave energy- in each frequency band of 0.011
Hertz, (one Hertz is one cycle/second). A
linear frequency scale is shown at the bottom
of each graph and a nonlinear period scale
at the top of each graph.
3-14
3.3 WA\"E FIELD
3.31 DBTLOPMEJiT OF k MkVE FIELD
Various descriptions of the aechanisa of wave gîeratioo by vind have

been given, and significant progress in explaining the aechanisa nas re-
ported by Miles C1957) and Phillips (19S7) . Inteir^r-: : ; cnssions of the
'
results of aany of the aore pr«»inrait descripticr.s :f i -er.eraticHi by

wind are given by Kinsaan (1965), Phillips (1966), "i I 1971).
Laboratory studies, (Hidy and Plate, 1966^ ^-i S-tzci:. i

1966), carefully designed to Batch the assusipti:-.5 r=.de by Miles and by
Phillips show reasonably good arrTe-^-" -:'- -'- z - rtretical rrefirtir::-
SusBaries of various filed studies .- -i .1> ->^7) deacr^trate "ir.a":
.
theory provides a reasonable frirt-:r- f:r -.r liI sis of observations.
The Nliles-Riillips theor>- as t -.ttiti ^- i i:m.-.i- :. t;;-eriBental

data pemits the foiBulation of a. differential equzticfa governing tie
growth of wave energ>-. This equat::- e written in a variety of ways.
(Inoue, 1966, 1967) and Bamett , 1.-: is approadi will cot be disr^s:
in detail because it requires = itrzi lii^zity co^pater i- i -:rT i-i.rz-
logical data than is likely to :e f:_-z r ztt": i- = -a;;r f:rT:iit :e--er.
A brief discussion of the physical concepi; tiloj-ed in tie ci-r-ter

wave forecast, however, is presented to show tT rtcfings a- i -Trits
of simpler procedures that can be used in wave irT.asting.
Growth and dissipation of wave energ)- are very sensitive to wave

frequency and wave direction relative to the wind direction. Ihus it is
desirable to consider each narrow band of directi(ms and frequencies
separately. A change in wave energy depends on the advecticn of energy
into and out of a region; trans forBatioo of the wind's kinetic energy
into the energ>- of water waves; dissipation of wave energy into tmbaleiice
and by fricticm, \-iscosity and breaking; and trans foraatirr. cf vave energy
at one frequency into wave energ>- at other frequencies.
Mave energy- is discussed in Section 2.258. Wave Energj- and Power.

-Althou^ it is known that ener5>' transfers from one band of wave frequen-
cies to another do take place, this trocess is secondary- to the transfer
of energ>- froa the atBosx)here to the sea, and is not yet well enon^ onder-
stood to justify its consideration in a practical wave prediction sdieae.
Riillips (1957) shewed that the turbulence assrciated with thf fl:-
of wind near the water v.c>uld create traveling pressure pulses. Thtir
pulses generate waves traveling at a speed appropriate to the diaensirn;
of the pressure pulse. Wave grc«t;h ry t-i5 7r::e55 i; - - f-
"
the waves are short and ^liien their speed is luenticu. _i-. : .;:
. t
of the wind velocity in the directirn cf wave travel. The err in; ii:a
1
analyzed by Inoue (1966, 1967) indicate- \' - r effect of : r: lei

-.'-
pressure pulses is real, but is only ii:_i ;r. -i eniieth as _^rge ^s tîe
original theor>' indicated.
3-15
Miles (1957) showed that the waves on the sea surface must be
matched by waves on the bottom surface of the atmosphere. The speed
of air and water must be equal at the water surface. Under most meteoro-
logical conditions, the air speed increases from near to 60 - 90 percent
of the free air value within 66 feet (20 meters) of the water surface.
Within a shear zone of this type, energy is extracted from the mean flow
of the wind and transferred to the waves. The magnitude of this transfer
at any frequency is proportional to the wave energy already present at
that frequency. Growth is normally most rapid at high frequencies. The
energy transfer is also a complex function of the wind profile, the
turbulence of the air stream, and the vector difference between wind and
wave velocities.
The theories of Miles and Phillips predict that waves grow most
rapidly when the component of the wind speed in the direction of wave
propagation is equal to the speed of wave propagation.
The wave generation process discussed by Phillips is very sensitive

to the structure of the turbulence. This is affected significantly by
any existing waves, and the temperature gradient in the air near the
water surface. The turbulence structure in an offshore wind is also
affected by land surface roughness near the shore.
The wave generation process discussed by Miles is very sensitive to

the vertical profile of the wind. This is determined largely by turbulence
in the wind stream, the temperature profile in the air, and by the rough-
ness of the sea surface.
Shorter waves grow most rapidly. Those waves which propagate obliquely
to the wind are favored, for they are better matched to the component of
the wind velocity in the direction of wave propagation than those moving
parallel to the wind. Thus, the first wave pattern to appear for short
fetches and durations consists of two wave trains forming a rhombic
pattern with one diagonal along the direction of the mean wind.
There is a limit to the steepness to which a wave can grow without

breaking. Shorter waves reach their limiting growth rather quickly;
longer waves, which grow more slowly but can obtain greater heights,
then become more prominent. Thus, the apparent direction of propagation
of the two wave trains tends to coalesce with increasing fetch and duration.
The length of the region in which a rhombic pattern is apparent may extend
from a few meters to a few kilometers depending on the width of the basin,
the wind speed, and previously existing waves.
Wave growth is significantly affected by any preexisting waves. The

empirical data analyzed by Inoue (1966, 1967) indicated that the magnitude
of the effect of seas already present is about eight times the value given
in the original Miles (1957) theory. Neglecting this effect in early wave
prediction theories has led to large errors in computing the duration
required for a fully arisen sea. There are many situations in which the
largest waves and the waves growing most rapidly are not being propagated
in the wind direction.
3-16
3.32 VERIFICATION OF WAVE HINDCASTING
Inoue (1967) prepared hindcasts for Weather Station J (located near

53°N, 18°W), for the period 15-28 December, 1959, using a differential
equation embodying the Miles-Phillips theory to predict wave growth. A
comparison of significant wave heights from shipboard observations and by
hindcasting at two separate locations near the weather ships is shown in
Figure 3-7. The location of Ocean Weather Ship J, the mesh points used in
the numerical calculations, and four other locations discussed below are
are shown in Figure 3-8. The calculations required meteorological data
from 519 grid points over the Atlantic Ocean as shown in Figure 3-8. The
agreement between observed and computed values seems to justify a high
level of confidence in the basic prediction model. Observed meteorological
data were interpolated in time and space to provide the required data,
thus these predictions were hindcasts.
Bunting and Moskowitz (1970) and Bunting (1970) have compared fore-
cast wave heights with observations, using the same model with comparable
results.
By 1970, it was generally believed that the major remaining difficulty

in wind wave prediction was the determination of the surface wind field over
the ocean. (Pore and Richardson, 1967), and (Bunting, 1970). It is partly
because of the difficulty in obtaining a satisfactory specification of the
wind field over the sea that simpler wave prediction systems are still
being used operationally. (Pore and Richardson, 1969), (Shields and Burdwell,
1970), and (Francis, 1971.)
3.33 DECAY OF A WAVE FIELD
Wind-energy can be transferred directly to the waves only when the

component of the surface wind in the direction of wave travel exceeds the
speed of wave propagation. Winds may decrease in intensity, pass over
land, or change in direction to such an extent that wave generation ceases,
or the waves may propagate out of the generation area. When any of these
events occurs, the wave field begins to decay. Wave energy travels at a
speed which increases with the wave period. Thus the energy packet leaving
the generating area spreads out over a larger area with increasing time.
The apparent period at the energy front increases and the wave height
decreases. If the winds subside before the sea is fully arisen, the
longer waves may begin to decay while the shorter waves are still growing.
This possibility is recognized in advanced wave prediction techniques.
The hindcast spectra, computed by the Inoue (1967) model and published by
Guthrie (1971) show many examples of this for low swell, as do the aerial
photographs and spectra given by Harris (1971). (See Figures 3-2 and 3-6.)
This swell is frequently overlooked in visual observations and even in the
subjective analysis of pen and ink records from coastal wave gages.
Most coastal areas of the United States are so situated that most of
the waves reaching them are generated in water so deep that depth has no
effect on wave generation. In many of these areas, wave characteristics
3-17
o o
77^
. • \ •
may be determined by first analyzing meteorological data to find deep-
water conditions. Then, by analyzing refraction (Section 2.32, General -
Refraction by Bathymetry.), the changes in wave characteristics as the
wave moves through shallow water to the shore may be found. In other
areas, in particular along the North- Atlantic coast, wKere the bathymetry
is complex, refraction procedure results are frequently difficult to inter-
pret, and the conversion of deepwater wave data to shallow-water and near-
shore data becomes laborious and sometimes inaccurate.
Along the Gulf coast and in many inland lakes, generation of waves by
wind is appreciably affected by water depth. In addition, the nature and
extent of transitional and shallow-water regions complicate ordinary re-
fraction analysis by introducing a bottom-friction factor and associated
wave energy dissipation.
3.4 WIND INFORMATION NEEDED FOR WAVE PREDICTION
Wave prediction from first principles, as described above, requires

very detailed specification of the wind field near the water surface. This
is generally developed in two steps: (1) Estimation of the mean fvee air
wind speed and direction, (This step may be omitted for reservoirs and small
lakes if surface wind observations are available.), and (2) Estimation of
the mean surfaae wind speed and direction.
When the full wave generation process is considered, a large capacity

computer must be used for the calculations, and fairly complex procedures
may be used for determining the wind field. Engineers who require wave
hindcasts for only a few locations, and perhaps for only a few dates must
employ simpler techniques. A brief discussion of the processes involved
in determining the surface wind and techniques suitable for use in deter-
mining the characteristics of the wind field needed for the simplified wave
prediction model described in Section 3.5, Simplified Wave Prediction Models,
are given in this section. These procedures will be accurate (within 20
percent) about two-thirds of the time. The following discussion provides
guidance for recognizing cases in which the simplified procedures are not
appropriate. Errors resulting from disregarding the exceptional situations
tend to be random. Thus climatological summaries, based on hindcast data,
may be much more accurate than the individual values that go into them.
Wind reports from ships at sea are generally estimates based on the
appearance of the waves, the drifting of smoke, or the flapping of flags
although some are anemometer measurements. Actually, even if all ships
were equipped with several aneometers, the wind field over the sea would
still not be known in sufficient detail or precision to permit full
exploitation of modem theories for wave generation.
Fortunately, estimates of the surface wind field that are usefully

accurate most of the time can be based on the isobaric pattern of synoptic
weather charts
3-20
Horizontal pressure gradients arise in the atmosphere primarily
because of density differences, which in turn are generated primarily by
temperature differences. Wind results from nature's efforts to eliminate
the pressure gradients, but is modified by many other factors.
The pressure gradient is nearly always in approximate equilibrium with

the acceleration produced by the rotation of the earth. The geostrophia
wind is defined by assuming that exact equilibrium exists, and is given by
1 dp
U = -T
* pt /
an
. (3-19)
where U^ is the wind speed, p^ the density of the air, f the coriolis
parameter, f = 2w sine}), where to = 7.292 x 10"^ radians/second and (f)
is the latitude, and dp/dn is the horizontal gradient of atmospheric

pressure. A graphic solution of this equation is given in Figure 3-11,
Section 3.41, Estimating the Wind Characteristics. The geostrophic wind
blows parallel to the isobars with low pressure to the left, when looking
in the direction toward which the wind is blowing, in the Northern
Hemisphere, and low pressure to the right in the Southern Hemisphere.
Geostrophic wind is usually the best simple estimate of the true wind in
the free atmosphere.
When the trajectories of air particles are curved, equilibriiim wind

speed is called gradient wind. Gradient wind is stronger than geostrophic
wind for flow around a high pressure area, and weaker than geostrophic wind
for flow around low pressure. The magnitude of the difference between
geostrophic and gradient winds is determined by the curvature of the tra-
jectories. If the pressure pattern does not change with time and friction
is neglected, trajectories are parallel with the isobars. The isobar curva-
ture can be measured from a single weather map, but at least two maps must
be used to estimate trajectory curvature. There is a tendency by some
analysts to equate the isobars and trajectories at all times, and to
compute the gradient wind correction from the isobar curvature. When the
curvature is small, but the pressure is changing, this tendency may lead
to incorrect adjustments. Corrections to the geostrophic wind that can-
not be determined from a single weather map are usually neglected, even
though they may be more important than the isobaric curvature effect.
The equilibrium state is further disturbed near the surface of the

earth by friction. Friction causes the wind to cross the isobars toward
low pressure at a speed lower than the wind speed in the free air. Over
water, the average surface wind speed is generally about 60 to 75 percent
of the free air value, and wind crosses the isobars at an angle of 10 to
20 degrees. In individual situations, the magnitude of the ratio between
the surface wind speed and the computed free air speed may vary from 20 to
more than 100 percent, and the crossing angle may vary from 0° to more than
90°. The magnitude of these changes is determined by the vertical tempera-
ture profile and the turbulent viscosity in the atmosphere.
3-21
3.41 ESTIMATING THE WIND CHARACTERISTICS
To predict wave properties from meteorological data by any of the

simplified techniques, it is necessary to:
(a) Estimate the mean surface wind speed and direction, as dis-
cussed in Section 3.4, Wind Information Needed for Wave Prediction;
(b) delineate a fetch over which the wind is reasonably constant

in speed and direction, and measure the fetch length, and
(c) estimate wind duration over the fetch.
These determinations may be made in many ways depending on the loca-

tion and the type of meterological data available. For restricted bodies
of water, such as lakes, the fetch length is often the distance from the
forecasting point to the opposite shore measured along the wind direction.
There is no decay distance, and it is often possible to use observational
data to determine wind speeds and durations.
When forecasting for oceans or other large bodies of water, the most
common form of meteorological data used is the synoptic surface weather
chart. (^Synoptic means that the charts are drawn by analysis of many
individual items of meteorological data obtained simultaneously over a
wide area.) These charts depict lines of equal atmospheric pressure,
called isobars. Wind estimates at sea, based on an analysis of the sea-
level atmoshperic pressure are generally more reliable than wind observa-
tions because pressure, unlike wind, can be measured accurately on a
moving ship. Pressures are recorded in millibars, 1,000 dynes per square
centimeter. One thousand millibars (a bar) equals 29.53 inches of mercury
and is 98.7 percent of normal atmoshperic pressure.
A simplified surface chart for the Pacific Ocean is shown in Figure

3-9, which is drawn for 27 October 1950 at 0030Z (0030 Greenwich mean
time). Note the area labelled L in the right center of the chart, and
the area labelled H in the lower left corner of the chart. These are
low- and high-pressure areas; the pressures increase moving out from L
(isobars 972, 975, etc.) and decrease moving out from H (isobars 1026,
1023, etc.).
Scattered about the chart are small arrow shafts with a varying num-
ber of feathers or barbs. The direction of a shaft shows the direction
of the wind; each one-half feather represents a unit of 5 knots (2.5
meters/second) in wind speed. Thus, in Figure 3-9 near the point 35°N.
latitude, 135°W. longitude, there are three such arrows, two with 5h
feathers which indicate a wind force of 31 to 35 knots (15 to 17.5 meters/
second) , and one with 3 feathers indicating a force of 26 to 30 knots
(13 to 15 meters/second)
On an actual chart, much more meteorological data than wind speed and
direction are shown for each station. This is accomplished by the use of
3-22
o
LO
U
(U
£i
O
t->
O
O
CNJ
M
O
tn
o
o
(-1
3
CO
to
(U
•rt
3-23
coded symbols, letters, and numbers placed at definite points in relation
to the station dot. A sample model report, showing the amount of informa-
tion possible to report on a chart, is shown in Figure 3-10. Not all of
the data shown on this plot are included in each report, and not all of
the data in the report are plotted on each map.
Figure 3-11 may be used to facilitate computation of the geostrophic

wind speed. A measure of the average pressure gradient over the area is
required. Most synoptic charts are drawn with either a 3- or 4-millibar
spacing. Sometimes when isobars are crowded, intermediate isobars are
omitted. Either of these standard spacings is adequate as a measure of
the geographical distance between isobars. Using Figure 3-11, the distance
between isobars on a chart is measured in degrees of latitude (an average
spacing over a fetch is ordinarily used), and the latitude position of the
fetch is determined. Using the spacing as ordinate and location as abscissa,
the plotted or interpolated slant line at the intersection of these two
values gives the geostrophic wind speed. For example, in Figure 3-9, a
chart with 3-millibar isobar spacing, the average isobar spacing (measured
normal to the isobars) over F2, located at 37°N. latitude, is 0.70° of
latitude. Using the scales on the bottom and right side of Figure 3-11,
a geostrophic wind of 67 knots is found.
Geostrophic wind speeds are generally higher than surface wind speeds.
The following instructions, U.S. Fleet Weather Facility Manual (1966), are
recommended for obtaining estimates of the surface wind speeds over the
open sea from the geostrophic wind speeds:
(a) For moderately curved to straight isobars - no correction is

applied.
(b) For great anticyclonic (clockwise movement about a high pressure

center in Northern Hemisphere and counter-clockwise in Southern Hemisphere)
curvature - add 10 percent to the geostrophic wind speed.
(c) For great cyclonic (counter-clockwise movement about a low

pressure center in Northern Hemisphere and clockwise in Southern Hemisphere)
curvature - subtract 10 percent from the geostrophic wind speed.
Frequently the curvature correction can be neglected since isobars

over a fetch are often relatively straight. The gradient wind can always
be computed if more refined computations are desired.
To correct for air mass stability, the sea-air temperature difference

must be computed. This can be done from ship reports in or near the fetch
area, aided by climatic charts of average monthly sea surface temperatures
when data are too scarce. The correction to be applied is given in Table
3-1. (U.S. Fleet Weather Facility Manual, 1966.)
Over oceans, the surface winds generally cross the isobars toward low
pressure at an angle of 10° to 20°.
3-24
Wind speed ( 23 to 27 knots ) Cloud type (Altostrotus)
ff
Cloud type (Dense cirrus
Itches)
Barometric pressure in
True direction from whict) tenths of millibars reduced

wind is blowing to sea level. Initial 9or 10
^^
and the decimal point ore
Current oir temperoture ppp omitted (247= 1024.7 mb)
(31° F) jj
Pressure change in 3 hours
Total onnount of cloud proceeding observotion
( Completely covered )
_(28 : 2.8mb)
|^
Visibility in miles and Characteristic of barograph

froction (3/4 mile) trace (Foiling or steody,
VV-»-^^^ then rising, or rising, then
_rising more quickly)
Present weottier.
Continuous ligtit snow WW Plus or minus sign showing
whether pressure is higher
Temperature of dew _or lower then 3 hours ago
point ( 30° F) T(J
Time precipitation begon
_or ended (4 = 3104 hrs.ogo)
Cloud type ( Froctocumulus) C
''
sky covered I
yu Post weather (Rain)
Height at base of cloud iddle cloud
(2 = 300 to 599 feet) ^ (6 = 7 or 8 tenths) Amount of precipitation
pp (45= 0.45 inches)
NOTE : The letter symbols for eoch weather

element ore shown above.
Courtesy United Stotes Weather Bureau
abridged from W.M.O. Code
Figure 3-10. Sample Plotted Report
3-25
I
Ap
Ur
/'nf An
For T : lO^C where

Ap : 3 mb and 4mb - angular velocity of earth,
An : isobar spacing measured in
0.2625 rodians/hour
degrees latitude - latitude in degrees
P 1013.3 mb
1. 247X10"' gm/cm'
f : Coriolis parameter : 2 wsin
0.3
30 35 40 45 50 55 60 70
Degrees Latitude
(offer Bretschneider, 1952a)
Figure 3-11. Geostrophic Wind Scale
3-26
Table 3-1. Correction for Sea-Air Temperature^
Sea Temperature minus
Air Temperature
with this precision. For practical wave predictions it is usually satis-
factory to regard the wind speed as reasonably constant if variations do
not exceed 5 knots (2.5 meters/second) from the mean. A coastline upwind
from the point of interest always limits a fetch. An upwind limit to the
fetch may also be provided by curvature or spreading of the isobars as
indicated in Figure 3-12 (Shields and Burdwell, 1970), or by a definite
shift in wind direction. Frequently the discontinuity at a weather front
will limit a fetch, although this is not always so.
1004
1004 '9°° 996
996
1000
1004
., 1012 1016
1020 1016
1006
1020 ' 1016 '006

1012
1012
Figure 3-12. Possible Fetch Limitations
Estimates of the duration of the wind are also needed for wave predic-
tion. Computed results, especially for short durations and high wind speeds
may be sensitive to differences of only a few minutes in the duration. Com-
plete synoptic weather charts are prepared only at 6-hour intervals. Thus
interpolation between charts to determine the duration may be necessary.
Linear interpolation is adequate for most uses, and, when not obviously
incorrect, is usually the best procedure.
3-28
3.43 FORECASTS FOR LAKES, BAYS, AND ESTUARIES
3.431 Wind Data . The techniques referred to for determination of wind

speeds and directions from isoharic patterns apply generally to ocean
areas. The friction that causes winds to spiral when crossing isobars
and to have a velocity lower than geostrophic or gradient winds is more
variable over land areas. When a fetch is close to land, this variability
will alter anticipated wind directions and velocities. In enclosed or
semienclosed bodies of water, such as lakes and bays, wind speeds and
directions should be taken from actual weather station reports whenever
possible.
In enclosed bodies of water, or in other areas where the wind blows

off the land, differing frictional effects of land and water should be
considered, and indicated wind speeds should be adjusted for these effects.
Studies by Myers (1954) and Graham and Nunn (1959) indicate recommended
adjustments in wind speeds. (See Table 3-2.) The adjustment factor may
vary considerably depending on the shoreline frictional characteristics.
This adjustment is used only for short fetches such as those in reservoirs
and small lakes.
Often, over small or well-defined fetch areas, it is not convenient

or even possible to utilize surface charts to determine wind characteris-
tics. Where wind records exist for locations in or near a fetch area,
these may be utilized. The accuracy of the forecast will depend on the
completeness of the records, the extent of fetch, and the wave prediction
technique employed. Where wind duration records are not available, local
wind speed reports may still be utilized to forecast waves assuming un-
limited durations, that is, wave growth is limited by the available fetch.
Wave characteristics deduced in this way are only qualitative.
Table 3-2.
Saville (1954) proposed a method to determine the effect of fetch
width on wave generation. Figure 3-13, based on this method, indicates
the effective fetch for a relatively uniform fetch width. The following
problem demonstrates the use of Figure 3-13.
************** EXAMPLE PROBLEM *************

GIVEN Consider a channel with a fetch length F = 20 miles, a width
:
W = 5 miles, an average depth d = 35 feet, and a windspeed U = 50 mph

along the long axis.
FIND: Estimate the significant wave height Hg, and the significant
wave period Tg.
SOLUTION : Compute W/F = 5/20 =0.25

From Figure 3-13 for W/F = 0.25, F^/F =0.45
Compute F^. = 0.45 X 20 = 9 miles or 47,500 feet.
Using the forecasting relations given in Section 3.6, Wave Forecasting

for Shallow Water, for a fetch of 47,500 feet and a wind speed of
50 mph and an average uniform depth of 35 feet, the significant wave
height may be determined from Figure 3-27 to be Hg = 5.2 feet, say
5 feet and the significant wave period will be T = 4.6 seconds, say
5 seconds
The preceding example presents a simplified method of determining the

effective fetch. Shorelines are usually irregular, and the uniform-width
method indicated in Figure 3-13 is not applicable. A more general method
must be applied. This method is based on the concept that the width of a
fetch in reservoirs normally places a very definite restriction on the
length of the effective fetch; the less the width-length ratio, the shorter
the effective fetch. A procedure for determining the effective fetch
distance is illustrated in Figure 3-14. It consists of constructing 15
radials from the wave station at intervals of 6° (limited by an angle of
45° on either side of the wind direction) and extending these radials
until they first intersect the shoreline. The component of length of
each radial in a direction parallel to the wind direction is measured and
multiplied by the cosine of the angle between the radial and the wind
direction. The resulting values for each radial are summed and divided
by the sum of the cosines of all the individual angles. This method is
based on the following assumptions:
(a) Wind moving over a water surface transfers energy to the

water surface in the direction of the wind and in all directions within
450 on either side of the wind direction.
(b) The wind transfers a unit amount of energy to the water along
the central radial in the direction of the wind and along any other radial
an amount modified by the cosine of the angle between the radial and the
wind direction.
3-30
( U.S. Army, B.E.B. Tech. Memo No 132,1962)
Figure 3-14. Computation of Effective Fetch for Irregular Shoreline
3-32
(c) Waves are completely absorbed at shorelines.
Fetch distances determined in this manner usually are less than those
based on maximum straight- line distances over open water. This is true
because the width of the fetch places restrictions on the total amount of
energy transferred from wind to water until the fetch width exceeds twice
the fetch length.
While 6° spacing of the radials is used in this example, any other

angular spacing could be used in the same procedure.
3.5 SIMPLIFIED WAVE-PREDICTION MODELS
Use of the wave-prediction models discussed in Section 3.3, Wave

Field, requires an enormous computational effort and more meteorological
data than one is likely to find outside of a major forecasting center.
The Fleet Numerical Weather Center, Monterey, California began using this
model on an experimental basis for a small part of the globe early in 1972.
Expansion to larger regions is planned. Wave prediction begins with a com-
putation of the existing wave field (often called a zero-time prediction),
and continues with a calculation of the effects of predicted winds on the
waves. A few years after this system is operational, it should be possible
to supply the needs for wave-hindcast statistics by compilations of zero-
time predictions. In the meantime, engineers who require wave statistics
derived by hindcasting techniques for design consideration must accept
simpler techniques.
Computational effort required for the model discussed in Section 3.31,

Development of a Wave Field, can be greatly reduced by the use of simpli-
fied assumptions with only a slight loss in accuracy for wave height cal-
culations, but sometimes with significant loss of detail on the distribu-
tion of wave energy with frequency. One commonly used approach is to
assume that both duration and fetch are large enough to permit an equi-
librium state between the mean wind, turbulence, and waves. If this
condition exists, all other variables are determined by the wind speed.
Pierson and Moskowitz (1964) consider three analytic expressions which

satisfy all of the theoretical constraints for an equilibrium spectrum.
Empirical data, described by Moskowitz (1964) were used to show that the
most satisfactory of these is
-flCw^/w*)
E(co)da; = (ag^/co*) e " dco ,
(3-20)
where a and 3 are dimensionless constants, a = 8.1 x 10 3, g = 0.74

and too - g/U, where g is the acceleration of gravity and U is the
wind speed reported by weather ships, and o) is the wave frequency
considered.
Equation 3-20 may be expressed in many other forms. Bretschneider

(1959, 1963) gave an equivalent form, but with different values for a
and 3. A similar expression was also given by Roll and Fischer (1956).
The condition in which waves are in equilibrium with the wind is called a
3-33
fully arisen sea. The assumption of a universal form for the fully arisen
sea permits the computation of other wave characteristics such as total
wave energy, significant wave height, and period of maximum energy. The
equilibrium state between wind and waves rarely occurs in the ocean, and
may never occur for higher wind speeds.
A more general model may be constructed by assuming that the sea is
calm when the wind begins to blow. Integration of the equations governing
wave growth then permits the consideration of changes in the shape of the
spectrum with increasing fetch and duration. If enough wave and wind
records are available, empirical data may be analyzed to provide similar
information. Pierson, Neumann, and James (1955) introduced this type of
wave prediction scheme based almost entirely on empirical data. Inoue
(1966, 1967) repeated this exercise in a manner more consistent with the
Miles-Phillips theory using a differential equation for wave growth. Inoue
was a member of Pierson 's group when this work was carried out, and his
prediction scheme may be regarded as a replacement for the earlier
Pierson-Neuman-James (PNJ) wave prediction model. The topic has been
extended by Silvester and Vongvisessomjai (1971) and others.
These simplified wave prediction schemes are based on the implicit
assumption that the waves being considered are due entirely to a wind
blowing at constant speed and direction for an overwater distance called
the fetch and for a time period called the duration.
In principle it would be possible to consider some effects of variable
wind velocity hy tracing each wave train. Once waves leave a generating
area and become swell, the wave energy is then propagated according to the
group velocity. The total energy at a point, and the square of the signif-
icant wave height could be obtained by adding contributions from individual
wave trains. Without a computer, this procedure is too laborious, and
theoretically inaccurate.
A more practical procedure is to relax the restrictions implied by
derivation of these schemes. Thus wind direction may be considered con-
stant if it varies from the mean by less than some finite value, say 30°.
Wind speed may be considered constant if it varies from the mean by less
than ± 5 knots (2.5 meters/second) or h barb on the weather map. This
assumption is not much greater than the uncertainty inherent in wind
reports from ships. In this procedure, average values are used and are
assumed constant over the fetch area, and for a particular duration.
The theoretical spectra for the partially arisen sea can be used
to develop formulas for such wave parameters as total energy, significant
wave height and period of maximum energy density.
Similar formulas can also be developed empirically from wind and wave
observations. A quasi-empirical - quasi-theoretical procedure was used by
Sverdrup and Munk (1947) to construct the first widely used wave predic-
tion system. The Sverdrup-Munk prediction curves were revised by
Bretschneider (1952a, 1958) with additional empirical data. Thus, this
prediction system is often called the Sverdrup-Munk-Bretschneider (SMB)
method. It is the most convenient wave prediction system to use when a
limited amount of data and time are available.
3-34
3.51 SMB METHOD FOR PREDICTING WAVES IN DEEP WATER
Revisions of earlier SMB forecasting curves are seen in Figures 3-15

and 3-16. The curves represent dimensional plots from the empirical
equations.
0.42
^ = 0.283 tanh 0.0125 (3-21)
U^
,0.25
'gF
= 1.20 tanh 0.077 (3-22)
2nU
and.
^
= Kexp B In + Db (3-23)
U
where
exp (x) =
In loge
K 6.5882
A 0.0161
B 0.3692
C 2.2024
and
D = 0.8798
With these relationships, the significant wave height H^, and significant
wave period T^, at the end of a fetch may be estimated when wind speed,
fetch length, and duration of wind over the fetch are given. Estimation
of wind parameters is discussed in Section 3.4, Wind Information Needed
for Wave Prediction, following the evaluation of simplified prediction
techniques. In using Figures 3-15 or 3-16 the estimated wind velocity U,
the fetch length F, and the estimated wind duration t, when a fetch
first appears on a weather chart, are tabulated. Figure 3-15 or 3-16 is
then entered with the value of U, using the scale on the left if U is
in knots, or the scale on the right if U is in statute miles per hour.
This U line is then followed from the left side of the graph across to
its intersection with the fetch length or F line, or the duration t
line, whichever comes first.
3-35
( jnoH J8d s»|iW^ P88ds puiiw
~ in <\t o a> u> ~
^
( jnoH Jad S8|iw ) paads puim
f s
************** EXAMPLE PROBLEM *************
GIVEN : A wind speed U = 35 knots (40 mph) and a duration t = 10 hours
FIND The significant wave height

: H^, and the significant period T^^,
at the fetch front for:
(a) A fetch length, F = 200 nautical miles and
(b) a fetch length, F = 80 nautical miles.
SOLUTION :
(a) Enter Figure 3-15 from the left side at U = 35 knots and move
horizontally across the figure from the left toward the right, until
intersecting the dashed line for a duration of 10 hours that comes
before the line indicating a fetch length of 200 nautical miles. At
the 10-hour duration line F = 92 nautical miles; this is the minimum
fetch F^ for this case.
With U = 35 knots, t = 10 hours, and F = 200 nautical miles then

H^ = 13.1 feet, T^ = 8.0 seconds, t^ equals 10 hours, and F^ = 92
nautical miles.
(b) Entering Figure 3-15 as above, when F = 80 nautical miles

and t = 10 hours, then the heights, periods, minimum duration and
fetch would be H^ = 12.6 feet, T^ = 7.8 seconds, and t^ = 9.0 hours.
The minimum duration t^ is 9 hours, corresponding to the miles
which limit generation, and comes before a duration of 10 hours.
In this example problem, the wave pattern in (a) is limited by the

duration; the wave pattern in (b) is limited by the fetch.
When a series of surface synoptic weather charts (Fig. 3-9) are used
to determine wave patterns, the values of U, F and t can be tabulated
for the first chart. For the same fetch on a later chart drawn for a time
Z, after the first chart, U, F, and t are again tabulated. Using
the subscript 2 to refer to those of the second chart and subscript 1
to refer to those of the first chart, if U2 = U-^ the above procedures
,
should be followed using either t2 = tî + Z or F2. If, however, U2 f U^

certain additional assumptions must be made before using the forecasting
curves
A change in wind speed from Uj^ to U2 in a time Z between charts may

be assumed to take place instantaneously at a time Z/2. Waves due to Ui
may then be calculated by assuming that the first chart minimum duration
time has been lengthened by an amount Z/2 or that its minimum fetch has
been changed by AF/2, where AF represents the change in fetch length
between weather charts. Since at the assumed abrupt change in wind speed.
3-38
the energy imparted to the waves by U^ , with a minimiim duration t^^ + Z/2
for a minimum fetch F^^ + F/2, does not change, U2 will be assumed to
impart energy to waves which already contain energy due to Uj^.
Plotted on Figures 3-15 and 3-16 are dotted lines of constant H^T'^ which
are considered lines of constant wave energy. To a first approximation,
deepwater wave energy is given by
If energy had been imparted to the waves by

"88
- ^^ = M2pg(m:
U2
.
acting alone, these waves

would be of length and height given in Figures 3-15 or 3-16 by the inter-
section of the U2 ordinate with the constant energy line (plotted or
interpolated) corresponding to energy imparted by Ui with a minimum
duration of t^-^ + Z/2 or a minimum fetch F^^ + F/2. By increasing
the minimum duration at this point by an amount Z/2 or by changing the
minimum fetch by an amount AF/2, wave conditions under U2 at the
time of the second chart may be approximated.
For example, if the wind speed increases so that U2 = 40 knots, and

with Ui = 35 knots, t^-^ = 10 hours, F^^ = 92 nautical miles, t^^ + Z/2 =
13 hours; an interpolated (by eye) dotted line of constant H^T^ would be
followed up to the LI2 = 40-knot line where the duration = 6.5 hours. To
this value Z/2 or 3 hours is added and then moving horizontally along
the line U2 = 40 knots to t = 6.5 + 3.0 = 9.5 hours, it is found that
H^2 - 15.6 feet, T^2 ~ ^-^ seconds, t^2 ~ ^-^ hours, and F^2 - ^^ nautical
miles. If the measured fetch F2 had been less than 95 nautical miles,
this length of fetch would limit the growth of waves. Although the pre-
ceding discussion would indicate that AF should be calculated, in practice
this need not be done; the results obtained through calculation of AF
would be found by reading off wave heights at the intersection of U2 and
F2 if F2 is limiting. Therefore, if F2 had equalled 85 nautical miles,
in this case less than 95 miles and therefore limiting, at the intersection
of U2 = 40 knots, then Hp2 = 15.0 feet, T^2 =8.5 seconds, T^2 = 8. 8 hours,
and F^2 - 85 nautical miles. Note this important distinction: t;^, F^ and Fj
are calculated by use of Figure 3-15. Some of the measured and calculated
values will be the same, but not all of them.
If the wind velocity U2 is less than Ui , the procedures followed

are nearly the same. From the intersection of U-^ and t^-^ + Z/2 a
constant energy line is followed to its intersection, if there is one.
3-39
with either U2 or F2 whichever comes first from the left side of the
figure. If U comes first, Z/2 is added to the duration at this point,
and the U2 ordinate is followed to either this new duration or to the
F2 whichever is first from the left side of the chart. (Compare with the
preceding paragraph.) At this point, Hp2' ^F2' hn2> ^^^ ^m2 inclusive,
are read off. If the constant energy line had intersected F2 before U2,
it is only necessary to drop down along the F2 abscissa to its intersec-
tion ^'ith U2, and at this point read H^2> '^F2> ^2 ^^^ ^m2- (This pro-
cedure could be used for many cases in which U2 is greater than Uj.)
The major differences in technique are used when U2 is less than Ui

and the H^T^ = constant line from the intersection of Ui and tmi + Z/2
does not intersect either U^ or F Forecasting theory used here pre-
.
dicts that waves due to a constant wind blowing over an unlimited fetch
for an unlimited duration will eventually reach limiting height and period
distributions beyond which growth will not continue. In Figure 3-16 the
limit of this state is delineated by the line labelled maximum condition.
To the right of this line, it is assumed that any energy transport to the
waves by the wind is compensated by wave breaking, hence no wave growth
occurs
3.52 EFFECTS OF MOVING STORMS AND A VARIABLE WIND SPEED AND DIRECTION
In principle, it should be possible to extend the Inoue differential

equation for wave growth to highly irregular conditions, but no experi-
mental verification of this concept has been published. Kaplan (1953)
and Wilson (1955) have proposed techniques for applying the simplified
prediction techniques to variable wind fields and changing fetches. The
procedures appear reasonable and these techniques are used, although no
statistics are available for verification.
3.53 VERIFICATION OF SIMPLIFIED WAVE HINDCAST PROCEDURES
Comparisons of hindcast wave heights and observed wave heights,

similar to Figure 3-7, have been given by Jacobs (1965) for the PNJ wave-
prediction system, by Bates (1949) and Isaacs and Saville (1949) for the
early Sverdrup-Munk method, by Kaplan and Saville (1954), for the early
SMB method, and by Bretschneider (1965) for a later revision. The basic
data from which the prediction curves were derived, siommarized by
Bretschneider (1951), also indicate the range of variation that may be
expected.
It is generally believed that much of the discrepancy between observed

and predicted waves is random, and that statistical summaries of observa-
tions and predicted values will agree much better than the individual
observations. Saville (1954) and Pierson, Neumann and James (1955) give
summaries of results from a systematic program for deepwater hindcasting
waves at the four locations shown in Figure 3-17. The U.S. Naval Weather
3-40
STATION A^
oo;
:CAPE COD, 42=

66'
STATION B
4I°50'
69° 30'
AREA 4
.NEW YORK/
40"
STATION C| 69"
72"
40° 15'
73°45* "AREA 5
AREA 6
^
.V
-AREA 8
A\
AV
Figure 3-17. Location of Wave Hindcasting Stations and Summary of

Synoptic Meterological Observations (SSMO] Areas
3-4
Service Command (1970) provides summaries of shipboard wave observations.
Summary of Synoptic Meteorological Observations (SSMO) for the hatched
areas indicated in Figure 3-17. Cumulative distribution functions for
wave heights as determined by both hindcasting techniques and the ship-
board observations are given in Figure 3-18. The average of the two
forecasting methods agrees reasonably well with the shipboard observations.
3.54 ESTIMATING WAVE DECAY IN DEEP WATER
Figures 3-19 and 3-20 are used to estimate wave characteristics after
the waves have left the fetch area but are still travelling in deep water.
With Figure 3-19, and given H^, T^, F^ and D (the decay distance), it is
possible to compute the ratios
—z —
decayed wave height
fetch wave height
= — D
Hj,
, and
decayed wave
—-:
fetch
period
—
wave period
= —D
Tp
With Figure 3-20, it is possible to compute wave travel time between

a fetch and a coast, knowing the decayed wave period T^ and the decay
distance D.
This tvavet time t^ is determined by dividing the decay distance

by the deepwater group velocity for waves having a period equal to the
decayed period Ij^. These values enable the estimation of arrival times
for waves at the end of the decay distance.
Waves, after leaving a generating area, will generally follow a great-

circle path toward a coast. However, sufficient accuracy is usually
obtained by assuming wave travel in a straight line on the synoptic chart.
Decay distance is found by measuring the straight line distance between
the front of a fetch and the point for which the forecast is being made.
If a forecast is being made for a coastal area, the effects of shoaling,
refraction, bottom friction and percolation will have to be considered in
translating the deepwater forecast to the shore.
3.6 WAVE FORECASTING FOR SHALLOW WATER
3.61 FORECASTING CURVES
Water depth affects wave generation. For a given set of wind and
fetch conditions, wave heights will be smaller and wave periods shorter if
generation takes place in transitional or shallow water rather than in deep
water. Several forecasting approaches have been made; the method given by
Bretschneider as modified using the results of Ijima and Tang (1966) is
presented here. Bretschneider and Reid (1953) consider bottom friction
and percolation in the permeable sea bottom.
There is no single theoretical development for determining the actual

growth of waves generated by winds blowing over relatively shallow water.
The numerical method presented here is based on successive approximations
3-42
(sja)9y^) igbidH 3aom
OloOr^^DlOTrOCVJ— Oe
PT
aoiasd 3AVM sAiivtau
aOIU3d 3AV* 3MlV\3ti'*l/^l J.H9I3H 3AVM 3Ali V13U '

'h/Î
3-44
4600
4400
4200
4000
3800
3600
3400 .^
E
3200 ^
3000 i2
2800 »
2600 -z
2400 S-
2200 "
2000
1800
1600
1400
1200
1000
800
600
400
200
10 12 14 16 22
Wave period T^ , at end of decay (seconds)
Figure 3-20. Travel Time of Swell Bosed on to = D/Cg
3-45
in which wave energy is added due to wind stress and subtracted due to
bottom friction and percolation. This method uses deepwater forecasting
relationships originally developed by Sverdrup and Munk (1947) and revised
by Bretschneider (1951) to determine the energy added due to wind stress.
Wave energy lost due to bottom friction and percolation is determined from
the relationships developed by Bretschneider and Reid (1953). Resultant
wave heights and periods are obtained by combining the above relationships
by numerical methods. The basic assumptions applicable to development of
deepwater wave-generation relationships (Bretschneider, 1952b) as well as
development of relationships for bottom friction loss (Putnam and Johnson,
1949) and percolation loss (Putnam, 1949) apply.
The choice of an appropriate bottom friction factor f^ for use in

the forecasting technique is a matter of judgement; a value of f^ = 0.01
has been used for the preparation of Figures 3-21 through 3-30 which are
forecasting curves for shallow-water areas of constant depth. These curves,
which may be used like Figures 3-15 and 3-16, are given by the equations:
^ = 0.283 tanh 0.530

1^^
,0.75
tanh (3-25)
U^
and
gd \ 0.375
= 1.20 tanh 0.833 tanh ,(3-26)
2-nU ,U^
which in deep water reduce to Equations 3-21 and 3-22 respectively.
************** EXAMPLE PROBLEM **************

GIVEN Fetch, F = 80,000 feet, wind speed, U = 50 mph , water depth,
:
.
d = 35 feet (average constant depth), bottom friction factor f^ = 0.01

(assumed)
FIND : Wave height H and wave period T.
3-46
-H = 2.5ft. Tr- 4 ô
^ 25 sec
—
^1
1 I
! I^-L M
15 2 25 3 4 5 6 7 8 9 10 1.5 2 25 3 4 5 6 7 8 9 10
X 1,000 Fetch (F) feet X 10,000
Figure 3-21. Forecasting Curves for Shallow-Water Waves

Constant Depth = 5 feet.
T-5 25sec^
H;4.5ft.~.^ T = 5.05ec^ /
1.5 2 2.5 3 4 5 6 7 8 9 10 1.5 2 2.5 3 4 5 6 7 8910

XI, 000 Fetch (F) feet X 10,000

3-47
H = 70ft
1=6 sec.
2 2.5 3 4 5 6 7 8 9 10 1.5 2 2.5 3 4 5 6 7 8 9 10

X 1,000 Fetch (F) feet X 10,000
Figure 3-23. Forecasting Cr.rves for Shallow-Water Wave-;

H = 5 75 ft.
1.5 2 2.5 3 4 5 6 7 8 9 10 1.5 2 2.5 3 4 5 6 7 8 9 10

X 1,000 Fetch (F) feet X 10,000

3-48
/T=4 5sec. T = 5.5 sec.\ T = 6.0 sec.^ ,H = 8.0 ft.
= 4.0 secV 1=5 setv.,^^ T = 6.5 sec.'
1.5 2 2.5 3 4 5 6 7 8 9 10 15 2 2.5 3 4 5 6 7 8 9 10

X 1,000 Fetch (F) feet X 10,000

H-8.0 ft.>
1.5 2 2.5 3 4 5 6 7 8 9 10 1.5 2 2.5 3 4 5 6 7 8 9 10

X 1,000 Fetch (F) feet X 10,000

3-49
,H-90 ft. H = IOft
= 7 sec
15 2 2 5 3 4 5 6 7 8 9 10 15 2 2 5 3 4 5 6 7 8 9 10
X 1,000 Fetch (F) feet X 10,000
l-'igure 3-27. Forecasting Curves for Shallow-Water Waves

H-IOOft-, H-II.Oft.
T = 6-5 sec.^ fT= 7 1
l'i;iure 3-28. Forecasting Curves for Shallow-Water Waves

Constnnt Pcpth = -10 feet.
3-50
T-7.0 secs
T=6 sec. H = ILO ft.N \
't:6 5 sec

figure 3-30. Forecasting Curves for Shallow-Water Waves

3-51
SOLUTION : From Figure 3-27 for constant depth, d = 35 feet,
for
F = 80,000 feet,
and
U = 50 mph.
Then
H = 5.9 feet, say 6 feet,
and
T = 5.1 seconds, say 5 seconds.
***************************** ***k**f*
3.62 DECAY IN LAKES, BAYS, AND ESTUARIES
Section 3.33. Decay of A Wave Field applies to water aroa^^ contijiuous

with land as well as those in the open ocean. Most fetches in inltu'.d
waters will be limited at the front and at the rear by ^ l.ind m;i<s .uul
decay distances will usually be relatively small or nonexistent.
3.7 HURRICANE WAVES
When predicting wave generation by hurricanes, the determination of

fetch and duration from a wind field is more difficult than fur more normal
weather conditions discussed earlier. The largo changes in wind speed and
direction with both location and time cause the difficulty. Estimation of
the free air wind field must be approached through matliematical models,
because of the scarcity of observations in severe storms. However, the
vertical temperature profile and atmospheric turbulence characte;-istics
associated with hurricanes differ less from one storm to another than for
other types of storms. Thus the relation between the free air winds and
the surface winds is less variable for hurricanes than for other storms.
3.71 DESCRIPTION OF HURRICANE WAVES
In hurricanes, fetch areas in which wind speed and direction remain

reasonably constant are always small; a fully arisen sea state never
develops. In the high-wind zones of a storm, however, long-period waves
which can outrun the storm may be developed within fetches of 10 to 20
miles and over durations of 1 to 2 hours. The wave field in front, or to
either side, of the storm center will consist of a locally generated sea,
and a swell from other regions of the storm. Samples of wave spectra,
obtained during hurricane Agnes, 1972, are shown in Figure 3-31. Most
of the spectra display evidence of two or three distinct wave trains; thus,
the physical implications of a signifiaant wave period is not clear.
Other hurricane wave spectra computed with an analog spectrum analyzer

from wave records obtained during Hurricane Donna, 1959. have been published
by Bretschneider (1963) Most of these spectra also contained two distinct
.
peaks
3-52
+ 0.4
An indication of the distribution of waves throughout a hurricane can
be obtained by plotting composite charts of shipboard wave observations.
The position of a report is determined by its distance from the storm
center and its direction from the storm track. Changes in storm intensity
and shape are often small enough to permit all observations obtained during
a period of 24 to 36 hours to be plotted on a single chart. Several plots
of this type from Pore (1957) are given in Figure 3-32. Additional data
of the same type have been presented by Arakawa and Suda (1953) Pore (1957)
,
and Harris (1962)
Goodknight and Russell (1963) give a tabulation of the significant

height and period for waves recorded on an oil drilling platform in approxi-
mately 33 feet of water, 1.5 miles from shore near Burrwood, Louisiana
during hurricanes Audrey, 1957, and Ella, 1950, and tropical storms Bertha,
1957, and Esther, 1957. These wave records were used to evaluate the
applicability of the Rayleigh distribution function (Section 3.22. Wave
Height Variability) to hurricane statistics for wave heights and periods.
They concluded that the Rayleigh distribution function is adequate for
deriving the ratios between Hg, Hio, H, etc., with sufficient accuracy
for engineering design, but that its acceptance as a basic law for wave
height distributions is questionable.
3.72 MODEL WIND AND PRESSURE FIELDS FOR HURRICANES
Many mathematical models have been proposed, for use in studying

hurricanes. Each is designed to simulate some aspect of the storm as
accurately as possible without making excessively large errors in describ-
ing other aspects of the storm. Each model leads to a slightly different
specification of the surface wind field. Available wind data are suffi-
cient to show that some models duplicate certain aspects of the wind field
better than certain other models; but there are not enough data for a
determination of a best model for all purposes.
One of the simplest and earliest models for the hurricane wind field
is the Rankin vortex. For this model, it is assumed that
U = Kr for r < R ,
KR^ (3-27)
U = for r > R ,
r
where K is a constant, R is the radial distance from the storm center

to the region of maximum wind speed, and r is the radial distance from
the storm center to any specified point in the storm system.
This model can be improved by adding a translational component to

account for storm movement and a term producing cross-isobar flow toward
the storm center.
Extensions of this model are still being used in some engineering

studies (Collins and Viehman, 1971). This model gives an artificial
discontinuity in the horizontal gradient of the wind speed at the radius
of maximum winds, and does not reproduce the well-known area of calm
winds near the storm center.
3-54
STORM DIR STORM DIR '\
.V
•
X \'
\10 \'»
,.V'
a CONNIE 1955 >. s
AUG n 0730E THRU
_!^'>- '5/^/1. JL'"^- C CONNIE 1955 AUG 12 0730E
..',
AUG 10 0130E THBU v A

AUG 11 0130E "'^v^:;vi;..^
PERIOD SCALE IN SEC -- ..'>
PERIOD SCAIE IN SEC '•-A.^,;\.

10 20 v'f^.
'h
?'^V Y
+-H— I
(tH
\
—I— t^'-
.^'
ivii:
^'' '»i 1 I
>K° ;^
r
/• , i to n
-^^^
•<:v i3'
X'A.
rtf;
/, \" '•
K'
'A . ...
STORM OIR STORM DIR

•r.
f DIANE 1955
\ ,./
AUG16. 1330E THRU
AUG 17 0730E
e DIANE 1955 .l>"/rt PERIOD SCAIE IN SEC

•^. ' "^
'l AUG 15
..._ .. 1330E THRU
--
V» A 10 20
A. . AUG 16 0730E af A
PERIOD SCAIE IN SEC -S"
10 20
'"
], \
•y fs
-P^^MtT -rh I I I I .,1 I % I

1
4»
I
*30
L
MO ->AiWO
1. /.
IBO laj 60 H \
y a, 10,
/
„ ".'Z /^v\* ^ -vx T.;v v^
"-•
Ml-"
il>4: "i/ +"/ •

'^
:\'^
'/
(offer Pore, 1957)
Figure 3-32. Compositive Wave Charts. The wave height in feet is plotted
beside the arrow indicating direction from which the waves
came. The length of the arrow is proportional to the wave
period. Dashed arrow indicates unknown period. Distances
are marked along the radii at intervals of 60 nautical miles.
3-55
A more widely used model was given by Myers (1954). A concise mathe-
matical description of this model is given by Harris (1958) as follows:
(3-28)
ZZ^ê-"/^
1,2
'0
10 20 30 40 50 60 70 10 20 30 40 SO (O
Distance From Pressure Center (Statute Miles) DIatonce From Wind Center (Statute Miles)
Hurricone August 26-27,1949 ^ ( from Horri,, 1958)
Figure 3-33. Pressure and Wind Distribution in Model Hurricane. Plotted

dots represents observations
where V„ is the velocity of the storm center, and Ugw(r) is the

convective term which is to be added vectorially to the wind velocity at
each value of r. Wilson (1955, 1961) and Bretschneider (1959, 1972) have
suggested other correction terms.
3.73 PREDICTION TECHNIQUE
For a slowly moving hurricane, the following formulas can be used to

obtain an estimate of the deepwater significant wave height and period at
the point of maximum wind:
RAp
0.208 a Vc
Ho = 16.5 e 100 1 + (3-31)
Vu,
and
RAp
0.104 a Vp
T, = 8.6 e
200 +
1 (3-32)
s
vuT
3-57
where
H^ = deepwater significant wave height in feet
Tg = the corresponding significant wave period in seconds
R = radius of maximum wind in nautical miles
Ap = p - p , where p„ is the normal pressure of 29.92 inches

of mercury, and p is the central pressure of the hurricane
V„ = The forward speed of the hurricane in knots
U^ = The maximum sustained wind speed in knots, calculated for

30 feet above the mean sea surface at radius R where
U^ = 0.865 U^^^ (For stationary hurricane) (3-33)
U^ = 0.865 U^^ + 0.5 V^ (For moving hurricane) (3-34)
^rmx ~ Maximum gradient wind speed in knots 30 feet above the

water surface
^max = 0.868 [73 (p„ - p^) 1/2 . R(0.575f)] (3-35)
f = Coriolis parameter = 2to sin(j), where w = angular velocity of

earth = 2ti/24 radians per hour
Latitude (()>) 25° 30° 35° 40°

f (rad/hr) 0.221 0.262 0.300 0.337
a = a coefficient depending on the forward speed of the

hurricane and the increase in effective fetch length,
because the hurricane is moving. It is suggested that
for slowly moving hurricane a = 1.0.
Once Hq is determined for the point of maximum wind from Equation

3-31 it is possible to obtain the approximate deepwater significant wave
height H^ for other areas of the hurricane by use of Figure 3-34.
The corresponding approximate wave period may be obtained from
T =2.13 y/H^ (in seconds) ,

(3-36)
where H^ is in feet (derived from empirical data showing that the wave
steepness H/T^ will be about 0.22).
3-58
-M to
Figure 3-34. Isolines of Relative Significant Wave

Height for Slow Moving Hurricane
3-59
************** EXAMPLE PROBLEM **************
GIVEN Consider a hurricane at latitude 350N. with R = 36 nautical miles,
:
Ap = 29.92 - 27.61 = 2.31 inches of mercury, and V^, forward speed, =

26 knots. Assume for simplicity that a = 1.0.
FIND : The deepwater significant wave height and period.
SOLUTION :
Using Equation 3-35
U 0.868 [73/p -p - R(0.575f)l

Y'
U max 0.868 [73(2.31)^^ - 36(0.575X0.300)1
^max =" 0-^^8 (110.95 - 6.23) = 90.9 knots.
Using Equation 3-34
Ur = 0.865 U^,^ + 0.5 V^
Uj^ = 0.865 (90.9) + 0.5 (26) = 91.6 knots.
Using Equation 3-31

RAp 0.208 aVc
H^ = 16.5 e 100 1 +
VuR
where the exponent
RAp _ 36(2.31)
= 0.832,
100 100
and
eO.832 = 2.30
then
0.208 X 1 X 26
H^ = 16.5 eO-832 1 +
V9I.6
Hq = 16.5 (2.30) (1.564) = 59.4 feet.
3-60
Using Equation 3-32
RAp 0.104 a Vjp

200
T = 8.6 e 1 +
Vu7
where the exponent
RAP 36(2.31)
= 0.416
200 200
0.104 X 1 X 26
0.416
Ts = 8.6 e 1 +
V91.6'
Tj = 8.6 (1.52) (1.282) 16.8 seconds.
Alternately, by Equation 3-36, it is seen that
T = 2.I3V59.4 = 16.4 seconds.
It should be noted that computing the values of wave height and period
to three significant figures does not imply the degree of accuracy of the
method; it is done to reduce the computational error.
Referring to Figure 3-34, H^ = 59.4 feet corresponds to the relative

significant wave height of 1.0 at r/R = 1.0, the point of maximum winds
located, for this example, 36 nautical miles to the right of the hurricane
center. At that point the wave height is about 60 feet, and the wave
period T is about 16 seconds. At r/R = 1.0 to the left of the hurricane
center, from Figure 3-34 the ratio of relative significant height is about
0.62, whence H^ = 0.62 (59.4) = 36.8 feet. This wave is moving in a direc-
tion opposite to that of the 59.4-foot wave The significant wave period
.
for the 36.8-foot wave is: Tg = 2.13 /36.8' = 12.9 seconds, say 13 seconds.
The most probable maximum wave is assumed to depend on the number of

waves considered applicable to the significant wave, H^ = 59.4 feet. This
number N depends on the length of the section of the hurricane for which
near steady state exists and the forward speed of the hurricane. It has
been found that maximum wave conditions occur over a distance equal to the
radius of maximum wind. The time it takes the radius of maximum wind to
pass a particular point is
36
t = = 1.38 hours = 4,970 seconds; (3-37)
26
the number of waves will be

4970
N = —t
= 303. (3-38)
T, 16.4
3-61
The most probable maximum waves can be obtained by using
Hn = 0-707 H^ - (3-39)
^ log^ .
The most probable maximum wave is obtained by setting n = 1, and

using Equation 3-39
j 30?
Hj = 0.707(59.4) 100.4 feet, say 100 feet.
Assuming that the 100-foot wave occurred, then the most probable
second highest wave is obtained by setting n = 2, the third from n = 3, etc.
30F
—
Hj = 0.707(59.4) •Jlogg
I
— = 94.1 feet, say 94 feet;
303
= 0.707(59.4)
H3
-3 v-,..^ ,
i
./log
\l^î>e -^
= 90.2 feet, say 90 feet.
The problem now is to determine the changes in the deepwater waves as

they cross the Continental Shelf, taking into account the combined effects
of bottom friction, refraction, the continued action of the wind, and the
forward speed of the hurricane. This requires numerical integration, using
Table 3-3, Figure 3-35, and refraction diagrams. It is also necessary to
obtain an effective fetch length, by use of
H,
Fe = (3-40)
0.0555 U,
where
F is the effective fetch in nautical miles,
H^ is the deepwater significant wave height in feet,
and
Uj^ is the maximum sustained wind speed in knots.
For this example, using Equation 3-40
59.4
Fe = = (11.69) = 137 nautical miles.
0.0555(91.6)
3-62
10
9 ^r"^] 1
1-
8
7
X
<
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
For the remaining part of this problem, either the value of F^„ equal
to 220 nautical miles as determined from Figure 3-15 for U^ = 91.6,
Hq = 59.4 can be used with the deepwater forecasting curves, or else
Equation 3-40 can be used, as modified,
H^ = 0.0555 U^ >JfJ + AF"
along with Equation 3-36
To = 2.13 V?^ .
Fg is defined below.
The latter being a numerical method is easier to use and more accurate
than the graphical method of using forecasting curves
The procedure for computing wind waves over the Continental Shelf
will be illustrated by using the bottom profile off the mouth of the
Chesapeake Bay and the standard project hurricane developed for the
Norfolk area. The storm surge computed for the standard project hurricane
and 2.5 feet of astronomical tide are added to the mean low water depths
to obtain the total water depth for wave generation. Refraction is
neglected in this example, i.e., K^ = 1.0. The results of these computa-
tions are given in Table 3-4 followed by examples and explanations.
Column 1 of Table 3-4 is the distance in nautical miles measured

seaward of the entrance to Chesapeake Bay, using increments of 5 nautical
miles for each section.
Column 2, d^, is the depth in feet referred to mean low water at the
shoreward end of each section, denoted by X of Column 1.
Colvimn 3 is the depth dj at the beginning of each section.
Column 4 is the depth 6.2 at the shoreward end of each section.

These depths are the water depths below MLW plus the 2.5-foot astro-
nomical tide plus the hurricane surge and are then rounded off to the
nearest foot.
Column 5, d„, is the average of Columns 3 and 4 to the nearest foot.
Column 6 is the effective fetch Fg (nautical miles), and is

obtained for the first step directly from Equation 3-40. For successive
steps, Fg = Fg + AF < 137 n.mi. where Fg is given in Column 14 one line
above in each case (e.g., line X = 40, Fg = 80.6 + 5.0 = 85.6) and AF is
5 n.mi. Fg is defined for Column 14.
3-65
o

o
c/)
>
OS
E
o
U
^'
Column 7 is the deepwater significant wave height H^ and is
obtained from Equation 3-40;
H^ = 0.0555 U^ y/F^ = 5.08 y/f^ , where U^ = 91.6 knots
and F is obtained from Column 6.
Column 8 is the deepwater significant wave period T^ and is

obtained from Equation 3-36:
T = 2.13 VH where H^ is obtained from Column 7.
Column 9 is T^/dy, or Column 8 squared over Column 5.
Column 10 is the shoaling coefficient (H/H' ) or K corresponding

to the value of T^/dy, Column 9, and is obtained from Table 3-3 Kg
versus T^/d.
Column 11 is the friction loss parameter and is equal to
frH K AX 0.01 H K, (5) (6,080) 304 H K
M' M' M'

where f^ is assumed as 0.01, AX = 5(6,080) = 30,400 ft., dj, is the
average water depth of the increment AX.
Column 12 is the friction factor K^ and is obtained from Figure

3-35 where Ky- is a function of T^/dy (Column 9) and
(fH K AX
^ ,- ,
= A (Column 11).
Column 13 is the equivalent deepwater wave height H^ and is

obtained from H' = H K (the product of Columns 7 and 12)
Column 14 is the equivalent effective fetch length for H^ and is

obtained from Equation 3-40
H^ H^
F ' = where U^j = 91.6 knots (moving hurricane).
(0.0555 U^ 5.08
Column 15 is obtained by using Equation 3-36 T^ = 2.13 vH^.
Column 16 is (T')^/d„, where d„ is the water depth at shoreward

end of section AX.
3-67
Column 17 is the shoaling coefficient Kg2 related to the values of
(To) /da (Column 16).
Column 18 is H = H^ x Kg2 (product of Columns 13 and 17)
Column 19 is obtained by using Equation 3-38
N = —T/ = '
Tj
,
where \ = R/Vp
P
=
26
= — 1.38 hours or 4,970 seconds.
Column 20 is H = 0.707 H /log N', H is from Column 18.

max ê
After one line of computations across is completed, the next line is

begun, using F = F' + AF <. 137 where F' is from Column 14 of the
preceding completed line. For example, consider the line corresponding to
X = 40 being completed. Then the computation for the next line X = 35
is as follows:
Fg = 66.4 from line X = 40, Column 14.
Column 6 Fg = 66.4 + 5 = 71.4 nautical miles for line X = 35 .
Compute
Column 7, H^ = 5.08 \ll\A'= 42.9 feet.
Column 8, T^ = 2.13 \/42.9~ = 14.0 seconds .
To (14)^
Column 9, =^ = = 1.73.
dj, 113
Column 10, K = 0.924 (Table 3-3) for values of J- = 1.73.
304 HK
Column 11, A = — _ = —rr7zzz~
(304 (42.9) (0.924)
(113)^
^^^,
= 0.943.
(d.p)^
Column 12, lO = 0.89 (from Figure 3-35) for values of =^ == 1.73 and
A = 0.943.
Column 13, H^ = H^ Ky = 42.9 (0.89) = 38.2 feet .
Column 14, %- 5.08

Column 15, t' = 2.13 >/38.2' = 13.2 seconds.
Column 16, —
d-
= 1.58
2
Column 17, K^^ = 0.919.
Column 18, H = 0.919 (38.2) = 35.1 feet, which is the shallow-water height
for depth d^ = 110 feet, corresponding to MLW of 104 feet.
4,970
Column 19, N = ^— = 377 ,
or the total
, ,
number r 1-
or waves applicable to
, ,
o
steady-state significant wave of H = 35.1 feet, say 35 feet
Column 20, H_^ = 35.1(0.707) Vlog 377' = 60.4 feet, say 60 feet
The moving fetch model of Wilson (1955) has been adapted for computer
usage by Wilson (1961) The basic equations were modified by Wilson (1966)
.
The Bretschneider (1959) model for hurricane wave prediction was modified
by Bretschneider (1972) Borgman (1972) used the results of Wilson (1957)
.
to develop an approach for estimating the maximum wave in a storm which

may be considered as an alternate to that presented here.
3.8 WATER LEVEL FLUCTUATIONS
The focus now changes from wave prediction to water level fluctuations
in oceans and other bodies of water which have periods substantially longer
than those associated with surface waves. Several known physical processes
combine to cause these longer-term variations of the water level.
The expression water level is used to indicate the mean elevation of

the water when averaged over a period of time long enough (about 1 minute)
to eliminate high frequency oscillations caused by surface gravity waves.
In the discussion of gravity waves the water level was also referred to as
the Stillwater level (SWL) to indicate the elevation of the water if all
gravity waves were at rest. In the field, water levels are determined by
measuring water surface elevations in a stilling well. Inflow and outflow
of the well is restricted so that the rapid responses produced by gravity
waves are filtered out, thus reflecting only the mean water elevation.
Water level fluctuations -- classified by the characteristics and
type of motion which take place —
may be identified as:
(a) astronomical tides
(b) tsunamis
(c) seiches
(d) wave setup
(e) storm surges
(f) climatological variations
(g) secular variations
3-69
The first five have periods that range from a few minutes to a few
days; the last two have periods that range from semi-annually to many
years. Although important in long-term changes in water elevations,
climatological and secular variations are not discussed here.
Forces caused by the graviational attraction between the moon, the

sun, and the rotating earth result in periodic level changes in large
bodies of water. The vertical rise and fall resulting from these forces
is called the tide or astronomical tide; the horizontal movements of
water are called tidal currents. The responses of water level changes to
the tidal forces are modified in coastal regions because of variations in
depths and lateral boundaries; tides vary substantially from place to
place. Astronomical tide generating forces are well understood, and can
be predicted many years in advance. The response to these forces can be
determined from an analysis of tide gage records. Tide predictions are
routinely made for many locations for which analyzed tide observations
are available. In the United States, tide predictions are made by the
National Ocean Survey, National Oceanographic and Atmospheric Administratioi
Tsunamis are generated by several mechanisms: submarine earthquakes,

submarine landslides, and underwater volcanos. These waves may travel
distances of more than 5,000 miles across an ocean with speeds at times
exceeding 500 miles per hour. In open oceans, the heights of these waves
are generally unknown but small; heights in coastal regions have been
greater than 100 feet
Seiches are long-period standing waves that continue after the

forces that start them have ceased to act. They occur commonly in
enclosed or partially enclosed basins.
Wave setup is defined as the superelevation of the water surface due

to the onshore mass transport of the water by wave action alone. Isolated
observations have shown that wave setup does occur in the surf zone.
Surges are caused by moving atmospheric pressure jumps and by the

wind stress accompanying moving storm systems. Storm systems are signifi-
cant because of their frequency and potential for causing abnormal water
levels at coastlines. In many coastal regions, maximum storm surges are
produced by severe tropical cyclones called hurricanes.
Prediction of water level changes is complex because many types of

water level fluctuations can occur simultaneously. It is not unusual
for surface wave setup, high astronomical tides, and storm surge to occur
coincidently at the shore on the open coast. It is difficult to determine
how much rise can be attributed to each of these causes. Although astro-
nomical tides can be predicted rather well where levels have been recorded
for a year or more, there are many locations where this information is not
available. Furthermore, the interaction between tides and storm surge in
shallow water is not well defined.
3-70
3.81 ASTRONOMICAL TIDES
Tide is a periodic rising and falling of sea level caused by the

gravitational attraction of the moon, sun, and other astronomical bodies
acting on the rotating earth. Tides follow the moon more closely than they
do the sun. There are usually two high and and two low waters in a tidal
or lunar day. As the lunar day is about 50 minutes longer than the solar
day, tides occur about 50 minutes later each day. Typical tide curves for
various locations along the Atlantic, Gulf, and Pacific coasts of the
United States are shown in Figures 3-36 and 3-37. Along the Atlantic
coast, the two tides each day are of nearly the same height. On the Gulf
coast, the tides are low but in some instances have a pronounced diurnal
inequality. Pacific coast tides compare in height with those on the
Atlantic coast but in most cases have a decided diurnal inequality.
(See Appendix A, Figure A-10.)
The dynamic theory of tides was formulated by Laplace (1775) and

special solutions have been obtained by Doodson and Warburg (1941) among
others. The use of simplified theories for the analysis and prediction
of tides has been described by Schureman (1941), Defant (1961) and Ippen
(1966). The computer program for tide prediction, currently being used
for official tide prediction in the United States is described by Pore
and Cummings (1967).
Data concerning tidal ranges along the seacoasts of the United States
are given to the nearest foot in Table 3-5. Spring ranges are shown for
areas having approximately equal daily tides; diurnal ranges are shown
for areas having either a diurnal tide or a pronounced diurnal inequality.
Detailed data concerning tidal ranges are published annually in Tide Tables,
U.S. Department of Commerce, National Ocean Survey.
3.82 TSUNAMIS
Long period gravity waves generated by such disturbances as earth-

quakes, landslides, volcano eruptions and explosions near the sea surface
are called tsunamis. The Japanese word tsunami has been adopted to replace
the expression tidal wave to avoid confusion with the astronomical tides.
Most tsunamis are caused by earthquakes that extend at least partly

under the sea, although not all submarine earthquakes produce tsunamis.
Severe tsunamis are rare events.
Tsunamis may be compared to the wave generated by dropping a rock in

a pond. Waves (ripples) move outward from the source region in every
direction. In general, the tsunami wave amplitudes decrease but the
number of individual waves increases with distance from the source region.
Tsunami waves may be reflected, refracted, or diffracted by islands, sea-
mounts, submarine ridges or shores. The longest waves travel across the
deepest part of the sea as shallow-water waves, and may obtain speeds of
several hundred knots. The travel time required for the first tsunami
disturbance to arrive at any location can be determined within a few
percent of the actual travel time by the use of suitable tsunami travel-
time charts.
3-71
DAr 10 II 12 13 14 15 16 17 18 19
LuMf dita: mu. S. declirntion, 9th; ipogee, lOlb: lul quirter, 13th: on equitor, 16th; new moon. 20Ui; peri|M,
22d: mu. N. dcdinition, 23d.
(from Notionol Ocean Survey, NOAA, Tide Tables)
Figure 3-36. Typical Tide Curves Along Atlantic and

Gulf Coasts
3-72
DAY 10 11 12 13 14 15 16 17 18 19 20
Ft
SAN FRANCISCO
^^^^^^^^^^^
6
4
2
-2
SEATTLE
12
10
8
6
4
2
-2
18
16
14
12
10
8
6
4
2
-2
32
30
28
26
24
22
20
18
16
14
12
10
-2
lunat (Jala: max S. declination, 9th, apogee, lOlh, last quartet, 13tli; on equalor, 16lh, new moon. 20th, perigee,
22d, max. N. declination, 23d.
(from Notional Ocean Survey, NOAA, Tide Tobies)
Figure 3-37. Typical Tide Curves Along Pacific Coasts of the

United States
3-73
Table 3-5. Tidal Ranges
Station
Tide Gage Record Showing Tsunami
HONOLULU. HAWAII
May 23-24, 1960
Approx. Hours G.M.T.

12 13 14 15 16 17 18 19 20 21
_l \ I I I I I I I l_

MOKUOLOE ISLAND. HAWAII
May 23-24. 1960

JOHNSTON ISLAND, HAWAII
May 23-24. 1960 A-^X.
(from Symons and Zelter, I960)
Figure 3-38. Sample Tsunami Records from Tide Gages
3-75
Theoretical and applied research dealing with tsunami problems has
been greatly intensified since 1960. Preisendorfer (1971) lists more than
60 significant theoretical papers published since 1960. The list does not
include observational papers concerned with the warning system.
3.83 LAKE LEVELS
Lakes have insignificant tidal variations, but are subject to sea-

sonal and annual hydrologic changes in water level and to water level
changes caused by wind setup, barometric pressure variations, and seiches.
Additionally some lakes are subject to occasional water level changes by
regulatory control works.
Water surface elevations of the Great Lakes vary irregularly from

year to year. During each year, the water surfaces consistently fall to
their lowest stages during the winter and rise to their highest stages
during the summer. Nearly all precipitation in the watershed areas during
the winter is snow or rainfall transformed to ice. When the temperature
begins to rise there is substantial runoff - thus the higher stages in the
summer. Typical seasonal and yearly changes in water levels for Lake Erie
are shown in Figure 3-39. The maximum and minimum monthly mean stages for
the lakes are summarized in Table 3-6.
Table 3-6. Fluctuations in Water Levels - Great Lakes System (1860 through 1973).
Lake
1 , 1 Nnr
the Great Lakes, is subject to greater wind-induced surface fluctuations,
that is, wind setup, than any other Lake. Wind setup is discussed in
Section 3.86, Storm Surge and Wind Setup.
In general, the maximum amount of these irregular changes in lake

level must be determined for each location under consideration. Table 3-7
shows short-period observed maximum and minimum water level elevations at
selected gage sites. More detailed data on seasonal lake levels and wind
setup may be obtained for specific locations from the Lake Survey Center,
National Oceanic and Atmospheric Administration, U.S. Department of
Commerce.
Table 3-7. Short-Period Fluctuations in Lake Levels at Selected Gage Sites

Lake and Gage Location
Free oscillations have periods that are dependent upon the horizontal
and vertical dimensions of the basin, the number of nodes of the standing
wave, that is, lines where deviation of the free surface from its undis-
turbed value is zero, and friction, ihe period of a true forced-wave
oscillation is the same as the period of the causative force. Forced
oscillations, however, are usually generated by intermittent external
forces, and the period of the oscillation is determined partly by the
period of the external force and partly by the dimensions of the water
basin and the mode of oscillation. Oscillations of this type have been
called forced seiches (Chrystal, 1905) to distinguish them from free
seiches in which the oscillations are free.
The fundamental and maximum period (T^ for n = 1) becomes
T =—If (3-43)
Equation 3-43 is called Merian's formula. (Sverdrup, Johnson and Fleming,

1942.)
In an open rectangular basin of length %' and constant depth d,

the simplest form of a one-dimensional, non-resonant, standing longitudi-
nal wave is one with a node at the opening, antinode at the opposite end,
and n' nodes in between. (See Figure 3-40(C).) The free oscillation
period T^/ in this case is
4V
For the fundamental mode (n' = 0), T^ becomes
T' =^1mL (3-45)
The basin's total length is occupied by one-fourth of a wave length.
This simplified theory must be modified for most actual basins,

because of the variation in width and depth along the basin axes.
Defant (1961) outlines a method to determine the possible periods for

one-dimensional free oscillations in long narrow lakes of variable width
and depth. Defant 's method is useful in engineering work because it permits
computation of periods of oscillation, relative magnitudes of the vertical
displacements along chosen axes, and the positions of nodal and antinodal
lines. This method, applicable only to free oscillations, can be used to
determine the nodes of oscillation of multinodal and uninodal seiches.
The theory for a particular forced oscillation was also derived by Defant
and is discussed by Sverdrup et al (1942). Hunt (1959) discusses some
complexities involved in the hydraulic problems of Lake Erie, and offers
an interim solution to the problem of vertical displacement of water at
the eastern end of the lake. More recently, work has been done by Simpson
and Anderson (1964), Platzman and Rao (1963), and Mortimer (1965).
Rockwell (1966) computed the first five modes of oscillation for each of
the Great Lakes by a procedure based on the work of Platzman and Rao (1965).
Platzman (1972) has developed a method for evaluating natural periods for
basins of general two-dimensional configuration.
3.85 WAVE SETUP
Field observations indicate that part of the variation in mean water

level near shore is a function of the incoming wave field. However these
3-80
observations are insufficient to provide quantitative trends. (Savage,
1957; Fairchild, 1958; Dorrestein, 1962; Galvin and Eagleson, 1965.) A
laboratory study by Saville (1961) indicated that for waves breaking on
a slope there would be a decrease in the mean water level relative to the
Stillwater level just prior to breaking, with a maximum depression or set-
down at about the breaking point. This study also indicated that from the
breaking point the mean water surface slopes upward to the point of inter-
section with the shore and has been termed wave setup. Wcwe setup is
defined as that super -elevation of the mean water level caused by wave
action alone. This phenomenon is related to a conversion of kinetic
energy of wave motion to a quasi-steady potential energy.
Theoretical studies of wave setup have been made by Dorrestein (1962),

Fortak (1962), Longuet-Higgins and Stewart (1960, 1962, 1963, 1964), Bowen,
Inman, and Simmons (1968), and Hwang and Divoky (1970). Theoretical devel-
opments can account for many of the principal processes, but contain
factors that are often difficult to specify in practical problems.
R.O. Reid (personal communication) has suggested the following approach

for estimating the wave setup at shore, using Longuet-Higgins and Stewart
(1963) theory for the setdown at the breaking zone and solitary wave theory.
The theory for setdown at the breaking zone indicates that
gi/2
Hj T
in which Sj^ is the setdown at the breaking zone, T is the wave period,
Hq is the deepwater significant wave height, dj^ is the depth of water
at the breaker point and g is gravity. The laboratory data of Saville
(1961) gives somewhat larger values than those obtained by use of Equation
3-46.
By using relations derived from solitary wave theory relating dh

to the breaker height of the significant wave, H^j, and d^/H^ to Ho/Lo,
the above relation can be converted to
3/2
0.536 H,
Su
'b
= —-^—
1/2
.
(3-47)
g J
Longuet-Higgins and Stewart (1963) show from an analysis of Saville's

data that the wave setup AS between the breaker zone and shore is given
approximately by AS = 0.15 dj. Assuming that djy = 1.28 Hj,, this becomes
AS = 0.19 Hfe.
The net wave setup at the shore is
Sj^, = AS + Sj, , (3-48)

or
Sh; = 0.19 1 - 2.82 Hy . (3-49)

gT^
3-81
Equation 3-49 provides a conservative estimate for wave setup at the
shore. The difference between laboratory data and theory, however, is not
likely to exceed the uncertainties of field data.
************** EXAMPLE PROBLEM **************
GIVEN : Hh = 20 feet, T = 12 seconds.
FIND : Wave setup, Sw

SOLUTION : Using Equation 3-49,
S^ = 0.19
gTV J
Sj^ = 0.19
landmass such as southern Florida and moves offshore. High offshore winds
in this case can cause the water level to drop several feet.
Setdown in semienclosed basins (bays and estuaries) also may be sub-

stantial, but the fall in water level is influenced by the coupling to the
sea. There are some detrimental effects as a result of setdown, such as
making water-pumping facilities inoperable due to exposure of the intake,
increasing the pumping heads of such facilities, and causing navigational
hazards because of decreased depths.
However, rises in water levels (setup rather than setdown) are of

most concern. Abnormal rises in water level in nearshore regions will not
only flood low-lying terrain, but provide a base on which high waves can
attack the upper part of a beach and penetrate farther inland. Flooding
of this type combined with the action of surface waves can cause severe
damage to low- lying land and backshore improvements.
Wind-induced surge, accompanied by wave action, accounts for most of

the damage to coastal engineering works and beach areas. Displacement of
stone armor units of jetties, groins and breakwaters, scouring around
structures, accretion and erosion of beach materials, cutting of new in-
lets through barrier beaches, and shoaling of navigational channels can
often be attributed to storm surge and surface waves. Moreover, surge can
increase hazards to navigation, impede vessel traffic, and hamper harbor
operations. A knowledge of the increase and decrease in water levels
that can be expected during the life of a coastal structure or project is
necessary to design structures that will remain functional.
3.862 Storms . A storm is an atmospheric disturbance characterized by

high winds which may or may not be accompanied by precipitation. Two
distinctions are made in classifying storms: a storm originating in the
tropics is called a tropical storm; a storm resulting from a cold and
warm front is called an extratropioat storm. Both of these storms can
produce abnormal rises in water level in shallow water near the edge of
water bodies. The highest water levels produced along the entire gulf
coast and from Cape Cod to the south tip of Florida on the east coast
generally result from tropical storms. High water levels are rarely
caused by tropical storms on the lower coast of California. Extreme
water levels in some enclosed bodies, such as Lake Okeechobee, Florida
can also be caused by a tropical storm. Highest water levels at other
coastal locations and most enclosed bodies of water result from extra-
tropical storms.
A severe tropical storm is called a hurricane when the maximum

sustained wind speeds reach 75 miles per hour (65 knots). Hurricane
winds may reach sustained speeds of more than 150 miles per hour (130
knots). Hiirricanes, unlike less severe tropical storms, generally are
well organized and have a circular wind pattern with winds revolving
around a center or eye (not necessarily the geometric center). The eye
is an area of low atmospheric pressure and light winds. Atmospheric
pressure and wind speed increase rapidly with distance outward from the
eye to a zone of maximum wind speed which may be anywhere from 4 to 60
3-83
nautical miles from the center. From the zone of maximum wind to the
periphery of the hurricane, the pressure continues to increase; however,
the wind speed decreases. The atmospheric pressure within the eye is the
best single index for estimating the surge potential of a hurricane. This
pressure is referred to as the centvdl pressure index (CPI) .Generally
for hurricanes of fixed size, the lower the CPI, the higher the wind speeds,
Hurricanes may also be characterized by other important elements, such as
the radius of maximum winds (R) which is an index of the size of the storm,
and the speed of forward motion of the storm system (Vj.) . A discussion of
the formation, development and general characteristics of hurricanes is
given by Dunn and Miller (1964)
Extratropical storms that occur along the northern part of the east
coast of the United States accompanied by strong winds blowing from the
northeast quadrant are called northeasters. Nearly all destructive north-
easters have occurred in the period from November to April; the hurricane
season is from about June to November. A typical northeaster consists of
a single center of low pressure and the winds revolve about this center,
but wind patterns are less symmetrical than those associated with hurri-
canes.
3.863 Factors of Storm Surge Generation The extent to which water

.
levels will depart from normal during a storm depends on several factors.
The factors are related to the:
(a) characteristics and behavior of the storm;

(b) hydrography of the basin;
(c) initial state of the system; and
(d) other effects that can be considered external to the system.
Several distinct factors that may be responsible for changing water levels
during the passage of a storm may be identified as:
(a) astronomical tides
(b) direct winds
(c) atmospheric pressure differences
(d) earth's rotation
(e) rainfall
(f) surface waves and associated wave setup
(g) storm motion effects
The elevation of setup or setdown in a basin depends on storm inten-
sity, path or track, overwater duration, atmospheric pressure variation,
speed of translation, storm size, and associated rainfall. Basin charac-
teristics that influence water-level changes are basin size and shape,
and bottom configuration and roughness. The size of the storm relative
to the size of the basin is also important. The magnitude of storm
surges is shown in Figures 3-41 and 3-42. Figure 3-41 shows the differ-
ence between observed water levels and predicted astronomical tide levels
during Hurricane Carla (1961) at several Texas and Louisiana coast tide
stations. Figure 3-42 shows high water marks obtained from a storm survey
made after Hurricane Carla. Harris (1963b) gives similar data from other
hurricanes
3-84
3.864 Initial Water Level . Water surfaces on the open coast or in en-
closed or semienclosed basins are not always at their normal level prior
to the arrival of a storm. This departure of the water surface from its
normal position in the absence of astronomical tides, referred to as an
initial water level, is a contributing factor to the water level reached
during the passage of a storm system. This level may be 2 feet above
normal for some locations along the U.S. Gulf coast. Some writers refer
to this difference in water level as a forerunner in advance of the storm
due to initial circulation and water transport by waves particularly when
the water level is above normal. Harris (1963b) on the other hand, indi-
cates that this general rise may be due to short-period anomalies in the
mean sea level not related to hurricanes. Whatever the cause, the initial
water level should be considered when evaluating the components of open-
coast storm surge. The existence of an initial water level preceeding the
approach of Hurricane Carla is shown in Figure 3-41 and in a study of the
synoptic weather charts for this storm. (Harris, 1963b.) At 0700 hours
(Eastern Standard Time) 9 September 1961, the winds at Galveston, Texas
were about 10 mph, but the open coast tide station (Pleasure Pier) shows
the difference between the observed water level and astronomical tide to
be above 2 feet. Rises of this nature on the open coast can also affect
levels in bays and estuaries.
There are other causes for departures of the water levels from normal
in semienclosed and enclosed basins, such as the effects of evaporation
and rainfall. Generally, rainfall plays a more dominant role since these
basins are affected by direct rainfall and can be greatly affected by
rainfall runoff from rivers. The initial rise caused by rainfall is due
to rains preceding the storm; rains during the passage of a storm have a
time-dependent effect on the change in water level.
3.865 Storm Surge Prediction . The design of coastal engineering works is

usually based on a life expectancy for the project and on the degree of
protection the project is expected to provide. This design requires that
the design storm have a specified frequency of occurrence for the partic-
ular area. An estimate of the frequency of occurrence of a particular
storm surge is required. One method of making this estimate is to use
frequency curves developed from statistical analyses of historical water
level data. Table 3-8, based on National Ocean Survey tide gage records,
indicates observed extreme storm surge water levels including wave setup
The water levels are those actually recorded at the various tide stations,
and do not necessarily reflect the extreme water levels that may have
occurred near the gages. Values in this table may differ from gage-station
values because of corrections for seasonal and secular anaomalies. The
frequency of occurrence for the highest and lowest water levels may be
estimated by noting the length of time over which observations were made.
The average yearly highest water level is an average of the highest water
level from each year during the period of observation. Extreme water
levels are rarely recorded by water level gages, partly because the gages
tend to become inoperative with extremely high waves, and partly because
the peak storm surge often occurs between tide gage stations. Post-storm
surveys showed water levels, as the result of Hurricane Camille, August
3-85
90*
Fort Point
Pier 21
Pleoture Pier
«S»J
-40° 40"
(from Harris, 1963 b)
Figure 3-41. Storm Surge and Observed Tide Chart. Hurricane C:irla,
7-12 September 1961. Insert Maps for Freeport and
Galveston, Texas, Areas
3-86
Time Eye of Hurricane
Neorest to Stotion
September, 1961
j
Table 3-8. Highest and Lowest Water Levels — Continued
Table
1969, in excess of 20 feet MSL over many miles of the open Gulf Coast,
with a peak value of 24 feet MSL near Pass Christian, Mississippi. High
water levels in excess of 12 feet MSL on the open coast and 20 feet within
bays were recorded along the Texas coast as the result of Hurricane Carla,
September, 1961. Water levels above 13 feet MSL were recorded in the
Florida Keys during Hurricane Donna, 1960.
Accumulation of data over many years in some areas, such as regions

near the North Sea, has led to relatively accurate empirical techniques
of storm surge prediction for some locations. However, these empirical
methods are not applicable to other locations. In general, not enough
storm surge observations are available in the United States to make accu-
rate predictions of storm surge. Therefore, it has been general practice
to use hypothetical design storms, and to estimate the storm-induced surge
by physical or mathematical models. Mathematical models are usually used
for predicting storm surge, since it is difficult to represent some of the
storm surge generating processes (such as the direct wind effects and
Coriolis effects) in physical laboratory models.
a. Hydrodynamic Equations . Equations that describe the storm surge

generation processes are the continuity equation expressing conservation
of mass and the equations of motion expressing Newton's second law. The
derivations are not presented here; references are cited below. The equa-
tions of motion and continuity given here represent a simplification of
the more complete equations. A more simplified form is obtained by verti-
cally integrating all governing equations and then expressing everything
in terms of either the mean horizontal current velocity or volume trans-
port. Vertically integrated equations are generally preferred in storm-
surge calculations since interest is centered in the free surface motion
and mean horizontal flow. Integration of the equations for the storm
surge problem are given by Haurwitz (1951), Welander (1961), Fortak (1962),
Platzman (1963), Reid (1964), and Harris (1967).
The equations given here are obtained by assuming:
(1) vertical accelerations are negligible,

(2) curvature of the earth and effects of surface
waves can be ignored,
(3) the fluid is inviscid, and
(4) the bottom is fixed and impermeable.
The notation and the coordinate scheme employed are shown schematic-
ally in Figure 3-43. D is the total water depth at time t, and is
given by D = d + S, where d is the undisturbed water depth and S is
the height of the free surface above or below the undisturbed depth re-
sulting from the surge. The Cartesian coordinate axes, x and y, are in
the horizontal plane at the undisturbed water level and the z axis is
directed positively upward. The x axis is taken normal to the shoreline
(positive in the shoreward direction), and the y axis is taken alongshore
(positive to the left when facing the shoreline from the sea).
3-92
W77777777777777777777777777Z777777777;77777777Z777Z
PROFILE
LAND
^ JZ.
JiL -^ SL J^
SHORELINE-
JU>.A^ .AJU^ >t/OC/
SEA
PLAN VIEW
Figure 3-43. Notation and Reference Frame
3-93
The differential equations appropriate for tropical or extratropical
storm surge problems on the open coast and in enclosed and semienclosed
basins are as follows;
»M^, »M^
H; ^ ^
^
as
^
^
a, ,„ il ,
^
II' _ :^ , W,P (3-50)
9t ax 3y 3x ax 3x p p ^
'
o s
c a
C «
" c
Hi = angular velocity of earth
(7.29 X 10"^ radians/second);
<f)
= geographical latitude;
Tg^, Tgj, = X and y components of surface wind stress;
"^bx' "^bii
~ ^ ^"^ y components of bottom stress;
p = mass density of water;

W^, Wj^ = X and y components of wind speed;
5 = atmospheric pressure deficit in head of water;
I, = astronomical tide potential in head of water;
u, V = X and y components, respectively, of current

velocity;
P = precipitation rate (depth/time)
g = gravitational acceleration; and
9 = angle of wind measured counterclockwise from

the X axis.
Equations 3-50 and 3-51 are approximate expressions for the equations
of motion and Equation 3-52 is the continuity relation for a fluid of
constant density. These basic equations provide, for all practical pur-
poses, a complete description of the water motions associated with nearly
horizontal flows such as the storm surge problem. Since these equations
satisfactorily describe the phenomenon involved, a nearly exact solution
can only be obtained by using these relations in complete form.
It is possible to obtain useful approximations by ignoring some terms

in the basic equations when they are either equivalent to zero or are
negligible, but accurate solutions can be achieved only by retaining the
full two-dimensional characteristics of the surge problem. Various sim-
plifications (discussed later) can be made by ignoring some of the physi-
cal processes. These simplifications may provide a satisfactory estimate,
but they must always be considered as only an approximation.
In the past, simplified methods were used extensively to evaluate

storm surge because it was necessary to make all computations manually.
Manual solutions of the complete basic equations in two dimensions were
prohibitively expensive because of the enormous computational effort.
With high speed computers, it is possible to resolve the basic hydro-
dynamic relations efficiently and economically. As a result of computers,
several workers have recently developed useful mathematical models for
computing storm surge. These models have substantially improved accuracy,
and provide a means for evaluating the surge in the two horizontal dimen-
sions. These more accurate methods are not covered here, but are highly
recommended for resolving storm-surge problems where more exactness is
3-95
warranted by the size or importance of the problem. These methods are
recommended only if a computer is available. A brief description of
these methods and references to them follows.
Solutions to the basic equations given can be obtained by the tech-

niques of numerical integration. The differential equations are approxi-
mated by finite differences resulting in a set of equations referred to
as the numerical analogs. The finite-difference analogs, together with
known input data and properly specified boundary conditions, allow evalua-
tion at discrete points in space of both the fields of transport and water
level elevations. Because the equations involve a transient problem,
steps in time are necessary; the time interval required for these steps is
restricted to a value between a few seconds and a few minutes depending on
the resolution desired and the maximum total water depth. Thus solutions
are obtained by a repetitive process where transport values and water-level
elevations are evaluated at all prescribed spatial positions for each time
level throughout the temporal range.
These techniques have been applied to the study of long-wave propa-

gation in various water bodies by numerous investigators. Some investi-
gations of this type are listed below. Mungall and Matthews (1970) devel-
oped a variable-boundary, numerical tidal model for a fjord inlet. The
problem of surge on the open coast has been treated by Miyazaki (1963),
Leendertse (1967), and Jelesnianski (1966, 1967, and 1970). Platzman
(1958) developed a model for computing the surge on Lake Michigan result-
ing from a moving pressure front, and also developed a dynamical wind tide
model for Lake Erie. (Platzman, 1963.) Reid and Bodine (1968) developed
a niimerical model for computing surges in a bay system taking into account
flooding of adjacent low lying terrain and overtopping of low barrier
islands.
b. Simplified Techniques for Determining Storm Surge . The tech-

niques described here for the determination of storm surge are simple, and
it is possible to carry out all storm surge calculations manually, using a
desk calculator or slide rule. In most cases, however, it is desirable to
employ a digital computer for the computations to reduce the effort and to
improve accuracy. It is sometimes possible to estimate surge with satis-
factory accuracy using a set of simplified equations, if the particular
problem is not too complex, and if the simplified technique can be verified
from actual prototype field data. Simpler schemes for computing storm
surge are obtained by including only those phenomena that appear signifi-
cant to the investigation; thus some of the less important terms are
omitted from Equations 3-50, 3-51 and 3-52.
(1) Storm Surge on the Open Coast . Ocean basins are large and
deep beyond the shallow waters of the Continental Shelf. The expanse of
ocean basins permits large tropical or extratropical storms to be situated
entirely over water areas allowing tremendous energy to be transferred
from the atmosphere to the water. Wind-induced surface currents, when
moving from the deep ocean to the coast, are impeded by the shoaling
bottom, causing an increase in water level over the Continental Shelf.
3-96
Onshore winds, cause the water level to begin to rise at the edge of the
Continental Shelf. The amount of rise increases shoreward to a maximum
level at the shoreline. Storm surge at the shoreline can occur over long
distances along the coast. The breadth and width of the surge will depend
on the storm's size, intensity, track and speed of forward motion as well
'
as the shape of the coastline, and the offshore bathymetry. The highest
water level reached at a location along the coast during the passage of a
storm is called the maximum surge for that location; the highest maximum
surge is called the peak surge. Maximum water levels along a coast do not
necessarily occur at the same time. The time of the maximum surge at one
location may differ by several hours from the maximum surge at another
location. The variation of maximum surge values and their time for many
locations along the east coast during Hurricane Carol, 1954, are shown in
Figure 3-44. This hurricane moved a long distance along the coast before
making landfall, and altered the water levels along the entire east coast.
The location of the peak surge relative to the location of the landfall
where the eye crosses the shoreline depends on the seabed bathymetry, wind-
field, configuration of the coastline, and the path the storm takes over
the shelf. For hurricanes moving more or less perpendicular to a coast
with relatively straight bottom contours, the peak surge will occur close
to the point where the region of maximum winds intersects the shoreline,
approximately at a distance R, to the right of the storm center. Peak
surge is generally used by coastal engineers to establish design water
levels at a site.
Attempts to evaluate storm surge on the open coast, and also in bays
and estuaries, when obtained entirely from theoretical considerations,
require verification particularly when simplified models are used. The
surge is verified by comparing the theoretical system response and computed
water levels with those observed during an actual storm. The comparison is
not always simple, because of the lack of field data. Most water-level
data obtained from past hurricanes were taken from high water marks in low-
lying areas some distance inland from the open coast. The few water-level
recording stations along the open coast are too widely separated for satis-
factory verification. Estimates of the water level on the open coast from
levels observed at inland locations are unreliable, since corrective adjust-
ments must be made to the data, and the transformation is difficult.
Systematic acquisition of hurricane data by a number of organizations

and individuals began during the first quarter of this century. Harris
(1963b) presented water-level data and synoptic weather charts for 28
hurricanes occurring from 1926 to 1961. Such data are useful for verifying
surge prediction techniques.
Because of the limited availability of observed hurricane surge data

for the open coast, design analysis for coastal structures is not always
based on observed water levels. Consequently a statistical approach has
evolved that takes into account the expected probability of the occurrence
of a hurricane with a specific CPI at any particular coastal location.
Statistical evaluation of hurricane parameters based on detailed analysis
of many hurricanes, have been compiled for coastal zones along the Atlantic
and Gulf coasts of the U.S. The parameters evaluated were the radius of
3-97
90" 80<» 70°
'^•B'ar Hartior
'"Portland
'Portsmouth
•"^Boston
NewLondorL, Woods Hole

Willets Point Newport
Batteryj; Montauk
— 40<> I
I
Sandy Hook
Philadelphia*
Baltimore*
Annapolis*'
31
Atlantic City
+ 40° —
Solomon Breakwater Harbor
Old Point Comfort^

Hampton Roads!
/
[Little Creek
Portsmouth^
Morehead Citv
Wilmington,
Southport
Charleston.
30
/
Savannah*.
29
./
S"
;302
+ 27 + 30° —
Key:
Development Stage ••••••
Tropical Stoge ^^î
Hurricane Stoge •^-^
Frontal Stage «»*^»
OPosition at 0700 EST
•Position at 1900 EST
90" 70°
I
(from Harris, 1963 b)
Figure 3-44. Storm Surge Chart. Hurricane Carol, 30, 31 August 1951,
Insert Map for New York Harbor
3-98
Aug 30 Aug 31 Aug 28 Aug. 29 Aug 30 Aug. 31 SepI
Figure 3-44. Storm Surge Chart. Hurricane Carol, 30, 31 August 1951.
Insert Map for New York Harbor -- Continued
3-99
maximum wind R; the minimiom central pressure of the hurricanes p ;
the forward speed of the hurricane V„, while approaching or crossing
the coast; and the maximum sustained wind speed W, 30 feet above the
mean water level.
Based on this analysis, the U.S. Weather Bureau (now the National
Weather Service) and U.S. Army Corps of Engineers jointly established
specific storm characteristics for use in the design of coastal struc-
tures. Because the parameters characterizing these storms are specified
from statistical considerations and not from observations, the storms
are termed hypothetical huvvioanes or hypo-hurriaanes . The parameters
for such storms are assumed constant during the entire surge generation
period. Graham and Nunn (1959) have developed criteria for a design storm
where the central pressure has an occurrence probability of once in 100
years. This storm is referred to as the standard project hurr-ioane (SPH)
The mathematical model used for predicting the wind and pressure fields in
the SPH is discussed in Section 3.72, Model Wind and Pressure Fields for
Hurricanes. The SPH is defined as a "hypo-hurricane that is intended to
represent the most severe combination of hurricane parameters that is
reasonably oharaateristia of a region excluding extremely rare combina-
tions." Most coastal structures built by the U.S. Army Corps of Engineers
that are designed to withstand or protect against hurricanes are based on
design water associated with the SPH.
The construction of nuclear-powered electric generating stations in

the coastal zone made necessary the definition of an extreme hurricane
called the probable maximum hurricane (PMH). The PMH has been adopted by
the Atomic Energy Commission for design purposes to ensure public safety
and the safety of nuclear-power facilities. Procedures for developing a
PMH for a specific geographical location are given in U.S. Weather Bureau
Interim Report HUR 7-97 (1968). The PMH was defined as "A hypothetical
hurricane having that combination of characteristics which will make the
most severe storm that is reasonably possible in the region involved, if
the hurricane should approach the point under study along a critical path
and at an optimum rate of movement."
Selection of hurricane parameters and the methods used for developing
overwater wind speeds and directions for various coastal zones of the
United States are discussed in detail by Graham and Nunn (1959) and in
HUR 7-97 (1968) for the SPH and PMH. The basic design storm data should
be carefully determined, since errors may significantly affect the final
results.
Two simple methods are presented here for estimating storm surge on
the open coast: one a quasi-static numerical scheme and the other a nomo-
graph method. These methods should never be used for estimating the surge
in bays, estuaries, rivers or in low-lying regions landward of the normal
open-coast shoreline. Neither method is entirely satisfactory for all
cases, but for many problems both appear to give reasonable solutions.
The use of each method is illustrated by estimating the peak open- coast
storm surge for an actual hurricane. The peak surges thus calculated are
compared to the surge computed by a complete two-dimensional numerical
model for the same storm.
3-100
(a) Quasi-Static Method for Prediction of Hurricane Surge .
TTiismethod for determining open-coast storm surge is based on theoretical
approximations of the governing hydrodynamic equations originally proposed
by Freeman, Baer, and Jung (1957). The term quasi-static method is used
here to emphasize that this method should be restricted to slow-moving,
large-scale storm systems. This method is called the Bathys trophic Storm
Tide Theory and, unlike earlier one-dimensional theories, some of the
effects of longshore flow and the apparent Coriolis force are considered.
Such an approximation of the theory can be described as a quasi-static
method in which a numerical solution is obtained by successively integrat-
ing wind stresses over the Continental Shelf from its seaward edge to the
shore for a predetermined interval of time.
This simplified method assumes that storm surge responds instanta-

neously to the onshore wind stresses, advection of momentum can be ignored,
longshore sea surface is uniform, and no flow is assumed normal to the
shore which is treated as a seawall. Barometric effects and precipita-
tion also can be neglected. Setup due to atmospheric pressure difference
can be estimated from another source, and added to the final design water
level. Based on the preceding assumptions. Equations 3-50 and 3-51 reduce
to
gD —
3x
= fV + -^ (3-53)
ay _ ^sy - '^by
(3-54)
3t p
Conservation of mass is not considered because, (1) there is no flow

perpendicular to shore, (2) the longshore flow is assumed independent of
y and, (3) the water level is assumed slowly changing. The forces (ex-
pressed in mass times acceleration per unit area) involved and the corres-
ponding response of the sea for the bathystrophic approximation are shown
in Figure 3-45. As indicated in the figure, the surface shear force act-
ing in the x-direction Xg^ and the apparent Coriolis force is balanced
by the hydrostatic pressure force pgA(9S/3x). Moreover, the surface shear
force acting in the y-direction t is balanced by the bottom shear
force T^jy and the inertial force p(9V/9t).
Bretschneider and Collins (1963) developed a computational model

based on this Bathystrophic Theory and applied it to open- coast surge
problems for the region around Corpus Christi, Texas. Marinos and
Woodward (1968) modified the Bretschneider and Collins model and calibra-
ted it for various reaches along the Texas coast by using three hurricanes
of record, and also made parametric studies of hypo-hurricane surges for
the entire coast of Texas.
In some cases the underlying assumptions made in the development of

this theory are not satisfied. Thus, as a consequence of assuming that
the onshore wind stresses cause an instantaneous change of the water level,
i.e., U = 0, the traverse line (the line over the Continental Shelf along
which computations are carried out) must always be taken perpendicular to
3-101
» o
A —
> o
o ° o
•H
D _ 4->
m » O
2 »- "- §
O o o
X
o
> TJ Q. (^
o u
z •H
o
p
M
>s
'P
cd
CQ
^1
.2
(A
in
n
o
p<
in
iJ
g
tn
Q>
u
^
o
u,
(4-1
0)
•H
[L,
2 §
II n
I- ». Q- o« ^
3-102
the bottom contours for valid computations. The above assumption also
implies that there is no flooding landward of the shore; thus, there is
a deficiency in the method when substantial flooding occurs.
The bottom contours of the actual seabed are rarely straight and para-
llel; however, the traverse line can often be oriented so that it is nearly
perpendicular to the contours in an average way. For complex offshore
bathymetry, such an approximation would be invalid. For storms moving
more or less perpendicular to a coastline, the traverse line can be taken
through, or anywhere to the right of, the region of maximum winds, but
never to the left of this region. Many other factors such as the angle
of approach of a storm, the coastline configuration, and inertial effects,
limit the use of such a simple approach.
The computation model given here is based on Bathystrophic Storm Tide

Theory as described by Bodine (1971). Although Bodine applied both manual
and digital computer calculation methods to the open coast storm surge
problem, only the manual method is presented.
The bottom and surface shear stresses are assumed to vary according
to:
T
^by KVIVI
(bottom shear stress) (3-55)
sx
^-
~p = kW^ cos e
(wind shear stress) (3-56)

T
^ = kW^ sin e
in which K is a dimensionless bottom friction coefficient, k is a

dimensionless surface friction coefficient, W is the wind speed, and
6 is the angle between the x-axis and the local wind vector. The bottom
friction coefficient K is related to the coefficient of Chezy C and
the Darcy-Weisbach friction factor f^ as follows:
Typical bottom conditions result in a value of K that lies in the

range between 2 x 10"^ and 5 x 10" ^. For a first estimate, a value
of K = 2.5 x 10"^ may be assumed. This coefficient is used in cali-
brating the model. It not only accounts for energy dissipation at
the bed, but may be used to adjust for inexact modeling and deficien-
cies caused by ignoring some of the hydrodynamic processes involved.
3-103
The wind stress coefficient is based on one given by Van Dom (1953) and
others which is assiomed to be a function of wind speed; thus,
k = X ,
for W < W^ (3-58)
as
ax
are used to denote discrete points in space and time, respectively. The
quantities Ax and At are allowed to be nonuniform spacings along the
traverse line and time increments, respectively. The finite difference
forms of Equations 3-61, 3-62, and 3-63 are then given as follows:
Ax n+ l
(3-65)
^^^- If +f v"""^
(As,)r.v! \
(3-66)
1
+
B.- + B.>i)" + (B.- + B.^,) n l
At + V + 'A
(3-67)
l + K|V«,^JAt(D-)-^^
where A and B are the kinematic forms of the wind stress given by:
A = k W cos (3-68)
B == k W^ sin (3-69)
The time step specified by the ordinal number n represents a time

level at which AS^, AS^, and V are known, while n+l represents the
new time level for which the quantities AS^, ^Sy and V are to be
determined.
The total water depth at the mid-interval between two time steps is
given by:
„n „n + l
*
Â Â
+ Js
+ (s. + s),;,^
i 2 ^
(3-70)
<^ (%),• + {Âp)i + l]" + [(%). + {Âp)i + lY^']
where
Sg = initial setup
S^ = astronomical tide
S^p — atmospheric pressure setup.
3-105
The initial setup refers to the water level at the time the storm surge
computations are started. An approximate relationship giving the atmos-
pheric pressure setup in feet when pressure is expressed in inches of
mercury is
Sap = 1-14 (p„-pj d-e-'^) (3-71)
where p„ is the pressure at the periphery of the storm, and r is the

radial distance from the storm center to the computation point on the
traverse line.
The total depth at the new time level is evaluated by the relation:
d,.+ d,.^,
Dn++'A1 + s, + s^"^+ (s. + sj;,,'A
(3-72)
n+l
(Sap), + (s^p),.,]
The total water level rise at the coast is the sum of all component water
level changes resulting from the meteorological storm plus those components
which are not related to the storm. Hence the total setup is given by
s. + S3, + s^p + s, + S^ + S^ + S^ (3-73)
where Sj/ is a component due to the wave setup in the region shoreward of
the breaker line and is related to the breaker height by Equation 3-49.
(See Section 3.85, Wave Setup.) The setup component S^ accounts for the
setup resulting from local conditions such as bottom configuration, coast-
line shape, or other flows influencing the system, such as flows from bay
inlets or rivers. This component can only be estimated from a full under-
standing of the influence of topographic and hydrographic features not
considered in the numerical calculations. Contributions made by the
various setup components to the total surge are shown in Figure 3-46.
Equations 3-65, 3-66, and 3-67 can be rewritten in a more useful form
by introducing dimensional constants for terms frequently appearing in the
equations. Hence,
C,1 Ax
n+l
(As.)r;4 A. + ^.l) (3-74)
i+'A
C2 Ax
(-^)::i (sm0),. + (sin0).^^] V^^,)^ (3-75)
D'
n+l
i+'A
(B, + B.,,)«+(B, + B.,,)«-^](At)+ V",,

n+ l
(3-76)
î+'A
l+C3lV«,^JAtK(D-)r;^^
3-106
22
20 \ /
18
E
ô 16
O
w 14
z
5 12
o
o
o
>
100
200
300
a.
a>
O
400
E
o
mo 500
600
The values of the dimensional constants Cj, C2, and C3 in Equations
3-74, 3-75 and 3-76 depend on the units used in performing the calculations.
Table 3-9 gives the dimensions of the variables used in the numerical
scheme in three systems of units and the corresponding value of the
constants for each system. The first column of units is given in the
metric system while the other two are given in mixed units in the English
system.
Table 3-9. Systems of Units for Storm Surge Computations
Parameters
It is customary to assume that the system is initially in a state of
equilibrium and V = 0, and S is uniform along the traverse line at the
start of the computational scheme. Thus, computations should be initiated
for conditions prior to the arrival of strong winds over the Continental
Shelf. Although the real system would seldom, if ever, be in a complete
initial state of equilibrium, errors in assuming it to be are of little
consequence in the computational scheme after several time steps in the
calculations, because the effects of the forcing functions will eventually
predominate
To demonstrate the computational procedures, the storm surge in the

Gulf of Mexico resulting from Hurricane Camille (1969) is calculated.
Hurricane Camille was an extremely severe storm that crossed the eastern
part of the Gulf of Mexico with the eye of the storm making landfall at
Bay St. Louis, Mississippi, at about 0500 Greenwich Mean Time (GMT) on
18 August 1969. Unusually high water levels were experienced along the
gulf coast during the passage of Camille because of the intense winds and
relatively shallow water depths which extend far offshore. The storm surge
is calculated and compared with the peak surge generated by Camille.
Information published by the U.S. Weather Bureau for Hurricane Camille

in HUR 7-113 (1969) indicates that R = 14 nautical miles and Wp = 13 knots
would be representative of these values which are assumed to be invariant
while the storm moved over the Continental Shelf. The wind data and track
for Hurricane Camille have been published by the Weather Bureau in HUR
7-113A (1970). The track, together with the traverse line used in the
present calculations, is shown in Figure 3-47. Overwater wind speeds and
directions are shown in Figure 3-48. A profile of the seabed along the
traverse line is shown in Figure 3-49.
Tide records from the region affected by the storm show the mean water
level to be about 1.2 feet above normal before being affected by the storm.
This value is taken to represent the initial water level. The range of
astronomical tides in this location is about 1.6 feet. A constant value of
0.8-foot above MSL is used in the computations; the final surge hydrograph
at the coast can be subsequently corrected to account for the predicted
variations of the tide. Variations in the initial water level and astro-
nomical tides may be added algebraically to the storm surge calculations
without seriously affecting the final results. The atmospheric pressure
difference, P„ - P„. is needed for evaluating the pressure setup com-
ponent, S^p. Data given in HUR 7-113 (1969) suggest that p^ = 26.73
inches of mercury and p^ = 29.92 inches of mercury are representative for
the hurricane. All of these values are assumed to remain constant for the
calculations
The wind data, basin profile, and hurricane characteristics provide

the basic information needed in making an estimate of the peak storm surge
associated with Hurricane Camille. The time intervals, distance increments
along the traverse line, etc., are given in the computational steps to
follow.
3-109
3-110
a m
o
o
u
3 in
O,
0)
3
ti
%
to
n o (D
JJ ON
<
in
D
E
o
V.^ ^ i
1
o
o
This quasi-linear computational scheme can he used for manual com-
putations; however, since the calculations are repetitive, they can be
performed more efficiently by us,ing a digital computer. When carried out
manually, the technique is laborious, tedious and subject to error.
Bodine (1971) gives a computer program based on the numerical scheme
presented here. Other programs bas«d on similar niomerical schemes using
the quasi-static methods have been developed.
When manual computation of storm surge is necessary, a systematic,

tabular procedure must be adopted to permit stepping through all of the
discrete computational points in space for each time increment. Table
3-10 represents a recommended procedure. One table is required for each
time increment. Table 3-10 corresponds to the time of peak surge for
Hurricane Camille; preceding tables required to bring the calculations to
this point are not included since those calculations are similar. The
first table in the series must reflect the initial conditions; thus V is
taken to be zero and S is assumed uniform over the system.
Manual surge calculations for Hurricane Camille give a peak surge of

25.03 feet, say 25 feet (MLW) or 24.2 feet (MSL) . The bottom friction
coefficient selected for this particular example was K = 0.003 and the
surge was found to be insensitive to small changes in the friction coeffi-
cient. Computer calculations using a friction coefficient of 0.003 result-
ed in a peak surge of 25.19 feet and a bottom friction coefficient of
0.0025 resulted in a peak surge of 25.40 feet. For some basins and storm
systems, the bottom shear stresses are more significant in determining
water levels. Therefore, it is important to select a bottom friction
coefficient by verification (i.e., by comparing calculated results with
observed water levels). After such verification, the model may be used
to estimate the storm surge from hypothetical hurricanes for the same
geographical region.
The surge hydrograph (water level as a function of time) for Hurricane

Camille is shown in Figure 3-50 for the most landward computational point
on the traverse line. This figure shows that the water level rose for
about the first 8 hours, hut then began to fall gradually until about 27
hours of computational period had elapsed, then began to rise rapidly.
A study of the local wind fields during this period shows that the winds
had an onshore component in the early stages of the storm, then the winds
began blowing offshore for several hours before the principal rise at the
coast.
(b) Nomograph Method . A simplified method for obtaining

a first approximation to the peak storm surge of a hurricane can be based
on an empirical analysis of past records, an empirical analysis of a sys-
tematic set of calculations with numerical models, or a combination of the
two. Jelesnianski (1972) combined empirical data from Harris (1959) with
his theoretical calculations to produce a set of nomograms that permit the
rapid estimation of peak surge for any geographical location when a few
parameters characterizing a storm are known.
The first nomogram (Fig. 3-51) permits an estimate of the peak surge
Sj, generated by an idealized hurricane with specified CPI and radius of
3-115
3ur|
O *-4 _
T f f f f f
'"J- 00 r^ cxD
'•^ <*i -^ in
13 I I I I
IS
<
+ t4-
°S IS
s
o"
S.S-
II
^ m -'
« ^ „ c^
<
^ -^ »o
o •— c>i
il < '
< E
« PJ ^ -H
^1 r- ^
vo
\f)
ph
r-
^h
~T~ t r^
>3
via 6
-» ^ so
— . ^ <s
n Q
o r-
o^ — ^ ^ *
tn .-' «
— r- ts
+
^5
Q O tn c^
w -: w
c^ <s o
a;
. -q * -^
.- « o
D O .X ri-
M -- "
sun
3-117
VD
in
o
•H
a-
1/1
uO
c
o
I
to
a;
•H
(4i) |9Aa-| jaiDAA |D|Oi
18
30 40 50 60 70 80 90 100 110 120 130 140
Ap (Pressure Drop)mbs. (from jeiesnionski, 1972)
Figure 3-51. Preliminary Estimate of Peak Surge
3-119
maximum wind that is moving perpendicular to a shoreline with a speed of
15 MPH. This nomogram indicates there is a critical storm size as reflec-
ted by the radius of maximum winds, R. For a given pressure drop greater
than zero, the highest peak surge is produced for a critical value of R
equal to 30 miles and any value of R greater or less than this value
results in a lesser value of the peak surge.
A second factor F^ given in Figures 3-52 and 3-53 adjusts for the
effects of variations in bathymetric characteristics along the gulf and
Atlantic coasts. A third factor Fm given in Figure 3-54, adjusts for
the effects of storm speed and the angle with which the storm track
intercepts the coast
The predicted peak storm surge Sp is then given by
Sp = h Fs ^M (3-78)
Jelesnianski (1972) applied the scheme to the 43 storms given by Harris

(1959) that entered land south of New England during the period from 1893
to 1957. The peak surges reported by Harris are plotted against the peak
surges predicted by the nomograph method in Figure 3-55. The two-
dimensional hurricane model and storm surge prediction model described by
Jelesnianski (1967) was used for all calcualtions without adjustment for
local variations in friction coefficient or other efforts to calibrate
the model for individual storms. For many of the hurricanes, the post-
storm surveys conducted were of limited scope and probably did not disclose
the true peak surge. Thus, at least a part of the spread between observed
and computed values must be due to the observed data. In addition to the
peak surge, other nomograms for computing other storm surge parameters are
given by Jelesnianski (1967).
An example problem illustrating the use of the nomogram method follows:
************** EXAMPLE PROBLEM **************

GIVEN : Parameters for Hurricane Camille are:
Ap = 3.19 inches of mercury (in. Hg.)
Vp = 13 knots
R = 14 nautical miles (n.m.)
FIND : Estimate peak open-coast surge produced by Hurricane Camille.
3-120
laaj ui ijidso
o
o
at
,082
9092
oOfrE
4 6 8 10 12
Computed Peak Surge, Sp (Feet above MSL)
Figure 3-55. Comparison of Observed and Computed Peak Surges

(for43 storms with a landfall south of New England
from 1893-1957)
3-124
SOLUTION: Because of the units used in the various nomograms, the units
of the parameters are converted to
3.19(1000)
Ap^ = 3.19 in. He.
^
= = 108 milhbars .
29.53
Note: 29.53 inches of Hg. « lOOO millibars
Yp = 13 knots * 14.5 mph
R = 14 n.m. * 15.6 miles.
Using Figure 3-51 with values of Ap and R,
Sj «s 22 feet (MSL)
Based on the approximate landfall location (west of Biloxi) of

Camillc (Fig. 3-47), the shoaling factor F
rricane Camille
Hxorricane is determined
s
from Figure 3-52,
F^ ^ 1.23 .
To evaluate the correction for storm motion Fm the angle of storm

approach to the coast must be first
tj; determined. The definition of
the angle is shown schematically in
\l)
the insert in Figure 3-54.
From Figure 3-47 ^ is estimated to be 102°. From Figure 3-54 for
4) = 102° and Vp = 14.5 mph,
^M * 0.97 .
The peak surge given by Equation 3-78
Sp = (22) (1.23) (0.97)
Sp = 26.2 feet, say 26 feet (MSL)
3-125
Jelesnianski (1972) has calculated the open coast surge for
Hurricane Camilla with a full two-dimensional mathematical model. The
maximum surge envelope along the coast, based on computations from this
model, is shown in Figure 3-56 where the zero distance corresponds to
the point of landfall of the hurricane eye. This figure shows that
the peak surge estimated by this method is about 25+ feet MSL.
Thus, by three independent estimates, it has been found that the peak
surge is about 25 feet MSL which corresponds approximately to that
observed (U.S. Geological Survey) at Pass Christian, Mississippi of
24.2 feet MSL. It would be expected that a slightly higher peak water
elevation occurred because Pass Christian is located a few miles left of
the position where the maximum winds made landfall.
It is rare that such a close agreement is found when estimating the

peak surge with these dissimilar models. Normally, because of the
difference in these predictive schemes, it can be expected that peak
surge estimates may deviate by as much as 25 percent. For well-
formulated schemes properly applied, there is usually a trade-off
between reliability of the estimate and the computational effort.
*************************************
(c) Predicting Surge for Storms other than Hurricanes .
Although the basic equations for water motion in response to atmospheric
stresses are equally valid for nonhurricane tropical and extratropical
storms, the structures of these storms are not nearly so simple as that
of a hurricane. Because the storms display much greater variability in
structure, it is difficult to derive a proper wind field. Moreover, no
system of classifying these storms by parameters has been developed
similar to hurricane classification by such parameters as radius to
maximum winds, forward motion of the storm center, and central pressure.
Criteria however have been established for a Standard Project North-

easter for the New England coast north of Cape Cod as given by Peterson
and Goodyear (1964). Specific standard-project storms other than hurri-
canes are not presently available for other coastal locations. Estimates
of design-stoim wind fields can be made by meteorologists working directly
with climatological weather maps, and by use of statistical wind records
on land and assuming that they blow toward shore for a significant dura-
tion over a long, straight line fetch.
Once the wind field is determined, estimation of the storm surge may
be determined by methods based on the complete basic formulas or the quasi-
static method given. The nomogram method cannot be used, since this scheme
is based on the hurricane parameters.
(2) Storm Surge in Enclosed Basins . An example of an inclined

water surface caused by wind shearing stresses over an enclosed body of
water occurred during passage of the hurricane of 26-27 August 1949 over
the northern part of Lake Okeechobee, Florida. After the lake level was
inclined by the wind, the wind direction shifted 180° in 3 hours, resulting
3-126
o
in a turning of the height contours of the lake surface. However, the
turning of the contours lagged behind the wind so that for a time the
wind blew parallel to the water level contours instead of perpendicular
to them. Contour lines of the lake surface from 1800 hours on 26 August
to 0600 hours on 27 August 1949 are shown in Figure 3-57. The map con-
tours for 2300 hours on 26 August show the wind blowing parallel to the
highest contours at two locations. (Haurwitz, 1951), (Saville, 1952),
(Sibul, 1955), (Tickner, 1957), and (U.S. Army, Corps of Engineers, 1955).)
Recorded examples of wind setup on the Great Lakes are available from
the U.S. Lake Survey, National Ocean Survey, and NOAA. These observations
have been used for the development of theoretical methods for forecasting
water levels during approaching storms and for the planning and design of
engineering works. As a result of the need to predict unusually high
stages on the Great Lakes, numerous theoretical investigations have been
made of wind setup for that area. (Harris, 1953), (Harris and Angelo,
1962), (Platzman and Rao, 1963), (Jelesnianski, 1958), (Irish and
Platzman, 1962), and (Platzman, 1958, 1963, 1965, and 1967),)
Water level variations in an enclosed basin cannot be estimated satis-
factorily if a basin is irregularly shaped, or if natural barriers such as
islands affect the horizontal water motions. However, if the basin is
simple in shape and long compared to width, then water level elevations
may be reasonably calculated using the hydrodynamic equations in one
dimension. Thus if the motion is considered only along the x-axis (major
axis), and advection of momentum, pressure deficit, astronomical effects
and precipitation effects are neglected, then Equations 3-50 and 3-52
reduce to
3U 98 1
9s aiu
(3-80)
at
~ ax
If it is further assumed that steady state exists, then Equation 3-79 becomes
dS 1
i^s+'^b)
(3-81)
dx pgD
The bottom stress is taken in the same direction as the wind stress, since
for equilibrium conditions the flow near the bottom is opposite to flow
induced by winds in the upper layers. Theoretical development of this
wind setup equation was given by Hellstrom (1941), Keulegan (1951), and
others. The mechanics of the various determinations have differed some-
what, but the resultant equation has been about the same. This wind
setup equation is expressed as:
k'np^W'F
AS = COS0 (3-82)
PgD
3-128
53 - *'
0»
3? > So - 3 . !iifi
ssi'
where
k' = a numerical constant as 0.003
= air density
p^
W = Wind velocity
F = Fetch length
G = angle between wind and the fetch
n = 1 + '^s/'^b; 1.15 < n < 1.30
AS = Wind setup. This value represents

the difference in water level
between the two ends of the fetch
D = average depth of the fetch
An approximate expression of Equation 3-82 is given by
CW^F
AS = cos 6 (3-83)
D
where is a coefficient having dimensions of time squared per unit

C
length. Saville (1952) in a comprehensive investigation of setup data
obtained from Lake Okeechobee found that C is approximately 1.165 x 10"^
when W is given miles per hour, F in miles and D and AS in feet.
This coefficient is almost identical with that of the Netherlands Zuider
Zee formula (1.25 x 10"^).
Equation 3-83 is often useful in making the first approximation of

the setup in an enclosed basin. Its advantage is that the setup can be
evaluated with fewer computations. The surge can be estimated more satis-
factorily by segmenting an enclosed basin into reaches and using a numeri-
cal integration procedure to solve Equation 3-81 for the various reaches
in the basin. Bretschneider (Ippen, 1966) presented solutions in parameter
form for Equation 3-81, and compiled these solutions in tables for different
conditions. These tables can be used to estimate the storm surge for a
rectangular channel of constant depth with either an exposed bottom or a
nonexposed bottom and a basin of regular shape. Such solutions may provide
an estimate of the water level variation in some basins.
A more complete scheme than those described above follows. This

method accounts for the time dependence of the problem; more specifically
Equations 3-79 and 3-80 are used. Although this more complete method
provides a better approximation of the surge problem, it is done at the
expense of increasing the computations.
3-130
When a basin enclosed with vertical sides has a well-defined major
axis and a gradually varying cross section, it is possible to use a
dynamic one-dimensional, computational model for evaluating fluctuations
in water level resulting from a forcing mechanism such as wind stress,
and also to account for some of the effects of a varying cross section.
The validity of such a model depends on the behavior of the storm system
and the geometric configuration of the basin.
Equations 3-79 and 3-80, when the varying width b of the basin is
introduced, can be written in terms of the volume flow rate Q(x,t) as
9Q 9S b ,
as _ 1 aq (3-85)
9t b dx
where x is taken along the major axis of the basin, and for any time t,
A is the cross-section area, and D is the average depth A/b or
D = d + S.
Various schemes have been proposed for evaluating the water level
changes in an enclosed body of water by using the differential equations.
(Equations 3-84 and 3-85.) The formulation of the problem and the numer-
ical scheme given here is from Bodine, Herchenroder and Harris (1972)
The surface stress and bottom stress terms are taken identical to the
terms given for the quasi-static method for open-coast surge. (See
Equations 3-55 and 3-56, Section 3-865b(l) (a) .) The stress terms, in
terms of the volume flow rate, become
^ _ (3-86)
A finite difference representation of Equations 3-88 and 3-85 may be
expressed in the form
Qi'l = ^ [qi, + f (b,,,/, +b,,3/,) (kw^ cose):';^
(3-89)
At n+ 1
(3-90)
where
G = 1 + (3-91)
{Î'A-'^ls/S (V/.+V3A)
The value of G is greater than unity for most flow conditions except
for the case when the flow vanishes (Q = 0). The subscripts and super-
scripts i and n are used to denote discrete points in space and time,
respectively. A schematic of the grid system used is shown in Figure 3-58.
It is seen that S at the new time level (t + At) is first evaluated
based on the known values of Q, D, S, A and b(x) lying on triangle (1)
and subsequently followed by an evaluation of S at the new time level
based on known values lying on triangle (2). The solutions at successive
time levels are obtained by a marching process with Q evaluated at the
new time level for all integer steps along x; S is evaluated for all
mid-integer steps along x. Width as a function of x, b(x), is taken
constant for all t and the total depth D is permitted to vary with
time. Thus the cross-section area of the basin A is a function of
distance along the major axis of the basin and a function of time.
The computational scheme is a combined boundary and initial value

problem. At each end of the basin, it is assumed that there is no flow
across the boundary, thus Q = at the boundary. The initial conditions
assumed are that Q = and S is uniform throughout the basin.
The scheme requires that for numerical stability, the time increment
specified for successive calculations At be taken less than Ax/i/gâx>
where Dmax is the maximum depth (d + S) anticipated in the basin during
passage of a storm system. (Abbott, 1966.)
The restriction imposed by the criterion for numerical stability

results in a trade-off between the resolution obtained in the solution
and the number of calculations involved. Decreasing Ax gives better
resolution of the surge, but requires a smaller At and increases the
number of computational steps. It is important to choose Ax small
enough so that reasonable resolution of the surge is obtained, but large
enough to reduce the computations. The choice of a Ax depends on the
problem involved.
3-132
CM
0>
O
I
X
e
c
D
~
— e
o
6
0)
+J
M
X
CO
X
•H
00
I
4)
HE
<
X
<a
3-133
For such a recursive type calculation, computer computations reduce
the effort required. However, since this is a one-dimensional formulation
of the problem, it is possible to carry out the necessary computations
manually.
The March 1955 storm on Lake Erie is used to demonstrate the scheme.
A plan view of Lake Erie is shown in Figure 3-59. The width and cross-
section area of the Lake are shown in Figure 3-60, mean bottom profile in
Figure 3-61, and wind and wind direction data for the March 1955 storm in
Figure 3-62.
In the following example problem computations are made at a single

spatial point where Q is evaluated at a distance of 3Ax from Buffalo
(Fig. 3-59) at a time when the wind speeds are approximately maximum.
Tables similar to Table 3-10 should be used in manual calculations.
*********** EXAMPLE PROBLEM
GIVEN: X
o
_l
<
to
3-135
20 40 60 80 100 120 140 160 180 200 220 240 260
Distonce from Toledo ( Statute Miles)
(after Plotzmon ond Roo, 1963)
Figure 3-60. Cross-Section Area and Surface Width -Lake Erie
3-136
20 40 60 80 100 120 140 160 180 200 220 240 260
Distance from Toledo ( Statute Miles)
Figure 3-61. Mean Bottom Profile of Lake Erie
3-137
FIND: Wind setup at the new time level.
SOLUTION : The wind stress coefficiei t given by Equation 3-59 is
k = K, + K, (l - = 1-1 X 1«" + 2.5 X 10- (l-^^j

^)
= 2.27 X 10"*
From Equation 3-91
4KAt|Q;'^^ I
G = 1 +
(D?.:^ + D?+3/J"{b,.^j,+b,>3/J
4 (0.003) (0.2) 0.0707

I I
= i
oi
*^ ~ "^ "^
[(0.0162)2 + (0.0132)2] (26.3 + 21.0)
The two terms needed in the evaluation of Q, (Equation 3-89) are

given by
(b.>'^+b,,3/^) (kW^ COS 0):';/

f
0-2
(26.3 + 21.0) (2.27 XIO'^) (50.5)^ (1) = 0.0274
and
gAt
'^
2Ax {î+'A î+3/2) {î+'A ^1 + 3/2)"
79,000 (0.2) (0.426 + 0.278) (3.32-3.76)

= - 0.0463
(2) (10) (5,280)
The volume flow rate from Equation 3-89 is
^^ =
QLV
1.0129
~ [0.070 + 0.0274-0.0463] = 0.0513 mi? /hr.
3-139
and finally, the water level at this discrete point in space and time
is given by Equation 3-90 as
At / _ _ \n+l
^"4 = C«-^(Q,-Q«)' i+'A
(0.2) (0.0915 -0.0513) (5280)

= 3.32 +
(10) (26.3)
= 3.48 feet, say 3.5 feet .
The significant digits indicated in the above computations do not

reflect the accuracy of the numerical procedure, but are retained to
reduce the accumulation of round-off errors.
The wind setup hydrograph for the ends of the lake as determined by
computer is shown in Figure 3-63. The storm winds blew in the general
direction toward Buffalo at the northeastern end of the lake as indi-
cated by the setdown at Toledo and the setup at Buffalo. The spatial
steps Ax taken are quite large; smaller increments in Ax would give
a more accurate estimate. Calculations with the mathematical model
were initiated with the system in a calm state, i.e., Q = S = 0.
The wind setup profile along the major axis of the lake determined
by the numerical scheme is shown in Figure 3-64 for three time periods
during the storm. The nodal point where the computed water surface
crosses the Stillwater level occurs near the center of the lake, but
this nodal point can vary with time.
Although the comparison of the computed and observed wind setup is

not in complete agreement, particularly at the beginning and late stages
of the storm, the method gives reasonable results for the wind setup
amplitudes. To engineers, it is frequently the maximum departure of
the water level from its normal position that is of greatest concern.
Results of the simplified model should be interpreted with care, since
many of the physical processes which may be significant have been neg-
lected. Wind and bottom stress laws, in particular, are oversimplified
for the Lake Erie problem. Better agreement can be expected with two-
dimensional schemes such as the one developed by Platzman (1963) , since
they more accurately model for the physical processes involved.
(3) Storm Surge in Semienclosed Basins It is generally im-

.
possible to make reliable estimates of storm surge in semienclosed basins

(bays, and estuaries) with less exact procedures such as those described
for specific problems within enclosed basins or on the open coast. This
is because bays and estuaries are nearly always irregular in shape, and
basin geometry is often further complicated by the presence of islands,
navigational channels, and harbors. Moreover, many of these basins have
3-140
in
o
t—
o
e
I—
O
-^-
en
I
o
T3
(1>
T3
C
o
o
o
M—
CD
O
t—
o
o
X
CO
T3
c
ro
CO
ro
(âaj) dnias pujM
3-141
1/3
o
•p
CO
mu
0)
•H
o
a>
CO
to
O
U
3
bo
•H
[I.
oo (O <\j O cvj ^ CO o
I I I I
(499j)dnias puiM
3-142
extensive low- lying areas surrounding them which may be subject to exten-
sive flooding during severe storm conditions. Also, these basins are
usually shallow, and on the windward side of the basin substantial bottom
areas can become exposed. Thus, during the passage of a storm system,
many semienclosed basins have time-dependent moving boundaries. The
water present at any time in a bay or estuary is also dependent upon the
sea state because of the interrelation between these two bodies of water.
Because of the above characteristics of semienclosed basins, a simplified
approach does not usually provide a satisfactory estimate of water motions,
and methods based on full two-dimensional dynamic approaches should be
employed.
3-143
ABBOTT, M.B., An Introduction to the Method of Charaateristias^ American

Elsevier, New York, 1966, 243 pp.
ARAKAWA, H., and SUDA, K. , "Analysis of Winds, Wind Waves, and Swell Over
the Sea to the East of Japan During the Typhoon of September 26, 1935,"
Monthly Weather Review ^ Vol. 81, No, 2, Feb. 1953, pp. 31-37.
BAER, J., "An Experiment in Numerical Forecasting of Deep Water Ocean

Waves," Lockheed Missile and Space Co., Sunnyvale, Calif., 1962.
BARNETT, T.P., "On the Generation, Dissipation, and Prediction of Ocean

Wind Waves," Journal of Geophysical Research^ Vol. 73, No. 2, 1968,
pp. 531-534.
BATES, C.C, "Utilizations of Wave Forecasting in the Invasions of

Normandy, Burma, and Japan," Annals of the New York Aaademy of
Sciences J Vol. 51, Art. 3, "Ocean Surface Waves," 1949, pp. 545-5 72.
BELLAIRE, F.R., "The Modification of Warm Air Moving Over Cold Water,"
Proceedings 3 Eight Conference on Great Lakes Research, 1965.
BODINE, B.R., "Storm Surge on the Open Coast: Fundamentals and Simpli-
fied Prediction," TM-35, U.S. Army, Corps of Engineers, Coastal
Engineering Research Center, Washington, D.C., May 1971.
BODINE, B.R., HERCHENRODER, B.E., and HARRIS, D.L., "A Numerical Model
for Calculating Storm Surge in a Long Narrow Basin with a Variable
Cross Section," Report in preparation, U.S. Army, Corps of Engineers,
Coastal Engineering Research Center, Washington, D.C. , 1972.
BORGMAN, L.E., "The Extremal Rayleigh Distribution: A Simple Approxima-

tion for Maximum Ocean Wave Probabilities in a Hurricane with Varying
Intensity," Stat. Lab Report No. 2004, University of Wyoming, Laramie,
Wyo. , Jan. 1972.
BOWEN, A.J.,INMAN, D.L., and SIMMONS, V.P., "Wave 'Setdown' and Set-up,"
Journal of Geophysical Research, Vol. 73, No. 8, 1968.
BRETSCHNEIDER, C.L., "Revised Wave Forecasting Curves and Procedures,"

Unpublished Manuscript, Institute of Engineering Research, University
of California, Berkeley, Calif, , 1951.
BRETSCHNEIDER, C.L. "Revised Wave Forecasting Relationships," Proceedings

of the Second Conference on Coastal Engineering, ASCE, Council on
Wave Research, 1952a.
BRETSCHNEIDER, C.L., "The Generation and Decay of Wind Waves in Deep

Water," Transactions of the American Geophysical Union, Vol. 33,
1952b, pp. 381-389.
3-145
BRETSCHNEIDER, C.L., "Hurricane Design-Wave Practice," Journal of the
Waterways and Harbors Division^ ASCE, Vol. 83, WW2 , No. 1238, 1957.
BRETSCHNEIDER, C.L., "Revisions in Wave Forecasting; Deep and Shallow

Water," Proceedings of the Sixth Conferenoe on Coastal Engineering,
ASCE, Council on Wave Research, 1958.
BRETSCHNEIDER, C.L., "Wave Variability and Wave Spectra for Wind-

Generated Gravity Waves," TM-118, U.S. Army, Corps of Engineers,
Beach Erosion Board, Washington, D.C., Aug. 1959.
BRETSCHNEIDER, C.L., "A One Dimensional Gravity Wave Spectriim," Ooean

Wave Spectra, Prentice-Hall, New York, 1963.
BRETSCHNEIDER, C.L., "Investigation of the Statistics of Wave Heights,"

Journal of the Waterways and Harbors Division, ASCE, Vol. 90, No. WWl,
1964, pp. 153-166.
BRETSCHNEIDER, C.L., "Generation of Waves by Wind; State of the Art,"

Report SN-134-6, National Engineering Science Co., Washington, D.C.,
1965.
BRETSCHNEIDER, C.L., Department of Ocean Engineering, University of

Hawaii, Honolulu, Hawaii, Consultations, 1970-71.
BRETSCHNEIDER, C.L., "A Non- Dimensional Stationary Hurricane Wave Model,"

Offshore Technology Conference, Preprint No. OTC 1517, Houston, Tex.,
May, 1972.
BRETSCHNEIDER, C.L., and COLLINS, J. I., "Prediction of Hurricane Surge; An

Investigation for Corpus Christi, Texas and Vicinity," Technical Report
No. SN-120, National Engineering Science Co., Washington, D.C., 1963.
(for U.S. Army Engineering District, Galveston, Tex.)
BRETSCHNEIDER, C.L., and REID, R.O., "Change in Wave Height Due to Bottom
Friction, Percolation and Refraction," S4th Annual Meeting of American
Geophysical Union, 1953.
BUNTING, D.C., "Evaluating Forecasts of Ocean-Wave Spectra," Journal of

Geophysical Research, Vol. 75, No. 21, 1970, pp. 4131-4143.
BUNTING, D.C., and MOSKOWITZ, L.I., "An Evaluation of a Computerized

Numerical Wave Prediction Model for the North Atlantic Ocean,"
Technical Report No. TR 209, U.S. Naval Oceanographic Office,
CHRYSTAL, G., "Some Results in the Mathematical Theory of Seiches," Pro-

ceedings of the Royal Society of Edinburgh, Session 1903-04, Robert
Grant and Son, London, Vol. XXV, Part 4, 1904, pp. 328-337.
3-146
CHRYSTAL, G., "On the Hydrodynamical Theory of Seiches, with a Biblio-
graphical Sketch," Transactions of the Royal Society of Edinburgh,
Robert Grant and Son, London, Vol. XLI, Part 3, No. 25, 1905.
COLE, A.L., "Wave Hindcasts vs Recorded Waves," Supplement No. 1 to Final

Report 06768, Department of Meteorology and Oceanography, University
of Michigan, Ann Arbor, Mich., 1967.
COLLINS, J.I., and VIEW4AN, M.J., "A Simplified Empirical Model for Hurri-
cane Wind Fields," Offshore Technology Conference, Vol. 1, Preprints,
Houston, Tex., Apr. 1971.
CONNER, W.C., KRAFT, R.H., and HARRIS, D.L., "Empirical Methods of Fore-
casting the Maximum Storm Tide Due to Hurricanes and Other Tropical
Storms," Monthly Weather Review, Vol. 85, No. 4, Apr. 1957, pp. 113-116.
COX, D.C., ed., Proceedings of the Tsunami Meetings Associated with the
10th Paoifia Science Congress, Honolulu, Hawaii, 1961.
DARWIN, G.H., The Tides and Kindred Phenomena in the Solar System,
W.H. Freeman, San Francisco, 1962, 378 pp.
DAWSON, W. BELL, "The Tides and Tidal Streams," Department of Naval

Services, Ottawa, 1920.
DEFANT. A., "Gezeitlichen Probleme des Meeres in Landnahe," Probleme

der Kosmischen Physik, VI, Hamburg, 1925.
DEFANT, A., Ebbe und Flut des Meeres, der Atmosphare, und der Erdfeste,
Springer, Berlin-Gottingen-Heidelberg, 1953.
DEFANT, A., Physical Oceanography, Vol. 2, Pergamon Press, New York,

1961, 598 pp.
DOODSON, A.T., and WARBURG, H.D., Admiralty Manual of Tides, Her Majesty's
Stationery Office, London, 1941, 270 pp.
DORRESTEIN, R., "Wave Setup on a Beach," Proceedings, Second Technical

Conference on Hurricanes, National Hurricane Research Report No. 50,
1962, pp. 230-241.
DRAPER, L., "The Analysis and Presentation of Wave Data - A Plea for Uni-
formity," Proceedings of the 10th Conference on Coastal Engineering,
ASCE, New York, 1967.
DUNN, G.E., and MILLER, B.I., Atlantic Hurricanes y Rev. ed. , Louisiana
State University Press, New Orleans, 1964.
EWING, J. A., "The Generation and Propagation of Sea Waves," Dynamic Waves
in Civil Engineering, Wiley, New York, 1971, pp. 43-56.
3-147
EWING, M., PRESS, F., and DONN, W.L., "An Explanation of the Lake Michigan
Wave of 26 June, 1954," Science, Vol. 120, 1954, pp. 684-686.
FAIRCHILD, JoC, "Model Study of Wave Setup Induced by Hurricane Waves at

Narragansett Pier, Rhode Island," The Bulletin of the Beach Erosion
Board, Vol. 12, July 1958, pp. 9-20.
FORTAK, H., "Concerning the General Vertically Averaged Hydrodynamic Equa-

tions with Respect to Basic Storm Surge Equations," National Hurri-
cane Research Project, Report No. 51, U.S. Weather Bureau, Washington,
D.C. 1962„
FRANCIS, G.W., "Weather Routing Procedures in the United States," The

Marine Observer, Vol. XLI , No. 232, Apr. 1971.
FREEMAN, J.C, JR., BAER, L., and JUNG, C.H., "The Bathystrophic Storm
Tide," Journal of Marine Research, Vol. 16, No. 1, 1957.
GALVIN, C.J., JR., and EAGLESON, P.S., "Experimental Study of Longshore

Currents on a Plane Beach," TM-10, U.S. Army, Corps of Engineers,
Coastal Engineering Research Center, Washington, D.C, Jan. 1965.
GOODKNIGHT, R.C, and RUSSELL, T.L., "Investigation of the Statistics of

Wave Heights," Journal of the Waterways and Harbors Division, ASCE,
Vol. 89, No. WW2, Proc. Paper 3524, May 1963, pp. 29-54.
GRAHAM, H.E., and NUNN, D.E., "Meteorological Considerations Pertinent to

Standard Project Hurricane, Atlantic and Gulf Coasts of the United
States," National Hurricane Research Project, Report No. 33, U.S.
Weather Bureau, Washington, D.C. , 1959.
GROVES, G.W., McGraw-Hill Encyclopedia of Science and Technology, McGraw-

Hill, New York, 1960, pp. 632-640.
GUTHRIE, R.C, "A Study of Ocean Wave Spectra for Relation to Ships Bending
Stresses," Informal Report No. IR 71-12, U.S. Naval Oceanographic Office,
HARRIS, R.A., "Currents, Shallow-Water Tides, Meteorological Tides, and

Miscellaneous Matters" Manual of Tides, Part V, App. 6, Superintendent
of the Coast and Geodetic Survey, Washington, D.C, 1907.
HARRIS, D.L. "Wind Tide and Seiches in the Great Lakes," Proceedings of
the Fourth Conference on Coastal Engineering, ASCE, 1953, pp. 25-51.
HARRIS, D.Lc, "The Effect of a Moving Pressure Disturbance on the Water

Level in a Lake," Meteorological Monographs, Vol. 2, No. 10, "Inter-
action of Sea § Atmosphere: a Group of Contributions," 1957, pp. 46-57.
HARRIS, D.L., "Meteorological Aspects of Storm Surge Generation," Journal

of the Hydraulics Division, ASCE, No. HY7, Paper 1859, 1958, 25 pp.
3-148
HARRIS, D.L., "An Interim Hurricane Storm Surge Forecasting Guide,"
National Hurricane Research Project, Report No. 32, U.S. Weather
Bureau, Washington, D.C. , 1959.
HARRIS, D.L., "Wave Patterns in Tropical Cyclones," Mariners Weather

Log, Vol. 6, No. 5, Sept., 1962, pp. 156-160.
HARRIS, D.L., "Coastal Flooding by the Storm of March 5-7, 1962," Pre-
sented at the Amerioan Meteorological Society Annual Meeting, Unpub-
lished Manuscript, U.S. Weather Bureau, 1963a.
HARRIS, D.L., "Characteristics of the Hurricane Storm Surge," Technical

Paper No. 48, U.S. Dept. of Commerce, Washington, D.C, 1963b.
HARRIS, D.L., "A Critical Survey of the Storm Surge Protection Problem,"
The 11th Symposium of Tsunami and Storm Surges, 1967, pp. 47-65.
HARRIS, D.L., "The Analysis of Wave Records," Proceedings of the 12th

Coastal Engineering Conference, ASCE, Washington, D.C, 1970.
HARRIS, D.L., "Characteristics of Wave Records in the Coastal Zone,"

Unpublished Manuscript, U.S. Army, Corps of Engineers, Coastal
Engineering Research Center, Washington, D. C. , 1971.
HARRIS, D.L., and ANGELO, A.T., "A Regression Model for the Prediction
of Storm Surges on Lake Erie," Proceedings, Fifth Conference on Great
Lakes Research, Toronto, 1962.
HASSE L., and WAGNER, V., "On the Relationship Between Geostrophic and
Surface Wind at Sea," Monthly Weather Review, Vol. 99, No. 4, Apr.
1971, pp. 255-260.
HAURWITZ, B., "The Slope of Lake Surfaces under Variable Wind Stresses,"
TM-25, U.S. Army, Corps of Engineers, Beach Erosion Board, Washington,
D.C, Nov. 1951.
HELLSTROM, B., "Wind Effects on Lakes and Rivers," Nr. 158, Handlingar,
Ingeniors Vetenskaps Akadamien, Stockholm, 1941.
HIDAKA, K., "A Theory of Shelf Seiches, and Seiches in a Channel,"

Memoirs of the Imperial Marine Observatory , Kobe, Japan, 1934-35.
HIDY, CM., and PLATE, E.J., "Wind Action on Water Standing in a Labora-
tory Channel," Journal of Fluid Mechanics, Vol. 25, 1966, pp. 651-687.
HONDA, K., et al., "Secondary Undulations of Oceanic Tides," Journal,

College of Science, Imperial University, Tokyo, Japan, Vol. 24, 1908.
HUNT, I. A., "Design of Seawalls and Breakwaters," Proceedings, American

Society of Civil Engineers, ASCE, Waterways and Harbors Division,
Vol. 85, No, WW3, Sept. 1959.
3-149
HWANG, L., and DIVOKY, D. , "Breaking Wave Setup and Decay on Gentle
Slopes," Prooeedings of the 12th Coastal Engineering Conferenae^
ASCE, Vol. 1, 1970.
IJIMA, T., and TANG, F.L.W., "Nximerical Calculation of Wind Waves in

Shallow Water," Frooeedings of 10th Conferenoe on Coastal Engineering
ASCE, Tokyo, Vol. 2, 1966, pp. 38-45.
INOUE, T. , "On the Growth of the Spectrum of a Wind Generated Sea Accord-
ing to a Modified Miles-Phillips Mechanism," Geophysical Sciences Lab-
oratory Report No. TR-66-6, Department of Meteorology and Oceanography,
New York University, N.Y. , 1966.
INOUE, T., "On the Growth of the Spectrum of a Wind Generated Sea Accord-
ing to a Modified Miles-Phillips Mechanism and its Application to Wave
Forecasting," Geophysical Sciences Laboratory Report No. TR-67-5,
Department of Meteorology and Oceanography, New York University,
N.Y., 1967.
IPPEN, A.T., "Estuary and Coastline Hydrodynamics," Engineering Sooieties

Monographs^ McGraw-Hill, New York, 1966.
IRISH, S.M., "The Prediction of Surges in the Southern Basin of Lake

Michigan," Part II," A Case Study of the Surge of August 3, 1960,"
Monthly Weather Review ^ Vol. 93, 1965, pp. 282-291.
IRISH, S.M., and PLATZMAN, G.W., "An Investigation of the Meteorological

Conditions Associated with Extreme Wind Tides on Lake Erie," Monthly
Weather Review, Vol. 90, 1962, pp. 39-47.
ISAACS, J.D., and SAVILLE, T. JR., "A Comparison Between Recorded and
Forecast Waves on the Pacific Coast," Annals of the New York Academy
of Sciences , Vol. 51, Art. 3, "Ocean Surface Waves," 1949, pp.
502-510.
JACOBS, S.L., "Wave Hindcasts vs. Recorded Waves," Final Report 06768,
Department of Meterology and Oceanography, University of Michigan,
Ann Arbor, Mich., 1965.
JELESNIANSKI, C.P., "The July 6, 1954 Water Wave on Lake Michigan

Generated by a Pressure Jump Passage," Unpublished Master of
Science Thesis, 1958.
JELESNIANSKI, C.P. "Numerical Computations of Storm Surges Without Bottom

Stress," Monthly Weather Review, Vol. 94, No. 6, 1966, pp. 379-394.
JELESNIANSKI, C.P., "Numerical Computations of Storm Surges with Bottom

Stress," Monthly Weather Review, Vol. 95, No. 11, 1967, pp. 740-756.
JELESNIANSKI, C.P., "Bottom Stress Time-History in Linearized Equations

of Motion of Storm Surges," Monthly Weather Review, Vol. 98, No. 6,
1970, pp. 462-478.
3-150
JELESNIANSKI, C.P., "SPLASH-Special Program to List Amplitudes of Surges
from Hurricanes," National Weather Service, Unpublished Manuscript,
National Oceanic and Atmospheric Administration, Rockville, Md. , 1972.
KAPLAN, K., "Analysis of Moving Fetches for Wave Forecasting," TM-35,

Mar. 1953.
KAPLAN, K., and SAVILLE, T. JR., "Comparison of Hindcast and Observed

Waves Along the Northern New Jersey Coast for the Storm of Nov. 6-7,
1953," The Bulletin of the Beach Erosion Boards Vol. 8, No. 3, July
1954, pp. 13-17.
KEULEGAN, G.H. , "Wind Tides in Small Closed Channels," Journal of

Researdhy National Bureau of Standards, RP2207, Vol. 46, No. 5, 1951.
KINSMAN, B., Wind Waves, Prentice-Hall, Englewood Cliffs, N.J,, 1965.
KRECKER, F.H., "Periodic Oscillations in Lake Erie," Contribution No. 1,

Franz Theodore Stone Laboratory, Ohio State University Press, Columbus,
Ohio, 1928.
LAPLACE, P.S., "Recherches sur plusieurs points du systeme du monde,"

Memoirs^ Aaademy of Royal Soienoes, Vol. 88, 1775, pp. 75-185.
LEENDERTSE, J.J., "Aspects of a Computational Model for Long-Period Water

Wave Propagation," Memorandum RM-5294-PR, prepared for U.S. Air Force,
Project Rand, 1967.
LONGUET-HIGGINS, M.S., and STEWART, R.W. , "Changes in the Form of Short-

Gravity Waves on Long Waves and Tidal Currents," Journal of Fluid
MechaniaSy Vol. 8, No. 565, 1960.
LONGUET-HIGGINS, M.S., and STEWART, R.W., "Radiation Stress and Mass

Transport in Gravity Waves, with Application to Surf Beat," Journal
of Fluid Meahaniasj Vol. 13, 1962, pp. 481-504.
LONGUET-HIGGINS, M.S., and STEWART, R.W., "A Note on Wave Setup," Journal
of Marine Researoh, Vol. 21(1), 1963, pp. 4-10.
LONGUET-HIGGINS, M.S., and STEWART, R.W., "Radiation Stress in Water

Waves; A Physical Discussion with Application," Deep-Sea Research,
Vol. 11, 1964, pp. 529-562.
MARINOS, G., and WOODWARD, J.W., "Estimation of Hurricane Surge Hydro-

graphs," Journal of the Waterways and Harbors Division, ASCE, Vol.
94, WW2, Proc. Paper 5945, 1968, pp. 189-216.
MARMER, H.A., The Tide, D. Appleton, New York, 1926, 2 82 pp.
3-151
MASCH, F.D., et al , "A Numerical Model for the Simulation of Tidal Hydro-
dynamics in Shallow Irregular Estuaries," Technical Report HYD 12,6901
Hydraulic Engineering Laboratory, University of Texas, Austin, Tex.,
1969.
MILES, J.W., "On the Generation of Surface Waves by Shear Flows," Journal
of Fluid Meohaniosj Vol. 3, 1957, pp. 185-204.
MIYAZAKI, M., "A Numerical Computation of the Storm Surge of Hurricane
Carla 1961 in the Gulf of Mexico," Technical Report No. 10, Depart-
ment of Geophysical Sciences, University of Chicago, Chicago, 111.,
1963.
MORTIMER, C.H., "Spectra of Long Surface Waves and Tides in Lake Michigan
and at Green Bay, Wisconsin," Proceedings, Eight Conference on Great
Lakes Eesearohj 1965.
MOSKOWITZ, L., "Estimates of the Power Spectrums for Fully Developed Seas
for Wind Speeds of 20 to 40 Knots," Journal of Geophysical Research,
Vol. 69, No. 24, 1964, pp. 5161-5179.
MUNGALL, J.C.H. , and MATTHEWS, J.B., "A Variable - Boundary Numerical

Tidal Model," Report No. R70-4, Office of Naval Research, Arlington,
Va. , 1970.
MUNK, W.H., "Proposed Uniform Procedure for Observing Waves and Interpret-
ing Instrument Records," Wave Project, Scripps Institute of Oceanogra-
phy, LaJolla, Calif., 1944.
MYERS, V.A., "Characteristics of United States Hurricanes Pertinent to

Levee Design for Lake Okeechobee, Florida," Hydrometeorological Report
No. 32, U.S. Weather Bureau, Washington, D.C., 1954.
PATTERSON, M.M., "Hindcasting Hurricane Waves in the Gulf of Mexico,"

Offshore Technology Conference, Houston, Tex., Preprint, Vol. 1, 1971.
PETERSON, K.R., and GOODYEAR, H.V., "Criteria for a Standard Project North-
easter for New England North of Cape Cod," National Hurricane Research
Project, Report No. 68, U.S. Weather Bureau, Washington, D.C., 1964.
PHILLIPS, O.M., "On the Generation of Waves by Turbulent Wind," Journal

of Fluid Mechanics y Vol. 2, 1957, pp. 417-445.
PHILLIPS, O.M., "The Dynamics of the Upper Ocean," Cambridge University

Press, Cambridge, 1966.
PIERSON, W.J., JR., and MOSKOWITZ, L,, "A Proposed Spectral Form for
Fully Developed Wind Seas Based on the Similarity Theory of S.A.
Kitaigorodskii," Journal of Geophysical Research, Vol. 69, No. 24,
1964, pp. 5181-5190.
PIERSON, W.J., JR., NEUMANN, G., and JAMES, R.W., "Practical Methods for
Observing and Forecasting Ocean Waves by Means of Wave Spectra and
Statistics," Publication No. 603, U.S. Navy Hydrographic Office,
3-152
PLATZMAN, G.W., "A Numerical Computation of the Surge of 26 June 1954 on
Lake Michigan," Geophysics j Vol. 6, 1958.
PLATZMAN, G.W., "The Dynamic Prediction of Wind Tides on Lake Erie,"

Meteopologiadl Monographs^ Vol. 4, No. 26, 1963, 44 pp.
PLATZMAN, G.W., "The Prediction of Surges in the Southern Basin of Lake

Michigan," Part I, "The Dynamical Basis for Prediction," Monthly
Weather Review y Vol. 93, 1965, pp. 275-281.
PLATZMAN, G.W., "A Procedure for Operational Prediction of Wind Setup

on Lake Erie," Technical Report to Environmental Science Services
Administration, U.S. Weather Bureau, 1967.
PLATZMAN, G.W., "Two-Dimensional Free Oscillations in National Basins,"

Journal of Physical Oceanography ^ Vol. 2, 1972, pp. 117-138.
PLATZMAN, G.W. , and RAO, D.B., "Spectra of Lake Erie Water Levels,"
Journal of Geophysical Research^ Vol. 69, Oct. 1963, pp. 2525-2535.
PLATZMAN, G.W., and RAO, D.B., "The Free Oscillations of Lake Erie,"
Studies on Oceanography ^ University of Washington Press, Seattle,
Wash., 1965, pp. 359-382.
PORE, N.A., "Ocean Surface Waves Produced by Some Recent Hurricanes,"

Monthly Weather Review, Vol. 85, No. 12, Dec. 1957, pp. 385-392.
PORE, N.A., and CUMMINGS, R.A., "A Fortran Program for the Calculation of
Hourly Values of Astronomical Tide and Time and Height of High and Low
Water," Technical Memorandum WBTM TDL-6, U.S. Department of Commerce,
Environmental Science Services Administration, Washington, D.C., 1967.
PORE, N.A., and RICHARDSON, W.S., "Interim Report on Sea and Swell Fore-
casting," Technical Memorandum TDL-13, U.S. Weather Bureau, 1967.
PORE, N.A., and RICHARDSON, W.S. , "Second Interim Report on Sea and Swell
Forecasting," Technical Memorandum TDL-17, U.S. Weather Bureau, 1969.
PREISENDORFER, R.W., "Recent Tsunami Theory," NOAA-JTRE-60, HIG-71-15,

Hawaii Institute of Geophysics, University of Hawaii, Honolulu,
Hawaii 1971.
PROUDMAN, J., and DOODSON, A.T., "Tides in Oceans Bounded by Meridians,"

Philosophical Transactions of the Royal Society (London), Parts I and
II, Vol. 235, No. 753, 1936, pp. 273-342; Part III, Vol. 237, No. 779,
1938, pp. 311-373.
PUTNAM, J. A., "Loss of Wave Energy Due to Percolation in a Permeable Sea

Bottom," Transactions of the American Geophysical Union Vol. 30, No. 3,
1949, pp. 349-357.
3-153
PUTNAM, J. A., and JOHNSON, J.W., "The Dissipation of Wave Energy by
Bottom Friction," Transactions of the American Geophysical Union^
Vol. 30, No, 1, 1949, pp. 67-74.
REID, RoO., "Short Course on Storm Surge," Summer Lectures, Texas A^M
University, College Station, Tex., 1964.
REID, R.O., and BODINE, B.R., "Numerical Model for Storm Surges Galveston
Bay," Journal of the Waterways and Harbors Division^ ASCE, Vol. 94,
No. WWl, Proc. Paper 5805, 1968, pp. 33-57.
RICHARDS, T.L., DRAGERT, H., and MacINTYRE, D.R., "Influence of Atmos-

pheric Stability and Over-Water Fetch on Winds Over the Lower Great
Lakes,'' Monthly Weather Review^ Vol. 94, No. 1, 1966, pp. 448-453.
ROCKWELL, D.C., "Theoretical Free Oscillations of the Great Lakes," Pro-

aeedingSy Ninth Conference on Great Lakes Research^ 1966, pp. 352-368.
ROLL, H.U., and FISHER, F., "Eine Kritische Bemerkung zum Neumann-
Spektrum des Seeganges," Deut. Hydrograph. Z, , Vol. 9, No. 9, 1956.
SAVAGE, R.P., "Model Tests of Wave Runup for the Hurricane Protection
Project," The Bulletin of the Beach Erosion Boards Vol. 11, July 1957,
pp. 1-12.
SAVILLE, T., JR., "Wind Setup and Waves in Shallow Water," TM-27, U.S.
Army, Corps of Engineers, Beach Erosion Board, Washington, D.C.,
June 1952.,
SAVILLE, T., JR., "The Effect of Fetch Width on Wave Generation," TM-70,
Dec. 1954.
SAVILLE, T. JR., "Experimental Determination of Wave Setup," Proceedings,

Second Technical Conference on Hurricanes , National Hurricane Research
Project, Report No. 50, 1961, pp. 242-252.
SCHUREMAN, P., "Manual of Harmonic Analysis and Prediction of Tides,"

Special Publication No. 98, U.S. Department of Commerce, Coast and
Geodetic Survey, Washington, D.C. , 1941.
SENTER, P.K., "Design of Proposed Crescent City Harbor, California,

Tsunami Model," Report H-71-2, U.S. Army, Corps of Engineers,
Waterways Experiment Station, Vicksburg, Miss., 1971.
SHEMDIN, O.H., and HSU, E.T., "The Dynamics of Wind in the Vicinity of
Progressive Water Waves," Technical Report No. 66, Department of
Civil Engineering, Stanford University, Stanford, Calif., 1966.
SHIELDS, G.C., and BURDWELL, G.B., "Western Region Sea State and Surf
Forecaster's Manual," Technical Memorandum WR-51, U.S. Weather
Bureau, 1970.
3-154
SIBUL, 0., "Laboratory Study of Wind Tides in Shallow Water," TM-61, U.S.
Army, Corps of Engineers, Beach Erosion Board, Washington, D.C., Aug.
1955.
SILVESTER, R. , and VONGVISESSOMJAI , S., "Computation of Storm Waves and

Swell," The Institution of Civil Engineers, Vol. 48, 1971, pp. 259-283.
SIMPSON, R.B., and ANDERSON, D.V., "The Periods of the Longitudinal Sur-
face Seiche of Lake Ontario," Proaeedings , Seventh Confevenoe on Great
Lakes Research , 1964.
SVERDRUP, H.U., JOHNSON, M.W., and FLEMING, R.H., The Oceans; Their
Physios t Chemistry, and General Biology; Prentice-Hall, New York, 1942.
SVERDRUP, H.U., and MUNK, W.H., "Wind, Sea, and Swell: Theory of Rela-
tions for Forecasting," Publication No. 601, U.S. Navy Hydrographic
Office, Washington, D.C. , Mar. 1947.
SYMONS, T.M., and ZELTER, B.D., "The Tsunami of May 22, 1960 as Recorded
at Tide Stations, Preliminary Report," U.S. Dept. of Commerce, Coast
and Geodetic Survey, Washington, D.C, 1960.
TICKNER, E.G., "Effect of Bottom Roughness on Wind Tide in Shallow Water,"

1^1-95, U.S. Army, Corps of Engineers, Beach Erosion Board, Washington,
D.C, May, 1957.
TUCKER, M.J., "Simple Measurement of Wave Records," Proceedings of the

Conference on Wave Recording for Civil Engineers, National Institute
of Oceanography, 1961.
U.S. ARMY, CORPS OF ENGINEERS, "Shore Protection, Planning and Design,"

Technical Report No. 4, 3rd ed. , Coastal Engineering Research Center,
U.S. ARMY, CORPS OF ENGINEERS, "Waves and Wind Tides in Shallow Lakes
and Reservoirs, Summary Report, Project CW-167," South Atlantic-
Gulf Region, Engineer District, Jacksonville, Fla. , 1955.
U.S. ARMY, CORPS OF ENGINEERS, "Hilo Harbor, Hawaii: Report on Survey

for Tidal Wave Protection and Navigation," Hawaii Region, Engineer
District, Honolulu, Hawaii, 1960.
U.S. ARMY, CORPS OF ENGINEERS, "Waves in Inland Reservoirs, (Summary

Report on Civil Works Investigation Projects CW-164 § CW-165),"
TM-132, Beach Erosion Board, Washington, D.C, Nov. 1962.
U.S. DEPARTMENT OF COMMERCE, COAST AND GEODETIC SURVEY, "Annotated Bibli-

ography on Tsunamis," International Union of Geodesy and Geophysics,
Paris, 1964.
3-155
U.S. DEPARTMENT OF COMMERCE, NATIONAL OCEANIC AND ATMOSPHERIC ADMINI-
STRATION, Tide Tables 3 East Coast, North and South America, National
Ocean Survey, Rockville, Md. , 1973.
UoS. DEPARTMENT OF COMMERCE, NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRA-

TION, Tide Tables, West Coast, North and South America, National Ocean
Survey, Rockville, Md. , 1973.
U.S. FLEET WEATHER FACILITY, "Forecasting Ocean Waves and Surf," U.S.
Naval Air Station, Oceanographic Services Section, Operations Depart-
ment, San Diego, Calif., 1966.
U.S. NAVAL WEATHER SERVICE COMMAND, "Summary of Synoptic Meteorological

Observations, North American Coastal Marine Areas," Vols. 1-10, 1970.
U.S. WEATHER BUREAU, "Meteorological Characteristics of the Probable

Maximum Hurricane, Atlantic and Gulf Coasts of the United States,"
Hurricane Research Interim Report, HUR 7-97 and HUR 7-97A, U.S.
Weather Bureau, Washington, D.C. , 1968.
U.S. WEATHER BUREAU, "Preliminary Analysis of Surface Wind Field and Sea-
Level Pressures of Hurricane Camille (August 1969)," Hurricane Research
Report, HUR 7-113, U.S. Weather Bureau, Washington, D.C, 1969.
U.S. WEATHER BUREAU, "Revised Surface Wind Field (30 ft) of Hurricane
Camille in Gulf of Mexico (August 1969)," Hurricane Research Report,
HUR 7-113A, U.S. Weather Bureau, Washington, D.C, 1970.
U.S. WEATHER BUREAU, "Synoptic Code," WB-242, U.S. Weather Bureau,

VAN DORN, W.C, "Wind Stress on an Artificial Pond, Journal of Marine

Research, Vol. 12, 1953.
VERMA, A. P., and DEAN, R.G., "Numerical Modeling of Hydromechanics of Bay

Systems," Proceedings of the Conference on Civil Engineering in the
Oceans II, ASCE, 1969.
WELANDER, P., "Numerical Prediction of Storm Surges," Advances in Geo-

physics, Vol. 8, 1961, pp. 316-379.
WIEGEL, R.L., Oceanographical Engineering, Prentis-Hall , Englewood Cliffs,

N.J., 1964.
WILSON, B.W, "Graphical Approach to the Forecasting of Waves in Moving

Fetches," TM-73, U.S. Army, Corps of Engineers, Beach Erosion Board,
Washington, D.C, Apr. 1955.
WILSON, B.W., "Hurricane Wave Statistics for the Gulf of Mexico," TM-98,
U.S. Array, Corps of Engineers, Beach Erosion Board, Washington, D.C,
June, 1957.
3-156
WILSON, B.W., "Deep Water Wave Generations by Moving Wind Systems,"
Journal of the Waterways and Harbors Division^ ASCE, Vol. 87,
No. WW2, Proc. Paper 2821, May, 1961, pp. 113-141.
WILSON, B.W., "Propagation and Run-up of Tsunami Waves," National Engi-

neering Science Co., Washington, D.C. , 1964.
WILSON, B.W., "Design Sea and Wind Conditions for Offshore Structures,"
Proaeedings of Offshore Exploration Conferenoe^ Long Beach, Calif.,
1966, pp. 665-708.
WILSON, B.W., WEB, L.M., and HENDRICKSON, J. A., "The Nature of Tsunamis,
Their Generation and Dispersion in Water of Finite Depth," National
Engineering Science Co., Washington, D.C, 1962.
3-157
CHAPTER 4
LITTORAL
PROCESSES
'^
\
'./•»-
,^^
.'
NORTH CAROLINA COAST - 13 February 1973
CHAPTER 4
LITTORAL PROCESSES
4.1 INTRODUCTION
Littoral processes result from the interactions among winds, waves,

currents, tides, sediments, and other phenomena in the littoral zone. The
purpose of this chapter is to discuss those littoral processes which involve
sediment motion. Shores erode, accrete, or remain stable, depending on the
rates at which sediment is supplied to and removed from the shore. Excessive
erosion or accretion may endanger the structural integrity or functional
usefulness of a beach, or other coastal structures. Therefore an under-
standing of littoral processes is needed to predict erosion or accretion
rates. An aim of coastal engineering design is to maintain a stable shore-
line where sediment supplied to the shore balances that which is removed.
This chapter presents information needed for understanding the effects of
littoral processes on coastal engineering design.
4.11 DEFINITIONS
In describing littoral processes, it is necessary to use clearly

defined terms. Commonly used terms, such as "beach" and "shore," have
specific meanings in the study of littoral processes, as shown in the
Glossary. (See Appendix A.)
4.111 Beach Profile Profiles perpendicular to the shoreline have char-

.
acteristic features that reflect the action of littoral processes. (See

Figure 1-1, Chapter 1, and Figures A-1 and A-2 of the Glossary for spe-
cific examples.) At any given time, a profile may exhibit only a few
specific features, but usually a dune, berm and beach face can be identi-
fied.
Profiles across a beach adapt to imposed wave conditions in ways as

illustrated in Figure 4-1 by a series of profiles taken between February
1963 and November 1964 at Westhampton Beach, N.Y. The figure shows how
the berm built up gradually from February through August 1963, then was
cut back in November through January, and was rebuilt in March through
September 1964. This process is typical of a cyclical process of storm
erosion in winter and progradation with the lower, and often longer,
waves in summer.
4.112 Areal View . Figure 4-2 shows three generalized charts of different
U.S. coastal areas, all to the same scale. Figure 4-2a shows a rocky coast,
well-indented, where sand is restricted to local pocket beaches. Figure
4-2b shows a long straight coast with an uninterrupted sand beach. Figure
4-2c shows short barrier islands interrupted by inlets. These are some of
the different coastal configurations which reflect differences in littoral
processes and the local geology.
4.12 ENVIRONMENTAL FACTORS
4.121 Waves . Water waves are the principal cause of most shoreline
changes. Without wave action on a coast, most of the coastal engineering
4-1
Datum = MTL = Mean Tide Level
Initial Survey 11 October 1962
- MTL-Beach
Intersect
(Surveys did not always reach MTL)
20
a>
a>
o 10
-300-200 -100 100 200 300

Distance from Mean Water Line of Initial Survey, feet
Figure 4-1. Typical Profile Changes with Time, Westhampton

Beach, New York
4-2
Rhode Island
(See USCaCS Chart I2IQ).
OCEAN
a. Rocky Coast with Limited Beaches
Northeastern Florida
(See use 8 GS Chart 1246)
BAY
OCEAN
b. Straight Barrier Island Shoreline
Southern New Jersey

(See use acs Chart 1217)
BAY
c, Short Barrier Island Shoreline
Figure 4-2. Tiiree Types of Shoreline
4-3
problems involving littoral processes would not occur. A knowledge of
incident waves, or of surf, is essential for coastal engineering planning,
design, and construction.
Three important aspects of a study of waves on beaches are: the

theoretical description of wave motion; the climatological data for waves
as they occur on a given segment of coast; and the description of how
waves interact with the shore to move sand.
The theoretical approach can provide a useful description of water

motion caused by waves when the limiting assumptions of the theory are
satisfied. Surprisingly, the small -amplitude theory (Section 2.23) and
aspects of solitary wave theory (Section 2.27) have proved useful beyond
the limits assumed in their derivations.
The theoretical description of water-wave motion provides estimates

of water motion, longshore force, and energy flux due to waves. These
estimates are useful in understanding the effect of waves on sediment
transport, but currently (1973) the prediction of wave-induced sediment
motion for engineering purposes relies heavily on empirical coefficients
and judgement rather than on theory.
Statistical distributions of wave characteristics along a given shore-

line provide a basis for describing the wave climate of a coastal segment.
Important wave characteristics affecting sediment transport near the beach
are height, period, and direction of breaking waves. Breaker height is
significant in determining the quantity of sand in motion; breaker direc-
tion is a major factor in determining longshore transport direction and
rate. Waves affect sediment motion in the littoral zone in two ways:
they initiate sediment movement, and they drive current systems that trans-
port the sediment once motion is initiated.
4.122 Currents. Water waves induce an orbital motion in the fluid beneath
them. (See Section 2.23.) These are not closed orbits, and the fluid
experiences a slight wave-induced drift, or mass transport. Magnitude and
direction of mass transport are functions of elevation above bottom and
wave parameters (Equation 2-55), and are also influenced by wind and tem-
perature gradients. The action of mass transport, extended over a long
period, can be important in carrying sediment onshore or offshore, par-
ticularly seaward of the breaker position.
As waves approach breaking, wave-induced bottom motion in the water

becomes more intense, and its effect on sediment becomes more pronounced.
Breaking waves create intense local currents (turbulence) that move sedi-
ment. As waves cross the surf zone after breaking, the accompanying fluid
motion is mostly uniform horizontal motion, except during the brief pass-
age of the breaker front where significant turbulence occurs. Since wave
crests at breaking are usually at a slight angle to the shoreline, there
is usually a longshore component of momentum in the fluid composing the
breaking waves. This longshore component of momentum entering the surf
4-4
zone is the principal cause of longshore currents - currents that flow
parallel to the shoreline within the surf zone. These longshore currents
are largely responsible for the longshore sediment transport.
There is some mean exchange between the water flowing in the surf
zone and the water seaward of the breaker zone. The most easily seen of
these exchange mechanisms are the rip currents (Shepard and Inman, 1950),
which are concentrated jets of water flowing seaward through the breaker
zone.
4.123 Tides and Surges . In addition to wave-induced currents, there are

other currents affecting the shore that are caused by tides and storm
surges. Tide-induced currents can be impressed upon the prevailing wave-
induced circulations, especially near entrances to bays and lagoons and
in regions of large tidal range. (Notices to Mariners and the Coastal
Pilot often carry this information.) Tidal currents are particularly
important in transporting sand in shoals and sand waves around entrances
to bays and estuaries.
Currents induced by storm surges (Murray, 1970) are less well known
because of the difficulty in measuring them, but their effects are un-
doubtedly significant.
The change in water level by tides and surges is a significant factor

in sediment transport, since, with a higher water level, waves can then
attack a greater range of elevations on the beach profile. (See Figure
1-7.) The appropriate theory for predicting storm surge levels is dis-
cussed in Section 3.8.
4.124 Winds . Winds act directly on beaches by blowing sand off the
beaches (deflation) and by depositing sand in dunes. (Savage and Wood-
house, 1968.) Deflation usually removes the finer material, leaving
behind coarser sediment and shell fragments. Sand blown seaward from
the beach usually falls in the surf zone; thus it is not lost, but is
introduced into the littoral transport system. Sand blown landward from
the beach may form dunes, add to existing dunes, or be deposited in
lagoons behind barrier islands.
For dunes to form, a significant quantity of sand must be available

for transport by wind, as must features that act to trap the moving sand.
Topographic irregularities, the dunes themselves, and vegetation are the
principal features that trap sand.
The most important dunes in littoral processes are foredunes - the

line of dunes immediately landward of the beach. They usually form be-
cause beachgrasses growing just landward of the beach will trap sand blown
landward off the beach. Foredunes act as a barrier to prevent waves and
high water from moving inland, and provide a reservoir of sand to replen-
ish the nearshore regime during severe shore erosion.
The effect of winds in producing currents on the water surface is

well documented, both in the laboratory and in the field. (Bretschneider,
4-5
1967; Keulegan, 1951; and van Dom, 1953.) These surface currents drift
in the direction of the wind at a speed equal to 2 to 3 percent of the
wind speed. In hurricanes, winds generate surface currents of 2 to 8
feet per second. Such wind-induced surface currents toward the shore
cause significant bottom return flows which may transport sediment sea-
ward; similarly, strong offshore winds can result in an offshore surface
current, and an onshore bottom current which can aid in transporting
sediment landward.
4.125 Geologic Factors . The geology of a coastal region affects the

supply of sediment on the beaches and the total coastal morphology. Thus,
geology determines the initial conditions for littoral processes, but geo-
logic factors are not usually active processes affecting coastal engineer-
ing.
One exception is the rate of change of sea level with respect to

land which may be great enough to influence design, and should be exam-
ined if project life is 50 years or more. On U.S. coasts, typical rates
of sea level rise average about 1 to 2 millimeters per year, but changes
range from -13 to +9 millimeters per year. (Hicks, 1972.) (Plus means
a rise in sea level with respect to local land level.)
4.126 Other Factors . Other principal factors affecting littoral processes

are the works of man and activities of organisms native to the particular
littoral zone. In engineering design, the effects on littoral processes
of construction activities, the resulting structures, and structure main-
tenance must be considered. This consideration is particularly necessary
for a project that may alter the sand budget of the area, such as jetty
or groin construction. In addition biological activity may be important
in producing carbonate sands, in reef development, or (through vegetation)
in trapping sand on dunes.
4.13 CHANGES IN THE LITTORAL ZONE
Because most of the wave energy is dissipated in the littoral zone,

this zone is where beach changes are most rapid. These changes may be
short-term due to seasonal changes in wave conditions and to occurrence
of intermittent storms separated by intervals of low waves, or long-term
due to an overall imbalance between the added and eroded sand. Short-term
changes are apparent in the temporary redistribution of sand across the
profile (Fig. 4-1); long-term changes are apparent in the more rtearly per-
manent shift of the shoreline. (See Figures 4-3, 4-4, and 4-5.)
Maximum seasonal or storm- induced changes across a profile, such as

those shown in Figure 4-1, are typically on the order of a few feet verti-
cally and from 10 to 100 feet horizontally. (See Table 4-1.) Only during
extreme storms, or where the available sand supply is restricted, do un-
usual changes occur over a short period.
Typical seasonal changes on southern California beaches are shown in

Table 4-1. (Shepard, 1950.) These data show greater changes on the beach
4-6
m
u
1-5
s
ID
2
e
o
t->
+->
o
•H
(U
2
C
o
•H
in
o
<L>
to
I
•H
4-7
m
U
(U
s
z
c
0)
>
o
ca
o
H
tn
o
U
C
o
•H
P

o
CO
-s
w
3
•H
4-9
Table 4-1 . Seasonal
than are typical of Atlantic coast beaches. (Urban and Galvin, 1969;
Zeigler and Tuttle, 1961.) Available data indicate that the greatest
changes on the profile are in the position of the beach face and of the
longshore bar - two relatively mobile elements of the profile. Beaches
change in plan view as well. Figure 4-6 shows the change in shoreline
position at seven east coast localities as a function of time between
autumn 1962 and spring 1967.
Comparison of beach profiles before and after storms suggests ero-

sion of the beach above MSL from 10,000 to 50,000 cubic yards per mile
of shoreline during storms expected to recur about once a year. (DeWall,
et al., 1971; and Shuyskiy, 1970.) While impressive in aggregate, such
sediment transport is minor compared to longshore transport of sediment.
Longshore transport rates may be greater than 1 million cubic yards per
year.
The long-term changes shown in Figures 4-3, 4-4, and 4-5 illustrate
shorelines of erosion, accretion, and stability. Long-term erosion or
accretion rates are rarely more than a few feet per year in horizontal
motion of the shoreline, except in localities particularly exposed to
erosion, such as near inlets or capes. Figure 4-5 indicates that shore-
lines can be stable for a long time. It should be noted that the erod-
ing, accreting, and stable beaches shown in Figures 4-3, 4-4, and 4-5
are on the same barrier island within a few miles of each other.
Net longshore transport rates along ocean beaches range from near
zero to 1 million cubic yards per year, but are typically 100,000 to
500,000 cubic yards per year. Such quantities, if removed from a 10- to
20-mile stretch of beach year after year, would result in severe erosion
problems. The fact that many beaches have high rates of longshore trans-
port without unusually severe erosion suggests that an equilibrium condi-
tion exists on these beaches, in which the material eroded is balanced by
the material supplied; or in which seasonal reversals of littoral trans-
port replace material previously removed.
4.2 LITTORAL MATERIALS
Littoral materials are the solid materials (mainly sedimentary) in

the littoral zone on which the waves and currents act.
4.21 CLASSIFICATION
The characteristics of the littoral materials are a primary input

to any coastal engineering design. Median grain size is the most fre-
quently used design characteristic.
4.211 Size and Size Parameters . Littoral materials are classified by

grain size into clay, silt, sand, gravel, cobble, and boulder. Several
size classifications exist, of which two, the Unified Soil Classification,
(based on the Casagrande Classification) and the Wentworth Classification,
are most commonly used in coastal engineering. (See Figure 4-7.) The
o
•D
C
O
I -r
o
o
(O
o
J=
CO
>
_l
o
a>
W
c
o
a>
c
o
Figure 4-6. Fluctuations in Location of Mean Sealevel

Shoreline on Seven East Coast Beaches
4-12
Unified Soil Classification is the principal classification used by engi-
neers. The Wentworth classification is the basis of a classification
widely used by geologists, but is becoming more widely used by engineers
designing beach fills.
For most shore protection design problems, typical littoral mate-

rials are sands with sizes between 0.1 and 1.0 millimeters or, in phi
units, between 3.3 and phi. According to the Wentworth classification,
sand size is in the range between 0.0625 cind 2.0 millimeters (4 and -1
phi); according to the Unified Soil Classification, it is between 0.074
and 4.76 millimeters (3.75 and -2.25 phi). Within these sand-size ranges,
engineers commonly distinguish size classes by median grain size measured
in millimeters, based on sieve analyses.
Samples of typical beach sediment usually have a few relatively

large particles covering a wide range of diameters, and many small par-
ticles within a small range of diameters. Thus, to distinguish one
sample from another, it is necessary to consider the small differences
(in absolute magnitude) among the finer sizes more than the same differ-
ences among the larger sizes. For this reason, all sediment size classi-
fications exaggerate absolute differences in the finer sizes compared to
absolute differences in the coarser sizes.
As shown in Figure 4-7, limits of the size classes differ. The

Unified Soil Classification boundaries correspond to U.S. Standard Sieve
sizes. The Wentworth classification varies as powers of 2 millimeters;
that is, the size classes have limits, in millimeters, determined by the
relation 2 , where n is any positive or negative whole number, including
zero. For example, the limits on sand size in the Wentworth scale are
0.0625 millimeters and 2 millimeters, which correspond to 2 and 2 ^
"*
millimeters.
This property of having class limits defined in terms of whole number

powers of 2 millimeters led Krumbein (1936) to propose a phi unit scale
based on the definition:
Phi units (0) = — log (diameter in mm.) (4-1)
Phi unit scale is indicated by writing ^ or phi after the numerical

value. The phi unit scale is shown in Figure 4-7. Advantages of phi
units are:
(a) Limits of Wentworth size classes are whole numbers in phi

units. These phi limits are the negative value of the exponent, n, in
the relation 2 . For example, the sand size class ranges from +4 to -1,
in phi units.
(b) Sand size distributions typically are near lognormal, so that

a unit based on the logarithm of the size better emphasizes the small
significant differences between the finer particles in the distribution.
4-13
Wentworth Scale
(Size Description)
(c) The normal distribution is described by its mean and standard
deviation. Since the distribution of sand size is approximately lognormal,
then individual sand size distributions can be more easily described by
units based on the logarithm of the diameter rather than the absolute diam-
eter. Comparison with the theoretical lognormal distribution is also a
convenient way of characterizing and comparing the size distribution of
different samples.
Of these three advantages, only (a) is unique to the phi units. The
other two, (b) and (c) , would be valid for any unit based on the logarithm
of size.
Disadvantages of phi units are:
(a) Phi units increase as absolute size in millimeters decreases.
(b) Physical appreciation of the size involved is easier when the

units are millimeters rather than phi units.
(c) The median diameter can be easily obtained without phi units.
(d) Phi units are dimensionless, and are not usable in physically
related quantities where grain size must have units of length such as
grain-size, Reynolds number, or relative roughness.
Size distributions of samples of littoral materials vary widely.

Qualitatively, the size distribution of a sample may be characterized by
a diameter that is in some way typical of the sample, and by the way that
the sizes coarser and finer than the typical size are distributed. (Note
that size distributions are generally based on weight, rather than number
of particles.)
A size distribution is described qualitatively as well-sorted if all

particles have sizes that are close to the typical size. If all the par-
ticles have exactly the same size, then the sample is perfeatly sorted.
If the particle sizes are distributed evenly over a wide range of sizes,
then the sample is said to be well-graded. A well-graded sample is poorly
sorted; a well-sorted sample is poorly graded.
The median (Mj) and the mean (M) define typical sizes of a sample of
littoral materials. The median size, N^ in millimeters, is the most com-
mon measure of sand size in engineering reports. It may be defined as
M^ = d,„ (4-2)
where dso is the size in millimeters that divides the sample so that half
the sample, by weight, has particles coarser than the dso size. An equiv-
alent definition holds for the median of the phi size distribution, using
the symbol M^,), instead of M^f.
Several formulas have been proposed to compute an approximate mean

(M) from the cumulative size distribution of the sample. (Otto, 1939;
4-15
Inraan, 1952; Folk and Ward, 1957, McCammon, 1962.) These formulas are
averages of 2, 3, 5, or more symmetrically selected percentiles of the
phi frequency distribution, such as the formula of Folk and Ward.
M0 = (4-3)
^
where i|) is the particle size in phi units from the distribution curve
at the percentiles equivalent to the subscripts 16, 50 and 84 (Fig. 4-8);
4>^ is the size in phi units that is exceeded by x percent (by dry weight)
of the total sample. These definitions of percentile (after Griffiths, 1967,
p. 105) are known as graphic measures. A more complex method - the method
of moments - can yield more precise results when properly used.
To a good approximation, the median, Mj is interchangeable with the

mean, (M) , for most beach sediment. Since the median is easier to deter-
mine it is widely used in engineering studies. For example, in one CERC
study of 465 sand samples from three New Jersey beaches, the mean computed
by the method of moments averaged only 0.01 millimeter smaller than the
median for sands whose average median was 0.30 millimeter (1.74 phi).
(Ramsey and Galvin, 1971.)
The median and the mean describe the approximate center of the sedi-
ment size distribution. In the past, most coastal engineering projects
have used only this size information. However, for more detailed design
calculations of fill quantities required for beach restoration projects
(Sections 5.3 and 6.3), it is necessary to know more about the size dis-
tribution.
Since the actual size distributions are such that the log of the size
is approximately normally distributed, the approximate distribution can be
described (in phi units) by the two parameters that describe a normal dis-
tribution - the mean and the standard deviation. In addition to these
two parameters (mean and standard deviation), skewness and kurtosis
describe how far the actual size distribution of the sample departs from
this theoretical lognormal distribution.
Standard deviation is a measure of the degree to which the sample

spreads out around the mean and may be approximated using Inman's (1952)
definition, by
'84 ^16
oa.
'4>
= : ,
(4-4)
where iîqi^ is the sediment size, in phi units, that is finer than 84 per-
cent by weight, of the sample. If the sediment size in the sample actually
has a lognormal distribution, then a± is the standard deviation of the
4-16
99.9
99.8
99.5
o
oo
c
0)
o
k_
«
Q.
>
E
o
1.5 2.0 4.0

Diameter (phi)
I I I
J I I I I
1.5 1.0 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.15 0.1 0.08 0.06
Diameter (mm)
Figure 4-8. Example Size Distribution
4-17
sediment in phi units. For perfectly sorted sediment, 0^ = 0. For
typical well-sorted sediments, o, =» 0.5.
The degree by which the phi size distribution departs from symmetry
is measured by the skewness (Inman, 1952), as
^<t>
" ^d4> (4-5)
* a<l>
where M^ is the mean, I^cf) is the median, and a^ is the standard

deviation in phi units. For a perfectly symmetric distribution, the mean
equals the median, and the skewness is zero.
Extensive literature is available on the definition, use, and impli-

cation of a, a, and other measures of the size distribution. (Inman,
1952; Folk and Ward, 1957; McCammon, 1962; Folk, 1965, 1966; and
Griffiths, 1967), but despite a long history of investigation and a
considerable background of data, applications of size distribution infor-
mation "are still largely empirical, qualitative, and open to alternative
interpretations." (Griffiths, 1967, p. 104.)
Currently, median grain size is the most commonly reported sand size
characteristic, probably because there are only limited data to show the
usefulness of other size distribution parameters in coastal engineering
design. However, the standard deviation (Equation 4-4) must also be
given as a parameter for use in beach fill design. (See Section 5,3;
Krumbein and James, 1965; Vallianos, 1970; Berg and Duane, 1968.)
4.212 Composition . In addition to classification by size, littoral

material may be classified by composition. For shore protection pur-
poses, composition normally is not an important factor, since the domi-
nant littoral materials are quartz sands which are durable and chemically
inert for periods longer than typical project lifetimes. However, sedi-
ment composition is useful when the material departs from this expected
condition. Other than quartz, littoral material may be composed of car-
bonates (usually shell, coral, and algal material), organics (most often
peat), and clays and silts (marsh and tidal flat deposits).
4.213 Other Characteristics . In addition to size and composition, sedi-

ments have a number of other properties by which they may be classified.
Table 4-2 lists some density-related properties. Radioactive properties
of naturally occurring thorium minerals have been used as tracers in beach
sands. (Kamel, 1962; Kamel and Johnson, 1962, p. 324.) Other properties
more directly related to soil mechanics studies are found in soil mechanics
manuals.
4.22 SAND AND GRAVEL
By definition the word sand refers to a size class of material, but

sand also implies the particular composition, usually quartz (silica).
4-18
Table 4-2. Density of Littoral Materials
Specific Gravity (dimensionless)

Table 4-3. Minerals Occurring in Beach Sand
Common Dominant Constituents*

and marshes then outcrop along the shore, (e.g. Kraft, 1971.) Many barrier
islands along the Atlantic and Gulf coasts contain tidal and marsh deposits
at or near the surface of the littoral zone. The fine material is often
bound together by the roots of marsh plants to form a cohesive deposit
that may function for a time as beach protection.
4.24 CONSOLIDATED MATERIAL
Along some coasts, the principal littoral materials are consolidated

materials, such as rock, beach rock, and coral, rather than unconsolidated
sand. Such consolidated materials protect a coast and resist shoreline
changes.
4.241 Rock. Exposed rock along a shore indicates that the rate at which
sand is supplied to the coast is less than the potential rate of sand
transport by waves and currents. Reaction of a rocky shore to wave attack
is determined by the structure, degree of lithification, and ground water
characteristics of the exposed rock, and by the severity of the wave
climate. Protection of eroding cliffs is a complex problem involving
geology, rock mechanics, and coastal engineering. Two examples of these
problems are the protection of the cliffs at Newport, Rhode Island (U.S.
Army, Corps of Engineers, 1965) and at Gay Head, Martha's Vineyard,
Massachusetts. (U.S. Army, Corps of Engineers, 1970.)
Most rocky shorelines are remarkably stable, with individual rock

masses identified in photos taken 50 years apart. (Shepard and Grant,
1947.)
4.242 Beach Rock . A layer of friable to well-lithified rock often occurs

at or near the surface of beaches in tropical and subtropical climates.
This material consists of local beach sediment cemented with calcium car-
bonate, and it is commonly known as beach rock. Beach rock is important
to coastal engineers because it provides added protection to the coast,
greatly reducing the magnitude of beach changes (Tanner, 1960), and be-
cause beach rock may affect construction activities. (Gonzales, 1970.)
According to Bricker (1971), beach rock is formed when saline waters

evaporate in beach sjinds, depositing calcium carbonate from solution. The
present active formation of beach rock is limited to tropical coasts, such
as the Florida Keys, but rock resembling beach rock is common at shallow
depths along the east coast of Florida, on some Louisiana beaches, and re-
lated deposits have been reported as far north as the Fraser River Delta
in Canada. Comprehensive discussion of the subject is given in Bricker
(1971) and Russell (1970).
4.243 Organic Reefs . Organic reefs are wave-resistant structures reach-

ing to about mean sea level that have been formed by calcium carbonate
secreting organisms. The most common reef-building organisms are herma-
typic corals and coralline algae. Reef- forming corals are usually re-
stricted to areas having winter temperatures above about 18°C (Shepard,
1963, p. 351), but coralline algae have a wider range. On U.S.
4-21
coastlines, active coral reefs are restricted to southern Florida, Hawaii,
Virgin Islands and Puerto Rico. On Some of the Florida coast, reeflike
structures are produced by sabellariid worms. (Kirtley, 1971.) Organic
reefs stabilize the shoreline and sometimes affect navigation.
4.25 OCCURRENCE OF LITTORAL MATERIALS ON U.S. COASTS
Littoral materials on U.S. coasts vary from consolidated rock to

clays, but sand with median diameters between 0.1 and 1.0 millimeters
(3.3 and phi) is most abundant. General information on littoral mate-
rials is in the reports of the U.S. Army Corps of Engineers' National
Shoreline Study; information on certain specific geological studies is
available in Shepard and Wanless (1971); and information on specific
engineering projects is published in Congressional documents and is
available in reports of the Corps of Engineers.
4.251 Atlantic Coast . The New England coast is generally characterized

by rock headlands separating short beaches of sand or gravel. Exceptions
to this dominant condition are the sandy beaches in northeastern Massachu-
setts, and along Cape Cod, Martha's Vineyard, and Nantucket.
From the eastern tip of Long Island, New York, to the southern tip
of Florida, the littoral materials are characteristically sand with median
diameters in the range of 0.2 to 0.6 millimeter (2.3 to 0.7 phi). This
material is mainly quartz sand. In Florida, the percentage of calcium
carbonate in the sand tends to increase going south until, south of the
Palm Beach area, the sand becomes predominantly calcium carbonate. Size
distributions for the Atlantic coast, compiled from a number of sources,
are shown in Figure 4-9. (Bash, 1972.) Fine sediments and organic sedi-
ments are common minor constituents of the littoral materials on these
coasts, especially in South Carolina and Georgia. Beach rock and coquina
are common at shallow depths along the Atlantic coast of Florida.
4.252 Gulf Coast . The Gulf of Mexico coasts of Florida, Alabama, and
Mississippi are characterized by fine white sand beaches and by stretches
of swamp. The swampy stretches are mainly in Florida, extending from
Cape Sable to Cape Romano, and from Tarpon Springs to the Ochlockonee
River. (Shepard and Wanless, 1971, p. 163.)
The Louisiana coast is dominated by the influence of the Mississippi

River which has deposited large amounts of fine sediment around the delta
from which wave action has winnowed small quantities of sand. Tul" oC-q
has been deposited along barrier beaches offshore of a deeply indented
marshy coast. West of the delta is a 75-mile stretch of shelly sand
beaches and beach ridges.
The Texas coast is a continuation of the Louisiana coastal plain ex-

tending about 80 miles to Galveston Bay; from there a series of long, wide
barrier islands extends to the Mexican border. Littoral materials in this
area are dominantly fine sand, with median diameters between 0.1 and 0.2
millimeter (3.3 and 2.3 phi).
4-22
1.6 -0.68
1.4 -0.49
1.2 -0.26
0.0
0.8 0.32
0.6 0.64
0.4 .32
|l|i
0.2 irwi 2.32

'i"i
NEW JERSEY DELMARVA NORTH CAROLINA

E
0.8 0.32
« 0.6 0.64
E 0.4
o 1.32
0.2 !!\i,
2.32 =
•a

1
1.0
0.8
0.6
0.4 —
0.2
4.253 Pacific Coast . Sands on the southern California coast range in
size from 0.1 to 0.6 millimeter (3.3 to 0.7 phi). (Emery, 1960, p. 190.)
The northern California coast becomes increasingly rocky, and coarser
material becomes more abundant. The Oregon and Washington coasts include
considerable sand (Bascom, 1964), with many rock outcrops. Sand-sized
sediment is contributed by the Columbia River and other smaller rivers.
4.254 Alaska . Alaska has a long coastline (47,300 miles), and is corres-
pondingly variable in littoral materials. However, beaches are generally
narrow, steep, and coarse-grained; they commonly lie at the base of sea-
cliffs. (Sellman, et al., 1971, p. D-10.) Quartz sand is less common and
gravel more common here than on many other U.S. coasts.
4.255 Hawaii . Much of the Hawaiian islands is bounded by steep cliffs,

but there are extensive beaches. Littoral materials consist primarily of
bed rock, and white sand formed from calcium carbonate produced by marine
invertebrates. Dark colored basaltic and olivine sands are common where
river mouths reach the sea. (Shepard and Wanless, 1971, p. 497, U.S. Army,
Corps of Engineers, 1971.)
4.256 Great Lakes . The U.S. coasts of the Great Lakes vary from high
bluffs of clay, shale, and rock, through lower rocky shores and sandy
beaches, to low marshy clay flats. (U.S. Army Corps of Engineers, National
Shoreline Study, August 1971, North Central Division, p. 13.) The litto-
ral materials are quite variable. Specific features are discussed, for
example, by Bowman (1951); Hulsey (1962); Davis (1964-65); Bajorunas and
Duane (1967); Berg and Duane (1968); Saylor and Upchurch (1970); Hands
(1970); and Corps of Engineers (1953, 1971).
4.26 SAMPLING LITTORAL MATERIALS
Sampling programs are designed to provide information about littoral

materials on one or more of the following characteristics:
(a) typical grain size (usually median size),
(b) size distribution,
(c) composition of the littoral materials,
(d) variation of (a), (b) , and (c) , with horizontal and vertical
position on the site, and
(e) possible variation in (a), (b) , (c) , and (d) with time.
A sampling program will depend on the intended purpose of the samples,

the time and money available for sampling, and an inspection of the site
to be sampled. A brief inspection will often identify the principal vari-
ations in the sediment and suggest the best ways to sample these variations.
Sampling programs usually involve beach and nearshore sands and potential
borrow sources.
4-24
The extent of sampling depends on the importance of littoral materials
as related to the total engineering problem. The sampling program should
specify:
(a) horizontal location of sample,
(b) spacing between samples,
(c) volume of sample,
(d) vertical location and type of sampled volume (e.g. surface layer
or vertical core)
(e) technique for sampling,
(f) method of storing and documenting the sample.
Beaches typically show more variation across the profile than along
the shore, so sampling to determine variation in the littoral zone should
usually be along a line perpendicular to the shoreline.
For reconnaissance sampling, a sample from both the wetted beach face
and from the dunes is recommended. More extensive samples could be ob-
tained at constant spacings across the beach or at different locations on
the beach profile. Spacings between sampling lines are determined by the
variation visible along the beach or by statistical techniques.
Many beaches have subsurface layers of peat or other fine material.

If this material will affect the engineering problem, vertical holes or
borings should be made to obtain samples at depth.
Sample volume should be adequate for analysis. For sieve analysis,

about 50 grams are required; for settling tube analysis, smaller quanti-
ties will suffice, but at least 50 grams are needed if other studies are
required later. A quarter of a cup is more than adequate for most uses.
Sand often occurs in fine laminae on beaches. However, for engineer-

ing applications it is rarely necessary to sample individual laminae of
sand. It is easier and more representative to take an equi dimensional
sample that cuts across many laminae. Experience at CERC suggests that
any method of obtaining an adequate volume of sample covering a few inches
in depth usually gives satisfactory results. Cores should be taken where
pile foundations are planned.
The sample is only as good as the information identifying it. The

following minimum information should be recorded at the time of sampling:
locality, date and time, position on beach, remarks, and initials of col-
lector. This information must stay with the sample, which is best ensured
by fixing it to the sample container or placing it inside the container.
Unless precautions are taken, the sample label may deteriorate due to
moisture, abrasion, or other causes. Improved labels result by using
4-25
ballpoint ink on plastic strip (plastic orange flagging commonly used by-
surveyors). Some information may be preprinted by rubber stamp on the
plastic strip using indelible laundry ink. The advantage is that the
label can be stored in the bag with the wet sample without the label
deteriorating or the information washing or wearing off.
4.27 SIZE ANALYSES
Three common methods of analyzing a beach sediment for size are:

visual comparison with a standard, sieve analysis, and settling tube
analysis.
The mean size of a sand sample can be estimated qualitatively by

visually comparing the sample with sands of known sizes. Standards can
be easily prepared by sieving out selected diameters, or by selecting
samples whose sizes are already known. The standards may be kept in
labeled treinsparent vials, or glued on cards. If glued, care is neces-
sary to ensure that the particles retained by the glue are truly repre-
sentative of the standard.
Good, qualitative, visual estimates of mean size are possible with

little previous experience. With experience, such visual estimates
become semiquantitative. Visual comparison with a standard is a useful
tool in reconnaisance, and in obtaining interim results pending a more
complete laboratory size analysis.
4.271 Sieve Analysis Sieves are graduated in size of opening according

.
to the U.S. Standard series. These standard sieve openings vary by a fac-
tor of 1.19 from one opening to the next larger (by the fourth root of 2,
or quarter phi intervals), e.g., 0.25, 0.30, 0.35, 0.42, and 0.50 milli-
meters (2.00, 1.75, 1.50, 1.25, 1.00 phi). The range of sieve sizes used,
and the size interval between sieves selected can be varied as required.
Typical beach sand can be analyzed adequately using sieves with openings
ranging from 0.062 to 2.0 millimeters (4.0 to -1.0 phi), in size incre-
ments increasing by a factor of 1.41 (half-phi intervals).
Sediment is usually sieved dry. However, for field analysis or for

size analysis of sediment with a high content of fine material, it may be
useful to wet-sieve the sediment. Such wet-sieve analyses are described
by (e.g., Lee, Yancy, and Wilde, 1970, p. 4).
Size analysis by sieves is relatively slow, but provides a widely

accepted standard of reference.
4.272 Settling Tube . Spherical sedimentary particles settle through

water at a speed that increases as the particle weight increases. Since
most sand is approximately spherical quartz, or calcium carbonate with a
specific gravity near quartz, particle size is proportional to particle
weight. Thus fall velocity can be used to measure size, (e.g., Colby and
Christensen, 1956; Zeigler and Gill, 1959; and Gibbs, 1972.) Figure 4-31
shows fall velocity for quartz spheres as a function of temperature.
4-26
There are numerous types of settling tubes; the most common is the
visual accumulation tube (Colby and Christensen, 1956), of which there
are also several types. The type now used at CERC (the rapid sediment
analyzer or RSA) works in the following way:
A 3- to 6-gram sample of sand is dropped through a tube filled with

distilled water at constant temperature. A pressure sensor near the
bottom of the tube senses the added weight of the sediment supported by
the column of water above the sensor. As the sediment falls past the
sensor, the pressure decreases. The record of pressure versus time is
empirically calibrated to give size distribution based on fall velocity.
(Zeigler and Gill, 1959.)
The advantage of settling tube analysis is its speed. With modem

settling tubes, average time for size analyses of bulk lots can be about
one-fifth the time required by sieves.
It is often claimed that a settling tube also provides a physically

more realistic size analysis than a sieve, since the fall velocity takes
into account the hydrodynamic effects of shape and density. However,
this claim has not been documented, and may be questioned in view of the
limited knowledge concerning the fluid mechanics of a sand sample falling
in a settling tube - the lead particles encounter effectively laminar flow,
the trailing particles encounter turbulent flow, and all particles inter-
act with each other.
Because of lack of an accepted standard settling tube, rapidly chang-

ing technology, possible changes in tube calibration, and the uncertainty
about fluid mechanics in settling tubes, it is recommended that all set-
tling tubes be carefully calibrated by running a range of samples through
both the settling tube and ASTM standard sieves. After thorough initial
calibration, the calibration should be spot-checked periodically by running
replicate sand samples of known size distribution through the tube.
4.3 LITTORAL WAVE CONDITIONS
4.31 EFFECT OF WAVE CONDITIONS ON SEDIMENT TRANSPORT
Waves arriving at the shore are the primary cause of sediment trans-
port in the littoral zone. Higher waves break further offshore, widening
the surf zone and setting more sand in motion. Changes in wave period or
height result in moving sand onshore or offshore. The angle between the
crest of the breaking wave and the shoreline determines the direction of
the longshore component of water motion in the surf zone, and usually the
lon^chore transport direction. For these reasons, knowledge about the
wave climate - the combined distribution of height, period, and direction
through the seasons - is required for an adequate understanding of the
littoral processes of any specific area.
4-27
4.32 FACTORS DETERMINING LITTORAL WAVE CLIMATE
The wave climate at a shoreline depends on the offshore wave climate,

caused by prevailing winds and storms, and on the bottom topography that
modifies the waves as they travel shoreward.
4.321 Offshore Wave Climate . Wave climate is the distribution of wave

conditions averaged over the years. A wave condition is the particular
combination of wave heights, wave periods, and wave directions at a given
time. A specific wave condition offshore is the result of local winds
blowing at the time of the observation and the recent history of winds
in the more distant parts of the same water body. For local winds, wave
conditions offshore depend on the wind velocity, duration, and fetch.
For waves reaching an observation point from distant parts of the sea,
a decay factor is added which preferentially filters out the higher,
shorter period waves with increasing distances. (Chapter 3 discusses
wave generation and decay.)
4.322 Effect of Bottom Topography . As waves travel from deep water, they
change height and direction because of refraction, shoaling, bottom fric-
tion, and percolation. Laboratory experiments indicate that height and
apparent period are also changed by nonlinear deformation of the waves in
shallow water.
Refraction is the bending of wave crests due to the slowing down of

that part of the wave crest which is in shallower water. (See Section
2.32.) As a result, refraction tends to decrease the angle between the
wave crest and the bottom contour. Thus, for most coasts, refraction re-
duces the breaker angle and spreads the wave energy over a longer crest
length.
Shoaling is the change in wave height due to conservation of energy

flux. (See Section 2.32). As a wave moves into shallow water the wave
height first decreases slightly, and then increases continuously to the
breaker position, assuming friction and refraction effects are negligible.
Bottom friction is important in reducing wave height where waves

must travel long distances in shallow water. (Bretschneider, 1954.)
There has been only limited field study of nonlinear deformation in

shallow water waves (Byrne, 1969), but because such deformation is common
in laboratory experiments (Calvin, 1972), it is expected that such phe-
nomena are also common in the field. An effect of nonlinear deformation
is to split the incoming wave crest into two or more crests affecting
both the resulting wave height and the apparent period.
Offshore islands, shoals, and other variations in hydrography also

shelter parts of the shore. In general, bottom hydrography has the
greatest influence on waves traveling long distances in shallow water.
Because of the effects of bottom hydrography, nearshore waves generally
have different characteristics than they had in deep water offshore.
4-28
Such differences are often visible on aerial photos. Photos may-
show two or more distinct wave trains in the nearshore area, with the
wave train most apparent offshore decreasing in importance as the surf
zone is approached, (e.g., Harris, 1972.) The difference appears to be
caused by the effects of refraction and shoaling on waves of different
periods. Longer period waves, which may be only slightly visible off-
shore, may become the most prominent waves at breaking, because shoaling
increases their height relative to the shorter period waves. Thus, the
wave period measured from the dominant wave offshore may be less than the
wave period measured from the dominant wave entering the surf zone when
two wave trains of unequal period reach the shore at the same time.
4.323 Winds and Storms . The relation of a shoreline locality to the

seasonal distribution of winds and to storm tracks is a major factor
in determining the wave energy available for littoral transport. For
example, strong winter winds in the northeastern United States usually
are from the northwest, but because they blow from land to sea, they do
not produce large waves at the shore. These northwest winds often imme-
diately follow a northeaster - a low pressure system with strong north-
east winds that generate high waves offshore.
A storm near the coastline will influence wave climate with storm
surge and high seas; a storm offshore will influence wave climate only
by swell. The relation between the meteorological severity of a storm
and the resulting beach change is complicated. (See Section 4.35.)
Storms are not uniformly distributed in time or space: storms vary
seasonally and from year to year; storms originate more frequently in
some areas than in others; and storms follow characteristic tracks deter-
mined by prevailing global circulation and weather patterns.
An investigation of 170 damaging storms affecting the east coast of

the U.S. from 1921-1962 (Mather, et al., 1964), classified the storms into
eight types based on origin, structure, and path of movement. Of these
eight types, although 33 percent were hurricanes, two types, comprising
only 19 percent of the total, characterized by weather fronts east and
south of the U.S. coasts produced more damage per storm because of long
fetches. (Damage is defined by Mather, et al., as "at best some water
damage," and includes "wave damage, coastal flooding, and tidal inunda-
tion," but specificially excludes wind damage.)
The probability that a given section of coast will experience storm

waves depends on its ocean exposure, its location in relation to storm
tracks, and the shelf bathymetry.
4.33 NEARSHORE WAVE CLIMATE
4.331 Mean Value Data on U.S. Littoral Wave Climates . Wave height and
period data for some localities of the U.S. are becoming increasingly
available (e.g., Thompson and Harris, 1972), but most localities still
lack such data. However, wave direction is difficult to measure, and
consequently direction data are rarely available.
4-29
The quality and quantity of available data often do not justify
elaborate statistical analysis. Even where adequate data are available,
a simple characterization of wave climate meets many engineering needs.
While mean values of height and, to a lesser degree, period are useful,
data on wave direction are generally of insufficient quality for even
mean value use. Table 4-4 compiles mean annual wave heights collected
from a number of wave gages and by visual observers along the coasts of
the United States. Visual observations were made from the beach of waves
near breaking. The gages were fixed in depths of 10 to 28 feet.
Wave data treated in this section are limited to nearshore observa-

tions and measurements. Consequently waves were fully refracted and had
been fully affected by bottom friction, percolation, and nonlinear changes
in wave form by shoaling. Thus, these data differ from data that would
be obtained by simple shoaling calculations based on the deepwater wave
statistics. In addition, data are normally lacking for the rarer, high-
wave events. For these reasons, the data should not be used for struc-
tural design, since they are only applicable to the particular site
where they were obtained. Normal design practice is based on deepwater
wave statistics which are then adjusted to the shallow-site conditions.
However, the nearshore data are useful in littoral transport calculations.
Mean wave height and period from a number of visual observations by

coastguardmen at shore stations are plotted by month in Figures 4-10 and
4-11, using the average values of stations within each of five coastal
segments. Table 4-4 and Figures 4-10 and 4-11 show average values char-
acteristic of the wave climate in exposed coastal localities. Visual
height data represent an average value of the higher waves just before
they first break. The data provide only approximate indications of the
height distributions, but mean values of these distributions are useful.
In Figure 4-10 the minimum monthly mean littoral zone wave height
averaged for the California, Oregon, and Washington coasts exceeds the
maxirmm mean littoral zone wave height averaged for the other coasts.
This difference greatly affects the potential for sediment transport in
the respective littoral zones, and should be considered by engineers
when applying experience gained in a locality with one nearshore wave
climate to a problem at a locality with another wave climate.
4.332 Mean vs Extreme Conditions . Section 3.22 contains a discussion

of wave height distributions and the relations between various wave
height statistics, such as the mean, significant, and RMS heights, and
extreme values. In general, a group of waves from the same record can be
approximately described by a Rayleigh distribution. (See Section 3.22.)
However, a different distribution appears necessary to describe the dis-
tribution of significant wave heights, where each significant wave height
is from a different wave record at the locality. (See Figure 4-12.)
Visual analysis of waves recorded on chart paper is discussed in

Section 3.22 and by Draper (1967), Tucker (1963), Harris (1970), and
Magoon (1970). Spectrum analysis of wave records is discussed in
4-30
fO OJ —
(4^) iM6!9H 9ADM dJogsJoaN Amiuo|/^ ud9^
0)
>
u
o
U
GO
c
•H
TJ
3
i-H
o
c
O
•H
>
o
to
cd V)
 CO
c
o -•
S rt
§So
0)
u
3
M
CVJ — O0>C0NU>lO ro
(39S) pOU9d 8ADM d.l0l)SJD9N ^m|UO^ UDd^
4-32
Tabic 44. Mean Wave Height at Coaslal Localities of Conterminous United States
o
o
O)
CC
i_
o
CD
>-
I
01
O)
en
o
O
>
to
o
o
o
e
o
X
>
o
o
c
CO
c
o
3
J3
to
3
C7>
00 to 't CM
i39^'il|6|3H 9ADM ;UD3!^|U5lS
4-34
Section 3.23 and by Kinsman (1965), National Academy of Sciences (1963),
and Neumann and Pierson (1966).
For the distribution of significant wave heights as defined by the

data reduction procedures at CERC (Thompson and Harris, 1972), the data
fit a modified exponential distribution of form
"s "s min
fH>h.) = ^ (4-6)
where Hg is the significant height, Hg the significant height of in-

terest, Hg Yrrin is the approximate "minimum significant height," and a
is the significant wave height standard deviation. This equation depends
on two parameters, Hg ^yi and a which are related to the mean height,
H = H . + a . (4-7)
I^ ^s min °^ ° ^^® "°^ available, but the mean significant height,

Hg is known, then an approximation to the distribution of (4-6) can be
obtained from the data of Thompson and Harris (1972, Table 1), which
suggest
H.^.-„ - 0.38 H^ . (4-8)
This approximation reduces Equation 4-6 to a one-parameter distribution

depending only on mean significant wave height
1.61 H —0.61 H
s s
f(h^>h> ". (4-9)
Equation 4-9 is not a substitute for the complete distribution function,

but when used with the wave-gage data on Figure 4-12, it provides an
estimate of higher waves with agreement within 20 percent. Greater
scatter would be expected with visual observations.
4.34 OFFICE STUDY OF WAVE CLIMATE
Information on wave climate is necessary for understanding local

littoral processes. Usually, time does not permit obtaining data from
the field, and it is necessary to compile information in an office study.
The primary variables of engineering interest for such a compilation are
wave height and direction.
Shipboard observations covering conterminous U.S. coasts and other

ocean areas are available as summaries (Summary of Synoptic Meterological
Observations, SSMO) through the National Technical Information Service,
Springfield, Va. 22151. See Harris (1972) for a preliminary evaluation
of this data for coastal engineering use.
4-35
When data are not available for a specific beach, the wave climate
can be estimated by extrapolating from another location, after correcting
for differences in coastal exposure, winds, and storms.
On the east, gulf, and Great Lakes coasts, local winds are often
highly correlated with the direction of longshore currents. Such wind
data are available in "Local Climatological Data" sheets published monthly
by the National Weather Service, National Oceanographic and Atmospheric
Agency (NOAA) for about 300 U.S. weather stations. Other NOAA wind-data
sources include annual summaries of the Local Climatological Data by sta-
tion (Local Climatological Data with Comparative Data), and weekly sum-
maries of the observed weather (Daily Weather Maps), all of which can be
ordered from the Superintendent of Documents, U.S. Government Printing
Office, Washington, D.C. 20402.
Local weather data are often affected by conditions in the neighbor-

hood of the weather station, so care should be used in extrapolating
weather records from inland stations to a coastal locality. However,
statistics on frequency and severity of storm conditions do not change
appreciably for long reaches of the coast. For example, in a study of
Texas hurricanes, Bodine (1969) felt justified in assuming no difference
in hurricane frequency along the Texas coast. In developing information
on the Standard Project Hurricane, Graham and Nunn (1959) divided the
Atlantic Coast into zones 200 miles long and the gulf coast into zones
400 miles long. Variation of most hurricane parameters within zones is
not great along straight open stretches of coast.
The use of weather charts for wave hindcasting is discussed in Sec-

tion 3.4. Computer methods for generating offshore wave climate are now
(1973) under test and development. However, development of nearshore
wave climate from hindcasting is usually a time-consuming job, and the
estimate obtained may suffer in quality because of the inaccuracy of
hindcast data, and the difficulty of assessing the effect of nearshore
topography on wave statistics. At the present time, if available at the
specific location, statistics based on wave-gage records are preferable
to hindcast statistics when wave data for the shallow-water conditions
are required.
Other possible sources of wave climate information for office stud-

ies include aerial photography, newspaper records, and comments from
local residents.
Data of greater detail and reliability than that obtained in an

office study can be obtained by recording the wave conditions at the
shoreline locality for at least 1 year by the use of visual observers
or wave gages. A study of year-to-year variation in wave height statis-
tics collected at CERC wave gages (Thompson and Harris, 1972), indicates
that six observations per day for 1 year gives a reliable wave height
distribution function to the 1 percent level of occurrence. At the gage
at Atlantic City, one observation a day for 1 year provided a useful
height-distribution function.
4-36
4.35 EFFECT OF EXTREME EVENTS
Infrequent events of great magnitude, such as hurricanes, cause sig-

nificant modification of the littoral zone, particularly to the profile
of a beach. An extreme event could be defined as an event, great in
terms of total energy expended or work done, that is not expected to occur
at a particular location, on the average, more than once every 50 to 100
years. Hurricane Camille in 1969 and the East Coast Storm of March 1962
can be considered extreme events. Because large storms are infrequent,
and because it does not necessarily follow that the magnitude of a storm
determines the amount of geomorphic change, the relative importance of
extreme events is difficult to establish.
Wolman and Miller (1960) suggested that the equilibrium profile of

a beach is more related to moderately strong winds that generate moderate
storm waves, rather than to winds that accompany infrequent catastrophic
events. Saville (1950) showed that for laboratory tests with constant
wave energy and angle of attack there is a particular critical wave steep-
ness at which littoral transport is a maximum. Under field conditions,
there is probably a similar critical value that produces transport out of
proportion to its frequency of occurrence. The winds associated with this
critical wave steepness may be winds generated by smaller storms, rather
than the winds associated with extreme events.
A review of studies of beach changes caused by major storms indi-

cates that no general conclusion that can be made concerning the signifi-
cance of extreme events. Many variables affect the amount of damage a
beach will sustain in a given storm.
Most storms move large amounts of sand from the beach offshore, but
after the storm, the lower waves that follow tend to restore this sand to
the face of the beach. Depending on the extent of restoration, the storm
may result in little permanent change. Depending on the path of the storm
and the angle of the waves, a significant amount of material can also be
moved alongshore. If the direction of longshore transport caused by the
storm is opposite to the net direction of transport, the sand will prob-
ably be returned in the months after the storm and permanent beach changes
effected by the storm will be small. If the direction of transport before,
during, and after the storm is the same, then large amounts of material
could be moved by the storm with little possibility of restoration. Suc-
cessive storms on the same beach may cause significant transport in oppo-
site directions, (e.g. Everts, 1973.)
There are some unique events that are only accomplished by catastro-
phic storms. The combination of storm surge and high waves allows water
to reach some areas not ordinarily attacked by waves. These extreme con-
ditions may result in the overtopping of dunes and in the formation of
washover fans and inlets. (Morgan, et al., 1958; Nichols and Marston,
1939.) Some inlets are periodically reopened by storms and then sealed
by littoral drift tremsported by normal wave action.
4-37
The wave climate at a particular beach also determines the effect
a storm will have. In a high-energy climate, storm waves are not much
larger than ordinary waves, and their effects may not be significant. An
example of this might be northeasters occurring at Cape Cod. In a low-
energy wave climate, where transport volumes are usually low, storm waves
can move significant amounts of sand, as do hurricanes on the gulf coast.
The type of beach sediment is also important in storm-induced changes.

A storm can uncover sediments not ordinarily exposed to wave action, and
thus alter the processes that follow the storm. (Morgan, et al., 1958.)
In sand-deficient areas where the beach is underlain by mud, the effects
of a storm can be severe and permanent.
The effects of particular storms on certain beaches are described in

the following paragraphs. These examples illustrate how an extreme event
may affect the beach.
In October 1963, the worst storm in the memory of the Eskimo people
occurred over an ice-free part of the Arctic Ocean, and attacked the coast
near Barrow, Alaska. (Hume and Schalk, 1967.) Detailed measurements of
some of the key coastal areas had been made just before the storm. Freeze-
up just after the storm preserved the changes to the beach until surveys
could be made the following July. Most of the beaches accreted 1 to 2
feet, although Point Barrow was turned into an island. According to Hume
and Schalk, "The storm of 1963 would appear to have added to the Point the
sediment of at least 20 years of normal longshore transport." Because of
the low-energy wave climate and the short season in which littoral pro-
cesses can occur at Barrow, this storm significantly modified the beach.
A study of two hurricanes, Carla (1961) and Cindy (1963), was made by
Hayes (1967). He concluded that "the importance of catastrophic storms
as sediment movers cannot be over-emphasized," and observed that, in low-
energy wave climates, most of the total energy is expended in the near-
shore zone as a series of catastrophes. In this region, however, the rare
"extreme" hurricane is probably not as significant in making net changes
as the more frequent moderate hurricanes.
Surprisingly, Hurricane Camille, with mEiximum winds of 200 mph, did

not cause significant changes to the beaches of Mississippi and Louisiana.
Tanner (1970) estimated that the sand transport along the beach appeared
to have been an amount equal to less than a year's amount under ordinary
conditions, and theorized that "the particular configuration of beach, sea
wall, and coastal ridge tended to suppress large scale transport."
Hurricane Audrey struck the western coast of Louisiana in June, 1957.

The changes to the beach during the storm were not extreme nor permanent.
However, the storm exposed marsh sediments in areas where sand was defi-
cient, and "set the stage for a period of rapid shoreline retreat follow-
ing the storm." (Morgan, et al., 1958.) Indirectly, then, the storm was
responsible for significant geomorphic change.
4-38
A hurricane (unnamed) coincided with spring tide on the New England
coast on 21 September 1938. Property damage and loss of life were both
high. A storm of this magnitude was estimated to occur about once every
150 years. A study of the beach changes along a 12-mile section of the
Rhode Island coast (Nichol and Marsten, 1939) showed that most of the
changes in the beach profile were temporary. The net result was some
cliff erosion and a slight retrogression of the beaches.
Beach changes from Hurricane Donna which hit Florida in September

1960 were more severe and permanent. In a study of the southwestern
coast of Florida before and after the storm. Tanner (1961), concluded
that "Hurricane Donna appears to have done 100 year's work, considering
the typical energy level thought to prevail in the area."
On 1 April 1946, a tsunami struck the Hawaiian Islands with runup in

places as high as 55 feet above sea level. (Shepard, et al., 1950.) The
beach changes were similar to those inflicted by storm waves although "in
only a few places were the changes greater than those produced during nor-
mal storm seasons or even by single severe storms." Because a tsunami is
of short duration, extensive beach changes do not occur, although property
damage can be quite high.
Several conclusions can be drawn from the above examples. If a beach

has a sufficient sand supply and fairly high dunes, and if the dunes are
not breached, little permanent modification will result from storms, except
for a brief acceleration of the normal littoral processes. This accelera-
tion will be more pronounced on a shore with low-energy wave conditions.
4.4 NEARSHORE CURRENTS
Nearshore currents in the littoral zone are predominantly wave-induced

motions superimposed on the wave-induced oscillatory motion of the water.
The net motions generally have low velocities, but because they transport
whatever sand is set in motion by the wave-induced water motions, they are
important in determining littoral transport.
There is only slight exchange of fluid between the offshore and the
surf zone. Onshore -offshore flows take place in a number of ways, which
at present are not fully understood.
4.41 WAVE- INDUCED WATER MOTION
In idealized deepwater waves, water particles have a circular motion

in a vertical plane perpendicular to the wave crest (Fig. 2-4, Section
2.235), but this motion does not reach deep enough to affect sediment on
the bottom. In depths where waves are affected by the bottom, the circu-
lar motion becomes elliptical, and the water at the bottom begins to move.
In shallow water, the ellipses elongate into nearly straight lines. At
breaking, particle motion becomes more complicated, but even in the surf
zone, the water moves forward and backward in paths that are mostly hori-
zontal, with brief, but intense, vertical motions produced by the passage
4-39
of the breaker crest. Since it is this wave-induced water particle motion
that causes the sediment to move, it is useful to know the length of the
elliptical path traveled by the water particles and the maximum velocity
and acceleration attained during this orbit.
The basic equations for water-wave motion before breaking are dis-
cussed in Chapter 2. Quantitative estimates of water motion are possible
from small -amplitude wave theory (Section 2.23), even near breaking where
assiomptions of the theory are not valid. (Dean, 1970; Eagleson, 1956.)
Equations 2-13 and 2-14, in Section 2.234 give the fluid-particle velocity
components u, w in a wave where small -amplitude theory is applicable.
(See Figure 2-3 for relation to wave phase and water particle accelera-
tion.)
For sediment transport, the conditions of most interest are those

when the wave is in shallow water. For this condition, and making the
small-amplitude assumption, the horizontal length 2A, of the path moved
by the water particle as a wave passes in shallow water is approximately
HT^gd
2A = -^^^ , C4-10)
and the maximum horizontal water velocity is
"^max = -TTT- ' (4-11)

2d
The term under the radical is the wave speed in shallow water.
************** EXAMPLE PROBLEM ***************

GIVEN A wave 1 foot high with a period of 5 seconds is progressing shore-
:
ward in a depth of 2 feet.
FIND ;
(a) Calculate the maximum horizontal distance 2A the water particle

moves during the passing of a wave.
(b) Determine the maximum horizontal velocity ^cuc °^ ^ water

particle.
(c) Compare the maximum horizontal distance 2A with the wavelength

in the 2- foot depth.
(d) Compare the maximum horizontal velocity ^cuc» with the wave
speed, C.
4-40
solution:
(a) Using Equation 4-10, the maximum horizontal distance is
HT Vgd^
2A =
2ffd
1 (5) ^32.2 (2)^

2A = = 3.2 feet .
27r(2)
(b) Using Equation 4-11, the maximum horizontal velocity is
2d
1^32.2(2)^
umax = TTT;
T '^^^
= per second
2.0 feet ^
2(2)
(c) Using the relation L = Ti/gd to determine the shallow-water

wavelength.
L = 5v/32.2(2)' = 40.1 feet .
From (a) above, the maximiam horizontal distance 2A is 3.2 feet

therefore the ratio 2A/L is
—
2A =
L
3.2
40.1
= 0.08 .
(d) Using the relation C = v'^ (Equation 2-9) to determine the

shallow-water wave speed
C = y^32.2 (2)" =8.0 feet per second .
From (b) above the maximum horizontal velocity âx> ^^ ^'^

feet per second. Therefore the ratio âx/^ i^
^max 2.0
-^^^ = = 0.25.
C 8.0
*************************************
Although small -amplitude theory gives a fair understanding of many
wave-related phenomena, there are important phenomena that it does not
predict. Observation and a more complete analysis of wave motion show
that particle orbits are not closed. Instead, the water particles
advance a little in the direction of the wave motion each time the wave
4-41
passes. The rate of this advance is the mass transport vetoaity; (Equa-
tion 2-55, Section 2.253). This velocity becomes important for sediment
transport, especially for sediment suspended above ripples seaward of the
breaker.
For conditions evaluated at the bottom (z = -d) , the maximum bottom

velocity, umoxr. j').
given by Equation 2-13 determines the average bottom
mass transport velocity, u(_j). obtained from Equation 2-55, according to
the equation
("wax/ J))
"(-d)='-^' . (4-12)
where C is the wave speed given by Equation 2-3. Equation 2-55, and
thus Equation 4-12, does not include allowance for return flow which
must be present to balance the mass transported in the direction of wave
travel. In addition, the actual distribution of the time-averaged net
velocity depends sensitively on such external factors as bottom character-
istics, temperature distribution, and wind velocity. (Mei, Liu, and Carter,
1972.) Most observations show the time-averaged net velocity near the
bottom is directed toward the breaker region from both sides. (See Inman
and Quinn, (1952), for field measurements in surf zone; Galvin and Eagle-
son, (1965) for laboratory observations; and Mei, Liu and Carter (1972,
p. 220), for comprehensive discussion.) However, both field and labora-
tory observations have shown that wind-induced bottom currents may be
great enough to reverse the direction of the shoreward time-averaged
wave-induced velocity at the bottom when there are strong onshore winds.
(Cook and Gorsline, 1972; and Kraai, 1969.)
4.42 FLUID MOTION IN BREAKING WAVES .
During most of the wave cycle in shallow water, the particle velo-
city is approximately horizontal and constant over the depth, although
right at breaking there are significant vertical velocities as the water
is drawn up into the crest of the breaker. The maximum particle velocity
under a breaking wave is approximated by solitary wave theory (Equation
2-66) to be
Hmax = '^ = v/g (H + d)' , (4-13)
where (H+d) is the distance measured from crest of the breaker to the
bottom.
Fluid motions at breaking cause most of the sediment transport in

the littoral zone, because the bottom velocities and turbulence at break-
ing suspend more bottom sediment. This suspended sediment can then be
transported by currents in the surf zone whose velocities are normally
too low to move sediment at rest on the bottom.
Tlie mode of breaking may vary significantly from spilling to plung-

ing to collapsing to surging, as the beach slope increases or the wave
4-42
steepness (height-to-length ratio) decreases. (Galvin, 1967.) Of the
four breaker types, spilling breakers most closely resemble the solitary
waves whose speed is described by Equation 4-13. (Galvin, 1972.) Spill-
ing breakers differ little in fluid motion from unbroken waves (Divoky,
LeMehaute, and Lin, 1970), and thus tend to be less effective in trans-
porting sediment than plunging or collapsing breakers.
The most intense local fluid motions are produced by plunging break-
ers. As the wave moves into shallower depths, the front face begins to
steepen. When the wave reaches a mean depth about equal to its height,
it breaks by curling over at the crest. The crest of the wave acts as a
free-falling jet that scours a trough into the bottom. At the same time,
just seaward of the trough, the longshore bar is formed, in part by sedi-
ment scoured from the trough and in part by sediment transported in rip-
ples moving from the offshore.
The effect of the tide on nearshore currents is not discussed, but

tide-generated currents may be superimposed on wave-generated nearshore
currents, especially near estuaries. In addition, the changing elevation
of the water level as the tide rises and falls may change the area and
the shape of the profile through the surf zone, and thus alter the near-
shore currents.
4.43 ONSHORE-OFFSHORE CURRENTS
4.431 Onshore-Offshore Exchange . Field and laboratory data indicate

that water in the nearshore zone is divided by the breaker line into
two distinct water masses between which there is only a limited exchange
of water.
The mechanisms for the exchange are: mass transport velocity in

shoaling waves, wind-induced surface drift, wave-induced setup, currents
induced by irregularities on the bottom, rip currents, and density cur-
rents. The resulting flows are significantly influenced by, and act on,
the hydrography of the surf and nearshore zones. Figure 4-13 shows the
nearshore current system measured for particular wave conditions on the
southern California coast.
At first observation, there appears to be extensive exchange of

water between the nearshore and the surf zone. However, the breaking
wave itself is formed largely of water that has been withdrawn from the
surf zone after breaking. (Galvin, 1967.) This water then reenters the
surf zone as part of the new breaking wave, so that only a limited amount
of water is actually transferred offshore. This inference is supported
by the calculations of Longuet-Higgins (1970, p. 6788) which show that
little mixing is needed to account for observed velocity distributions.
Most of the exchange mechanisms indicated act with speeds much slower
than the breaking-wave speed, which may be taken as an estimate of the
maximum water particle speed in the littoral zone indicated by Equation
4.13.
4-43
Scripps Canyon
2 December 1948
Wave Period 15 Seconds
Wave from WNW
-.25
.25-.50KN
.50- 1.0 KN
>l Knot
— Hh = 4.5
Observed Current (not measured)

Starting Position of Surface Float Scripps
Breaker Height stitution
Hb =
Float Recovery Area
k
n
<^y\/^---^ SCALE IN FEET

500 1000
t-l l-l i-l t-l l-l I
s^^"
(from Shepard and Inman, 1950)
Figure 4-13. Nearshore Current System Near La Jolla Canyon, California
4-44
4.432 Diffuse Return Flow . Wind- and wave-induced water drift, pres-
sure gradients at the bottom due to setup, density differences due to
suspended sediment and temperature, and other mechanisms produce patterns
of motion in the surf zone that vary from highly organized rip currents
to broad diffuse flows that require continued observation to detect. Dif-
fuse return flows may be visible in aerial photos as fronts of turbid
water moving seaward from the surf zone. Such flows may be seen in the
photos reproduced in Sonu (1972, p. 3239).
4.433 Rip Currents . Most noticeable of the exchange mechanisms between

offshore and surf zone are rip currents. (See Figure 4-14, and Figure
A-7, Appendix A.) Rip currents are concentrated jets that carry water
seaward through the breaker zone. They appear most noticeable when long,
high waves produce wave setup on the beach. In addition to the classi-
cal rip currents, there are other localized currents directed seaward
from the shore. Some are due to concentrated flows down gullies in the
beach face, and others can be attributed to interacting waves and edge
wave phenomena. (Inman, Tait, and Nordstrom, 1971, p. 3493.) The origin
of rip currents is discussed by Arthur (1962), and Sonu (1972).
Three-dimensional circulation in the surf is documented by Shepard and

Inman (1950), and this complex flow needs to be considered, especially in
evaluating the results of laboratory tests for coastal engineering purposes.
However, at present, there is no proven way to predict the conditions that
produce rip currents or the spacing between rips. In addition, data are
lacking that would indicate quantitatively how important rip currents are
as sediment transporting agents.
4.44 LONGSHORE CURRENTS
4.441 Velocity and Flow Rate . Longshore currents flow parallel to the
shoreline, and are restricted mainly between the zone of breaking waves
and the shoreline. Most longshore currents are generated by the long-
shore component of motion in waves that obliquely approach the shoreline.
Longshore currents typically have mean values of 1 foot per second

or less. Figure 4-15 shows a histogram of 5,591 longshore current veloc-
ities measured at 36 sites in California during 1968. Despite frequent
reports of exceptional longshore current speeds, most data agree with
Figure 4-15 in showing that speeds above 3 feet per second are unusual.
A compilation of 352 longshore current observations, most of which appear
to be biased toward conditions producing high speed, showed that the maxi-
mum observed speed was 5.5 feet per second, and that the highest observa-
tions were reported to have been wind-aided. (Calvin and Nelson, 1967.)
Although longshore currents generally have low speeds, they are important
in littoral processes because they flow along the shore for extended peri-
ods of time, transporting sediment set in motion by the breaking waves.
The most important variable in determing the longshore current veloc-

ity is the angle between the wave crest and the shoreline. However, the
volume rate of flow of the current and the longshore transport rate depend
4-45
Figure 4-14. Typical Rip Currents, Ludlam
Island, New Jersey
4-46
mostly on breaker height. The outer edge of the surf zone is determined
by the breaker position. Since waves break in water depths approximately
proportional to wave height, the width of the surf zone on a beach in-
creases with wave height. This increase in width increases the cross
section of the surf zone.
2400 T T T
2000
10
c
o
Total of 5591 Observations
ô 1600
March -December 1968
>
w
O 1200
Xi 800
E
400 NORTH SOUTH
ij -3 -2 -I 1 2 3 4 5
Longshore Current Velocity, (feet per sec)
Figure 4-15. Distribution of Longshore Current Velocities. Data taken

from CERC California LEO Study (See Szuwalski 1970).
If the surf zone cross section is approximated by a triangle, then

an increase in height increases the area (and thus the volume of the flow)
as the square of the height, which nearly offsets the increase in energy
flux (which increases as the 5/2 power of height). Thus, the height is
important in determining the width and volume rate of longshore current
flow in the surf zone. (Calvin, 1972.)
Longshore current velocity varies both across the surf zone (Longuet-
Higgins, 1970b) and in the longshore direction (Calvin and Eagleson, 1965).
Where an obstacle to the flow, such as a groin, extends through the surf
zone, the longshore current speed downdrift of the obstacle is low, but
it increases with distance downdrift. Laboratory data suggest that the
current takes a longshore distance of about 10 surf widths to become fully
developed. These same experiments (Calvin and Eagleson, 1965) suggest that
the velocity profile varies more across the surf zone at the start of the
flow than it does downdrift where the flow has fully developed. The ratio
of longshore current speed at the breaker position to longshore current
speed averaged across the surf zone varied from about 0.4 where the flow
started to about 0.8 or 1.0 where the flow was fully developed.
4-47
4.442 Velocity Prediction . The variation in longshore current velocity
across the surf zone and along the shore, and the uncertainties in vari-
ables such as the surf zone hydrography, make prediction of longshore
current velocity uncertain. There are three equations of possible use
in predicting longshore currents: Longuet-Higgins (1970); an adaptation
from Bruun (1963); and Galvin (1963). All three equations require co-
efficients identified by comparing measured and computed velocities, and
all three show about the same degree of agreement with data. Two sets of
data (Putnam, et al., 1949, field data; Galvin and Eagleson, 1965, labora-
tory data) appear to be the most appropriate for checking predictions.
The radiation stress theory of Longuet-Higgins (1970a, Equation 62),

as modified by fitting it to the data, is the one recommended for use
based on its theoretical foundation. The other two semiempirical equa-
tions may provide a check on the Longuet-Higgins prediction. Written
in common symbols (m is beach slope; g is acceleration of gravity; H^
is breaker height; T is wave period; and a^j is angle between breaker
crest and shoreline), these equations are:
a. Longuet-Higgins .
vj, = M, m(gHj,)*/^ sin2aj, , (4-14)
where
'2
0.694 r(2i3)
"^
M, = / (4-15)
y
According to Longuet-Higgins (1970a, p. 6788) , v^ is the longshore cur-
rent speed at the breaker position, r is a mixing coefficient which
ranges between 0.17 (little mixing) and 0.5 (complete mixing), but is
commonly about 0.2; g is the depth-to-height ratio of breaking waves
in shallow water taken to be 1.2 and fj? is the friction coefficient,
taken to be 0.01. Using these values, M,= 9.0.
Applying equation 4-14 to the two sets of data yields predictions

that average about 0.43 of the measured values. In part, these predicted
speeds are lower because v^j, as given in Equation 4-14 is for the speed
at the breaker line, whereas the measured velocities are mostly from the
faster zone of flow shoreward of the breaker line. (Galvin and Eagleson,
1965.) Therefore, Equation 4-14 multiplied by 2.3 leads to the modified
Longuet-Higgins equation for longshore current velocity:
V = 20.7 m(gHj,)'/^ sin 2ay ,

(4-16)
used in Figure 4-16. Further developments in the Longuet-Higgins' (1970b

and 1971) theory permit calculation of velocity distribution, but there
is no experience with these predictions for longshore currents flowing
on erodible sand beds.
4-48
'
o
a>
T3
a>
a>
o.
CO
o
C7>
o
0)
CO
O
a>
b. Bruun (1963 as Modified) .
^b = ^2 (^^bf ["'^b (si" 2aj,)/T]'^^ , (4-17)
where v^ is the mean velocity in the surf zone where the flow is fully
developed, and M2 involves a friction factor of the Chezy kind (see
Galvin, 1967, p. 297.)
c. Galvin (1963) .
vj, = KgmTsin2aj, , (4-18)
where v^ is the mean velocity in the surf zone where the flow is fully
developed, and K is a coefficient depending on breaker height-to-depth
ratio and the ratio of trough depression on breaker height. To a good
approximation, K may be taken as 1.0. (Galvin and Eagleson, 1965.)
4.45 SUMMARY
The major currents in the littoral zone are wave-induced motions

superimposed on the wave-induced oscillatory motion of the water. The
net motions generally have low velocities, but because they transport
whatever sand is set in motion by the wave-induced water motions, they
are important in determining littoral transport.
Evidence indicates that there is only slight exchange of fluid

between the offshore and the surf zone.
Longshore current velocities are most sensitive to changes in breaker

angle, and to a lesser degree, to changes in breaker height. However, the
volume rate of flow of the longshore current is most sensitive to breaker
height, probably proportional to H^. The modified Longuet-Higgins equation
(4-16) is recommended for predicting mean longshore current velocity of
fully developed flows, and the two semiempirical equations (4-17 and 4-18)
are available as checks on the Longuet-Higgins equation.
4.5 LITTORAL TRANSPORT
4.51 INTRODUCTION
4.511 Importance of Littoral Transport Sediment motions indicated by

.
the shoreline configuration in Figure 4-17 are aspects of littoral trans-

port. If the coast is examined on satellite imagery as shown in Figure
4-17, only its general characteristics are visible. At this elevation,
the shore consists of bright segments that are straight or slightly curved.
The brightness is evidence of sand, the most common material along the
shore. The straightness often is evidence of sediment transport.
In places, the straight segments of shoreline cut across preexisting

topography. Elsewhere, the shoreline segments are separated by wide la-
goons from the irregular mainland. The fact that the shore is nearly
straight across both mainland and irregular bays is evidence of headland
4-50
erosion, accompanied by longshore transport which has carried sand along
the coast to supply the barriers and spits extending across the bays.
The primary agent producing this erosion and transport is the action of
waves impinging on the shore.
Littoral transport is the movement of sedimentary material in the

littoral zone by waves and currents. The littoral zone extends from the
shoreline to just beyond the most seaward breakers.
Littoral transport is classified as onshore-offshore transport or as

longshore transport. Onshore-offshore transport has an average net direc-
tion perpendicular to the shoreline; longshore transport has an average
net direction parallel to the shoreline. The instantaneous motion of
sedimentary particles has both an onshore-offshore and a longshore com-
ponent. Onshore-offshore transport is usually the most significant type
of transport in the offshore zone, except in regions of strong tidal
currents. Both longshore and onshore-offshore transport are significant
in the surf zone.
Engineering problems involving littoral transport generally require

answers to one or more of the following questions:
(a) What are the longshore transport conditions at the site?

(Needed for the design of groins, jetties, navigation channels, and
inlets.)
Cb) What is the trend of shoreline migration over short and long
time intervals? (Needed for design of coastal structures, including
navigation channels.)
(c) How far seaward is sand actively moving? (Needed in the design
of sewage outfalls and water intakes.)
(d) What is the direction and rate of onshore-offshore sediment

motion? (Needed for sediment budget studies and beach fill design.)
(e) What is the average shape, and the expected range of shapes,
for a given beach profile? (Needed for design of groins, beach fills,
navigation structures and flood protection.)
(f) What effect will a postulated structure or project have on

adjacent beaches and on littoral transport? (Needed for design of all
coastal works.)
This section presents recommended methods for answering these and

related questions. The section indicates accepted practice based on
field observations and research results. Section 4.52 deals with onshore-
offshore transport, presenting material pertinent to answering questions
(b) through (f). Section 4.53 deals with longshore transport, presenting
material pertinent to questions (a) , (b) , and (f)
4-52
4.512 Zones of Transport . Littoral transport occurs in two modes: hed-
toad transport J the motion of grains rolled over the bottom by the shear
of water moving above the sediment bed; and suspended-load transport j the
transport of grains by currents after the grains have been lifted from
the bed by turbulence.
Both modes of transport are usually present at the same time, but it
is hard to distinguish where bedload transport ends and suspended-load
transport begins. It is more useful to identify two zones of transport
based on the type of fluid motion initiating sediment motion: the off-
shore zone where transport is initiated by wave-induced motion over rip-
ples, and the surf zone where transport is initiated by the passing break-
er. In either zone, net sediment transport is the product of two pro-
cesses: the periodic wave-induced fluid motion that initiates sediment
motion, and the superimposed currents (usually weak) which transport the
sediment set in motion.
a. Offshore Zone. Waves traveling toward shallow water eventually

reach a depth where the water motion near the bottom begins to affect the
sediment on the bottom. At first, only low-density material (such as sea-
weed and other organic matter) moves. This material oscillates back and
forth with the waves, often in ripple-like ridges parallel to the wave
crests. For a given wave condition, as the depth decreases, water motion
immediately above the sediment bed increases until it exerts enough shear
to move sand particles. The sand then forms ripples with crests parallel
to the wave crests. These ripples are typically uniform and periodic,
and sand moves from one side of the crest to the other with the passage
of each wave.
As depth decreases to a value several times the wave height, the veloc-
ity distribution with time changes from approximately sinusoidal to a dis-
tribution that has a high shoreward component associated with the brief
passage of the wave crest, and lower seaward velocities associated with
the longer time interval occupied by the passage of the trough. As the
shoreward water velocity associated with the passing crest decreases and
begins to reverse direction over a ripple, a cloud of sand erupts upward
from the lee (landward) side of the ripple crest. This cloud of sand
drifts seaward with the seaward flow under the trough. At these shallow
depths, the distance traveled by the cloud of suspended sediment is two
or more ripple wavelengths, so that the sand concentration at a point
above the ripples usually exhibits at least two maximums during the pass-
age of the wave trough. These maximums are the suspension clouds shed by
the two nearest upstream ripples. The approach of the next wave crest
reverses the direction of the sand remaining suspended in the cloud. The
landward flow also drags material shoreward as bedload.
For the nearshore profile to be in equilibrium with no net erosion or

accretion, the average rate at which sand is carried away from a point on
the bottom must be balanced by the average rate at which sand is added.
Any net change will be determined by the net residual currents near the
bottom which transport sediment set in motion by the waves. These currents,
4-53
the subject of Section 4.4, include longshore currents and mass-transport
currents in the on shore -offshore direction. It is possible to have ripple
forms moving shoreward while residual currents above the ripples carry
suspended sediment clouds in a net offshore direction. Information on the
transport of sediment above ripples is given in Bijker (1970), Kennedy and
Locher (1972), and Mogridge and Kamphuis (1972).
b. Surf Zone. The stress of the water on the bottom due to turbu-
lence and wave- induced velocity gradients moves sediment in the surf zone
with each passing breaker crest. This sediment motion is both bedload
and suspended- load transport. Sediment in motion oscillates back and
forth with each passing wave, and moves alongshore with the longshore
current. On the beach face, the landward termination of the surf zone,
the broken wave advances up the slope as a bore of gradually decreasing
height, and then drains seaward in a gradually thinning sheet of water.
Frequently, the draining return flows in gullies and carries sediment to
the base of the beach face.
In the surf zone, ripples cause significant sediment suspension, but

here there are additional eddies caused by the breaking wave. These eddies
have more energy and are larger than the ripple eddies. The greater energy
suspends more sand in the surf zone than offshore. The greater eddy size
mixes the suspended sand over a larger vertical distance. Since the size
is about equal to the local depth, significant quantities of sand are sus-
pended over most of the depth in the surf zone.
Since breaking waves suspend the sediment, the amount suspended is

partly determined by breaker type. Data from Fairchild (1972, Figure 5),
show that spilling breakers usually produce noticeably lower suspended
sediment concentrations than do plunging breakers. See Fairchild (1972)
and Watts (1953) for field data; Fairchild (1956 and 1959) for lab data.
Typical suspended concentrations of fine sand range between 20 parts per
million and 2 parts per thousand by weight in the surf zone, and are about
the same near the ripple crests in the offshore zone.
Studies of suspended sediment concentrations in the surf zone by

Watts (1953) and Fairchild (1972) indicate that sediment in suspension in
the surf zone may form a significant portion of the material in longshore
transport. However, present understanding of sediment suspension, and
the practical difficulty of obtaining and processing sufficient suspended
sediment samples have limited this approach to predicting longshore trans-
port.
4.513 Profiles . Profiles are two-dimensional vertical sections showing

how elevation varies with distance. Coastal profiles (See Figs. 4-1 and
4-18) are usually measured perpendicular to the shoreline, and may be
shelf profiles, nearshore profiles, or beach profiles. Changes on near-
shore and beach profiles are interrelated, and are highly important in
the interpretation of littoral processes. The measurement and analysis
of combined beach and nearshore profiles is a major part of most engineer-
ing studies of littoral processes.
4-54
30
V
20
10
a. Shelf Profiles. The shelf profile is typically a smooth, concave-
up curve showing depth to increase seaward at a rate that decreases with
distance from shore. (bottom profile in Figure 4-18.) The smoothness of
the profile may be interrupted by other superposed geomorphic features,
such as linear shoals. (Duane, et al., 1972.) Data for shelf profiles
are usually obtained from charts of the National Ocean Survey (formerly,
U.S. Coast and Geodetic Survey).
The measurable influence of the shelf profile on littoral processes

is largely its effect on waves. To an unknown degree, the shelf may also
serve as a source or sink for beach sand. Geologic studies show that
much of the outer edge of a typical shelf profile is underlain by rela-
tively coarse sediment, indicating a winnowing of fine sizes. (Dietz,
1963; Milliman, 1972; and Duane, et al . , 1972.) Landward from this resi-
dual sediment, sediment often becomes finer before grading into the rela-
tively coarser beach sands.
b. Nearshore Profiles . The nearshore profile extends seaward from

the beach to depths of about 30 feet. Prominent features of most near-
shore profiles are longshore bars; see middle profile of Figure 4-18 and
Section 4.525. In combination with beach profiles, repetitive nearshore
profiles are used in coastal engineering to estimate erosion and accre-
tion along the shore, particularly the behavior of beach fill, groins,
and other coastal engineering structures. Data from nearshore profiles
must be used cautiously. (see Section 4.514.) Under favorable condi-
tions nearshore profiles have been used in measuring longshore transport
rates. (Caldwell, 1956.)
c. Beach Profiles. Beach profiles extend from the foredunes, cliffs,

or mainland out to mean low water. Terminology applicable to features of
the beach profile is in Appendix A (especially Figures A-1 and A-2). The
backshore extends seaward to the foreshore, and consists of one or more
berms at elevations above the reach of all but storm waves. Berm sur-
faces are nearly flat and often slope landward at a slight downward angle.
(See Figure 4-1.) Berms are often bounded on the seaward side by a break
in slope known as the berm crest.
The foreshore is that part of the beach extending from the highest
elevation reached by waves at normal high tide seaward to the ordinary
low water line. The foreshore is usually the steepest part of the beach
profile. The boundary between the backshore and the foreshore may be
the crest of the most seaward berm, if a berm is well developed. The
seaward edge of the foreshore is often marked by an abrupt step at low
tide level.
Seaward from the foreshore, there is usually a low-tide terrace which

is a nearly horizontal surface at about mean low tide level. (Shepard,
1950; and Hayes, 1971.) The low-tide terrace is commonly covered with sand
ripples and other minor bed forms, and may contain a large bar-and-trough
system, which is a landward-migrating sandbar (generally parallel to the
shore) common in the nearshore following storms. Seaward from the low-tide
4-56
terrace (seaward from the foreshore, if the low-tide terrace is absent)
are the longshore troughs and longshore bars.
4.514 Profile Accuracy . Beach and nearshore profiles are the major
source of data for engineering studies of beach changes; sometimes lit-
toral transport can be estimated from these profiles. Usually, beach
and nearshore profiles are measured at about the same time, but differ-
ent tecliniques are needed for their measurement. The nearshore profile
is usually measured from a boat or amphibious craft, using an echo sounder
or leadline, or from a sea sled. (Kolessar and Reynolds, 1965-66; and
Reimnitz and Ross, 1971.) Beach profiles are usually surveyed by standard
leveling and taping techniques.
The accuracy of profile data is affected by four types of error:

sounding error, spacing error, closure error, and error due to temporal
fluctuations in the sea bottom. These errors are more significant for
nearshore profiles than for beach profiles.
Saville and Caldwell (1953) discuss sounding and spacing errors.

Sounding error is the difference between the measured depth and the
actual depth. Under ideal conditions, average sonic sounding error may
be as little as 0.1 foot, and average leadline sounding error may be
about twice the sonic sounding error. (Saville and Caldwell, 1953.) (This
suggests that sonic sounding error may actually be less than elevation
changes caused by transient features like ripples. Experience with suc-
cessive soundings in the nearshore zone indicates that errors in practice
may approach 0.5 foot.) Sounding errors are usually random and tend to
average out when used in volume computations, unless a systematic error
due to the echo sounder or tide correction is involved. Long-period
water level fluctuations affect sounding accuracy by changing the water
level during the survey. At Santa Cruz, California, the accuracy of
hydrographic surveys was ±1.5 feet due to this effect. (Magoon, 1970.)
Spacing error is the difference between the actual volume of a seg-

ment of shore and the volume estimated from a single profile across that
segment. Spacing error is potentially more important than sounding error,
since survey costs of long reaches usually dictate spacings between near-
shore profiles of thousands of feet. For example, if a 2-mile segment of
shore 4,000 feet wide is surveyed by profiles on 1,000-foot spacings, then
the spacing error is about 9 cubic yards per foot of beach front per survey,
according to the data of Saville and Caldwell (1953, Figure 5). This error
equals a major part of the littoral budget in many localities.
Closure error arises from the assumption that the outer ends of
nearshore profiles have experienced no change in elevation between two
successive surveys. Such an assumption is often made in practice, and
may result in significant error. An uncompensated closure error of 0.1
foot, spread over 1,000 feet at the seaward end of a profile, implies a
change of 3.7 cubic yards per time interval per foot of beach front where
4-57
the time interval is the time between successive surveys. Such a volume
change may be an important quantity in the sediment budget of the litto-
ral zone.
A fourth source of error comes from assuming that the measured beach
profiles (which are only an instantaneous picture), represent a long-term
condition. Actually, beach and nearshore profiles change rapidly in re-
sponse to changing wave conditions, so that differences between succes-
sive surveys of a profile may merely reflect temporary differences in
bottom elevation caused by storms and seasonal changes in wave climate.
Such fluctuations obliterate long-term trends during the relatively short
time available to most engineering studies. This fact is illustrated for
nearshore profiles by the work of Taney (1961a, Appendix B) who identified
and tabulated 128 profile lines on the south shore of Long Island that had
been surveyed more than once from 192 7 to 1956. Of these, 47 are on
straight shorelines away from apparent influence by inlets, and extend
from Mean Low Water (MLW) to about -30 feet MLW. Most of these 47 pro-
files were surveyed three or more times, so that 86 separate volume
changes are available. These data lead to the following conclusions:
(a) The net volume change appears to be independent of the time

between surveys, even though the interval ranged from 2 months to 16
years. (See Figure 4-19.)
(b) Gross volume changes (the absolute sums of the 86 volume changes)
are far greater than net voliome changes (the algebraic sums of the 86 vol-
lome changes). The gross volume change for all 86 measured changes is 8,113
cubic yards per foot; the net change is -559 cubic yards per foot (loss in
volume)
(c) The mean net change between surveys, averaged over all pairs of
surveys, is -559/86 or -6.5 cubic yards per foot of beach. The median
time between surveys is 7 years, giving a nominal rate of volume change
of about -1 cubic yard per year per foot.
These results point out that temporary changes in successive surveys

of nearshore profiles are usually much larger than net changes, even when
the interval between surveys is several years. These data show that care
is needed in measuring nearshore profiles if results are to be used in
engineering studies. The data also suggest the need for caution in inter-
preting differences obtained in two surveys of the same profiles.
The positions of beach profiles must be marked so that they can be

recovered during the life of the project. The profile monuments should
be tied in by survey to local permanent references. If there is a long-
term use for data at the profile positions, the monuments should be ref-
erenced by survey to a state coordinate system or other reference system,
so that the exact position of the profile may be recovered in the future.
Even if there is no anticipated long-term need, future studies in any
coastal region are likely, and will benefit greatly from accurately sur-
veyed, retrievable benchmarks.
4-58
500-
>>
to
>»
V
(jt>- -^
>
k.
3
c oo o
«
I I I I ll I I I I I I I
100 200
J.
.c 8
u
«
E
_3
O
>
c
13
500-
Time Between Surveys, months
( based on doto from Toney, 1961 a )
Figure 4-19. Unit Volume Change Versus Time Between Surveys for Profiles
on South Shore of Long Island. Data from Profiles Extending
from MSL to about the -30 depth Contour.
4-59
For coastal engineering, the accuracy of shelf profiles is usually
less critical than the accuracy of beach and nearshore profiles. Gener-
ally, observed depth changes between successive surveys of the shelf do
not exceed the error inherent in the measurement. However, soundings
separated by decades suggest that the linear shoals superposed on the
profile do show small but real shifts in position. (Moody, 1964, p. 143.)
Charts giving depths on the continental shelves may include soundings
that differ by decades in date.
Plotted profiles usually use vertical exaggeration or distorted

scales to bring out characteristic features. This exaggeration may lead
to a false impression of the actual slopes. As plotted, the three pro-
files in Figure 4-18 have roughly the same shape, but this sameness has
been obtained by vertical exaggerations of 2x, lOx, and 50x.
Sand level changes in the beach and nearshore zone may be measured
quite accurately from pipes imbedded in the sand. (Inman and Rusnak, 1956;
Urban and Galvin, 1969; and Gonzales, 1970.)
4.52 ONSHORE -OFF SHORE TRANSPORT
4,521 Sediment Effects . Properties of individual particles which have

been considered important in littoral transport include: size, shape,
immersed specific gravity, and durability. Collections of particles
have the additional properties of size distribution, permeability, and
porosity. These properties determine the forces necessary to initiate
and maintain sediment motion.
For typical beach sediment, size is the only property that varies
greatly. However, quantitative evaluation of the size effect is usually
lacking. A gross indication of a size effect is the accumulation of
coarse sediment in zones of maximum wave energy dissipation, and deposi-
tion of fine sediment in areas sheltered from wave action. (e.g. King,
1972, pp. 302, 307, 426.) Sorting by size is common over ripples (Inman,
1957) and large longshore bars (Saylor and Hands, 1970). Field work on
size effects in littoral transport does not permit definite conclusions.
(King, 1972, p. 483; Inman, Komar, and Bowen, 1969; Castanho, 1970; Ingle,
1966, Figure 112; Yasso, 1962; and Zenkovich, 1967a.)
The shape of most littoral materials is approximately spherical;

departures from spherical are usually too slight to affect littoral
transport.
Immersed specific gravity (specific gravity of sediment minus spec-

ific gravity of fluid) is theoretically an important physical property of
the sediment particle. (Bagnold, 1963.) However, the variation in immersed
specific gravity for typical littoral materials in water is small since
most beach sediments are quartz (immersed specific gravity = 1.65), and
most of the remainder are calcium carbonate (immersed specific gravity
= 1.9). Thus, little variation in littoral transport is expected from
variation in immersed specific gravity.
4-60
Durability (resistance to abrasion, crushing, and solution) is usu-
ally not a factor within the lifetime of an engineering project. (Kuenen,
1956; Rusnak, Stockman, and Hofmann, 1966; and Thiel, 1940.) Possible
exceptions may include basaltic sands on Hawaiian beaches (Moberly, 1968),
some fragile carbonate sands which may be crushed to finer sizes when sub-
ject to traffic, (Duane and Meisburger, 1969, p. 44), and carbonate sands
which may be soluble imder some conditions. (Bricker, 1971.) In general,
recent information lends further support to the conclusion of Mason (1942)
that, "On sandy beaches the loss of material ascribable to abrasion . . .
occurs at rates so low as to be of no practical importance in shore pro-

tection problems."
Size distribution and its relation to sediment sorting may be impor-

tant for design of beach fills. (See Sections 5.3 and 6.3.) Permeability
and porosity affect energy dissipation (Bretschneider and Reid, 1954; Bret-
schneider, 1954) and wave runup. (See Section 7.21; and Savage, 1958.)
Sediment properties are physically most important in determining fall

velocity and the hydraulic roughness of the sediment boundary. Fall velo-
city effects are important in onshore-offshore transport. Hydraulic rough-
ness effects have been insufficiently studied, but they appear to affect
initiation of sediment transport and energy dissipation. This may be par-
ticularly true for flow of the swash on the plane surface of the foreshore.
(Everts, 1972.)
4.522 Initiation of Sediment Motion Considerable hydraulic and coastal

.
engineering research has been devoted to the initiation of sediment motion

under moving water. From this research has come general agreement (Graf,
1971, Chapter 6; Hjulstrom, 1939; and Everts, 1972) that the initiation of
motion on a level bed of fine or medium sand requires less shear (lower
velocities) than the initiation of motion on a level bed of silt or gravel;
(Figure 4-7 for size classes). It is also generally agreed that critical
entraining velocities for sand are usually less than 1 foot per second.
Velocities of wave-induced water motion in the offshore zone can be

estimated fairly well from the equation of small -amplitude theory. (See
Chapter 2.) This theory leads to Equation 2-13 which can be transformed
into a dimensionless expression for velocity at the sand surface (z = -d)
max ( j-> TT , ^
y '^>
,
(4-19)
H sinh 2^<y^
which is plotted in dimensionless form in Figure 4-20 and for common values
of wave period in Figure 4-21. Figures 4-20 and 4-21 give maximum bottom
particle velocity, îmax{-d) > ^^ ^ function of depth, wave period, and
local wave height. This prediction by linear theory for the offshore zone,
and the related results from solitary-wave theory for the zone near break-
ing, agree fairly well with measurement in the field. (Inman and Nasu,
1956; and Cook and Gorsline, 1972, Figures 5 and 6.).
4-61
I.Oi)|- II 1 !
1
; t 1
'Q
0.8
0.6
0.5
0.4
0.3
0.2
0.1
0.08
0.061-
J>
0.05
- 0.04g
o 0.03
>
I 0.02
0.01
0.0081-
0.006
0.005
0.004
0.003 =
0.002
0.001
5000
4000
3000
18 20 22 24
(dimensionless)
Figure 4-21. Maximum Bottom Velocity from Small Amplitude Theory
4-63
************** EXAMPLE PROBLEM ***************
GIVEN: A wave in depth d = 200 feet, with period T = 9 seconds, and a
maximum bottom velocity umaxr-d) 2: 1.0 foot per second.
FIND : The minimxim wave height that will create the given bottom velocity.
SOLUTION : Calculate
Lo =
lit
= 5.12(9)2
= 415 feet .
and
200
Lo 415
= 0.482
Entering Figure 4-20 with d/Lo = 0.482, determine
u T
= 0.30
H ,
"ma.C(_jjT
H
0.30
H =(l)(2) = 3ofeee.
0.30
Thus a 30-foot minimum wave height with a 9-second wave period is needed
to create a bottom velocity equal to or greater than 1 foot per second
in 200 feet of water. Alternatively, a curve for a 9-second period can
be interpolated in Figure 4-21 and uma«(-tf)T/H can be read from the
curve's intersection with the 200-foot depth.
*************************************
As a wave moves into shallower water, both bottom velocity and size
of water-particle orbit increase. For some low velocities where the hori-
zontal orbit is small, sand moves as individual grains rolling across the
surface, but for most conditions, once quartz sand begins to move, ripples
form (Kennedy and Falcon, 1965; Carstens, et al., 1969; Cook, 1970). Thus,
the initiation of sediment motion is usually indicated by the formation
of sediment ripples.
4-64
Figure 4-22, from Inman (1957) and including data from two earlier
studies, shows the velocity needed to start motion in a sediment bed of
a given grain size. These results are in general agreement with other
studies relating critical velocity to grain size. Also shown in this
figure are maximum velocities above which ripples tend to be smoothed
off, in qualitative agreement with conditions for bed forms in unidirec-
tional flows. (Southard, 1972.)
From Figure 4-22, it appears that maximum wave-induced bottom veloc-

ities between 0.4 and 1.0 foot per second are sufficient to initiate sand
motion imder waves. In field studies, Inman (1957) found that ripples are
always present whenever computed maximum velocities exceed 0.33 foot per
second, and Cook and Gorsline (1972) report ripples above a velocity range
of 0.5 to 0.6 foot per second. Equation 4-19 can be used to determine the
combination of wave conditions and depth that produces any given critical
value of Umax at the bottom. Figure 4-23 shows the relation between depth
and wave height, for given wave periods, for a critical velocity, vmax =
0.5 foot per second.
4.523 Seaward Limit of Significant Transport . Figures 4-20 through 4-23

and a knowledge of offshore wave climate suggest that waves can move bot-
tom sediments over most of the Continental Shelf (to depths of 100 to 400
feet or more) during some time of the year. Geologic studies indicate
that fine material has been winnowed from the surficial sediments over
much of the shelf. (Shepard, 1963; and Dietz, 1963.) The question is,
what is the maximum depth to which the rate of sand movement is signifi-
cant in coastal engineering? This section discusses field data that
supply partial answers to this question.
a. Bathymetry. Dietz (1963) and others point out that waves rework
nearshore sands, smoothing out irregularities by longshore and onshore-
offshore transport. This smoothing produces a quasi -equilibrium surface
in the nearshore zone which forms relatively straight contours, nearly
parallel to the shoreline.
Most bathymetric charts with closely spaced contours as illustrated

by Figures 4-24 and 4-25 show that isobaths near the shore run parallel
to the shoreline; further offshore, the contours may indicate linear
shoals (Duane, et al., 1972), or other irregular submarine features.
Following the idea of Dietz (1963), the depth below which shore-
parallel contours give way to irregular contours is assumed to mark the
local transition between the nearshore zone where sands are moved by the
waves in significant quantities and the offshore zone where sand is moved
in lesser quantities. Possible exceptions to this shore-parallel contour
rule are the contours around river deltas and inlets, or where reefs and
ledges occur in the nearshore zone.
4-65
o
•H
4-1
^
(1)
rH
CX
ex.
•H
m
o
C
o
•H
•H
4J
•H
C
CM
CM
01
U
3
(08S/U) Tl /îlooiSA
4-66
1000
' 1
i
—
r-
800 E
600
500 : :.. .
400
300
200
100
80
- 60
S 50
"'-
.C
40
i- 30
20
10
6
5
4
2
e5'>4e
30'I0
Contour Intervol 5 feet (from Dietz, 1963, p. 984)

up to 40, 2 feet thereofter Feet
2000 4000
Figure 4-24. Nearshore Bathymetry with Shore-Parallel

Contours off Panama City, Florida
Bathymetry, such as that in Figures 4-24 and 4-25, suggests that the
depth to the deepest shore-parallel contour is usually constant along the
shore for distances of several miles, but that this depth may vary with
longshore distances of about 10 miles. (See Figure 4-25.) The depth to
the deepest shore-parallel contour may depend on the contour spacing, but
this is not important if contour intervals are small relative to the total
depths involved. In general, the deepest shore-parallel contour is between
15 and 60 feet. In most localities, this depth is somewhat deeper than
that at which nearshore profiles are presumed to close-out. These near-
shore contours probably reflect longer term adjustment to extreme storms
that occur rarely during the typical time interval between repetitive
nearshore profile surveys.
b. Size Distribution Geologic studies (Milliman, 1972; and Curray,

.
1965) suggest that littoral sands grade seaward into finer materials before
the relatively coarse sands of the shelf are reached. In some places the
boundary between the coarser shelf sediment and this finer material is quite
sharp. (Pilkey and Frankenberg, 1964.) The finer material is currently
4-68
4-69
interpreted as bounding the seaward edge of sediment moved by waves in sig-
nificant quantities. This band of finer material suggests that there is
little exchange between littoral and shelf sands in most places.
c. Sand Budget Balancing . Onshore sand transport has been suggested

as a source of sand for several coastal localities that lack other obvious
sand sources. (Shepard, 1963, pp. 176-177; and Pierce, 1969.) For example.
Pierce suggests that the offshore must supply 440,000 cubic yards per year
of sand to the southern segment of the Outer Banks of North Carolina be-
cause other known sources do not balance the budget.
d. Transport in Nearshore Zone . Although theoretical, experimental,

and field data show that waves move sand some of the time over most of the
Continental Shelf, most of the data suggest that sand from the shelf is not
a significant contributor to the sediment budget of the littoral zone.
Sand transport from the nearshore zone is more likely. Surveys show
that sand in the nearshore zone does move, although it is difficult to meas-
ure direction of motion. The presence of shore-parallel contours along most
open shores is evidence that the waves actively mold the sand bottom in the
nearshore zone. However, the time scale of transport in the nearshore zone
may be relatively slow.
In tests at Santa Barbara, California, and at Atlantic City and Long

Branch, New Jersey, dredged sands were dumped offshore in depths ranging
from 15 to 40 feet, but no measurable onshore migration of the sand re-
sulted for times of about 1 year. (Hall and Herron, 1950.) Radioactive
tracers have shown that gravel moves slowly landward in 30 feet of water
at a rate of about 0.5 cubic yard per year per foot of beach. (Crickmore
and Waters, 1972.)
At shallower depths in the nearshore zone, onshore sand transport

following storms is well documented. Transport of sediment suspended
over ripples by the mass transport velocity is more than adequate to
return sand eroded from the beach or to transport sand eroded from the
nearshore bottom to the beach.
e. Summary on Seaward Limit The deepest shore-parallel contour

.
appears to be a usable estimate of the maximum depth where significant

sand transport can be expected. This depth varies from 15 feet to per-
haps 60 feet or more along U.S. coasts. This choice may be modified for
specific wave conditions using Figure 4-23 to find the depth where maxi-
mum wave-induced velocity first exceeds 0.5 foot per second. If this
depth is less than the depth of the deepest shore-parallel contour, it
should be used as the seaward limit for the given wave conditions,
4.524 Beach Erosion and Recovery .
a. Beach Erosion . Beach profiles change frequently in response to

winds, waves, and tides. Profiles are also affected by events in the long-
shore direction that influence the longshore transport of sand. The most
4-70
notable rapid rearrangement of a profile is by storm waves, especially
during storm surge (Section 3.8) which enables the waves to attack at
higher elevations on the beach, (see Figure 1-7.)
The part of the beach washed by runup and runback is the beach face.
Under normal conditions, the beach face is contained within the fore-
shore, but during storms, the beach face is moved shoreward by the cut-
ting action of the waves on the profile. The waves during storms are
steeper, and the runback of each wave on the beach face carries away more
sand than is brought to the beach by the runup of the next wave. Thus
the beach face migrates landward, cutting a scarp into the berm. (See
Figure 1-7.)
In moderate storms, the storm surge and accompanying steep waves will
subside before the berm has been significantly eroded. In severe storms, or
after a series of moderate storms, the backshore may be completely eroded,
after which the waves will begin to erode the coastal dunes, cliffs, or main-
land behind the beach.
The extent of storm erosion depends on wave conditions, storm surge,

the stage of the tide and storm duration.
Potential damage to property behind the beach depends on all these

factors and on the volume of sand stored in the beach-dune system when a
storm occurs.
For planning and design purposes, it is useful to know the magnitude

of beach erosion to be expected during severe storms. Table 4-5 tabulates
the effect of four notable extratropical storms along the Atlantic coast
of the U.S. This table provides information on typical observed order-of-
magnitude values for beach erosion above mean sea level (MSL) from single
storms.
For the storm of 17 December 1970, information is available from

seven localities (Column 2 of Table 4-5). (DeWall, et al. , 1971.) The
three other storms include two closely spaced storms affecting Jones Beach,
New York, in February 1972 (Everts, 1972), and a storm that affected the
northern New Jersey coast in November 1953. (Caldwell, 1959.) Character-
istics that distinguish one storm from another are duration and storm
surge. (See Columns 9 and 10, Table 4-5.) Storm waves lasted about 1 day
for the 17 December 1970 storm, and about 2 days for the other three
storms. (See Column 9.) Storm surge elevation varied from a low of 2.8
feet to a high of 6 feet in the New York Bight area. The November 1953
storm combined longer duration and high storm surge; the 17 December 1970
storm had short duration and moderate storm surge; and the February 1972
storms both had longer duration, one with moderate storm surge and the
other with low storm surge.
Duration and storm surge (Columns 9 and 10) are consistent with storm
damage data (Columns 11 through 16, although the effect is influenced by
the choice of profiles included in each study. The December 1970 storm
4-71
includes all profiles surveyed. The two February 1972 storms at Jones
Beach include all profiles away from the influence of inlets, reducing
the number of profiles from 15 in the December 1970 storm to 10 in the
February 1972 storms. The November 1953 storm gives relatively high
storm damage which may partly result from the long time interval between
pre-storm survey and the storm. (See Columns 1 and 3.) These results
are also affected by the fact that they omit some profiles "believed to
be influenced by the presence of a seawall or a bulkhead," (Caldwell,
1959, p. 4.)
Although the data in Table 4-5 are not completely comparable, the
results do suggest that the average volume of sand eroded above mean sea
level from beaches about 5 or more miles long has a certain range of val-
ues. A moderate storm may remove 4 to 10 cubic yards per foot of beach
front above MSL; an extreme storm (or a moderate storm that persists for
a long time) may remove 10 to 20 cubic yards per foot; rare storms that
are most destructive in beach erosion due to a combination of intensity,
duration, and orientation may remove 20 to 50 cubic yards per foot. These
values are average for beaches 5 to 10 miles or more long, and they are
compatible with other, less complete, data for notable storms. (Caldwell,
1959; Shuyskiy, 1970; and Harrison and Wagner, 1964.) For comparative
purposes, a berm 100 feet wide at an elevation of 10 feet MSL contains 37
cubic yards per foot of beach front, a quantity that would be adequate
except for extreme storms.
In terms of horizontal changes rather than the volume changes in

Table 4-5, a moderate storm can erode a typical beach 75 to 100 feet or
more, and leave it exposed to greater erosion if a second storm follows
before the beach has recovered. This possibility should be considered
in design and placement of beach fills and other protective measures.
Extreme values of erosion may be more useful than mean values for
design. Column 17 of Table 4-5 suggests that the ratio of the most ero-
ded above-MSL-profile to the average profile for east coast beaches ranges
from about 1.5 to 6. If the average erosion per profile is based only on
those profiles showing net erosion, then this ratio is probably between
1.5 and 3.
Although the dominant result of storms on the above MSL part of

beaches is erosion, most post-storm surveys show that the storm produces
local accretion as well. Of the 90 profiles from Cape Cod, Massachusetts,
to Cape May, New Jersey, surveyed immediately after the December 1970
storm, 16 showed net accretion above mean sea level. (Compare Columns 4,
11, and 12 in Table 4-5.) Similar results are indicated for a number of
more severe storms. (Caldwell, 1959.)
The storm surveys also show that the shoreline on many beaches may
prograde seaward even though the profile as a whole loses volume, or
vice versa. This possibility suggests caution in interpreting aerial
photos of storm damage. (Everts, 1973.)
4-74
b. Beach Recovery . The typical beach profile left by a severe storm
is a simple, con cave -upward curve extending seaward to low tide level or
below, (See top of Figure 4-26.) The sand that has been eroded from the
beach is deposited mostly as a ramp or bar in the surf zone that exists at
the time of the storm. Immediately after the storm, beach repair begins
by a process that has been documented in detail, (e.g., Hayes, 1971; Davis,
et al., 1972; Davis and Fox, 1972; and Sonu and van Beek, 1971.) Sand that
has been deposited seaward of the shoreline during the storm begins moving
landward as a sandbar with a gently sloping seaward face and a steeper land-
ward face. (See Figure 4-26.) These bars have associated lows (riinnels)
on the landward side and occasional drainage gullies across them. (King,
1972, p. 339.) These systems are characteristic of post-storm beach accre-
tion under a wide range of wave, tide, and sediment conditions. (Davis,
et al., 1972.) They are sometimes called ridge -and- runnel systems.
The processes of accretion occur as follows. Sand is transported

landward over the nearly flat seaward face of the bar by the waves. At
the bar crest, the sand avalanches down the landward slip face. If the
process continues long enough, the bar reaches the landward limit of
storm erosion where it is "welded" onto the beach. (e.g. Davis, et al.,
1972.) Further accretion continues by adding layers of sand to the top
of the bar which, by then, is a part of the beach. (See Figure 4-27.)
Berms may form immediately on a post-storm profile without an inter-

vening bar-and-trough, but the mode of berm accretion is quite similar
to the mode of bar-and-trough growth. Accretion occurs both by addition
of sand laminae to the beach face (analogous to accretion on the seaward-
dipping top of the bar in the bar-and-trough) and by addition of sand on
the slight landward slope of the berm surface when waves carrying sedi-
ment overtop the berm crest (analogous to accretion on the landward dip-
ping slip face of the bar). This process of berm accretion is also illus-
trated in Figure 4-1.
The rate at which the berm builds up or the bar migrates landward to
weld onto the beach varies greatly, apparently in response to: wave con-
ditions, beach slope, grain size, and the length of time the waves work
on the bars. (Hayes, 1971.) Compare the slow rate of accretion at Crane
Beach in Figure 4-26 (mean tidal range 9 feet, spring range 13 feet) with
the rapid accretion on the Lake Michigan shore in Figure 4-2 7 (tidal range
less than 0.25 foot).
Post-storm studies by CERC show that the rate of post-storm replenish-

ment by bar migration and berm building is usually rapid immediately after
the storm. T^his rapid buildup is important in evaluating the effect of
severe storms because (unless surveys are made within hours after the
storm) the true extent of erosion during the storm is likely to be ob-
scured by the post-storm recovery. Lack of surveys before the start of
post-storm recovery may explain some survey data that show MSL accretion
on profiles that have lost volume.
4-75
Summer Accretion 29 May -7 September 1967
Station CBA, Crane Beach
Ipswich, Massachusetts
^»
-29 Moy
\ -I 3 June
\
^ __M_LW_
**
— I July
\ .
>•
,\ I July
-20 July
\
-24 July
-"3 Aug
\ ^.
_8 Aug
14 Aug
\
\
\ 22 Aug
10 n 30 Aug
;
5 —
'•
7Sept„LW
100 200 300 400 500

Feet (from Hayes, 1972)
Figure 4-26. Slow Accretion of Ridge- and- Runnel at Crane

Beach, Massachusetts
4-76
o
The ideal result of post-storm beach recovery is a wide backshore
that will protect the shore from the next storm. Beach recovery may be
prevented when the period between successive storms is too short. Main-
tenance of coastal protection requires knowledge of the necessary width
and elevation of the backshore appropriate to local conditions, and
adequate surveillance to determine when this natural sand reservoir is
diminished to a point where it may not protect the backshore during the
next storm.
4.525 Bar-Berm Prediction . High, steep waves scour the beach, eroding the
foreshore into a simple con cave -upwards profile. The material eroded from
the beach is deposited offshore as a longshore bar. Waves of low steepness
tend to push sand onto the beach, usually as migrating longshore-bar sys-
tems which eventually become part of the beach. In contrast to the concave-
up eroded profile discussed previously, the accreted profile is concave-
downward. Idealized eroded and accreted profiles (measured in a prototype-
scale wave tank) across the beach and nearshore zone are shown in Figure
4-28.
To design a beach that contains a reservoir of sand in the backshore

sufficient to survive a design storm, a minimum requirement is the ability
to distinguish wave conditions that cause eroded profiles from those that
cause accretion. Usually, it is assiomed that a berm characterizes an ac- :
creted profile and that a bar characterizes an eroded profile. (See Figure
4-28.) Tliis picture is somewhat idealized. A sharp berm crest between
backshore and foreshore is often lacking, and on some beaches the berm is
absent, so that the top of the foreshore reaches the dune or cliff line.
Berms are illustrated in Figures 4-1 and 4-27.
Similarly, the idealized longshore bar seaward of an eroded beach,

(middle profile of Figure 4-18) is often absent, and in its place there
may be several subdued bars or a platform extending to the breaker line
at nearly constant depth.
a. Longshore Bars . The term bar has been applied to a number of

quite different coastal features, including barrier islands (the "off-
shore bar" of Johnson, 1919), ridge-and-runnel systems, and linear shoals.
(Duane, et al., 1972.) Longshore bars are unrelated to any of these fea-
tures. They appear to most nearly resemble a ridge-and-runnel system,
but differ in that longshore bars are located at the breaker position and,
at least in part, are eroded out of the bottom by the falling breaker,
whereas the ridge-and-runnel system is an accretionary feature migrating
landward across the surf zone.
The typical longshore bar, as described from the observations of

Shepard (1950) and the experiments of Keulegan (1948) , is a ridge of sand
parallel to the shore and formed at the breaking position of large plung-
ing breakers. Longshore bars seem most directly related to the height of
larger breakers (not necessarily of maximum height). The depth to the sea-
ward bar increases with the height of the larger breakers along the Pacific
coast. (Shepard, 1950.) Bars form readily in tidal seas, but seem better
4-78
, .
developed in tideless seas such as Lake Michigan (Fig. 4-27). (Saylor
and Hands, 1970.) Keulegan (1948) found that the ratio of depth of long-
shore trough to depth of bar was approximately 1.69 in laboratory experi-
ments, but most field measurements showed less depth difference, averag-
ing about 1.23 (based on MLLW) for 276 measurements from the Scripps pier.
(Shepard, 1950.) According to Shepard, bars are not significant on slopes
steeper than 4° (1 on 14).
There is evidence that longshore bars, as described, are formed by a

transient condition when waves of a given height and period plunge on a
relatively plane sand slope. Shepard found that steep storm waves elimi-
nate the bars rather than build them, and that bars form after the largest
storm wave subsided. Such a relation is consistent with plunging breaker
conditions predicted by the breaker type parameter (Galvin, 1972). Be-
cause bars are formed by high waves, they may persist through long inter-
vals of low waves. Once formed, bars may trigger the breaking of higher
waves, dissipating the wave energy and thus reducing beach erosion. (Davis
and Fox, 1972; and Zwambom, Fromme, and Fitzpatrick, 1970.)
Laboratory observations show that longshore bars form when waves

plunge, and that these bars are absent when waves spill. With constant
wave conditions, a wave may plunge initially on a steep sand slope and
form a bar. The beach then erodes forming a flatter slope, which changes
the breaker type to spilling, which eliminates the pronounced longshore
bar. For other constant wave conditions, a wave may spill initially on
a gentle sand slope, and no pronounced bar forms. Later in the test, the
breaker position migrates closer to the steeper foreshore, the breaker
begins to plunge, and a longshore bar forms.
b. Steepness Effect . The distinction between profiles with pro-

nounced berms (usually without bars) and profiles with eroded foreshores
(often with longshore bars) is well known. (Nayak, 1970, and Johnson,
1956.) Early laboratory results suggested that the shape of the profile
depends on deepwater steepness, H^/Lo, of the waves reaching the beach.
Between 1936 and 1956, laboratory experiments were made which led to the
assumption that beach profiles generally eroded if Hq/ô exceeded 0.025
and accreted if ho/ho was less than about 0.02.
However, neither field data nor prototype-size laboratory experiments

support this widely used criterion. (Saville, 1957.) Field and prototype-
size laboratory data of Saville showed that beaches eroded at signifi-
cantly lower deepwater steepness than the value of 0.025 derived from
model laboratory experiments. Saville (1957) concluded that the absolute
size of wave height was probably as important as steepness in determining
the profile.
4-80
c. Dimensionless Fall Time. Prediction of accreted (berm) or erod-
ed (bar) profiles is possible using a dimensionless fall-time parameter
Ho
F^
O
= ,
(4-20)
VrT
where H^ is deepwater wave height, T is wave period, and Vj? is the

fall velocity of the beach sediment. (Dean, 1973.)
The fall-time parameter Vq, is plotted against deepwater steep-

ness, Ho/Lo, in Figure 4-29 using the profile data of Rector (1954, Table
1, Column 25) and Saville (1957, unpublished). These data include wave
heights ranging from about 0.05 foot to 5.0 feet, a range compatible with
field conditions. They also include a range of initial slopes.
In Figure 4-29, the line of demarcation between deposition offshore

and deposition onshore is approximately at the value F^ = 1. More com-
plete separation is possible when F^ is plotted against H^/Mj. (See
Figure 4-30.)
Values of F^ > 1 indicate that the time required by the particle

to fall a distance about equal to the maximum depth in the surf zone is
greater than the time available between arriving wave crests. Thus,
values of F,^ significantly greater than 1 suggest significant concen-
trations of suspended sediment, which are expected to diffuse seaward
and deposit offshore.
Since values of V^ range from about 0.066 foot per second (2 centi-
meters per second) for fine sand to 0.49 foot per second (15 centimeters
per second) for coarse sand (Fig. 4-31), ¥q ranges from about 0.25 for
low swell on coarse sand beaches to 10 or more for storm waves on fine
sand beaches. Such values suggest that typical field and laboratory con-
ditions define a range of conditions within which the importance of sus-
pended sediment in the surf zone may vary from significant to negligible.
The effect of temperature on fall velocity (Fig. 4-31) is important

enough to be critical under some conditions. It appears that temperature
by itself, in its effect on fall velocity, can change a profile from erod-
ing to accreting.
To summarize results on berm-bar criteria, the dimensionless fall

time, Vq = H^/(Vy?T), Equation 4-20 provides an estimate of the separa-
tion between berm and bar-type profiles. A value of V^ between 1 and 2
appears to be the critical value, with F^ = 2 being more appropriate for
prototype-size waves. For ¥q less than the critical value, the beach
accretes above MLW. The effect of fall velocity is important. Deepwater
wave steepness by itself is an unreliable criterion for prototype condi-
tions.
4-81
0Q t^^^H?^:r':T[-i-t-^4-::||;::ra i -<i'r: l i
ll;j» !
0.001 0.002 0.004 0.01 0.02 0.04 0.07 0.1 0.2 0.4 0.7 I
( Kohler and Galvin,l973)

Deepwater Steepness, Hq/Lq
Figure 4-29. Berm-Bar Criterion Based on Diraensionless Fall

Time and Deep Water Steepness
4-82
\KJKJ
a|039 dAdi9 pjopuois Jd|Aj^
^m (DrôoOkOCM^too^wcMinciifloOin
— — — — cvj<Mc\jm»o^îo*o
4.526 Slope of the Foreshore . The foreshore is the steepest part of the
beach profile. The equilibrium slope of the foreshore is a useful design
parameter, since this slope, along with the berm elevation, determines
minimum beach width.
The slope of the foreshore tends to increase as the grain size in-
creases. (U.S. Army, Beach Erosion Board, 1933; Bascom, 1951; and King,
1972, p. 324.) This relationship between size and slope is modified by-
exposure to different wave conditions (Bascom, 1951; and Johnson, 1956);
by specific gravity of beach materials (Nayak, 1970; and Dubois, 1972);
by porosity and permeability of beach material (Savage, 1958); and prob-
ably by the tidal range at the beach. Analysis by King (1972, p. 330)
suggests that slope depends dominantly on sand size, and also signifi-
cantly on an unspecified measure of wave energy.
Figure 4-32 shows trends relating slope of the foreshore to grain

size along the Florida Panhandle, New Jersey-North Carolina, and the
U.S. Pacific coasts. Trends shown on the figure are simplifications of
actual data, which are plotted in Figure 4-33. The trends show that,
for constant sand size, slope of the foreshore usually has a low value
on Pacific beaches, intermediate value on Atlantic beaches, and high
value on Gulf beache*:.
This variation in foreshore slope from one region to another appears

to be related to the mean nearshore wave heights. (See Figures 4-10, 4-11,
and Table 4-4.) The gentler slopes occur on coasts with higher waves. An
increase in slope with decrease in wave activity is illustrated by data
from Half Moon Bay (Bascom, 1951), and is indicated by the results of King
(1972, p. 332).
The inverse relation between slope and wave height is partly caused
by the relative frequency of steep or high eroding waves which produce
gentle foreshore slopes and low accretionary post-storm waves which pro-
duce steeper beaches. (See Figures 4-1, 4-26, and 4-2 7.)
The relation between foreshore slope and grain size shows greater
scatter in the laboratory than in the field. However, the tendency for
slope of the foreshore to increase with decreasing mean wave height is
supported by laboratory data of Rector (1954, Table 1). In this labora-
tory data, there is an even stronger inverse relation between deepwater
steepness, Wq/Lo, and slope of the foreshore than between Wq a^d the
slope.
To summarize the results on foreshore slope for design purposes, the

following statements are supported by available data:
(a) Slope of the foreshore on open sand beaches depends principally

on grain size, and (to a lesser extend) on nearshore wave height.
(b) Slope of the foreshore tends to increase with increasing median

grain size, but there is significant scatter in the data.
4-85
adois
lO
o
CVJ
CO
"~
adois
o
OJ
lO
00
d
o
(c) Slope of the foreshore tends to decrease with increasing wave
height, again with scatter.
(d) For design of beach profiles on ocean or gulf beaches, use

Figure 4-32, keeping in mind the large scatter in the basic data on Fig-
ure 4-33, much of which is caused by the need to adjust the data to
account for differences in nearshore wave climate.
4.53 LONGSHORE TRANSPORT RATE
4.531 Definitions and Methods Littoral drift is the sediment (usually

.
sand) moved in the littoral zone imder action of waves and currents. The
rate, Q, at which littoral drift is moved parallel to the shoreline is
the longshore transport rate. Since this movement is parallel to the
shoreline, there are two possible directions of motion, right and left,
relative to an observer standing on the shore looking out to sea. Move-
ment from the observer's right to his left is motion toward the left, in-
dicated by the subscript It. Movement toward the observer's right is
indicated by the subscript rt.
Gross longshore transport rate, Qg, is the sum of the amounts of

littoral drift transported to the right and to the left, past a point on
the shoreline in a given time period.
Qg = Qrt + Qfit ' (4-21)
Similarly, net longshore transport rate, Q^, is defined as the

difference between the amounts of littoral drift transported to the right
and to the left past a point on the shoreline in a given time period.
%=%t- Q£. •
C4-22)
The quantities 0^.^, Q£t> On ^îd Q^ have engineering uses: for

example, Qg is used to predict shoaling rates in uncontrolled inlets;
Qj^ is used for design of protected inlets and for predicting beach ero-
sion on an open coast; Q^,^ and Q^^ are used for design of jetties and
impoundment basins behind weir jetties. In addition, Q^ provides an
upper limit on other quantities.
Occasionally, the ratio
7 = —
^1rt
, (4-23)
is known, rather than the separate values Q^^ and Then Q^ is

related to Q^ in terms of y by
Qj,^.
^
(1 +7)
—
Q„ = Q„
^^
(4-24)
This equation is not very useful when y approaches 1.
4-88
Longshore transport rates are usually given in units of volume per
time (cubic yards per year in the U.S.). Typical rates for oceanfront
beaches range from 10^ to 10^ cubic yards per year. (See Table 4-6.)
These volume rates typically include about 40 percent voids and 60 per-
cent solids.
At present, there are four basic methods to use for the prediction
of longshore transport rate:
1. The best way to predict longshore transport at a site is to

adopt the best known rate from a nearby site, with modifictions based
on local conditions.
2. If rates from nearby sites are unknown, the next best way to
predict transport rates at a site is to compute them from data showing
historical changes in the topography of the littoral zone (charts, sur-
veys, and dredging records are primary sources).
3. If neither Method 1 nor Method 2 is practical, then it is

accepted practice to use either measured or calculated wave conditions
to compute a longshore component of "wave energy flux" which is related
through an empirical curve to longshore transport rate. (Das, 1972.)
4. A recently developed empirical method (Galvin, 1972) is avail-

able to estimate gross longshore transport rate from mean annual near-
shore breaker height. The gross rate, so obtained, can be used as an
upper limit on net longshore transport rate.
Method 1 depends largely on engineering judgement and local data.

Method 2 is an application of historical data, which gives usable answers
if the basic data are reliable and available at reasonable cost, and the
interpretation is based on a thorough knowledge of the locality. By
choosing only a few representative wave conditions, I-tethod 3 can usually
supply an answer with less work than Method 2, but with correspondingly
less certainty. Because calculation of wave statistics in Method 3 follows
an established routine, it is often easier to use than researching the
hydrographic records and computing the changes necessary for Method 2.
Method 4 requires mean nearshore breaker height data. Section 4.532
utilizes Methods 3 and 4. Methods 1 and 2 are discussed in Section 4.8.
4.532 Energy Flux Method . Method 3 is based on the assumption that long-
shore transport rate, Q, depends on the longshore component of energy flux
in the surf zone. The longshore energy flux in the surf zone is approxi-
mated by assuming conservation of energy flux in shoaling waves, using
small -amplitude theory, and then evaluating the energy flux relation at
the breaker position. The energy flux per unit length of wave crest, or,
equivalently, the rate at which wave energy is transmitted across a plane
of unit width perpendicular to the direction of wave advance is (from
Section 2.238, combining Equations 2-39 and 2-40):
P = EC = ^ H2 C . (4-25)
4-89
Table 4-6. Longshore
If the wave crests make an angle, a with the shoreline, the energy flux
in the direction of wave advance per unit length of beaah is
P cos a = —— C
PgH^
cos a , (4-26)
8 *
and the longshore component is given by
—
?£ = P cos a sin a = Pg
-^ H 1 C cos a sin a ,
8 *
or since cos a sin a = 1/2 sin 2a
Pp = 77 H^ C sin 2a . (4-27)
* lo *
The surf-zone approximation of Pj^, is written as P^.^.
= (4-28)
Pfi.
ll^S '^2ai, .
Usable formulations of this surf- zone approximation can be obtained by

several methods. The principal approximation is in evaluating C^ and H
at the breaker position. It is standard practice to approximate the group
velocity, Cg, by the phase velocity, C, at breaking. The phase velo-
city may then be approximated by either linear wave theory (Equation 2-3)
or by solitary wave theory (Equation 4-13).
Figure 4-34 presents the longshore component of wave energy flux in

a dimensionless form, P^/pg^ H? T, as a function of breaker steepness,
Hi/gT^, and the angle the wave crest makes with the shoreline in either
deep water a^, or at the breaker line, ai^. Figure 4-34 is based on
Equation 4-28 using linear wave theory to determine C^ and assuming that
refraction is by straight parallel bottom contours. Figure 4-35 can be
used to determine the longshore component of wave energy flux when breaker
height, H^,, period, T, and angle, aj,, are known- -for example, for
surf observation data. The use of Figure 4-34 is illustrated by an
example problem below.
For linear theory, in shallow water, C^ « C and
P«. = ^ Hfc C sin 2aj, , (4-29)
where H^j and a^ are the wave height and direction at breaking and C
is the wave speed from Equation 2-3, evaluated in a depth equal to 1.28
Hi.
4-91
4" 6" 8' 10' 20° 60° 80° 100°
Oeepwater Wave Angle, Qo, degrees
Figure 4-34. Longshore Component of Wave Energy Flux in Dimensionless Form

as a Function of Breaker Conditions
4-92
************** EXAMPLE PROBLEM ***************
GIVEN A breaking wave with height, Hj, = 4 feet; period T = 7 seconds.
:
Surf observations indicate that the wave crest at breaking makes an

angle, ajy = 6° with the shoreline.
FIND ;
(a) The longshore component of wave energy flux.
(b) The angle the wave made with the shoreline when it was in deep
water, Uq.
SOLUTION : Calculate
Hfe 4.0
-— = r = 0.00254 .
gT^ 32.2 (7.0)2
Enter Figure 4-34 to the point where the line labeled aj, = 6" crosses
H2,/gT2 = 0.00254 and read Pji/pg^ "^ = 6.7 x lO"** from the left axis
"f
and a(-) = 18° from the bottom axis.
The longshore energy flux can then be calculated as,
Pg = 0.00067 pg2 Hi T ,
Pg = 0.00067 (2) (32.2)2 (4 q)2 (70) .
Pg = 155.6, say 160 ft.-lb./ft.-sec. ,
and
«o = 18° •
*************************************
For offshore conditions, the group velocity is equal to one-half the
deepwater wave speed C^, where C^ is given by Equation 2-7, and the
refraction coefficient, K^, can be determined by the methods of Section
2.32. Hence,
P£s = ^T(H^K^)2 3i„2a, . (4-30)
Figure 4-35 also presents the longshore component of wave energy flux
in a dimensionless form, P£,/(pg^ H^ T), as a function of deepwater wave steep-
ness, H^/gT^, and the angle the wave crest makes with the shoreline in
either deep water, a^, or at the breaker line, a^,. Refraction by
straight parallel bottom contours is again assumed. As illustrated by
4-93
E
oo
ID
(i,°H,H/''d
4-94
the example problem below. Figure 4-35 can be used to determine the
longshore component of wave energy flux when deepwater wave height, H^,
period, T, and deepwater wave angle, a^, are known.
************** EXAIIPLE PROBLEM

***************
GIVEN :A wave in deep water has a height, H^ = 5 feet and a period,
T = 7 seconds. While in deep water, the wave crest makes an angle.
a^ = 25 with the shoreline.
FIND:
(a) The longshore component of wave energy flux.
(b) The angle the wave makes with the shoreline when it breaks [assum-
ing refraction is by straight, parallel bottom contours).
SOLUTION: Calculate,
Ho 5.0
= 0.0032
gT^ 32.2 (7.0)
Enter Figure 4-35 with a^, = 25° and H^/gT^ = 0.0032, read P^/pg^H^ T =
1.5 X 10"^. This corresponds with a breaker angle a^, = 9.5° which is
obtained by interpolation between the dashed curves of constant aj,.
Tlierefore,
Pj = (0.0015)pg^H2T ,
Pj = (0.0015) (2) (32.2)2 (5)2 (7 q) ,
Pj = 544.3 , say 540 ft.-lb./ft.-sec. ,
and
afc = 9.5° .
**************************************
Equations 4-25 and 4-28 are valid only if there is a single wave
train with one period and one height. However, most ocean wave condi-
tions are characterized by a variety of heights with a distribution usu-
ally described by a Rayleigh distribution. (See Section 3.22.) For a
Rayleigh distribution, the correct height to use in Equation 4-28 or in
the formulas shown in Table 4-8 is the root-mean-square amplitude. How-
ever, most wave data are available as significant heights, and coastal
engineers are used to dealing with significant heights.
Significant height is implied in all equations for P^g. The value

of Pj^g computed using significant height is approximately twice the
4-95
value of the exact energy flux for sinusoidal wave heights with a Rayleigh
distribution. Since this means that P^g is proportional to energy flux
and not equal to it, P^s is referred to as the longshore energy flux
factor in Table 4-8 and the following sections.
Longshore energy flux in this general case (P^) is given by equa-

tion 4-27. This is an exact equation for the longshore component of
energy flux in a single small -amplitude, periodic wave. This equation
is good for any specified depth, but because the wave refracts, P^ will
have different values as the wave moves into shallower water. The value
of Pj, in Equation 4-27 can be manipulated through use of small -an^Jlitude
wave theory to obtain the four equivalent formulas for Pjj, shown in Table
4-7.
In order to use Pj, for longshore transport computations in the surf

zone, it is necessary to approximate P^ for conditions at the breaker
position. These approximations are shown as P^g in Table 4-8, evaluated
in foot-pound per second units. The bases for these approximations are
shown in Table 4-9. Measurements show that the longshore transport rate
depends on P^g. (See Figures 4-36 and 4-37.)
As implied by the definition of Pusy the energy flux factors in

Figures 4-36 and 4-37 are based on significant wave heights. The plotted
P£g values were obtained in the following manner. For the field data of
Watts (1953) and Caldwell (1956) , the original references give energy flux
factors based on significant height, and these original data (after unit
conversion) are plotted as Pj^g in Figures 4-36 and 4-37. Similarly, the
one field point of Moore and Cole (1960), as adopted by Saville (1962),
is assumed to be based on significant height. (See Figure 4-37.) Finally,
the field data of Komar (1969), are given in terms of root-mean-square
energy flux. This energy flux is multiplied by a factor of 2 (Das, 1972),
converted to consistent units, and then plotted in Figure 4-36 and 4-37.
For laboratory conditions (Fig. 4-36 only), waves of constant height

are assumed. When these heights are used in the equations of Table 4-8,
the result is an approximation of the exact longshore energy flux. In
order to plot the laboratory data in terms of an energy-flux factor con-
sistent with the plotted field data, this energy flux is multiplied by 2
before plotting in Figure 4-36.
For the purpose of this section, it is assumed that the shoaling co-
efficient. Kg, for nearshore breaking waves is equivalent to the breaker
height index, H^^/Ho', found from observation. (See Figure 2-65.)
The choice of equations to determine P^g depends on the data avail-

able. The right hand columns of Tables 4-7 and 4-8 tabulate the data
required to use each of the formulas. An example using the second P^g
formula is given in Section 4.533.
Possible changes in wave height due to energy losses as waves travel

over the Continental Shelf are not considered in these equations. Such
4-96
Table 4-7. Longshore Energy Flux.Pg , for a Single Periodic Wave in
Any Specified Depth. (Four Equivalent Expressions from
Small-Amplitude Theory)
Equation
Table 4-9. Assumptions for Pj Formulas in Table 4-8.
1. Formula 1 — Equation 4-35
a. Energy density at breaking is given by linear theory,
E = (pg4)/8 « SHg
b. Group velocity equals wave speed at breaking, and breaking speed is given by
solitary wave theory according to the approximation. (Galvin, 1967, Equation 11.)
C^ = C«(2gHj,)^/^= 8.02 (Hfe)^/^
c. a can be replaced by ay
a. Same as lb above
b. H, is related to H by refraction and shoaling coefficients, where the coefficients

are evaluated at the breaker position
c. Refraction coefficient K^^ given by small-amplitude theory; shoaUng coefficient

K assumed constant, so that
d. (Hj,)^/^ = 1.14 (cosaj'/^ h/^
if (cosaj,)'/^ = 1.0
and (Kj'/^ = 1.14
a. Refraction coefficient at breaking is given by small-amplitude theory.
a. Same as la above
b. Same as lb above
c. Same as 3a above
d. Cos ttjj = 1.0
NOTE: Constants evaluated for foot-pound-second units. Small-amplitude theory is assumed valid in
deep water. Nearshore contours are assumed to be straight and parallel to the shoreline.
4-98
10'
I
t• •
10'
o
O
ce ,0
rio^
o
a.
o For Design, use Figure 4-37

^10^
.c
V) O t^;
o>
c
o
°^ ^r O O
°
Oo
°%°
o
10' <b
o °
oo ô o
• Field Doto, Significont Height
° Loborofory Dato, Quartz Sand, Periodic Woves
10
-1 0'
10" 10 10 10' 10"
Longshore Energy Flux Factor, Pa, ft.-lbs/sec/ft. of beach front
Figure 4-36. Longshore Transport Rate Versus Energy Flux Factor for Field
and Lab Conditions
4-99
10^
a.
ro
T5
10*
o
O
o
(A
.c
I/)
c
o
10'
changes may reduce the value of P^g when deepwater wave height statis-
tics are used as a starting point for computing P^g. (Walton, 1972;
Bretschneider and Reid, 1954; and Bretschneider, 1954.)
The Equations 4-35 through 4-38 in Table 4-8 are related to the Equa-
tion for Eq previously recommended for use with this method (Caldwell,
1956, Equations 5 and 6; or the equations in Figure 2-22, page 175, CERC
Technical Report No. 4, 1966 edition) by a constant
E^ = (8.64 X 10'*)P£^ (4-39)
where E^ is in units of foot-pounds per foot per day and P^g is in

units of foot-pounds per foot per second.
The term in parenthesis for Equation 4-32 in Table 4-7 is identical

to the longshore force of Longuet-Higgins (1970a). This longshore force
also correlates well with the longshore transport rate.
The relation between Q and P^g in Figures 4-36 and 4-37 can be
approximated by
Q = (7.5X103)Pj^ (4-40)
Equation 4-40 tends to overestimate Q at the higher values of P^g

for the plotted field data, but it falls below the estimated rates com-
puted from the data of Johnson (1952). (See Das, 1972, Figure 6.) The
value of 7.5 x 10^ in Equation 4-40 is approximately twice the equiva-
lent value from the design curve of CERC Technical Report No. 4, 1966
edition, and is about 5 percent greater than the value estimated by Komar
and Inman (1970).
Judgement is required in applying Equation 4-40. Although the data

in Figures 4-36 and 4-37 appear to follow a smooth trend, the log- log
scale compresses the data scatter. For example, the average difference
between the plotted points from field data and the prediction given by
Equation 4-40 is at least 28 percent of the value of prediction (average
difference derived by Das (1972) is 42 percent). In addition, some in-
complete measurements suggest transport rates ranging from tv;o orders of
magnitude below the line (Thornton, 1969) to one order of magnitude above
the line, (Johnson, 1952.) These additional data are plotted by Das
(1972).
As an aid to computation. Figures 4-38 and 4-39 give lines of con-

stant Q based on Equation 4-40 and Equations 4-35 and 4-36 for Pis
given in Table 4-8. To use Figures 4-38 and 4-39 to obtain the longshore
4-101
transport rate, only the (H^j, a^) data and Figure 4-38, or the (H^, a^)
data and Figure 4-39 are needed. If the shoaling coefficient is signifi-
cantly different from 1.3, multiply the Q obtained from Figure 4-39 by
the factor 0.88 vlCg. (See Table 4-9, Assumption 2d.)
Figure 4-39 applies accurately only if o.q is a point value. If a^

is a range of values, for example a 45° sector implied by the direction
NE, then the transport evaluated from Figure 4-39 using a single value
of a^ for NE may be 12 percent higher than the value obtained by aver-
aging over the 45° sector implied by NE. The more accurate approach is
given in the example problem of the next section.
The imit for Q is a volume of deposited quartz sand (including

voids in the volume) per year. Bagnold (1963) suggests using immersed
weight instead of volume in the unit for longshore transport rate (Section
4.521), since immersed weight is the pertinent physical variable related
to the wave action causing the sediment transport. Use of an immersed
weight unit does eliminate the difference between lightweight material
and quartz that occurs if volume units are used. (Das, 1972.) However,
in coastal engineering design, it is the volume and not the immersed
weight of eroded or deposited sand that is important, and since beach
sand is predominantly quartz (specific gravity 2.65), volume is directly
proportional to immersed weight. On some beaches the sand may be calcium
carbonate which has a specific gravity ranging from 2.87 (calcite) to
2.98 (aragonite) in pure form. Naturally occurring oolitic aragonite
sand with a specific gravity of 2.88 (Monroe, 1969), has an immersed
weight 14 percent greater than pure quartz sand. Since the longshore
transport Equation 4-40, with one exception (Watts, 1953; and Das, 1971,
p. 14) , is based on quartz sand, then oolitic sand beaches may have
slightly lower longshore transport rates than is suggested by comparison
with data from quartz sand beaches. However, the scatter in the data
(Fig. 4-36) makes such a specific gravity effect difficult to detect.
4.533 Energy Flux Example (Method 3) . Assume that an estimate of the

longshore transport rate is required for a locality on the north-south
coastline along the west edge of an inland sea. The locality is in an
area where stronger winds blow out of the northwest and north, resulting
in a deepwater distribution of height and direction as listed in Table
4-10o Assume the statistics were obtained from visual observations
collected over a 2-year interval at a point 2 miles offshore by seamen
aboard vessels entering and leaving a port in the vicinity. This type
of problem, based on SSMO wave statistics (Section 4.34), is discussed
in detail by Walton, (1972), and Walton and Dean (1973). Shipboard
data are subject to uncertainty in their applicability to littoral trans-
port, but often they are the only data available. It is assumed that
shipboard visual observations are equivalent to significant heights.
(Cartwright, 1972; and Walton, 1972.)
4-102
Lij 1
Qo iQO 20° 30° 40° 50° 60° 70° 80° 90°
Deepwater Angle, Of
q > ( degrees )
( Do not use averaged angle )
Figure 4-39. Longshore Transport Rate as a Function of Deepwater

Height and Deepwater Angle
4-104
Table 4-10. Deepwater Wave Heights, in Percent by Direction, off East-Facing
Coast of Inland Sea
Compass Direction
sector of wave directions. If ^(.a^) is evaluated at a^ = 45° (NE or
SE in the example problem), it will have a value 12 percent higher than
the average value for FCa^) over 45° sector bisected by the NE or SE
directions. Thus, if the data warrant a high degree of accuracy. Equa-
tion 4-43 should be averaged by integrating over the sector of directions
involved.
Table 4-11. Computed Longshore

To illustrate how values of Qa^' Ho listed in Table 4-11 were
calculated, the value of Qa^, H^ '^ here calculated for H^ = 1 and
the north direction, the top value in the first column on Table 4-11.
The direction term, FCa^) is averaged over the sector from a = 67.5°
,
to a = 90°, i.e., from NNE to N in the example. The average value of

F(aci) is found to be 0.261. H^ to the 5/2 power is simply 1 for this
case. The frequency given in Table 4-10 for H^^ = 1 and direction =
north (NW to NE) is 9 percent or in decimal terms, 0.09. This is mul-
tiplied by 0.5 to obtain the part of shoreward directed waves from the
north sector (i.e., N to NE) resulting in f = 0.09 (0.5) = 0.045. Put-
ting all these values into Equation 4-42 gives
Qf^j = 1.373 X 10^ (0.045) (1)^2 (0.261)
= 1,610 yd.Vyr. (See Table 4-11)
Table 4-11 indicates the importance of rare high waves in determin-

ing the longshore transport rate. In the example, shoreward moving 8- foot
waves occur only 0.5 percent of the time, but they account for 12 percent
of the gross longshore transport rate. (See Table 4-11.)
Any calculation of longshore transport rate is an estimate of poten-

tial longshore transport rate. If sand on the beach is limited in quan-
tity, then calculated rates may indicate more sand transport than there
is sand available. Similarly, if sand is abundant, but the shore is
covered with ice for 2 months of the year, then calculated transport
rates must be adjusted accordingly.
The procedure used in this example problem is approximate and

limited by the data available. Equation 4-42, and the other approxi-
mations listed in Table 4-11, can be refined if better data are avail-
able. An extensive discussion of this type of calculation is given by
Walton (1972).
Although this example is based on shipboard visual observations of

the SSMO type (Section 4.34), the same approach can be followed with
deepwater data from other sources, if the joint distribution of height
and direction is known. At this level of approximation, the wave period
has little effect on the calculation, and the need for it is bypassed as
long as the shoaling coefficient (or breaker height index) reasonably
satisfies the relation (Kg)^/^ = 1.14. (See Assumption 2d, Table 4-9.)
For waves on sandy coasts, this relation is reasonably satisfied. (e.g.,
Bigelow and Edmondson, 1947, Table 33; and Goda, 1970, Figure 7.)
4.534 Enyirical Prediction of Gross Longshore Transport Rate (Method 4) .

Longshore transport rate depends partly on breaker height, since as
breaker height increases, more energy is delivered to the surf zone. At
the same time, as breaker height increases, breaker position moves off-
shore widening the surf zone and increasing the cross-section area through
which sediment moves.
4-107
Galvin (1972) showed that when field values of longshore transport
rate are plotted against mean annual breaker height from the same locality,
a curve
Q = 2 X 10= H^ ,
(4-44)
forms an envelope above almost all known pairs of (Q, Hj,) , as shown in
Figure 4-40. Here, as before, Q is given in units of cubic yards per
year; Hjy in feet.
Figure 4-40 includes all known (Q, Hj,) pairs for which both Q and
H2j are based on at least 1 year of data, and for which Q is considered
to be the gross longshore transport rate, Q^, defined by Equation 4-21.
Since all other known (Q, H^,) pairs plot below the line given by Equation
4-41, the line provides an upper limit on the estimate of longshore trans-
port rate. From the defining equations for Q^ and Q^, any line that
forms an upper limit to longshore transport rate must be the gross trans-
port rate, since the quantities Qrt> Q£t> ^^^ Qn» ^^ defined in Section
4.531, are always less than or equal to Q^^.
In Equation 4-44, wave height is the only independent variable, and

the physical explanation assumes that waves are the predominant cause of
transport. (Galvin, 1972.) Therefore, where tide-induced currents or
other processes contribute significantly to longshore transport. Equation
4-44 would not be the appropriate approximation. The corrections due to
currents may either add or subtract from the estimate of Equation 4-44,
depending on whether currents act with or against prevailing wave-induced
transport.
4,535 Method 4 Example (Empirical Prediction of Gross Longshore Transport

Rate. Near the site of the problem outlined in Section 4.533, it is de-
sired to build a small craft harbor. The plans call for an unprotected
harbor entrance, and it is required to estimate costs of maintenance
dredging in the harbor entrance. The gross transport rate is a first
estimate of the maintenance dredging required, since transport from
either direction could be trapped in the dredged channel. Wave height
statistics were obtained from a wave gage in 12 feet of water at the end
of a pier. (See Columns (1) and (2) of Table 4-12.) Heights are avail-
able as empirically determined significant heights. (Thompson and Harris,
1972.) (To facilitate comparison, the frequencies are identical to the
deepwater frequencies of onshore waves in Table 4-10 for the problem of
Section 4.533. That is, the frequency associated with each H„ in Table
4-12 is the sum of the frequencies of the shoreward Hq on the correspond-
ing line of Table 4-10.)
The breaker height, H^j, in the empirical Equation 4-44 is related

to the gage height, Hg, by a shoaling-coefficient ratio, (Ks)^/(Kg)^,
where (Kg)^, is the shoaling coefficient (Equation 2-44), (H/H^ in
4-108
0^
a> 10
o.
cr.
o
CO
CO 10
c
o
10
Table 4-12. Example Estimate of Gross Longshore Transport Rate for Shore of Inland Sea
(1)
Table C-1 Appendix C) evaluated at the breaker position and i^s^g i^
the shoaling coefficient evaluated at the wave gage:
H. - H (4-45)
Kg or H/H^ can be evaluated from small -amplitude theory, if wave-period

information is available from the wave gage statistics. For simplicity,
assume shoaling-coefficient ratios as listed in Column 4 of Table 4-12.
Such shoaling coefficient ratios are consistent with the shoaling co-
efficient of Kg=1.3 (between deepwater and breaker conditions) assumed
in deriving P^g (Table 4-9) ,and with the fact that waves on the inland
sea of the problem would usually be steep, locally generated waves.
Column 5 of the table is the product fHg (Kg)^/(Kg)g. The sum

(1.06 feet) of entries in this column is assumed equivalent to the aver-
age of visually observed breaker heights. Substituting this value in
Equation 4-44, the estimated gross longshore transport rate is 226,000
cubic yards per year. It is instructive to compare this value with the
value of 262,000 cubic yards per year obtained from the deepwater example.
(See Table 4-11.) The two estimates are not expected to be the same,
since the same wave statistics have been used for deep water in the first
problem and for a 12-foot depth in the second problem. However, the numer-
ical values do not differ greatly. It should be noted that the empirical
estimate just obtained is completely independent of the longshore energy
flux estimate of the deepwater example.
In this example, wave gage statistics have been used for illustrative
purposes. However, visual observations of breakers, such as those listed
in Table 4-4, would be even more appropriate since Equation 4-44 has been
"calibrated" for such observations. On the other hand, hindcast statis-
tics would be less satisfactory than gage statistics due to the uncertain
effect of nearshore topography on the transformation of deepwater statis-
tics to breaker conditions.
4.6 ROLE OF FOREDUNES IN SHORE PROCESSES
4.61 BACKGROUND
The cross section of a barrier island shaped solely by marine hydrau-

lic forces has three distinct subaerial features: beach, crest of island,
and deflation plain. (See Figure 4-41.) The dimensions and shape of the
beach change in response to varying wave and tidal conditions (Section
4.524), but usually the beach face slopes upward to the island crest - the
highest point on the barrier island cross section. From the island crest,
the back of the island slopes gently across the deflation plain to the
edge of the lagoon separating the barrier island from the mainland. These
three features are usually present on duneless barrier island cross sec-
tions; however, their dimensions may vary.
4-III
X
o
OS
O
c
o
•H
t^
6
•H
X
o
u
0)
CO
o
ex
•H
OQ
o
•H
•H
(laai ) nsi/^ aAoqo uouDAa|3
4-112
Island crest elevation is determined by the nature of the sand form-
ing the beach, and by the waves and water levels of the ocean. The beach
and waves interact to determine the elevation of the limit of wave runup -
the primary factor in determining island crest elevation. Normally the
island crest elevation is almost constant over long sections of beach.
However, duneless barrier island crest elevations vary with geographical
area. For example, the crest elevation typical of Core Banks, North
Carolina, is about +6 feet MSL; +4 feet MSL is typical for Padre Island,
Texas; +11 feet MSL is typical for Nauset Beach, Massachusetts.
Landward of the upper limit of wave uprush or berm crest are the back-
shore and the deflation plain. This area is shaped by the wind, and in-
frequently by the flow of water down the plain when the island crest is
overtopped by waves. (e.g., Godfrey, 1972.) Obstructions which trap
wind-transported sand cause the formation of dunes in this area. (See
discussion in Section 6.4 Sand Dunes.) Beachgrasses which trap wind-
transported sand from the beach and the deflation plain are the major
agent in creating and maintaining foredunes.
4.62 ROLE OF FOREDUNES
Foredunes, the line of dunes just behind a beach, have two primary
functions in shore processes. First, they prevent overtopping of the
island during some abnormal sea conditions. Second, they serve as a
reservoir for beach sand.
4.621 Prevention of Overtopping . By preventing water from overtopping,

foredunes prevent wave and water damage to installations landward of the
dune. They also block the water transport of sand from the beach area to
the back of the island and the flow (overwash) of overtopping sea water.
Large reductions in water overtopping are effected by small increases

in foredune crest elevations. For example, the hypothetical 4-foot dune
shown in Figure 4-41 raises the maximum island elevation about 3 feet to
an elevation of 6 feet. On this beach of Padre Island, Texas, the water
levels and wave runup maintain an island crest elevation of +4 feet MSL
(about 2 feet above MHW) . This would imply that the limit of wave runup
in this area is 2 feet (the island crest elevation of 4 feet minus the
MHW of 2 feet). Assuming the wave runup to be the same for all water
levels, the 4-foot dune would prevent significant overtopping at water
levels up to 4 ft MSL (the 6-foot effective island height at the dune
crest minus 2 feet for wave runup). This water level occurs on the aver-
age once each 5 years along this section of coast. (See Figure 4-42.)
Thus, even a low dune, one which can be built with vegetation and sand
fences in this area in 1 year (Woodard et al , 1971) provides consider-
.
able protection against wave overtopping. (See Section 5.3 and 6.3.)
Foredunes or other continuous obstructions on barrier islands may

cause unacceptable ponding from the land side of the island when the la-
goon between the island and mainland is large enough to support the needed
wind setup. (See Section 3.8.) There is little danger of flooding from
4-113
O—
this source if the lagoon is less than 5 miles wide. Where the lagoon
is wider (especially 10 miles or greater) flooding from the lagoon side by
wind setup should be investigated before large dune construction projects
are undertaken.
4.622 Reservoir of Beach Sand . During storms, erosion of the beach occurs
and the shoreline recedes. If the storm is severe, waves attack and erode
the foredunes and supply sand to the beach; in later erosion stages, sand
is supplied to the back of the island by overwash. (Godfrey, 1972.)
Volumes of sand eroded from beaches during storms have been estimated
in recent beach investigations. Everts (1973) reported on two storms dur-
ing February 1972 which affected Jones Beach, New York. The first storm
eroded an average of 27,000 cubic yards per mile above mean sea level for
the 9-mile study area; the second storm (2 weeks later) eroded an average
of 35,000 cubic yards of sand per mile above mean sea level at the same
site. Losses at individual profiles ranged up to 120,000 cubic yards per
mile. Davis (1972) reported a beach erosion rate on Mustang Island, Texas,
following Hurricane Fern (September 1971), of 12.3 cubic yards per linear
foot of beach for a 1,500-foot stretch of beach (about 65,000 cubic yards
per mile of beach). On Lake Michigan in July 1969, a storm eroded an aver-
age of 3.6 cubic yards per linear foot of beach (about 29,000 cubic yards
per mile) from an 800-foot beach near Stevensville, Michigan. (Fox, 1970.)
Because much of the eroded sand is usually returned to the beach by wave
action soon after the storm, these volumes are probably representative of
temporary storm losses.
Volumes equivalent to those eroded during storms have been trapped

and stored in foredunes adjacent to the beach. Foredunes constructed
along Padre Island, Texas, and Core Banks, North Carolina, (Section 6.43
and 6.447) contain from 30,000 to 80,000 cubic yards of sand per mile of
beach. Assuming the present rate of entrapment of sand continues for the
next 3 years at these sites, sand volumes ranging from 50,000 to 160,000
cubic yards per mile of beach will be available to nourish eroding beaches
during a major storm. Sand volumes trapped during a 30-year period by
European beachgrass at Clatsop Spit, Oregon, averaged about 800,000 cubic
yards per mile of beach. Tlius, within a few years, foredunes can trap and
store a volume of sand equivalent to the volumes eroded from beaches dur-
ing storms of moderate intensity.
4.623 Long-Term Effects . Dolan, (1972-73) advances the concept that a

massive, unbroken foredune line restricts the landward edge of the surf
zone during storms causing narrower beaches and thus increased turbulence
in the surf zone. The increased turbulence causes higher sand grain attri-
tion and winnowing rates and leads to accelerated losses of fine sand, an
erosive process that may be detrimental to the long-range stability of bar-
rier islands. However, as discussed in Section 4.521, the effects of sedi-
ment size are usually of secondary importance in littoral transport pro-
cesses - processes which are important in barrier island stability. In
addition, geographical location is probably more important in determining
beach sand size than dune effects, since both fine and coarse sand beaches
4-115
front major foredune systems in different geographical locations. For
example, fine sand beaches front a massive foredune system on Mustang
Island, Texas, and coarse sand beaches front dunes on the Cape Cod spits,
Godfrey (1972) discusses the effect of a foredune system on the

long term stability of the barrier islands of the Cape Hatteras and Cape
Lookout National Seashores, North Carolina. Important implicit asstimp-
tions of the discussion are that no new supply or inadequate new supplies
of sand are available to the barrier island system, and that rising sea
level is, in effect, creating a sand deficit by drowning some of the
available island volume. The point of the geomorphic discussion is that
under such conditions the islands must migrate landward to survive. A
process called "oceanic overwash" (the washing of sand from low foredunes
or from the beach over the island crest onto the deflation plain by over-
topping waves) is described as an important process in the landward migra-
tion of the islands. Since a foredune system blocks overtopping and pre-
vents oceanic overwash, foredunes are viewed as a threat to barrier island
stability.
Granted the implicit assumptions and a geologic time frame, the

geomorphic concept presented has convincing logic and probably has merit.
However, the assumptions are not valid on all barrier islands or at all
locations in most barrier islands or at all locations in most barrier
island systems. Too, most coastal engineering projects are based on
a useful life of 100 years or less. In such a short period, geologic
processes, such as sea-level rise, have a minor effect in comparison with
the rapid changes caused by wind and waves. Therefore, the island crest
elevation and foredune system will maintain their elevation relative to
the mean water level on stable or accreting shores over the life of most
projects. On eroding shores, the foredunes will eventually be eroded
and overwash will result in shoreward migration of the island profile;
sand burial and wave and water damage will occur behind the original
duneline. Therefore, planning for and evaluation of the probable suc-
cess of a foredune system must consider the general level of the area of
the deflation plain to be protected, the rate of sea level rise, and the
rate of beach recession.
4.7 SEDIMENT BUDGET
4.71 INTRODUCTION
4.711 Sediment Budget A sediment budget is based on sediment removal,

.
transportation and deposition, and the resulting excesses or deficiencies

of material quantities. Usually, the sediment quantities are listed
according to the sources, sinks, and processes causing the additions and
subtractions. In this chapter, the sediment is usually sand, and the
processes are either littoral processes or the changes made by man.
The purpose of a sediment budget is to assist the coastal engineer
by: identifying relevant processes; estimating volume rates required for
design purposes; singling out significant processes for special attention;
and, on occasion, through balancing sand gains against losses, checking
the accuracy and completeness of the design budget.
4-116
Sediment budget studies have been presented by Johnson (1959), Bowen
and Inman (1966), Vallianos (1970), Pierce (1969), and Caldwell (1966).
4.712 Elements of Sediment Budget . Any process that increases the

quantity of sand in a defined control volume is called a sauroe. Any
process that decreases the quantity of sand in the control volume is
called a sink. Usually, sources are identified as positive and sinks
as negative. Some processes (longshore transport is the most important)
function both as source and sink for the control volume, and these are
called conveoting processes.
Point sources or point sinks are sources or sinks that add or sub-
tract sand across a limited part of a control volume boundary. A tidal
inlet often functions as a point sink. Point sources or sinks are gener-
ally measured in units of volume per year.
Line sources or tine sinks are sources or sinks that add or subtract
sand across an extended segment of a control volume boundary. Wind trans-
port landward from the beaches of a low barrier island is a line sink for
the ocean beach. Line sources or sinks are generally measured in units of
volume per year per lonit length of shoreline. To compute the total effect
of a line source or sink, it is necessary to multiply this quantity by the
total length of shoreline over which the line source or sink operates.
The following conventions are used for elements of the sediment

budget:
Q^ is a point source
Q^ is a point sink
q^ is a line source
q^ is a line sink
These subscripted elements of the sediment budget are identified by name

in Table 4-13 according to whether the element makes a point or line con-
tribution to the littoral zone, and according to the boundary across which
the contribution enters or leaves. Each of the elements is discussed in
following sections.
The length of shoreline over which a line source is active is in-

dicated by hi and the total contribution of the line source or line
*+ *—
sink by Q^ or Q^ , so that in general
Q* = b,. q,. . (4-46)
4-117
Table 4-13. Classification of Elements in the Littoral Zone Sediment Budget
Location of
Source or Sink
4.713 Sediment Budget Boundaries .A sediment budget is used to identify
and quantify the sources and sinks that are active in a specified area.
By so doing, erosion or deposition rates are determined as the balance
of known sinks and sources. Boundaries for the sediment budget are deter-
mined by the area imder study, the time scale of interest, and study pur-
poses. In a given study area, adjacent sand budget compartments (control
volumes) may be needed with shore-perpendicular boundaries at significant
changes in the littoral system. For example, compartment boundaries may
be needed at inlets, between eroding and stable beach segments, and between
stable and accreting beach segments. Shore-parallel boundaries are needed
on both the seaward and landward sides of the control volumes. They may be
established wherever needed, but the seaward boundary is usually established
at or beyond the limit of active sediment movement, and the landward bound-
ary beyond the erosion limit anticipated for the life of the study. The
bottom surface of a control volume should pass below the sediment layer
that is actively moving, and the top boundary should include the highest
surface elevation in the control volume. Thus, the budget of a particular
beach and nearshore zone would have shore parallel boundaries landward of
the line of expected erosion and at or beyond the seaward limit of signifi-
cant transport. A budget for barrier island sand dunes might have a bound-
ary at the bay side of the island and the landward edge of the backshore.
A sediment budget example and analysis are shown in Figure 4-43.

This example considers a shoreline segment along which the incident wave
climate can transport more material entering from updrift. Therefore
the longshore transport in the segment is being fed by a continuously
eroding sea cliff. The cliff is composed of 50 percent sand and 50 per-
cent clay. The clay fraction is assumed to be lost offshore while the
sand fraction feeds into the longshore transport. The budget balances
the sources and sinks using the following continuity equation:
Sum of Known Sources - Sum of Known Sinks = Difference
An example calculation is shown in Figure 4-43.
4.72 SOURCES OF LITTORAL MATERIAL
4.721 Rivers . It is estimated that rivers of the world bring about 3.4
cubic miles or 18.5 billion cubic yards of sediment to the coast each year
(voliime of solids without voids). (Stoddard, 1969; from Strakhov, 1967.)
Only a small percentage of this sediment is in the sand size range that is
common on beaches. The large rivers which account for most of the volume
of sediment carry relatively little sand. For example, it is estimated
(Scruton, 1960) that the sediment load brought to the Gulf of Mexico each
year by the Mississippi River consists of 50 percent clay, 48 percent silt,
and only 2 percent sand. Even lower percentages of sand seem probable for
other large river discharges. (See Gibbs, 1967, p. 1218, for information
on the Amazon River.) But smaller rivers flowing through sandy drainage
areas may carry 50 percent or more of sand. (Chow, 1964, p. 17-20.) In
southern California, sand brought to the coast by the floods of small
rivers is a significant source of littoral material. (Handin, 1951; and
Norris, 1964.)
4-119
Offshore Sink
WATER
Longshore Tronsporf Longshore Transport

4.
Qa
I
*- Q4
IN
r r
Sediment Budget Boundary Top of Cliff
LAND
ERODING SHORELINE-PLAN VIEW

(Not to Scole)
Budget Boundory
ri Recession
\ \
I \
> \,*i-First Survey Profile
Most of the sediment carried to the coast by rivers is deposited
in comparatively small areas, often in estuaries where the sediment is
trapped before it reaches the coast. (Strakhov, 19670 The small frac-
tion of sand in the total material brought to the coast and the local
estuarine and deltaic depositional sites of this sediment suggest that
rivers are not the immediate source of sediment on beaches for much of
the world's coastline. Many sources of evidence indicate that sand-
sized sediment is not supplied to the coasts by rivers on most segments
of the U.S. Atlantic and Gulf coasts. Therefore, other sediment sources
must be important.
4.722 Erosion of Shores and Cliffs . Erosion of the nearshore bottom,

the beach, and the seaward edge of dunes, cliffs, and mainland (Fig. 4-44)
results in a sand loss. In many areas, erosion from cliffs of one area is
the principal source of sand for downdrift beaches. Kuenen (1950) esti-
mates that beach and cliff erosion along all coasts of the world totals
about 0.03 cubic mile or 160 million cubic yards per year. Although this
amount is only about 1 percent of the total solid material carried by
rivers, it is a major source in terms of sand delivered to the beaches,
since the sand fraction in the river sediments is usually small, and is
usually trapped before it reaches the littoral zone. Shore erosion is an
especially significant source where older coastal deposits are being ero-
ded, since these usually contain a large fraction of sand.
If an eroding shore maintains approximately the same profile above

the seaward limit of significant transport while it erodes, then the ero-
sion volume per foot of beach front is the vertical distance from dune
base or berm crest to the depth of the seaward limit (h) , multiplied by
the horizontal retreat of the profile. Ax. (See Figure 4-44.)
Figure 4-44 shows three equivalent volumes, all indicating a net ero-
sion of hAx„ To the right in Figure 4-44 is a typical beach profile.
The dashed line profile below it is the same as the solid line profile.
The horizontal distance between solid and dashed profiles is Ax, the
horizontal retreat of the profile due to (assumed) uniform erosion. The
unit volume loss, hAx, between dune base and depth to seaward limit is
equivalent to the unit volume indicated by the slanted parallelogram on
the middle of Figure 4-44. The unit volume of this parallelogram, hAx,
is equivalent to the shaded rectangle on the left of Figure 4-44. If the
vertical distance, h is 40 feet, and Ax = 1 foot of horizontal erosion,
then the unit volume lost is 40/27, or 1.5 cubic yards per foot of beach
front
4.723 Transport from Offshore Slope . An uncertain and potentially signifi-

cant source in the sediment budget is the contribution from the offshore
slopCo However, hydrography, sediment size distribution, and related evi-
dence discussed in Section 4.523 indicate that contributions from the con-
tinental shelf to the littoral zone are probably negligible in many areas.
Most shoreward moving sediment appears to originate in areas fairly close
to shoreo Significant onshore-offshore transport takes place within the
littoral zone due to seasonal and storm-induced profile changes and to
4-121
«
erosion of the nearshore bottom and beaches, but in the control volume
defined, this transport takes place within the control volume. Transport
from the offshore has been treated as a line source.
In some places, offshore islands or shoals may act as point sources

of material for the littoral zone. For example, the drumlin islands and
shoals in Boston Harbor and vicinity may be point sources for the nearby
mainland.
4.724 Windblown Sediment Sources . To make a net contribution to the

littoral zone in the time frame being considered, windblown sand must
come from a land source whose sand is not derived by intermediate steps
from the same littoral zone. On U.S. ocean coasts, such windblown sand
is not a significant source of littoral materials. Where wind is impor-
tant in the sediment budget of the ocean shore, wind acts to take away
sand rather than to add it, although local exceptions probably occur.
However, windblown sand can be an important source, if the control

volume being considered is a beach on the lagoon side of a barrier island.
Such shores may receive large amounts of windblown sand.
4.725 Carbonate Production . Dissolved calcium carbonate concentration

in the ocean is near saturation, and it may be precipitated under favor-
able conditions. In tropical areas, many beaches consist of calciiom car-
bonate sands; in temperate zones, calcium carbonate may be a significant
part of the littoral material. These calciiom carbonate materials are gen-
erally fragments of shell material whose rate of production appears to in-
crease with high temperature and with excessive evaporation. (See Hayes,
1967.) Oolitic sands are a nonbiogenic chemical precipitate of calcium
carbonate on many low latitude beaches.
Quantitative estimates of the production of calcium carbonate sedi-

ment are lacking, but maximum rates might be calculated from the density
and rate of growth of the principal carbonate-producing organisms in an
area. For example, following northeasters along the Atlantic coast of
the U.S., the foreshore is occasionally covered with living clams thrown
up by the storm from the nearshore zone. One estimate of the annual con-
tribution to the littoral zone from such a source would assume an average
shell thickness of about 0.04 foot completely covering a strip of beach
100 feet wide all along the coast. On an annual basis, this would be
about 0.15-cubic yard per year per foot of beach front. Such a quantity
is negligible under almost all conditions. However, the dominance of
carbonate sands in tropical littoral zones suggests that the rate of pro-
duction can be much higher.
4.726 Beach Replenishment . Beach protection projects often require

placing sand on beaches. The quantity of sand placed on the beach in
such beach-fill operations may be a major element in the local sediment
budget. Data on beach-fill quantities may be available in Corps of Engi-
neer District offices, in records of local government engineers, and in
dredging company records. The exact computation of the quantity of a
4-123
beach fill is subject to uncertainties: the source of the dredged sand
often contains significant but variable quantities of finer materials that
are soon lost to the littoral zone; the surveys of both the borrow area
and the replenished area are subject to uncertainty because sediment trans-
port occurs during the dredging activities; and in practice only limited
efforts are made to obtain estimates of the size distribution of fill
placed on the beach. Thus, the resulting estimate of the quantity of
suitable fill placed on the beach is uncertain. More frequent sampling
and surveys could help identify this significant element in many sedi-
ment budgets.
4.73 SINKS FOR LITTORAL MATERIALS
4.731 Inlets and Lagoons . Barrier islands are interrupted locally by

inlets which may be kept open by tidal flow. A part of the sediment moved
alongshore by wave action is moved into these inlets by tidal flow. Once
inside the inlet, the sediment may deposit where it cannot be moved sea-
ward by the ebb flow. (Brown, 1928.) The middleground shoals common to
many inlets are such depositional features. Such deposition may be reduced
when the ebb currents are stronger than the flood currents. (Johnson,
1956.)
It is evident from aerial photography (e.g., of Drum Inlet, N.C.,

Fig. 4-45) that inlets do trap significant quantities of sand. Caldwell's
(1966) estimate of the sand budget for New Jersey, calculates that 23
percent of the local gross longshore transport is trapped by the seven
inlets in southern New Jersey, or about 250,000 cubic yards per year for
each inlet. In a study of the south shore of Long Island, McCormick (1971)
estimated from the growth of the f loodtide delta of Shinnecock Inlet (shown
by aerial photos taken in 1955 and 1969) that this inlet trapped 60,000
cubic yards per year. This amounts to about 20 percent of the net long-
shore transport (Taney, 1961a, p. 46), and probably less than 10 percent
of the gross transport. (Shinnecock Inlet is a relatively small inlet.)
It appears that the rate at which an inlet traps sediment is higher imme-
diately after the inlet opens than it is later in its history.
4.732 Overwash . On low barrier islands, sand may be removed from the
beach and dune area by overwashing during storms. Such rates may average
locally up to 1 cubic yard per year per foot. Data presented by Pierce
(1969) suggest that for over half of the shoreline between Cape Hatteras
and Cape Lookout, North Carolina, the short term loss due to overwash was
0.6 cubic yard per year per foot of beach front. Figure 4-46 is an aerial
view of overwash in the region studied by Pierce (1969). Overwash does
not occur on all barrier islands, but if it does, it may function as a
source for the beach on the lagoon side.
4.733 Backshore and Dune Storage Sand can be temporarily withdrawn

.
from transport in the littoral zone as backshore deposits and dune areas
along the shore. Depending on the frequency of severe storms, such sand
may remain in storage for intervals ranging from months to years. Back-
shore deposition can occur in hours or days by the action of waves after
4-124
Figure 4-45. Sediment Trapped Inside Old Drim Inlet,
North Carolina
4-125
( I November 1971
Figure 4-46. Overwash on Portsmouth Island, North

Carolina
4-126
storms. Dune deposits require longer to form - months or years - because
wind transport usually moves material at a lesser rate than wave transport.
I£ the immediate beach area is the control volume of interest, and budget
calculations are made based on data taken just after a severe storm, allow-
ance should be made in budget calculations for sand that will be stored in
berms through natural wave action. (See Table 4-5.)
4.734 Offshore Slopes . The offshore area is potentially an important

sink for littoral material. Transport to the offshore is favored by:
storm waves which stir up sand, particularly when onshore winds create
a seaward return flow; turbulent mixing along the sediment concentration
gradient which exists between the sediment-water mixture of the surf zone
and the clear water offshore; and the slight offshore component of gravity
which acts on both the individual sediment particles and on the sediment-
water mixture.
It is often assumed that the sediment sorting toss that commonly

reduces the volume of newly placed beach fill is lost to the offshore
slopes. (Corps of Engineers, Wilmington District, 1970; and Watts,
1956.) A major loss to the offshore zone occurs where spits build into
deep water in the longshore direction. Sandy Hook, New Jersey, is an
example. (See Figure 4-47.) It has been suggested (Bruun and Gerritsen,
1959) that ebb flows from inlets may sometimes cause a loss of sand by
jetting sediment seaward into the offshore zone.
The calculation of quantities lost to the offshore zone is difficult,

since it requires extensive, accurate, and costly surveys. Some data on
offshore changes can be obtained by studies of sand level changes on rods
imbedded in the sea floor (Inman and Rusnak, 1956), but without extending
the survey beyond the boundary of the moving sand bed, it is difficult
to determine net changes.
4.735 Submarine Canyons . Probably the most frequently mentioned sinks

for littoral materials are submarine canyons. Shepard (1963) and Shepard
and Dill (1966) provide extensive description and discussion of the origin
of submarine canyons. The relative importance of submarine canyons in
sediment budgets is still largely unknown.
Of 93 canyons tabulated by Shepard and Dill (1966), 34 appear to be

receiving sediment from the coast, either by longshore transport or by
transport from river mouths. Submarine canyons are thought to be espe-
cially important as sinks off southern California. Herron and Harris
(1966, p. 654) suggest that Mugu Canyon, California, traps about 1 mil-
lion cubic yards per year of the local littoral drift.
The exact mechanism of transport into these canyons is not clear,

even for the La Jolla Canyon (California) which is stated to be the most
extensively studied submarine feature in the world. (Shepard and Buff-
ington, 1968.) Once inside the canyons, the sediment travels down the
floors of the heads of the canyons, and is permanently lost to the litto-
ral zone.
4-127
( 14 September 1969)
Figure 4-47. Growth of a Spit into Deep Water, Sandy-

Hook, New Jersey
4-128
4.736 Deflation . The loose sand that forms beaches is available to be
transported by wind. After a storm, shells and other objects are often
found perched on pedestals of sand left standing after the wind eroded
less protected sand in the neighborhood. Such erosion over the total
beach surface can amount to significant quantities. Unstabilized dunes
may form and migrate landward, resulting in an important net loss to the
littoral zone. Examples include some dunes along the Oregon coast (Coo-
per, 1958), between Pismo Beach and Point Arguello, California (Bowen
and Inman, 1966); central Padre Island (Watson, 1971); and near Cape Hen-
lopen, Delaware (Kraft, 1971). Typical rates of transport due to wind
range from 1 to 10 cubic yards per year per foot of beach front where
wind transport is noticeable. (Cooper, 1958; Bowen and Inman, 1966;
Savage and Woodhouse, 1968; and Gage, 1970.) However average rates prob-
ably range from 1 to 3 cubic yards per year per foot.
The largest wind-transported losses are usually associated with

accreting beaches that provide a broad area of loose sand over a period
of years. Sand migrating inland from Ten Mile River Beach in the vicin-
ity of Laguna Point, California, is shown in Figure 4-48.
Study of aerial photographs and field reconnaissance can easily

establish whether or not important losses or gains from wind transport
occur in a study area. However, detailed studies are usually required
to establish the importance of wind transport in the sediment budget.
4.737 Carbonate Loss . The abrasion resistance of carbonate materials

is much lower than quartz, and the solubility of carbonate materials is
usually much greater than quartz. However, there is insufficient evidence
to show that significant quantities of carbonate sands are lost from the
littoral zone in the time scale of engineering interest through either
abrasion or solution.
4.738 Mining and Dredging . From ancient times, sand and gravel have
been mined along coasts. In some countries, for example Denmark and
England, mining has occasionally had undesirable effects on coastal settle-
ments in the vicinity. Sand mining in most places has been discouraged by
legislation and the rising cost of coastal land, but it still is locally
important. (Magoon, et al., 1972.) It is expected that mining will become
more important in the offshore area in the future. (Duane, 1968, and Fisher,
1969.)
Such mining must be conducted far enough offshore so the mined pit
will not act as a sink for littoral materials, or refract waves adversely,
or substantially reduce the wave damping by bottom friction and percolation.
Material is also lost to the littoral zone when dredged from naviga-
ble waters (channels and entrances) within the littoral zone, and the
dredged material is dumped in some area outside of the littoral zone.
These dump areas can be for land fill, or in deep water offshore. This
action has been a common practice, because the first costs for some
dredging operations are cheaper when done this way.
4-129
(24 Moy 1972)
Figure 4-48. Dunes Migrating Inland Near Laguna Point, California
4-130
4.74 CONVECTION OF LITTORAL MATERIALS
Sources and sinks of littoral materials are those processes that

result in net additions or net subtractions of material to the selected
control volume. However, some processes may subtract at the same rate
that they add material, resulting in no net change in the volume of lit-
toral material of the control volume.
The most important convecting process is longshore sediment transport.

It is possible for straight exposed coastlines to have gross longshore
transport rates of more than 1 million cubic yards per year. On a coast
without structures, such a large Q^ can occur, and yet not be apparent
because it causes no obvious beach changes. Other convecting processes
that may produce large rates of sediment transport with little noticeable
change include tidal flows, especially around inlets, wind transport in
the longshore direction, and wave-induced currents in the offshore zone.
Since any structure that interrupts the equilibrium convection of

littoral materials will normally result in erosion or accretion, it is
necessary that the sediment budget quantitatively identify all processes
convecting sediment through the study area. This is most important on
shores with high waves.
4.75 RELATIVE CHANGE IN SEA LEVEL
Relative changes in sea level may be caused by changes in sea level

and changes in land level. Sea levels of the world are now generally ris-
ing. The level of inland seas may either rise or fall, generally depend-
ing on hydrologic influences. Land level may rise or fall due to tectonic
forces, and land level may fall due to subsidence. It is often difficult
to distinguish whether apparent changes in sea level are due to change in
sea level, change in land level, or both. For this reason, the general
process is referred to as relative change in sea level.
While relative changes in sea level do not directly enter the sedi-
ment budget process, the net effect of these elevation changes is to move
the shoreline either landward (relative rise in sea level) or seaward
(relative fall in sea level). It thus can result in the appearance of a
gain or loss of sediment volume.
The importance of relative change in sea level on coastal engineer-

ing design depends on the time scale and the locality involved. Its
effect should be determined on a case-by-case basis.
4.76 SUMMARY OF SEDIMENT BUDGET
Sources, sinks, and convective processes are summarized diagrammati-

cally in Figure 4-49 and listed in Table 4-14. The range of contributions
or losses from each of these elements is described in Table 4-14 measured
as a fraction of the gross longshore transport rate, or as a rate given
in cubic yards per year per foot of beach front. The relative importance
4-131
of elements in the sand budget varies with locality and with the bound-
aries of the particular littoral control volume. (These elements are
classified as point or line sources or sinks in Table 4-13, and the budget
is summarized in Equation 4-48.)
LAND OCEAN
Cliff, Dune and

Backshore Erosion Offshore Slope
(source or sink?)
Beach Replenishment
Submarine Canyon
Rivers
Dredging
Dune and Backshore
Storage
Winds
(supply and deflation)
Calcium Carbonate
(production and loss)
Inlets and Lagoons

Including Overwash
Mining
Longshore Currents
Tidal Currents
Longshore Winds
Figure 4-49. Materials Budget for the Littoral Zone
In most localities, the gross longshore transport rate significantly

exceeds other volume rates in the sediment budget, but if the beach is
approximately in equilibrium, this may not be easily noticed.
The erosion of beaches and cliffs and river contributions are the
principal known natural sources of beach sand in most localities. Inlets,
4-132
Table 4-14. Sand Budget of the Littoral Zone
lagoons and deep water in the longshore direction comprise the principal
known natural sinks for beach sand. Of potential, but usually unknown,
importance as either a source or a sink is the offshore zone seaward of
the beach.
The works of man in beach replenishment and in mining or dredging

may provide major sources or sinks in local areas. In a few U.S. locali-
ties, submarine canyons or wind may provide major sinks, and calcium car-
bonate production by organisms may be a major source.
************** EXAMPLE PROBLEM ***************

GIVEN:
(a) An eroding beach 4.4 miles long at root of spit that is 10 miles
long. Beaches on the remainder of the spit are stable. (See
Figure 4-50a.)
(b) A uniform recession rate of 3 feet per year along the eroding
4o4 miles.
(c) Depth of lowest shore parallel contour is -30 feet MSL, and
average dune crest elevation is 15 feet MSL.
(d) Sand is accumulating at the tip of the spit at an average rate of

400,000 cubic yards per year.
(e) The variation of y along the beaches of the spit is shown in

Figure 4-51. (y = Q.it/^rt'' Equation 4-23.)
(f) No sand accumulates to the right of the erosion area; no sand is

lost to the offshore.
(g) A medium-width jettied inlet is proposed which will breach the

spit as shown in Figure 4-50a.
(h) The proposed inlet is assumed to trap about 15 percent of the

gross transport, Q^.
(i) The 1.3-mile long beach to the right of the jettied inlet will
stabilize (no erosion) and realign with y changing to 3.5.
(j) The accumulation at the end of the spit will continue to grow
at an average annual rate of 400,000 cubic yards per year after
the proposed inlet is constructed.
4-134
(a) Site Sketch
Reach 4
10 -
Reach 4 Reaches 1-3

(average / = 9) (average / =4)
10
Distance from Spit Tip -Miles
Figure 4-51, Variation of y with Distance Along the Spit,

before Inlet Condition
FIND:
(a) Annual littoral drift trapped by inlet.
(b) After-inlet erosion rate of the beach to the left of the inlet.
(c) After-inlet nourishment needed to maintain the historic erosion

rate on the beach to the left of the inlet.
(d) After-inlet nourishment needed to eliminate erosion left of the

inlet.
SOLUTION Divide the beach under study into four sand budget compart-
:
ments (control volumes called reaches) as shown in Figure 4-50a. Shore

perpendicular boundaries are established where important changes in the
littoral system occur. To identify and quantify the hefore-inlet sys-
tem, the continuity of the net transport rate along the spit must be
established. The terminology of Figure 4-43 and Table 4-13 is used
4-136
for the sand budget calculation. The average annual volume of material
contributed to the littoral system per foot of eroding beach reaches
2 and 3 is:
n+ -
%(2) -
n+
^5(5)
-^-=
- 1,A
(15 + 30) (3)
Yl
= 5.0 yd?/yryft. .
Then, from Equation 4-46 the total annual contribution of the eroding
beaches to the system can be determined as:
^{(2)
"*"
^1(3)
" (1-3 f"i-
+ 3-1 "î-) (5280 ft./mi.) (5 yd?/yr./ft.)
= 1.16 X 105yd?/yr.
Since there is no evidence of sand accumulation or erosion to the right

of the eroding area, the eroding beach material effectively moves to the
left becoming a component of the net transport volume (Q„) toward the
end of the spit. Contiuity requires the erosion volume and Reach 1 Q^
must combine to equal the acretion at the end of the spit (400,000 cubic
yards per year). Thus, Q^^ at the root of the spit is
" 400,000 yd?/yr. - 116,000 yd?/yr.

%(1 2)
284,000 yd.3/yr.
\{1,2)^
Q^ across the boundary between Reaches 2 and 3 (Qj^r2 o)) is:
%{2,3) = Q«(i,2) + \2) (15(2))

^^
= 284,000 yd?/yr. + (1.3 mi.) 15,280— j (5 yd?/yr./ft.)
= 318,000 yd.Vyr.
Qyj across the boundary between Reaches 3 and 4 is:
^n{3,4)^^n{2,3) + î) «(i))

= 318,000 yd?/yr. + (3.1 mi.) 15,280 — | (5 yd?/yr/ft.)
= 4000,000 yd?/yr.
This Qri(.3 h) moves left across reach 4 with no additions or subtrac-

tions, and since the accretion rate at the end of the spit is 400,000
cubic yards per year, the budget balances. Knowing Q^ and y for
each reach, gross transport, Q^, transport to the right Q^,^ and
transport to the left Qg^^ can be computed using the following equa-
tions :
4-137
From Equation 4-24
then
Q..
=
'^
1+7 '
and
%> Q-rt' ^lt> ^^^ %. ^°^ each reach are shown in Figure 4-50b.
Now the after-inlet condition can be analyzed.
Q„(I2) = 284,000 yd?/yr. (same as "before inlet") ,
%{2,3) = Q«(i,2) = 284,000 yd?/yr. (reach 2 is stable) .
The gross transport rate across the inlet with the new y = 3.5 using
Equation 4-24, is:
Qn(2.5)(l+T)
= (284,000 yd?/yr.)
%{2,3) f^J ,
%{2,3)
= 512,000 yd?/yr. .
The inlet sink (Qp = 15 percent of (^(2,3)
Q~2 = 512,000 yd.Vyr. X 0.15 ,
Q2 = 77,000 yd?/yr.
The erosion value from Reach 3 now becomes
Reach 3 erosion = spit accretion + inlet sink - net littoral drift

right of inlet.
\3) «i3)) - %0,4) ^ ^2

- %i2,3)
'
^(3)
" 400,000 yd?/yr. + 77,000 yd?/yr. - 284,000 yd?/yr. ,
{^H3))
b(^) = 193,000 yd?/yr. .

(q^(^^)
4-138
Nourishment needed to maintain historic erosion rate on Reach 3 beach
is:
Reach 3 nourishment = Reach 3 erosion "after inlet" - Reach 3 erosion

"before inlet".
^5(5)
"
\3) ('i'3i3))
after inlet - b^^^ before inlet .
(qj^^^)
= 193,000 yd?/yr. - 82,000 yd?/yr.

^5(5) ,
= 11 1,000 yd?/yr. .
Qj^^^
If reach 3 erosion is to be eliminated, it will be necessary to provide

nourishment of 193,000 cubic yards per year.
Q^, Qj^^, and Q^^ for the after-inlet condition are computed using
Equation 4-24 and related equations. The after-inlet sand budget is
shown in Figure 4-50c.
*************************************
4.8 ENGINEERING STUDY OF LITTORAL PROCESSES
This section demonstrates the use of Chapter 4 in the engineering

study of littoral processes.
4.81 OFFICE STUDY
The first step in the office phase of an engineering study of litto-

ral processes is to define the problem in terms of littoral processes.
The problem may consist of several parts, especially if the interests of
local groups are in conflict. An ordering of the relative importance of
the different parts may be necessary, and a complete solution may not be
feasible. Usually, the problem will be stated in terms of the require-
ments of the owner or local interests. For example, local interests may
require a recreational beach in an area of limited sand, making it neces-
sary to estimate the potential rates of longshore and onshore-offshore
sand transport. Or a fishing community may desire a deeper channel in an
inlet through a barrier island, making it necessary to study those litto-
ral processes that will affect the stability and long-term navigability
of the inlet, as well as the effect of the improved inlet on neighboring
shores and the lagoon.
4.811 Sources of Data The next step is to collect pertinent data.

. If
the problem area is located on a U.S. coastline, the National Shoreline
Study may be consulted. This study can provide a general description of
the area, and may give some indication of the littoral processes occur-
ring in the vicinity of the problem area.
4-139
Historical records of shoreline changes are usually in the form of
charts, surveyed profiles, dredging reports, beach replenishment reports
and aerial photos. As an example of such historical data. Figure 4-52
shows the positions of the shoreline at Sandy Hook, New Jersey, during
six surveys from 1835 to 1932. Such shoreline changes are useful for
computing longshore transport rates. The Corps of Engineers maintains,
in its District and Division offices, survey, dredging, and other reports
relating to Corps projects. Charts may be obtained from various federal
agencies including the Defense Mapping Agency Hydrographic Center, Geolog-
ical Survey, National Ocean Survey, and Defense Mapping Agency Topographic
Center. A map called "Status of Aerial Photography," which may be obtained
from the Map Information Office, Geological Survey, Washington, D.C. 20242,
shows the locations and types of aerial photos available for the U.S., and
lists the sources from which the photos may be requested. A description of
a coastal imagery data bank can be found in the interim report by Szuwalski
(1972).
Other kinds of data usually available are wave, tide, and meteoro-
logical data. Chapter 3 discusses wave and water level predictions; Sec-
tion 4.3 discusses the effects of waves on the littoral zone; and Sub-
section 4.34 presents methods of estimating wave climate and gives pos-
sible sources of data. These referenced sections indicate the wave, tide,
and storm data necessary to evaluate coastal engineering problems.
Additional information can be obtained from local newspapers, court-

house records, and from area residents. Local people can often identify
factors that outsiders may not be aware of, and can also provide quali-
titive information on previous coastal engineering efforts in the area
and their effects.
4.812 Interpretation of Shoreline Position . Preliminary interpretation

of littoral processes is possible from the position of the shoreline on
aerial photos and charts. Stafford (1971) describes a procedure for uti-
lizing periodic aerial photographs to estimate coastal erosion. Used in
conjunction with charts and topographic maps, this technique may provide
quick and fairly accurate estimates of shoreline movement, although the
results can be biased by the short-term effects of storms.
Charts show the coastal exposure of a study site, and since exposure
determines the possible directions from which waves reach the coast, expo-
sure also determines the most likely direction of longshore transport.
Direction of longshore transport may also be indicated by the posi-

tion of sand accumulation and beach erosion around littoral barriers. A
coastal structure in the surf zone may limit or prevent the movement of
sand, and the buildup of sediment on one side of the littoral barrier serv-
es as an indicator of the net direction of transport. This buildup can be
determined from dredging or sand bypassing records or aerial photos. Fig-
ure 4-53 shows the accumulation of sand on one side of a jetty. But wave
direction and nearshore currents at the time of the photo indicate that
transport then was in the opposite direction. Thus, an erroneous conclusion
4-140
Figure 4-52. Growth of Sandy Hook, New Jersey, 1835-1932
4-141
about the net transport might be made, if only wave patterns of this photo
are analyzed. The possibility of seasonal or storm-induced reversals in
sediment transport direction should be investigated by periodic inspections
or aerial photos of the sand accumulation at groins and jetties.
(4 December 1967)
Figure 4-53. Transport Directions at New Buffalo Harbor Jetty on
Lake Michigan
The accumulation of sand on the updrift side of a headland is illus-

trated by the beach north of Point Mugu in Figure 4-54. The tombolo in
Figure 4-55 was created by deposition behind an offshore barrier (Grey-
hound Rock, California). Where a beach is fixed at one end by a struc-
ture or natural rock formation, the updrift shore tends to align perpen-
dicular to the direction of dominant wave approach. (See Figures 4-55,
and 4-56.) This alignment is less complete along shores with signifi-
cant rates of longshore transport.
Sand accumulation at barriers to longshore transport may also be

used to identify nodal zones. There are two types of nodal zones: diver-
gent and convergent. A divergent nodal zone is a segment of shore char-
acterized by net longshore transport directed away from both ends of the
zone. A convergent nodal zone is a segment of shore characterized by net
longshore transport directed into both ends of the zone.
Figure 4-56 shows a nodal zone of divergence centered around the

fourth groin from the bridge on the south coast of Staten Island, Outer
New York Harbor. Central Padre Island, Texas, is thought to be an example
4-142
(21 Moy 1972;
Figure 4-54. Sand Accumulation at Point Mugu, California
4-143
(29 August 1972)
Figure 4-55. Tombolo and Pocket Beach at Greyhound

Rock, California
4-144
of a convergent nodal zone. (Watson, 1971.) Nodal zones o£ divergence
are more common than nodal zones of convergence, because longshore trans-
port commonly diverges at exposed shores and converges toward major gaps
in the ocean shore, such as the openings of New York Harbor, Delaware Bay,
and Chesapeake Bay.
Nodal zones are usually defined by long-term average transport direc-

tions, but because of insufficient data, the location of the mid-point of
the nodal zone may be uncertain by up to 10' s of miles. In addition, the
short-term nodal zone most probably shifts along the coast with changes
in wave climate.
The existence, location, and planform of inlets can be used to inter-

pret the littoral processes of the region. Inlets occur where tidal flow
is sufficient to maintain the openings against longshore transport which
acts to close them. (e.g., Bruun and Gerritsen, 1959.) The size of the
inlet opening depends on the tidal prism available to maintain it. (O'Brien,
1969.) The dependence of inlet size on tidal prism is illustrated by Fig-
ure 4-57, which shows three bodies of water bordering the beach on the south
shore of Long Island, New York. The smallest of these (Sagaponack Pond)
is sealed off by longshore transport; the middle one (Mecox Bay) is partly
open; and the largest (Shinnecock Bay) is connected to the sea by Shinne-
cock Inlet, which is navigable.
( 14 September 19691
Figure 4-57. South Shore of Long Island, N.Y., Showing Closed,

Partially Closed, and Open Inlets
4-146
Detailed study of inlets through barrier islands on the U.S. Atlan-
tic and gulf coasts shows that the shape of the shoreline at an inlet can
be classified in one of four characteristic planforms. (See Figure 4-58,
adapted from Galvin, 1971,) Inlets with overlapping offset (Fig. 4-59)
occur only where waves from the updrift side dominate longshore transport.
Wliere waves from the updrift side are less dominant, the updrift offset
(Fig. 4-59) is common. Where waves approach equally from both sides, in-
lets typically have negligible offset. (See Figure 4-60.) IVhere the sup-
ply of littoral drift on the updrift side is limited, and the coast is
fairly well exposed, a noticeable downdrift offset is common as, for exam-
ple, in southern New Jersey and southern Delmarva. (See Hayes, et al.,
(1970.) These planform relations to littoral processes have been found
for inlets through sandy barrier islands, but they do not necessarily hold
at inlets with rocky boundaries. The relations hold regionally, but tem-
porary local departures due to inlet migration may occur.
4.82 FIELD STUDY
A field study of the problem area is usually necessary to obtain

types of data not found in the office study, to supplement incomplete
data, and to serve as a check on the preliminary interpretation and corre-
lations made from the office data. Information on coastal processes may
be obtained from wave gage data and visual observations, sediment sam-
pling, topographic and bathymetric surveys, tracer programs, and effects
of natural and manmade structures.
4.821 Wave Data Collection .A wave-gaging program yields height and

period data. However, visual observations may currently be the best
source of breaker direction data. Thompson and Harris (1972) deter-
mined that 1 year of wave-gage records provides a reliable estimate of
the wave height frequency distribution. It is reasonable to assume that
the same is true for wave direction.
A visual observation program is inexpensive, and may be used for

breaker direction and for regional coverage when few wave-gage records
are available. The observer should be provided with instructions, so
that all data collected will be uniform, and contact between observer
and engineer should be maintained.
4.822 Sediment Sampling . Sediment sampling programs are described in

Section 4.26. Samples are usually surface samples taken along a line
perpendicular to the shoreline. These are supplemented by borings or
cores as necessary. Complete and permanent identification of the sam-
ple is important.
4.823 Surveys . Most engineering studies of littoral processes require

surveying the beach and nearshore slope. Successive surveys provide
data on changes in the beach due to storms, or long-term erosion or
accretion. If beach length is also considered, an approximate volume
of sand eroded or accreted can be obtained which provides information
for the sediment budget of the beach. The envelope of a profile defines
4-147
Qrt Q/t
Overlapping Offset
Length of arrows indicotes

relative magnitude of
Qrt Q/t Longshore Tronsport Rate
Adequate
updrift source j^
Updrift Offset
Downdrift Offset
Inadequate
updrift source
Qrt Qit
Negligible Offset
(Galvin,l97l )
Figure 4-58. Four Types of Barrier Island Offset
4-148
(14 September 1969)
Figure 4-59. Fire Island Inlet, New York - Overlapping Offset
BAY
fluctuations of sand level at a site (Everts, 1973), and thus provides
data useful in beach fill and groin design.
Methods for obtaining beach and nearshore profiles, and the accu-
racy of the resulting profiles are discussed in Section 4.514.
4.824 Tracers . It is often possible to obtain evidence on the direction

of sediment movement and the origins of sediment deposits by the use of
tracer materials which move with the sediment. Fluorescent tracers were
used to study sand migration in and around South Lake Worth Inlet, Florida.
(Stuiver and Purpura, 1968.) Radioactive sediment tracer tests were con-
ducted to determine whether potential shoaling material passes through or
around the north and south jetties of Galveston Harbor. (Ingram, Cummins,
and Simmons, 1965.)
Tracers are particles which react to fluid forces in the same manner
as particles in the sediment whose motion is being traced, yet which are
physically identifiable when mixed with this sediment. Ideally, tracers
must have the same size distribution, density, shape, surface chemistry,
and strength as the surrounding sediment, and in addition have a physical
property that easily distinguishes them from their neighbors.
Three physical properties have been used to distinguish tracers:

radioactivity, color, and composition. Tracers may be either naturally
present or introduced by man. There is considerable literature on recent
investigations using or evaluating tracers including reviews and biblio-
graphy: (Duane and Judge, 1969; Bruun, 1966; Galvin, 1964; and Huston,
1963); models of tracer motion: (James, 1970; Galvin, 1964; Hubbell and
Sayre, 1965; and Duane, 1970); and use in engineering problems: (Hart,
1969; Cherry, 1965; Cummins, 1964; and Duane, 1970).
a. Natural Tracers. Natural tracers are used primarily for back-

ground information about sediment origin and transport directions, i.e.,
for studies which involve an understanding of sediment patterns over a
long period of time.
Studies using stable, nonradioactive natural tracers may be based

on the presence or absence of a unique mineral species, the relative
abundance of a particular group of minerals within a series of samples,
or the relative abundance and ratios of many mineral types in a series
of samples. Although the last technique is the most complex, it is often
used, because of the large variety of mineral types normally present in
sediments and the usual absence of singularly unique grains. The most
suitable natural tracers are grains of a specific rock type originating
from a localized specific area.
Occasionally, characteristics other than mineralogy are useful for

deducing source and movement patterns. Krinsley, et al., (1964) devel-
oped a technique for the study of surface textures of sand grains with
electron microscopy, and applied the technique to the study of sand trans-
port along the Atlantic shore of Long Island. Naturally occurring radio-
active materials in beach sands have also been used as tracers. (Kamel,
1962.)
4-150
One advantage of natural tracers is their tendency to "average" out
short-term trends and provide qualitatively accurate historical background
information on transport. Their use requires a minimum amount of field
work and a minimum number of technical personnel. Disadvantages include
the irregularity of their occurrence, the difficulty in distinguishing the
tracer from the sediment itself, and a lack of quantitative control on
rates of injection. In addition, natural tracers are unable to reveal
short-term changes in the direction of transport and changes in material
sources.
Judge (1970) found that heavy mineral studies were unsatisfactory as

indicators of the direction of longshore transport for beaches between
Point Conception and Ventura, California, because of the lack of unique
mineral species and the lack of distinct longshore trends which could be
used to identify source areas. North of Point Conception, grain size and
heavy mineral distribution indicated a net southward movement. Cherry
(1965) concluded that the use of heavy minerals as an indicator of the
direction of coastal sand movement north of Drakes Bay, California was
generally successful.
b. Artificial Tracers . Artificial tracers may be grouped into two

general categories: radioactive or nonradioactive. In either case, the
tracers represent particles that are placed in an environment selected
for study, and are used for relatively short-term studies of sediment
dispersion.
While particular experiments employ specific sampling methods and

operational characteristics, there are basic elements in all tracing
studies. These are: selection of a suitable tracer material, tagging
the particle, placing the particle in the environment, and detection of
the particle.
Colored glass, brick fragments and oolitic grains are a few examples
of nonradioactive particles that have been used as tracers. The most com-
monly used stable tracer is made by coating indigenous grains with bright
colored paint or flourescent dye. (Yasso, 1962; Ingle, 1966; Stuvier and
Purpura, 1968; Kidson and Carr, 1962; and Teleki, 1966.) The dyes make
the grains readily distinguishable among large sample quantities, but do
not significantly alter the physical properties of the grains. The dyes
must be durable enough to withstand short-term abrasion. The use of paints
and dyes as tracer materials offers advantages over radioactive methods in
that they require less sophisticated equipment to tag and detect the grains,
and do not require licensing or the same degree of safety precautions. How-
ever, less information is obtained for the same costs, and generally in a
less timely matter.
When using nonradioactive tracers, samples must be collected and

removed from the environment to be analyzed later by physically counting
the grains. For fluorescent dyes and paints, the collected samples are
viewed under an ultraviolet lamp and the coated grains counted.
For radioactive tracer methods, the tracer may be radioactive at the

time of injection or it may be a stable isotope capable of being detected
by activation after sampling. The tracer in the grains may be introduced
4-151
by a number of methods. Radioactive material has been placed in holes
drilled in a large pebble. It has been incorporated in molten glass
which, when hardened, is crushed and resized (Sato, et al., 1962; and
Taney, 1963). Radioactive material has been plated on the surface of
natural sediments. (Stephens, et al., 1968.) Radioactive gas (krypton
85 and xenon 133) has been absorbed into quartz sand. (Chleck, et al.,
1963; and Acree, et al., 1969.)
In 1966, the Coastal Engineering Research Center, in cooperation

with the Atomic Energy Commission, initiated a multiagency program to
create a workable radioisotopic sand tracing (RIST) program, for use in
the littoral zone. (Duane and Judge, 1969.) Tagging procedures (by
surface plating with gold 198-199), instrumentation, field surveys and
data handling techniques were developed which permit collection and
analysis of over 12,000 bits of information per hour over a survey track
about 18,000 feet long.
These recent developments in radioactive tracing permit in situ

observations and faster data collection over much larger areas (Duane,
1970) than has been possible using fluorescent or stable isotope tracers.
However, operational and equipment costs of radionuclide tracer programs
can be high.
Accurate determination of long-term sediment transport volume is not

yet possible from a tracer study, but qualitative data on sediment move-
ment useful for engineering purposes can be obtained.
Experience has shown that tracer tests can give information on

direction of movement, dispersion, shoaling sources, relative velocity
and movement in various areas of the littoral zone, means of natural by-
passing, and structure efficiency. Reasonably quantitative data on move-
ment or shoaling rates can be obtained for short-time intervals. It
should be emphasized that this type of information must be interpreted
with care, since the data are generally determined by short-term littoral
transport phenomena. However, tracer studies conducted repeatedly over
several years at the same location could result in estimates of longer
term littoral transport.
4.83 SEDIMENT TRANSPORT CALCULATIONS
4.831 Longshore Transport Rate . The example calculation of a sediment

budget in Section 4.76 is typical in that the magnitude of the longshore
transport rate exceeds by a considerable margin any other element in the
budget. For this reason, it is essential to have a good estimate of the
longshore transport rate in an engineering study of littoral processes.
A complete description of the longshore transport rate requires

knowledge of two of the five variables
(Q,,,Qe,,Q^,Q„,7).
defined by Equations 4-21, 4-22, and 4-23. If any two are known, the
remaining three can be obtained from the three equations.
4-152
Section 4.531 describes four methods for estimating longshore trans-
port rate, and Sections 4.532 through 4.535 describe in detail how to use
two of these four methods. (See Methods 3 and 4.)
One approach to estimating longshore transport rate is to adopt a

proven estimate from a nearby locality, after making allowances for local
conditions. (See Method 1.) It requires considerable engineering judge-
ment to determine whether the rate given for the nearby locality is a
reliable estimate, and, if reliable, how the rate needs to be adjusted
to meet the changed conditions at the new locality.
Method 2 is an analysis of historical data. Such data may be found

in charts, maps, aerial photography, dredging records, beach fill records,
and related information. Section 4.811 describes some of these sources.
To apply Method 2, it is necessary to know or assume the transport

rate across one end of the littoral zone being considered. The most suc-
cessful applications of Method 2 have been where the littoral zone is
bounded on one end by a littoral barrier which is assumed to completely
block all longshore transport. The existence of such a complete littoral
barrier implies that the longshore transport rate is zero across the
barrier, and this satisfies the requirement that the rate be known across
the end of the littoral zone being considered. Examples of complete lit-
toral barriers include large jetties immediately after construction, or
spits building into deep quiet water.
Data on shoreline changes permit estimates of rates of erosion and

accretion that may give limits to the longshore transport rate. Figure
4-50 is a shoreline change map which was used to obtain the rate of
transport at Sandy Hook, New Jersey. (Caldwell, 1966.)
Method 3 (the energy flux method) is described in Section 4.532

with a worked example in Section 4.533. Method 4 (the empirical pre-
diction of gross longshore transport rate) is described in Section 4.534
with a worked example in Section 4.535. The essential factor in Methods
3 and 4, and often in Method 1, is the availability of wave data. Wave
data applicable to studies of littoral processes are discussed in detail
in Section 4.3.
4.832 Onshore-Offshore Motion . Typical problems requiring knowledge of

onshore-offshore sediment transport are described in Section 4.511. Four
classes of problems are treated:
(1) The seaward limit of significant sediment transport. Avail-

able field data and theory suggest that waves are able to move sand dur-
ing some days of the year over most of the Continental Shelf. However,
field evidence from bathymetry and sediment size distribution suggest
that the zone of significant sediment transport is confined close to
shore where bathymetric contours approximately parallel the shoreline.
The depth to the deepest shore-parallel contour tends to increase with
average wave height, and typically varies from 15 to 60 feet.
4-153
(2) Sediment transport in the nearshore zone. Seaward of the
breakers, sand is set in motion by waves moving over ripples, either
rolling the sand as bed load, or carrying it up in vortices as suspended
load. The sand, once in motion, is transported by mean tidal and wind-
induced currents and by the mass transport velocity due to waves. The
magnitude and direction of the resulting sediment transport are uncer-
tain under normal circumstances, although mass transport due to waves
is more than adequate to return sand lost from the beach during storms.
It appears that bottom mass transport acts to keep the sand close to the
shore, but that some material, probably finer sand, escapes offshore as
the result of the combined wind- and wave-induced bottom currents.
(3) The shape and expectable changes in shape of nearshore and

beach profiles. Storms erode beaches to produce a simple concave-up
beach profile with deposition of the eroded material offshore. Rates of
erosion due to individual storms vary from a few cubic yards per foot to
10 's of cubic yards per foot of beach front. The destructiveness of the
storm in producing erosion depends on its intensity, duration, and orienta-
tion, especially as these factors affect the elevation of storm surge and
the wave height and direction. Immediately after a storm, waves begin to
return sediment to the eroded beach, either through the motion of bar-and-
trough (ridge-and-runnel) systems, or by berm building. The parameter,
Fo = Ho/iVfT), given by Equation 4-20 determines whether the beach erodes
or accretes under given conditions. If V^ is above critical value
between 1 and 2 the beach erodes. (See Figures 4-29, and 4-30.)
(4) The slope of the foreshore. There is a tendency for the fore-
shore to become steeper as grain size increases, and to become flatter
as mean wave height increases. Data for this relation exhibit much scat-
ter and quantitative relationships are difficult to predict.
4.833 Sediment Budget. Section 4.76 siommarizes material on the sediment

budget. Table 4-14 tabulates the elements of the sediment budget and in-
dicates the importance of each element. Table 4-13 classifies the ele-
ments of the sediment budget.
A sediment budget carefully defines the littoral control volume,

identifies all elements transferring sediment to or from the littoral
control volume, ranks the elements by their magnitude, and provides an
estimate of unknown rates by the balancing of additions against losses
(Equation 4-46)
If prepared with sufficient data and experience, the budget permits

an estimate of how proposed improvements will affect neighboring segments
of the littoral zone.
4-154
ACREE, E.H., et al., "Radioisotopic S'lnd Tracer Study (RIST) , Status

Report for May 1966-April 1968," OKNL-4341, Contract No. W-7405-eng-
26, Oak Ridge National Laboratory (operated by Union Carbide Corp.,
for the U.S. Atomic Energy Commission), 1969.
AKYUREK, M., "Sediment Suspension by Wave Action Over a Horizontal Bed,"

Unpublished Thesis, University of Iowa, Iowa City, Iowa, July 1972.
ARTHUR, R.S., "A Note on the Dynamics of Rip Currents," Journal of

Geophysioal Researah, Vol. 67, No. 7, July 1962.
AYERTON, H, "The Origin and Growth of Ripple Marks," Philosophical

Transaotions of the Royal Society of London, Ser. A, Vol. 84, 1910,
pp. 285-310.
BAGNOLD, R„A„, "Mechanics of Marine Sedimentation," The Sea, Vol. 3,

Wiley, New York, 1963, pp. 507-528.
BAJORUNAS, L„, and DUANE, D.B., "Shifting Offshore Bars and Harbor Shoal-
ing," Journal of Geophysioal Research, Vol. 72, 1967, pp. 6195-6205.
BASCOM, W.N., "The Relationship Between Sand Size and Beach-Face Slope,"
Transactions of the American Geophysical Union, Vol. 32, No. 6, 1951.
BASCOM, W.N. , Manual of Amphibious Oceanography, Vol. 2, Sec. VI, Univer-

sity of California, Berkeley, Calif., 1952.
BASCOM, W.N., "Waves and Beaches," Beaches, Ch. IX, Doubleday, New York,
1964, pp. 184-212.
BASH, B.F., "Project Transition; 26 April to 1 June 1972," Unpublished

Research Notes, U.S. Army, Corps of Engineers, Coastal Engineering
Research Center, Washington, D.C., 1972.
BERG, D.W., "Systematic Collection of Beach Data," Proceedings of the

11th Conference on Coastal Engineering, London, Sept. 1968.
BERG, D.W., and DUANE, D.B., "Effect of Particle Size and Distribution on
Stability of Artificially Filled Beach, Presque Isle Penninsula,
Pennsylvania," Proceedings of the 11th Conference Great Lakes Research,
April 1968.
BERTMAN, D„Y., SHUYSKIY, Y.D., and SHKARUPO I.V., "Experimental Study

of Beach Dynamics as a Function of the Prevailing Wind Direction and
Speed," Oceanology, Translation of the Russian Journal, Okeanologiia
(Academy of Sciences of the U.S.S.R.), Vol. 12, Feb. 1972.
BIGELOW, H.B., and EDMONDSON, W.T., "Wind Waves at Sea Breakers and Surf,"
H.Oo 602, U.S. Navy Hydrographic Office, Washington, D.C. , 1947.
4-155
BIJKER, E.W., "Bed Roughness Influence on Computation of Littoral Drift,"
Abstracts of the 12th Coastal Engineering Conference^ Washington, D.C.,
1970.
BODINE, B.R., "Hurricane Surge Frequency Estimated for the Gulf Coast of
Texas," Bl-26, U.S. Army, Corps of Engineers, Coastal Engineering
Research Center, Washington, D.C., Feb. 1969.
BOWEN, A.J., and INMAN, D.L., "Budget of Littoral Sands in the Vicinity
of Point Arguello, California," TM-19, U.S. Army, Corps of Engineers,
Coastal Engineering Research Center, Washington, D.C. , Dec. 1966.
BOWMAN, R.S., "Sedimentary Processes Along Lake Erie Shore: Sandusky Bay,
Vicinity of Willow Point," in "Investigations of Lake Erie Shore Ero-
sion," Survey 18, Ohio Geological Survey, 1951.
BRETSCHNEIDER, C.L., "Fundamentals of Ocean Engineering - Part 1 -

Estimating Wind Driven Currents Over the Continental Shelf," Ooean
Industry J Vol. 2, No. 6, June 1967, pp. 45-48.
BRETSCHNEIDER, C.L., "Field Investigations of Wave Energy Loss in

Shallow Water Ocean Waves," TM-46, U.S. Army, Corps of Engineers,
Beach Erosion Board, Washington, D.C, Sept. 1954.
BRETSCHNEIDER, C.L., and REID, R.O., "Modification of Wave Height Due

to Bottom Friction, Percolation, and Refraction," TM-45, U.S. Army,
Corps of Engineers, Beach Erosion Board, Washington, D.C, Oct. 1954.
BRICKER, O.P., ed.. Carbonate Sediments, No. 19, The Johns Hopkins
University Studies in Geology, 1971, 376 pp.
BROWN, Ed., "Inlets on Sandy Coasts," Proceedings of the American

Society of Civil Engineers^ ASCE, Vol. 54, 1928, pp. 505-553.
BRUUN, P., "Sea-Level Rise as a Cause of Shore Erosion," Journal of the

Waterways and Harbors Division, ASCE, Vol. 88, WWl , Feb. 1962, pp.
117-130.
BRUUN, P., "Longshore Currents and Longshore Troughs," Journal of Geo-

physical Research, Vol. 68, 1963, pp. 1065-1078.
BRUUN, P., "Use of Tracers in Coastal Engineering," Shore and Beach,

No. 34, pp. 13-17, Nuclear Science Abstract, 20:33541, 1966.
BRUUN, P„ and GERRITSEN, F., "Natural Bypassing of Sand at Inlets,"

Joujpnal of the Waterways and Harbors Division, ASCE, Dec. 1959.
BUMPUS, D.F., "Residual Drift Along the Bottom on the Continental Shelf
in the Middle Atlantic Bight," Limnology and Oceanography, 75th
Anniversary Vol., Supplement to Vol. X, 1965.
4-156
BYRNE, R„J., "Field Occurrences of Induced Multiple Gravity Waves,"
Journal of Geophysical Besearoh^ Vol. 74, No. 10, May 1969, pp.
2590-2596.
CALDWELL, J.M., "Wave Action and Sand Movement Near Anaheim Bay,
California," TM-68, U.S. Army, Corps of Engineers, Beach Erosion
Board, Washington, D.C., Feb. 1956.
CALDWELL, J„M., "Shore Erosion by Storm Waves," MP 1-59, U.S. Army,

Corps of Engineers, Beach Erosion Board, Washington, D.C., Apr.
1959.
CALDWELL, J„M., "Coastal Processes and Beach Erosion," Journal of the

Boston Society of Civil Engineers^ Vol. 53, No. 2, Apr. 1966, pp.
142-157,
CARSTENS, M.K., NEILSON, M. , and ALTINBILEK, H.D., "Bed Forms Generated

in the Laboratory Under an Oscillatory Flow: Analytical and Experi-
mental Study," TM-2 8, U.S. Army, Corps of Engineers, Coastal Engineer-
ing Research Center, Washington, D.C. , June 1969.
CARTWRIGHT, D.E., "A Comparison of Instrumental and Visually Estimated

Wave Heights and Periods Recorded on Ocean Weather Ships," National
Institute of Oceanography, Oct. 1972.
CASTANHO, J., "Influence of Grain Size on Littoral Drift," Abstracts of

the 12th Coastal Engineering Conference, Washington, D.C, 1970,
CHERRY, Jo, "Sand Movement along a Portion of the Northern California

Coast," TM-14, U.S. Army, Corps of Engineers, Coastal Engineering
Research Center, Oct. 1965.
CHLECK, D., et al., "Radioactive Kryptomates," International Journal of

Applied Radiation and Isotopes, Vol. 14, 1963, pp. 581-610.
CHOW, V,T., ed.. Handbook of Applied Hydrology, McGraw-Hill, New York,

1964.
COLBY, B.C., and CHRISTENSEN, R.P., "Visual Accumulation Tube for Size
Analysis of Sands," Journal of the Hydraulics Division, ASCE, Vol. 82,
NOo 3, June 1956.
COLONY, R„J., "Source of the Sands on South Shore of Long Island and the
Coast of New Jersey," Journal of Sedimentary Petrology, Vol. 2, 1932,
pp. 150-159.
COOK, D.O., "Sand Transport by Shoaling Waves," Ph.D. Thesis, University

of Southern California, University Microfilms, Ann Arbor, Mich., 1970.
COOK, D.Oo, and GORSLINE, D.S., "Field Observations of Sand Transport

by Shoaling Waves," Marine Geology, Vol. 13, No. 1, 1972.
4-157
COOPER, W.S., "Coastal Sand Dunes of Oregon and Washington," Memoir No.
72, Geological Society of America, June 1958.
COOPERATIVE FEDERAL INTER-AGENCY PROJECT, "Methods of Analyzing Sediment

Samples," Report No. 4, St. Paul U.S. Engineer District Sub-Office,
Hydraulic Laboratory, University of Iowa, Iowa City, 1941.
CRICKMORE, J.M., and WATERS, C.B., "The Measurement of Offshore Shingle

Movement," 23th International Conferenoe on Coastal Engineering,
Vancouver, B.C., Canada, July 1972.
CUMMINS, R.S., Jr., "Radioactive Sediment Tracer Tests, Cape Fear River,
North Carolina," MP No. 2-649, Waterways Experiment Station, U.S.
Army, Corps of Engineers, Vicksburg, Miss., May 1964.
CURRAY, J.R., "Late Quartemary History, Continental Shelves of the United

States," The Quartemary of the U.S., Princeton University Press,
Princeton, N.J., 1965, pp. 723-735.
DABOLL, J.M., "Holocene Sediments of the Parker River Estuary, Massa-

chusetts," Contribution No. 3-CRG, Department of Geology, University
of Massachusetts, June 1969.
DARLING, J.M., "Surf Observations Along the United States Coasts," Journal
of the Waterways and Harbors Division, ASCE, WWl, Feb. 1968, pp. 11-21.
DARLING, J.M., and DUMM, D.G., "The Wave Record Program at CERC," MP
1-67, U.S. Army, Corps of Engineers, Coastal Engineering Research
Center, Washington, D.C., Jan. 1967.
DAS, M.M., "Longshore Sediment Transport Rates: A Compilation of Data,"

MP 1-71, U.S. Army, Corps of Engineers, Coastal Engineering Research
Center, Washington, D.C., Sept. 1971.
DAS, M.M., "Suspended Sediment and Longshore Sediment Transport Data

Review," IZth International Conference on Coastal Engineering,
Vancouver, B.C., Canada, July 1972.
DAVIS, R.A., Jr., "Sedimentation in the Nearshore Environment, South-
eastern Lake Michigan," Thesis, University of Illinois, Urbana,
Illinois, 1964.
DAVIS, R.A., Jr., "Beach Changes on the Central Texas Coast Associated
with Hurricane Fern, September 1971," Vol. 16, Contributions in Marine
Science, University of Texas, Marine Science Institute, Port Aransas,
Tex., 1972.
DAVIS, R.A., Jr., and FOX, W.T., "Beach and Nearshore Dynamics in Eastern
Lake Michigan," TR No. 4, ONR Task No. 388-092/10-18-68, (414), Office
of Naval Research, Washington, D.C., June 1971.
DAVIS, R.A., Jr., and FOX, W.T., "Coastal Processes and Nearshore Sand
Bars," Journal of Sedimentary Petrology, Vol. 42, No. 2, June 1972,
pp. 401-412.
4-158
DAVIS, R.A., Jr., and FOX, W.T., "Four-Dimensional Model for Beach and
Inner Nearshore Sedimentation," The Jowmal of Geology , Vol. 80, No.
4, July 1972.
DAVIS, R.A., Jr., et al., "Comparison of Ridge and Runnel Systems in

Tidal and Non-Tidal Environments," Journal of Seidmentavy Petrology
Vol. 42, No. 2, June 1972, pp. 413-421.
DAVIS, R.A., Jr., and MCGEARY, D.F.R., "Stability in Nearshore Bottom

Topography and Sedimentary Distribution, Southeastern Lake Michigan,"
Proceedings of the Eighth Conference on Great Lakes Research, Pub. No.
13, Great Lakes Research Division, University of Michigan, Ann Arbor,
Mich., 1965.
DEAN, R.G., "Relative Validities of Water Wave Theories," Journal of the

Waterways and Harbors Division, ASCE, Vol. 96, Ml, 1970, pp. 105-119.
DEAN, R.G., "Storm Characteristics and Effects," Proceedings of Seminar

on Planning and Engineering in the Coastal Zone, published by Coastal
Plains Center for Marine Development Services, Wilmington, N.C., 1972.
DEAN, R.G., "Heuristic Models of Sand Transport in the Surf Zone," Con-
ference on Engineering Dynamics in the Coastal Zone, 1973.
DEWALL, A.E., "The 17 December 1970 East Coast Storm Beach Changes,"
Unpublished Manuscript, U.S. Army, Corps of Engineers, Coastal
Engineering Research Center, Washington, D.C., 1972.
DEWALL, A.E., PRITCHETT, P.C., and GALVIN, C.J., Jr., "Beach Changes
Caused by a Northeaster Along the Atlantic Coast," Abstracts from the
Annual Meeting of the Geological Society of America, Washington, D.C.,
1971.
DEWALL, A.E., and RICHTER, J.J. , "Beach Changes at Boca Raton, Florida,"
Annual Meeting of the American Shore and Beach Preservation Associa-
tion, (presentation only) Fort Lauderdale, Fla. , Aug. 1972.
DIETZ, R.S., "Wave Base, Marine Profile of Equilibrium, and Wave-Built

Terraces: A Critical Appraisal," The Geological Society of America
Bulletin, Vol. 74, No. 8, Aug. 1963, pp. 971-990.
DILL, R.F., "Sedimentation and Erosion in Scripps Submarine Canyon Head,"

Papers in Marine Geology, Shepard Commemorative Volume, Macmillan,
New York, 1964.
DIVOKY, D., LeMEHAUTE, B., and LIN, A., "Breaking Waves on Gentle Slopes,"
Journal of Geophysical Research, Vol. 75, No. 9, 1970.
DOLAN, R., "Barrier Dune System Along the Outer Banks of North Carolina:
A Reappraisal," Science, 1972, pp. 286-288.
DOLAN, R„, "Barrier Islands: Natural and Controlled," in "Coastal

Geomorphology," Proceedings of the Third Annual Geomorphology Symposia
Series, 1973, pp. 263-278.
4-159
DOLAN, R., PERM, J.C., and McARTHUR, L.S., "Measurements of Beach Process
Variables, Outer Banks, North Carolina," TR No. 64, Coastal Studies
Institute, Louisiana State University, Baton Rouge, La., Jan. 1969.
DRAPER, L., "Wave Activity at the Sea Bed Around Northwestern Europe,"
Marine Geology ^ Vol. 5, 1967, pp. 133-140.
DUANE, D.B., "Sand Deposits on the Continental Shelf: A Presently

Exploitable Resource," Trans actions of National Symposium on Oeean
Sciences and Engineering of the Atlantic Shelf^ Marine Technology
Society, Mar. 1968.
DUANE, D.B., "Synoptic Observations of Sand Movement," Proceedings of

the 12th Coastal Engineering Conference, Washington, D.C. September
1970.
DUANE, D.B., "Tracing Sand Movement in the Littoral Zone: Progress in

the Radioisotopic Sand Tracer (RIST) Study, July 1968- February 1969,"
MP 4-70, U.So Army, Corps of Engineers, Coastal Engineering Research
Center, Washington, D.C, Aug. 1970.
DUANE, D.B., et al., "Linear Shoals on the Atlantic Inner Continental

Shelf, Florida to Long Island," Shelf Sediment Transport, edited by
Swift, Duane, and Pilkey; Dowden, Hutchinson, and Ross, Stroudsburg,
Pao 1972.
DUANE, D.B., and JUDGE, C.W., "Radioisotopic Sand Tracer Study, Point
Conception, California," MP 2-69, U.S. Army, Corps of Engineers,
Coastal Engineering Research Center, Washington, D.C, May 1969.
DUANE, D.B., and MEISBURGER, E.P., "Geomorphology and Sediments of the

Nearshore Continental Shelf, Miami to Palm Beach, Florida," TM-29,
U.S. Army, Corps of Engineers, Coastal Engineering Research Center,
Washington, D.C, Nov. 1969.
DUBOIS, R.N., "Inverse Relation Between Foreshore Slope and Mean Grain
Size as a Function of the Heavy Mineral Content," Geological Society
of America Bulletin, Vol. 83, Mar. 1972, pp. 871-876.
DUNN, G.E., and MILLER, B.I., Atlantic Hurricanes, Louisiana State

University Press, Baton Rouge, 1964, pp. 204, 257-258.
EAGLESON, PoS., "Properties of Shoaling Waves by Theory and Experiment,"

Transactions of the American Geophysical Union, Vol. 37, 1956, pp. 565-
572.
EAGLESON, P.S., and DEAN, R.G., "Wave-Induced Motion of Discrete Bottom

Sediment Particles," Proceedings of the American Society of Civil
Engineers, Vol. 85, No. HYIO, 1959=
EAGLESON, P.S., and DEAN, R.G., "A Discussion of 'The Supply and Loss of
Sand to the Coast' by J. Wo Jolinson," Journal of the Waterways and
Harbors Division, ASCE, Vol. 86, No. WW2, June 1960.
4-160
EATON, R„0., "Littoral Processes on Sandy Coasts," Proceedings of the
First Conference on Coastal Engineering ^ Council on Wave Research,
Engineering Foundation, Oct. 1950.
EMERY, K,0., "The Sea off Southern California," Sediments, Ch. 6,

Wiley, New York, 1960, pp. 180-295.
EMERY, K.O., "A Simple Method of Measuring Beach Profiles," Limnology

and Oceanography y Vol. 6, No. 1, 1961, pp. 90-93.
EVANS, O.F., "The Classification of Wave-Formed Ripple-Marks," Journal of

Sedimentary Petrology, Vol. 11, No. 1, Apr. 1941, pp. 37-41.
EVERTS, C.H., "A Rational Approach to Marine Placers," Unpublished Ph.D.

Thesis, University of Wisconsin, Madison, Wis., 1971.
EVERTS, C„H., "Geometry of Profiles Across Some Inner Continental Shelves

of the United States," Unpublished Research Paper, U.S. Army, Corps of
Engineers, Coastal Engineering Research Center, Washington, D.C., 1972.
EVERTS, C.H., "Particle By-Passing on Loose, Flat Granular Boundaries,"

Transactions of the American Geophysical Union, Vol. 53, No. 4, Apr.
1972.
EVERTS, CH,, "Beach Profile Changes in Western Long Island," in "Coastal

Geomorphology," Proceedings of the Third Annual Geomorphology Symposia
Series, 1973, pp. 279-301.
FAIRCHILD, J.C., "Development of a Suspended Sediment Sampler for Labora-

tory Use Under Wave Action," Bulletin of the Beach Erosion Board, Vol.
10, No. 1, July 1956.
FAIRCHILD, J.C., "Suspended Sediment Sampling in Laboratory Wave Action,"

D.C., June 1959.
FAIRCHILD, J.C, "Laboratory Tests of Longshore Transport," Proceedings of

the 12th Conference on Coastal Engineering, ASCE, 1970.
-FAIRCHILD, J.C, "Longshore Transport of Suspended Sediment," Proceedings

of the 13th Coastal Engineering Conference, Vancouver, B.C., Canada,
July 1972.
FISHER, CM., "Mining the Ocean for Beach Sand," Proceedings of Civil
Engineering in the Oceans II, ASCE, Dec. 1969.
FOLK, R.L,, Petrology of Sedimentary Rooks, Hemphill's, Austin Tex., 1965.
FOLK, R.L., "A Review of Grain Size Parameters," Sedimentology , Vol. 6,

1966, pp. 73-93.
4-161
FOLK, R.L., and WARD, W.C., "Brazos River Bar. A Study in the Signifi-
cances of Grain Size Parameters," Journal of Sedimentcay Petyyology,
Vol. 27, 1957, pp. 3-26.
FOX, W.T., "Anatomy of a Storm on the Lake Michigan Coast," Unpublished

Paper, Confevenoe on Effects of Extreme Conditions on Coastal Environ-
mentSy Western Michigan University, Nov. 1970.
GAGE, B.O., "Experimental Dunes of the Texas Coast," MP 1-70, U.S. Army,
Corps of Engineers, Coastal Engineering Research Center, Washington,
D.C., Jan. 1970.
GALVIN, C.J., Jr., "Experimental and Theoretical Study of Longshore Cur-

rents on a Plane Beach," Unpublished Ph.D. Thesis, Department of Geolo-
gy and Geophysics, Massachusetts Institute of Technology, Cambridge,
Mass., 1963.
GALVIN, C.J., Jr., "A Theoretical Distribution of Waiting Times for Tracer
Particles on Sand Bed," Vol. 1, Bulletin and Summary of Research Pro-
gress, Fiscal Year 1964, U.S. Army, Corps of Engineers, Coastal
GALVIN, CoJ. Jr., "A Selected Bibliography and Review of the Theory and
Use of Tracers in Sediment Transport Studies," Vol. I, Bulletin and
Summary Report of Research Progress, Fiscal Year 1964, U.S. Army,
D.C., 1964.
GALVIN, C.J., Jr., "Longshore Current Velocity: A Review of Theory and

Data," Reviews of GeophysicSj Vol. 5, No. 3, Aug. 1967.
GALVIN, C.J., Jr., "A Gross Longshore Transport Rate Formula," Proceedings
of the 13th Coastal Engineering Conference^ Vancouver, B.C., Canada,
July 1972.
GALVIN, C.J., Jr., "Wave Breaking in Shallow Water," âves on Beaches and
Resulting Sediment Transport^ Academic Press, March 1972.
GALVIN, C.J., Jr., "Finite-Amplitude, Shallow Water-Waves of Periodically

Recurring Form," Proceedings of the Symposium on Long Waves ^ University
of Delaware, Sept. 1970.
GALVIN, C.J., "Wave Climate and Coastal Processes," Water Environments and
Human Needs^ A.T. Ippen, Ed., M.I.T. Parsons Laboratory for Water
Resources and Hydrodynamics, Cambridge, Mass., 1971., pp. 48-78.
GALVIN C.J., Jr., and EAGLESON, P.S., "Experimental Study of Longshore

Currents on a Plane Beach," TM-10, U.S. Army, Corps of Engineers,
Coastal Engineering Research Center, Washington, D.C. , Jan. 1965.
GALVIN, C.Jo, and NELSON, R.A., "A Compilation of Longshore Current Data,"
Center, Washington, D.C, Mar. 1967.
4-162
GALVIN, C.J., and SEELIG, W.N., "Surf on U.S. Coastline," Unpublished
Research Paper, U.S. Army, Corps of Engineers, Coastal Engineering
Research Center, Washington, D.C., Aug. 1969.
GALVIN, C.J., and SEELIG, W.N., "Nearshore Wave Direction from Visual
Observation," Transactions of the American Geophysical Union^ Vol. 52,
No. 4, Apr. 1971.
GIBBS, R.J., "The Geochemistry of the Amazon River System: Part I. The
Factors that Control the Salinity and the Composition and Concentration
of the Suspended Solids," Geological Society of America Bulletin^ Vol.
78, Oct. 1967, pp. 1203-1232.
GIBBS, R.J., "The Accuracy of Particle-Size Analysis Utilizing Settling

Tubes," Journal of Sedimentary Petrology^ Vol. 42, No. 1, Mar. 1972.
GILES, RoT., and PILKEY, O.H., "Atlantic Beach and Dune Sediments of the
Southern United States," Journal of Sedimentary Petrology ^ Vol. 35,
No. 4, Dec. 1965, pp. 900-910.
GODA, Y., "A Synthesis of Breaker Indices," Proceedings of the Japan

Society of Civil Engineers ^ No. 180, Aug. 1970.
GODFREY, P.J., and GODFREY, M.M., "Comparison of Ecological and Geomorphic

Interactions Between Altered and Unaltered Barrier Island Systems in
North Carolina," in "Coastal Geomorphology," Proceedings of the Third
Annual Geomorphology Symposia Series ^ 1972, pp. 239-258.
GOLDSMITH, V., COLONELL, J.M., and TURBIDE, P.W., "Forms of Erosion and
Accretion on Cape Cod Beaches," Proceedings of the 13th International
Conference on Coastal Engineering ^ Vancouver, B.C., July 1972.
GONZALES, W„R., "A Method for Driving Pipe in Beach Rock," Vol. Ill,
Bulletin and Summary of Research Progress, Fiscal Years 1967-69, U.S.
Army, Corps of Engineers, Coastal Engineering Research Center, Wash-
ington, D.C., July 1970.
GRAF, W.H., Hydraulics of Sediment Transport^ McGraw-Hill, New York, 1971.
GRAHAM, H.E., and NUNN, D.E., "Meteorological Considerations Pertinent

to Standard Project Hurricane, Atlantic and Gulf Coasts of the United
States," National Hurricane Research Project Report No. 33, U.S. Depart-
ment of Commerce, Weather Bureau, Washington, D.C., Nov., 1959.
GRIFFITHS, J.C, "Scientific Method in Analysis of Sediments," McGraw-Hill,

New York, 1967.
HALL, J. v., and HERRON, W.J., "Test of Nourishment of the Shore by Off-
shore Deposition of Sand," TM-17, U.S. Army, Corps of Engineers, Beach
Erosion Board, Washington, D.C., June 1950.
4-163
HALLERMEIER, R.J., and GALVIN, C.J., "Wave Height Variation Around
Vertical Cylinders," Transactions of the Amerioan Geophysical Union
53rd Annual Meeting, 1972, p. 397.
HANDIN, J.W., "The Source, Transportation, and Deposition of Beach

Sediment in Southern California," TM-22, U.S. Army, Corps of Engi-
neers, Beach Erosion Board, Washington, D.C., Mar. 1951.
HANDS, E.B., "A Geomorphic Map of Lake Michigan Shoreline," Proceedings

of the 13th Conference on Great Lakes Research^ International Associa-
tion Great Lakes Research, 1970, pp. 250-265.
HANDS, E.B., "Anomalous Sand Size-Swash Slope Relationship," Unpublished

MFR, U.S. Army, Corps of Engineers, Coastal Engineering Research Center,
Washington, D.C., Apr. 1972.
HARRIS, D.L., "The Analysis of Wave Records," Proceedings of the Conference

on Coastal Engineering, Washington, D.C., 1970, pp. 85-100.
HARRIS, D.L., "Characteristics of Wave Records in the Coastal Zone,"

leaves on Beaches and Resulting Sediment Transport, Academic Press, 1972.
HARRIS, D.L., "Wave Estimates for Coastal Regions," Shelf Sediment Trans-
port, edited by Swift, Duane, and Pilkey. Dowden, Hutchinson, and
Ross, Stroudsburg, Pa., 1972.
HARRISON, W. , et al., "Circulation of Shelf Waters off the Chesapeake

Bight," ESSA Professional Paper 3, U.S. Dept. of Commerce, Washing-
ton, D.C., June 1967.
HARRISON, W., and WAGNER, K.A., "Beach Changes at Virginia Beach, Virginia,"
Center, Washington, D.C., Nov. 1964.
HART, E.D., "Radioactive Sediment Tracer Tests, Houston Ship Channel,

Houston, Texas," H-69-2, U.S. Army Engineer Waterways Experiment
Station, Vicksburg, Miss. , 1969.
HAYES, M.O., "Hurricanes as Geological Agents: Case Studies of Hurricanes

Carla, 1961, and Cindy, 1963," Report 61, Bureau of Economic Geology,
University of Texas, Austin, Tex., 1967.
HAYES, M.O., "Relationship between Coastal Climate and Bottom Sediment on

the Inner Continental Shelf," Marine Geology, 1967, pp. 111-132.
HAYES, M.O., "Forms of Sediment Accumulation in the Beach Zone," Waves on

Beaches and Resulting Sediment Transport, Academic Press, New York,
Octo 1971.
HAYES, M.O., GOLDSMITH, V., and HOBBS, C.H. Ill, "Offset Coastal Inlets,"
Proceedings of the 12th Coastal Engineering Conference, Sept. 1970,
pp. 1187-1200.
4-164
HERRON, W.J., and HARRIS, R.L., "Littoral Bypassing and Beach Restoration
in the Vicinity of Port Huenerae, California," Proceedings of the 10th
Conferenae on Coastal Engineering^ Tokyo, 1966, ASCE, United Engineer-
ing Center, New York, 1967.
HICKS, S.D., "On Classifications and Trends of Long Period Sea Level
Series," Shore and Beach^ Apr. 1972.
HJULSTROM, F., "Transportation of Detritus by Moving Water," Recent Marine

Sediments^ American Association of Petroleum Geologists, 1939.
HODGES, T.Ko, "Sand Bypassing at Hillsboro Inlet, Florida," Bulletin of

the Beach Erosion Boards Vol. 9, No. 2, Apr. 1955.
HOUBOLT, J.J.H.C., "Recent Sediment in the Southern Bight of the North

Sea," Geologie en MijnbouWj Vol. 47 (4), 1968, pp. 245-273.
HSU, So "Coastal Air-Circulation System: Observations and Empirical

Model," Monthly Weather Review , VoU 98, No. 7, July 1970.
HUBBELL, D.W., and SAYRE, W.W., "Sand Transport Studies with Radioactive
Tracers," Journal of the Hydraulics Division^ ASCE, Vol. 90, No. HY3,
1965, pp. 39-68.
HULSEY, J.D., "Beach Sediments of Eastern Lake Michigan," Ph.D. Disserta-

tion, University Microfilming 62-6164, University of Illinois, Urbana,
111., 1962.
HUME, J.D., and SCHALK, M., "Shoreline Processes Near Barrow, Alaska: A
Comparison of the Normal and the Catastrophic," Arctic, Vol. 20, No. 2,
June 1967, pp. 86-103.
HUSTON, K.H., "A Critical Appraisal of the Technique of Using Naturally

Occurring Radioactive Materials as Littoral Tracers," HEL-4-1, Hydraulic
Engineering Institute, University of California, Berkeley, Calif., 1963.
INGLE, J.Co, "The Movement of Beach Sand," Bevel. Sediment, Vol. 5, Elsevier,
Amsterdam, 1966.
INGRAM, L„F., CUMMINS, R„S., and SIMMONS, H.B., "Radioactive Sediment

Tracer Tests Near the North and South Jetties, Galveston Harbor Entrance,"
MP 2-472, U.S. Army Engineer Waterways Experiment Station, Vicksburg,
Miss., Nov. 1965.
INMAN, DoLo, "Sorting of Sediments in the Light of Fluid Mechanics,"

Journal of Sedimentary Petrology, Vol. 19, 1949, pp. 51-70.
INMAN, D.L., "Measures for Describing the Size Distribution of Sediments,"

Journal of Sedimentary Petrology, Vol. 22, No. 3, Sept. 1952, pp. 125-
145.
4-165
INMAN, D.L., "Wave-Generated Ripples in Nearshore Sands," TM-100, U.S.
Army, Corps of Engineers, Beach Erosion Board, Washington, D.C., Oct.
1957.
INMAN, D.L., "Sediments: Physical Properties and Mechanics of Sedimenta-

tion," Submarine Geology, P.P. Shepard, 2nd ed. , Harper and Row, New
York, 1963.
INMAN, D.L., and Quinn, W.H., "Currents in the Surf Zone," Proceedings of
the Second Conference on Coastal Engineering, ASCE, Council on Wave
Research, Berkeley, Calif., 1952, pp. 24-36.
INMAN, D.L., and NASU, N., "Orbital Velocity Associated with Wave Action
Near the Breaker Zone," TM-79, U.S. Army, Corps of Engineers, Beach
Erosion Board, Washington, D.C., Mar. 1956.
INMAN, D.L., and RUSNAK, G.S., "Changes in Sand Level on the Beach and
Shelf at La Jolla, California," TM-82, U.S. Army, Corps of Engineers,
Beach Erosion Board, Washington, D.C., July 1956.
INMAN, D.L., and BOWEN, A.J., "Flume Experiments on Sand Transport by

Waves and Currents," Proceedings of the Eighth Conference on Coastal
Engineering^ ASCE, Council on Wave Research, 1963, pp. 137-150.
INMAN, D.L., and FRAUTSCHY, J.D., "Littoral Processes and the Development
of Shorelines," Proceedings of the Coastal Engineering Specialty Con-
ference (Santa Barbara) ^ ASCE, 1966, pp. 511-536.
INf4AN,D.L., KOMAR, P.D., and BOWEN, A.J., "Longshore Transport of Sand,"

Proceedings of the 11th Conference on Coastal Engineering , ASCE, London,
1969.
INMAN, D.L., TAIT, R.J., and NORDSTROM, C.E., "Mixing in the Surf Zone,"
Journal of Geophysical Research^ Vol. 76, No. 15, May 1971, p. 3493.
IVERSEN, H.W., "Waves and Breakers in Shoaling Water," Proceedings of the

Third Conference on Coastal Engineering^ Council on Wave Research, ASCE,
1952, pp. 1-12.
IWAGAKI, Y, and NODA, H., "Laboratory Study of Scale Effect in Two Dimen-
sional Beach Processes," Proceedings of the Eighth Conference on Coastal
Engineering, Ch. 14, ASCE, 1962.
JAMES, W.R., "A Class of Probability Models for Littoral Drift," Proceedings
of the 12th Coastal Engineering Conference, Washington, D.C., Sept. 1970.
JOHNSON, D.W., Shore Processes and Shoreline development, Hafner Publishing,

New York, 1919.
JOHNSON, J.W., "Sand Transport by Littoral Currents," lER, Technical Report

Series 3, Issue 338, Wave Research Laboratory, University of California,
Berkeley, Calif., June 1952.
4-166
JOHNSON, J.W., "Sand Transport by Littoral Currents," Prooeedings of the
Fifth Hydraulios Conference^ Bulletin 34, State University o£ Iowa,
Studies in Engineering, 1953, pp. 89-109.
JOHNSON, J.W., "Dynamics of Nearshore Sediment Movement," American Associa-

tion of Petroleum Geology Bulletin, Vol. 40, No. 9, 1956, pp. 2211-2232.
JOHNSON, J.W., "The Littoral Drift Problem at Shoreline Harbors," Joiomal

of the Waterways and Harbors Division, ASCE, 83, WWl, Paper 1211, Apr.
1957.
JOHNSON, J.W., "The Supply and Loss of Sand to the Coast," ASCE Journal,
Vol. 85, NOo WW3, Sept. 1959, pp. 227-251.
JUDGE, CoW., "Heavy Minerals in Beach and Stream Sediments as Indicators

of Shore Processes between Monterey and Los Angeles, California," TM-33,
Washington, D.C., Nov. 1970.
JUDSON, S., "Erosion of the Land or What's Happening to Our Continents,"

American Scientist, Vol. 56, No. 4, 1968, pp. 356-374.
KAMEL, A.M., "Littoral Studies near San Francisco Using Tracer Techniques,"
1^-131, U.S. Army, Corps of Engineers, Beach Erosion Board, Washington,
D.C., Nov. 1962.
KAMEL, A.M., and JOHNSON, J.W., "Tracing Coastal Sediment Movement by

Naturally Radioactive Minerals," Proceedings of the Eighth Coastal
Engineering Conference, ASCE, 1962, p. 324.
KAMPHUIS, J.W., "The Breaking of Waves at an Angle to the Shoreline and

the Generation of Longshore Currents (A Laboratory Investigation),"
Unpublished Thesis, Queens University, Kingston, Ontario, Apr. 1963.
KENNEDY, J.F., and FALCON, M., "Wave -Gene rated Sediment Ripples," Hydro-
dynamics Laboratory Report 86, Department of Civil Engineering,
Massachusetts Institute of Technology, Cambridge, Mass., 1965.
KENNEDY, J.F., and LOCHER, F.A., "Sediment Suspension by Water Waves,"

]fJaoes on Beaches and Resulting Sediment Transport, Academic Press,
New York, 1972.
KENNEDY, V.C., and KOUBA, D.L., "Fluorescent Sand as a Tracer of Fluvial

Sediment," Open-File Report, U.S. Geological Survey, Washington, D.C.,
1968.
KEULEGAN, G.H., "An Experimental Study of Submarine Sand Bars," TR-3,

Aug. 1948.
KEULEGAN, G.H., "Wind Tides in Small Closed Channels," Research Paper

No. 2207, National Bureau of Standards, Washington, D.C., 1951.
67
KIDSON, C, and CARR, A. P., "Marking Beach Materials for Tracing Experi-
ments," Jonxmdl of the Hydrcculias Division^ ASCE, Vol. 88, July 1962.
KING, C.A.M., Beaches and Coasts, Edward Arnold, Ltd., 1972.
KINSMAN, BLAIR, Wind Waves, Their Generation and Propagation on The Ocean
Surface, Prentice-Hall, Englewood Cliffs, N.J,, 1965.
KIRTLEY, D.W. , "Reef-Building Worms," Sea Frontiers, Vol. 17, No. 2, 1971.
KOHLER, R.R., and GALVIN, C.J., "Berm-Bar Criterion," Unpublished CERC

Laboratory Report, Aug. 1973, 70 p.
KOLESSAR, M.A., and REYNOLDS, J.L., "The Sears Sea Sled for Surveying
in the Surf Zone," Vol. II, Bulletin and Sununary Report of Research
Progress, Fiscal years 1965-66, U.S. Army, Corps of Engineers, Coastal
KOMAR, P.D., "The Longshore Transport of Sand on Beaches," Unpublished

Ph.D. Thesis, University of California, San Diego, Calif., 1969.
KOMAR, P.D., "The Mechanics of Sand Transport on Beaches," Journal of

Geophysical Research, Vol. 76, No. 3, Jan. 1971.
KOMAR, PAUL D., and INMAN, DOUGLAS L., "Longshore Sand Transport on
Beaches, Journal of Geophysical Research, V. 75, No. 30, Oct 20, 1970,
pp. 5914-5927.
KRAAI, P.T., "Comparison of Wind Wave and Uniform Wave Effects on a

Beach," HEL 1-13, Hydraulic Engineering Laboratory, College of
Engineering, University of California, Berkley, Calif., Aug. 1969.
KRAFT, J.C, "A Guide to the Geology of Delaware's Coastal Environments,"

Publication 2GL039, College of Marine Studies, University of Delaware,
Newark, Del., 1971.
KRINSLEY, D., et al , "Transportation of Sand Grains Along the Atlantic

.
Shore of Long Island, New York: An Application of Electron Micros-

copy," Marine Geology, Vol. 2, 1964, pp. 100-121.
KRUMBEIN, W.C., "Application of Logarithmic Moments to Size Frequency

Distribution of Sediments," Journal of Sedimentary Petrology, Vol. 6,
No. 1, 1936, pp. 35-47.
KRUMBEIN, W.C, "Shore Currents and Sand Movement on a Model Beach,"

D.C., Sept. 1944.
KRUMBEIN, W.C, "A Method for Specification of Sand for Beach Fills,"
114-102, U.S.Army, Corps of Engineers, Beach Erosion Board, Washington,
D.C., Oct. 1957.
4-168
KRUMBEIN, W.C, and JAMES, W.R., "A Lognormal Size Distribution Model for
Estimating Stability of Beach Fill Material," TM-16, U.S. Army, Corps
of Engineers, Coastal Engineering Research Center, Washington, D.C.,
Nov. 1965.
KRUMBEIN, W.C, and SLOSS, L.L., "Stratigraphy and Sedimentation," Ch. 4,

Properties of Sedimentary Rooks ^ W.H. Freeman § Company, 1963,
pp. 93-149.
KUENEN, P.H., Marine Geology, Wiley, New York, 1950.
KUENEN, P.H., "Experimental Abrasion of Pebbles, Rolling by Current,"

Journal of Geology ^ Vol. 64, 1956, pp. 336-368.
LEE, J., YANCY, T., and WILDE, P., "Recent Sedimentation of the Central
California Continental Shelf," Part A: Introduction and Grain Size
Data, HEL 2-28, College of Engineers, University of California,
Berkeley, Calif., Oct. 1970.
LeMEHAUTE, D., and LIN, "Internal Characteristics of Explosion-Generated

Waves on the Continental Shelf," Tetra Technical Report No. TC - 116,
1968.
LONGUET-HIGGINS, M.S., "Longshore Currents Generated by Obliquely Incident

Sea Waves, 1", Journal of Geophysical Research ^ Vol. 75, No. 33, Nov.
1970a, pp. 6788-6801.
LONGUET-HIGGINS, M.S., "Longshore Currents Generated by Obliquely Incident

Sea Waves, 2," Journal of Geophysical Research^ Vol. 75, No. 33, Nov.
1970b, pp. 6790-6801.
LONGUET-HIGGINS, M.S., "Recent Progress in the Study of Longshore Currents,"

Waves on Beaches and Resulting Sediment Transport ^ Academic Press, New
York, Oct. 1971.
McCAMMON, R.B., "Efficiencies of Percentile Measures for Describing the

Mean Size and Sorting of Sedimentary Particles," Journal of Geology^
Vol. 70, 1962, pp. 453-465.
McCORMICK, C.L., "A Probable Cause for Some Changes in the Beach Erosion
Rates on Eastern Long Island," Unpublished Paper, Southampton College,
Long Island University, New York, 1971.
McMASTER, R.L., "Petrography and Genesis of the N.J. Beach Sands,"

Geology Series, Vol. 63, New Jersey Department of Conservation §
Economic Development, 1954.
MAGOON, O.T., and SARLIN, W.O., "Effect of Long-Period Waves on Hydro-

graphic Surveys," Proceedings of the 12th Coastal Engineering Con
ferenoe^ Washington, D.C., September 1970.
4-169
MAGOON, O.T,, HAUGEN, J.C., and SLOAN, R.L., "Coastal Sand Mining in
Northern California, U.S.A.," Pvoaeedings of the ISth Coastal
Engineering Research Conference, Vancouver, B.C., Canada, July
1972.
MANOHAR, M,, "Mechanics of Bottom Sediment Movement Due to Wave Action,"

D.C., June 1955.
MARTENS, J.H.C., "Beaches of Florida," Annual Report (21st-22nd), Florida

State Geological Survey, 1928-1930.
MASON, M.A., "Abrasion of Beach Sands," TM-2, U.S. Army, Corps of Engineers,
Beach Erosion Board, Washington, D.C., Feb. 1942.
MATHER, J.R., ADAMS, H. , and YOSHIOKA, G.A. , "Coastal Storms of the

Eastern United States," Journal of Applied Meteorology , Vol. 3, 1964,
pp. 693-706.
MEADE, R., "Landward Transport of Bottom Sediments in Estuaries of the

Atlantic Coastal Plain, Journal of Sedimentary Petrology, Vol. 39,
No. 1, pp. 222-234, 1969.
MEI, C.C, LIU, P., and CARTER, T.G., "Mass Transport in Water Waves,
Part I: Theory - Part II: Experiments," Report No. 146, Ralph M.
Parsons Laboratory for Water Resources and Hydrodynamics, Massachu-
setts Institute of Technology, Apr. 1972.
MENARD, H.W., "Sediment Movement in Relation to Current Velocity,"

Journal of Sediment Petrology, Vol. 20, 1950, pp. 148-160.
MILLIMAN, J.D., "Atlantic Continental Shelf and Slope of the United

States-Petrology of the Sand Friction of Sediments, Northern New
Jersey to Southern Florida," Geological Survey Professional Paper
529-J, U.S. Government Printing Office, Washington, D.C., 1972.
MOBERLEY, R. , "Loss of Hawaiian Sand," Journal of Sedimentary Petrology

Vol. 38, 1968, pp. 17-34.
MOGRIDGE, G.R. , and KAMPHUIS, J.W., "Experiments on Bed Form Generation

by Wave Action," Proceedings of the 12th Conference on Coastal Engi-
neering, July 1972.
MONROE, F.F., "Oolitic Aragonite and Quartz Sand: Laboratory Comparison

Under Wave Action," MP 1-69, U.S. Army, Corps of Engineers, Coastal
Engineering Research Center, Washington, D.C., Apr. 1969.
NDODY, D.W., "Coastal Morphology and Processes in Relation to the Develop-

ment of Submarine Sand Ridges off Bethany Beach, Del.," Unpublished
Ph.D. Dissertation, The John Hopkins University, Baltimore, Md. , 1964.
4-170
MOORE, G.W., and COLE, A.Y., "Coastal Processes, Vicinity of Cape
Thompson, Alaska," in "Geologic Investigations of Cape Thompson, NW
Alaska - Preliminary Report," Trace Element Investigation Report 753,
U.S. Geological Survey, Washington, D.C., 1960.
MORGAN, J. P., NICHOLS, L.G., and WRIGHT, M. , "Morphological Effects of

Hurricane Audrey on the Louisiana Coast (Advance Summary and Con-
clusions) ," TR lOA, Coastal Studies Institute, Louisiana State
University, Baton Rouge, La., 1958.
MORSE, B., and GROSS, M.G., and BAJWES, C.A., "Movement of Seabed
Drifters Near the Columbia River," Journal of the Waterways and
Hcwbors Division, ASCE, Vol. 94, No. WWl, Paper No. 5817, Feb. 1968,
pp. 93-103.
MURRAY, S.P., "Bottom Currents Near the Coast During Hurricane Camille,"
Journal of Geophysical Research, Vol. 75, No. 24, Aug. 1970.
NAKAMICHI, M., and SHIRAISHI, N., "Recent Researches on Sand Drift in

Japan," Proceedings of the 20th International Navigation Congress,
Baltimore, 1961, pp. 101-124.
NATIONAL ACADEMY OF SCIENCES, "Ocean Wave Spectra," Proceedings of

National Research Covmcil Conference, Easton, Md., May 1-4, 1961,
Prentice-Hall, Englewood Cliffs, N.J., 1963.
NATIONAL MARINE CONSULTANTS, INC., "Wave Statistics for Seven Deep Water
Stations Along the California Coast," prepared for Department of the
Army, U.S. Army, Corps of Engineers Districts, Los Angeles and San
Francisco, Calif., Dec. 1960a.
NATIONAL MARINE CONSULTANTS, INC., "Wave Statistics for Ten Most Severe
Storms Affecting Three Selected Stations off the Coast of Northern
California, Ehirmg the Period 1951-1960," prepared for U.S. Army
Engineer District, San Francisco, Calif., Dec. 1960b.
NAYAK, loV., "Equilibrium Profiles of Model Beaches," Report HEL 2-25,

University of California, Berkeley, Calif., May 1970.
NICHOLS, R.L., and MARSTON, A.F., "Shoreline Changes in Rhode Island

Produced by Hurricane of Sept. 21, 1938," Bulletin of the Geological
Society of America, Vol. 50, Sep. 1939, pp. 1357-1370.
NORRIS, R.M., "Dams and Beach-Sand Supply in Southern California,"

Papers in Marine Geology, Shepard Commemorative Volume, MacMillan,
New York, 1964.
O'BRIEN, M.P., "Equilibrium Flow Areas of Inlets on Sandy Coasts,"

Journal of the Waterways and Harbors Division, ASCE, No. WWl, Feb.
1969, pp. 43-52.
OTTO, G.H., "A Modified Logarithmic Probability Graph for the Interpreta-
tion of Mechanical Analyses of Sediments," Journal of Sedimentary
Petrology, Vol. 9, 1939, pp. 62-76.
4-171
PETTIJOHN, F.J., Sedimentary Rooks, Harper and Brothers, New York, 1957,
p. 117.
PIERCE, J.W., "Sediment Budget Along a Barrier Island Chain," Sedimentary

Geology, Vol. 3, No. 1, 1969, pp. 5-16.
PIERCE, J.W., "Holocene Evolution of Portions o£ the Carolina Coast,"

Bulletin of the Geologic Society of America, Vol. 81, Dec. 1970.
PIERSON, WILLARD J,, JR., and NEUMANN, GERHARD, Princiiples of Physical

Oceanography, Prentice-Hall, Englewood Cliffs, N.J., 1966.
PILKEY, O.H., and FIELD, M.E., "Onshore Transportation of Continental

Shelf Sediment: Atlantic Southeastern United States," Shelf Sedi-
ment Transport, edited by Swift, Duane, and Pilkey; Dowden, Hutchin-
son, and Ross, Inc., Stroudsburg, Pa., 1972.
PILKEY, O.H., and FRANKENBERG, D., "The Relict-Recent Sediment Boundary

on the Georgia Continental Shelf," The Bulletin of the Georgia
Academy of Science, Vol. XXII, No. 1, Jan. 1964.
PRICE, W.A., "Reduction of Maintenance by Proper Orientation of Ship

Channels Through Tidal Inlets," Proceedings of the Second Conference
on Coastal Engineering, Council on Wave Research, University of
California, Berkeley, Calif., 1952, pp. 243-255.
PUTNAM, J. A., MUNK W.H. , and TRAYLOR, M.A. , "The Prediction of Longshore
Currents," Transactions of the American Geophysical Union, Vol. 30,
1949, pp. 337-345.
RAMSEY, M.D., "The Relationship Between Mean Sand Size vs Local Foreshore
Slope for 166 New Jersey Sand Samples," Unpublished MFR, U.S. Army,
D.C., Sept. 1971.
RAMSEY, M.D., and CALVIN, C.J., "Size Analysis of Sand Samples from
Three S. New Jersey Beaches," Unpublished Paper, U.S. Army, Corps
of Engineers, Coastal Engineering Research Center, Washington, D.C.,
Sept. 1971.
RAUDKIVI, A.J., Loose Boundary Hydraulics, Pergamon Press, New York,

Jan. 1965.
RECTOR, R.L., "Laboratory Study of Equilibrium Profiles of Beaches,"

TM-41, UoSo Army, Corps of Engineers, Beach Erosion Board, Washington,
D.C., Aug. 1954.
REIMNITZ, E., and ROSS, D.A., "The Sea Sled - A Device for Measuring
Bottom Profiles in the Surf Zone," Marine Geology, Vol. 11, 1971.
ROSS, D.A., "Atlantic Continental Shelf and Slope of the United States -
Heavy Minerals of the Continental Margin from Southern Nova Scotia to
Northern New Jersey," Professional Paper 529-G, U.S. Geological Survey,
U.S. Government Printing Office, Washington, D.C., 1970.
4-172
RUSNAK, G.A., STOCKMAN, R.W. , and HOFMANN, H.A., "The Role of Shell
Material in the Natural Sand Replenishment Cycle of the Beach and
Nearshore Area Between Lake Worth Inlet and the Miami Ship Channel,"
CERC Contract Report (DA-49-055-CIV-ENG-63-12) , Institute of Marine
Sciences, University of Miami, Coral Gables, Fla. , 1966.
RUSSELL, "R.J. , "Florida Beaches and Cemented Water-Table Rocks," Technical

Report No. 88, Coastal Studies Institute, Louisiana State University,
Baton Rouge, La., Oct. 1970.
SATO, S., IJIMA, T. , and TANAKA, N. , "A Study of Critical Depth and Mode
of Sand Movement Using Radioactive Glass Sand," Proceedings of the
Eighth Conferenoe on Engineeving , Mexico City, 1962, pp. 304-323.
SAUVAGE, M.G., and VINCENT, M.G., "Transport and Littoral Formation de

Flecheset de Tombolos," Proceedings of the Fifth Conference on Coastal
Engineering J 1955, pp. 296-328.
SAVAGE, R.P., "Wave Run-up on Roughened and Permeable Slopes," Journal of

the ASCE^ Vol. 84, No. WW3, May 1958.
SAVAGE, R.P., "Notes on the Formation of Beach Ridges," Bulletin of the

Beach Erosion Boards Vol. 13, July 1959, pp. 31-35.
SAVAGE, R.P., "Laboratory Determination of Littoral Transport Rates,"

Journal of the Waterways and Harbors Division^ ASCE, No. WW2, May 1962,
pp. 69-92.
SAVAGE, R.P., and WOODHOUSE, W.W., "Creation and Stabilization of Coastal

Barrier Dunes," Proceedings of the 11th Coastal Engineering Conference
London, Sept. 1968.
SAVILLE, T., Jr., "Model Study of Sand Transport Along an Infinitely Long
Straight Beach," Transactions of the American Geophysical Union, Vol.
31, 1950.
SAVILLE, T., Jr., "Scale Effects in Two Dimensional Beach Studies,"

Proceedings of the Seventh General Meeting, International Association
for Hydraulic Research, 1957.
SAVILLE, T., Jr., and CALDWELL, J.M., "Accuracy of Hydro graphic Surveying
In and Near the Surf Zone," TM-32, U.S. Army, Corps of Engineers, Beach
Erosion Board, Washington, D.C., Mar. 1953.
SAVILLE, T., JR., and SAVAGE, R.P., "Laboratory Determination of Littoral

Transport Rates," Journal of the Waterways and Harbors Division, ASCE,
V. 88, No. WW2, May 1962, pp. 69-92. Discussion: Nov. 1962, Feb. 1963,
Discussion by Thomdike Saville, Jr., Nov. 1962.
SAYLOR, J.H., and HANDS, E.B., "Properties of Longshore Bars in the Great
Lakes," Proceedings of the 12th Conference on Coastal Engineering, Vol.
2, 1970, pp. 839-853.
4-173
SAYLOR, J.H., and UPCHURCH, S. , "Bottom Stability and Sedimentary Processes
at Little Lake Harbors, Lake Superior," U.S. Army, Corps of Engineers,
Lake Survey District Detroit, Mich., 1970.
SCHULZ, E.F., WILDE, R.H., and ALBERTSON, M.L., "Influence of Shape on

the Fall Velocity of Sedimentary Particles," U.S. Army, Corps of
Engineers, Missouri River Division, 1954.
SCRUTON, P.C., "Delta Building and the Deltaic Sequence," Recent Sedi-
mentSy Northwest Gulf of Mexico^ American Association of Petroleum
Geologists, 1960, pp. 82-102.
SEELIG, W.N., "Beach Evaluation Program, Visual Wave Observation Program,"

Unpublished Paper, U.S. Array, Corps of Engineers, Coastal Engineering
Research Center, Washington, D.C., Mar. 1972.
SELLMAN, PAUL V., et al., "Investigations of Terrain and Coastal Conditions

on the Arctic Alaskan Coastal Plain," Draft of Special Report, U.S.
Army Cold Regions Research and Engineering Laboratories, Aug. 1971.
SHAY, E.A., and JOHNSON, J.W., "Model Studies on the Movement on Sand
Transported by Wave Action Along a Straight Beach," Issue 7, Ser. 14,
Institute of Engineering Research, University of California, Berkeley,
Calif., 1951.
SHEPARD, P.P., "Beach Cycles in Southern California," TM-20, U.S. Army,

Corps of Engineers, Beach Erosion Board, Washington, D.C., July 1950.
SHEPARD, P.P., "Gulf Coast Barriers," Recent Sediments, Northwest Gulf

of Mexico, American Association of Petroleum Geologists, 1960a,
pp. 197-220.
SHEPARD, P.P., "Rise of Sea Level Along Northwest Gulf of Mexico," Recent
Sediments , Northwest Gulf of Mexico, American Association of Petroleum
Geologists, 1960b, pp. 338-344.
SHEPARD, P.P., "Submarine Geology," Physical Properties of Sediments,

Ch. V, and Beaches and Related Shore Processes, Ch VII, Harper and
Row, New York, 1963, pp. 167-205.
SHEPARD, P.P., and BUFFINGTON, E.G., "La Jolla Submarine Fan Valley,"
Marine Geology, Vol. 6, 1968, pp. 107-143.
SHEPARD, P.P., and DILL, R.F., ^'Submarine Canyons and Other Sea Valleys,"
Rand McNally, Chicago, 1966.
SHEPARD, P.P., and GRANT, U.S., IV, "Wave Erosion Along the Southern
California Coast," Bulletin of the Geologic Society of America, Vol.
58, 1947, pp. 919-926.
SHEPARD, F.P., and INMAN, D.L., "Nearshore Water Circulation Related to

Bottom Topography and Wave Refraction," Transactions of the American
Geophysical Union, Vol. 31, No. 2, 1950, pp. 196-212.
4-174
SHEPARD, F,P., MacDONALD, G.A., and COX, D.C., "The Tsunami of April 1,
1946," Bulletin of the Soripps Institute of Oceanography, Vol. 5,
No. 6, 1950, pp. 391-528.
SHEPARD, P.P., and WANLESS, H.R. , Our Changing Coastlines, McGraw-Hill,

New York, 1971.
SHUYSKIY, Y.D., "The Effect of Strong Storms on Sand Beaches of the Baltic
Eastern Shore," Ooeanology, Vol. 9, No. 3, 1970, p. 388.
SONU, C.J., "Field Observation of Nearshore Circulation and Meandering

Currents," Journal of Geophysical Eesearahj Oceans and Atmospheres,
Vol. 77, No. 18, 1972.
SONU, C.J., and VAN BEEK, J.L., "Systemic Beach Changes on the Outer
Banks, North Carolina," Journal of Geology, Vol. 79, 1971, pp. 416-425.
SOUTHARD, J.B., "Representation of Bed Configurations in Depth-Velocity-

Size Diagrams," Journal of Sedimentary Petrology, Vol. 41, No. 4,
Dec. 1972.
STAFFORD, D.B., "An Aerial Photographic Technique for Beach Erosion Sur-
veys in North Carolina," TM-36, U.S. Army, Corps of Engineers, Coastal
Engineering Research Center, Washington, D.C., Oct. 1971.
STEPHENS, N.H., et al., "Evaluation of Four Methods Using Gold-198 for

Surface Labeling Sand and a New Technique for Simuluating Mass Label-
ing," ORNL-4338, Oak Ridge National Laboratory, 1968.
STODDARD, D.R„, "World Erosion and Sedimentation," Water, Earth, and Man,
Barnes and Noble, 1969.
STONE, R.O., and SUMMERS, H.J., "Final Report, Study of Subaqueous and
Subaerial Sand Ripples," ONR Project No. N00014-67-A-0269-0002, Task
No. NR- 388-085, Report No. USG Geology 72-1, 1972.
STRAKHOV, N.M., "Principles of Lithogene sis," Mol. 1, (Translation from

1962 Russian Edition by J, P. Fitzsimmons, Oliver, and Boyd), Edinburgh
and London, 1967, p. 245.
STUIVER, M., and PURPURA, J. A., "Application of Fluorescent Coastal Sand

in Littoral Drift and Inlet Studies," Proceedings of the 11th Conference
on Coastal Engineering, ASCE, 1968, pp. 307-321.
SVENDSEN, S., "Munch-Petersen* s Littoral Drift Formula," Bulletin of the

Beach Erosion Board, Vol. 4, No. 4, Oct. 1950.
SWIFT, DoJ.P., "Quartemary Shelves and the Return to Grade," Marine

Geology, Elsevier Publishing, Amsterdam, 1970.
SZUWALSKI, A., "Littoral Environment Observation Program in California

Preliminary Report," MP 2-70, U.S. Army, Corps of Engineers, Coastal
Engineering Research Center, Washington, D.C., Feb. 1970.
4-175
SZULWALSKI, A., "Coastal Imagery Data Bank: Interim Report," MP-3- 72,
Washington, D.C., Nov. 1972.
TANEY, NoE., "Geomorphology of the South Shore of Long Island, New York,"
D.C., Sept. 1961a.
TANEY, N.E., "Littoral Materials of the South Shore of Long Island, New
York," TM-129, U.S. Army, Corps of Engineers Beach Erosion Board,
Washington, D.C., Nov. 1961b.
TANEY, N.E., "Laboratory Applications of Radioisotopic Tracers to Follow

Beach Sediments," Proceedings of the Eighth Confevenoe on Coastal
Engineering^ Council on Wave Research, ASCE, University of California,
Berkeley, Calif., 1963, pp. 279-303.
TANNER, W.F., "Florida Coastal Classification," Transactions of the Gulf

Coast Association of Geological Societies ^ Vol. X, 1960, pp. 259-266.
TANNER, W.F., "Mainland Beach Changes Due to Hurricane Donna," Journal of

Geophysical Research ^ Vol. 66, No. 7, July 1961.
TANNER, W.F., "Significance of Camille," Southeastern Geology ^ Vol. 12,

No„ 2, 1970, pp. 95-104.
TELEKI, P., "Fluorescent Sand Tracers," Journal of Sedimentary Petrology

Volo 36, Jime 1966.
TELEKI, P., "Automatic Analysis of Tracer Sand," Journal of Sedimentary

Petrology J Vol. 37, Sept. 1967, pp. 749-759.
TERZAGHI, K., and PECK, R.B., Soil Mechanics in Engineering Practice^

Wiley, New York, 1967, p. 28.
THIEL, G.A., "The Relative Resistance to Abrasion of Mineral Grains of Sand

Size," Journal of Sedimentary Petrology^ Vol. 10, 1940, pp. 102-124.
THOMPSON, E.F., and HARRIS, D.L., "A Wave Climatology for U.S. Coastal
Waters," Proceedings of Offshore Technology Conference, Dallas, Texas,
May, 1972.
THOREAU, H.Do, Cape Cod, Thicknor and Fields, Boston, Mass., 1865.
THORNTON, E.B., "Longshore Current and Sediment Transport," TR-5, Depart-

ment of Coastal and Oceanographic Engineering, University of Florida,
Gainesville, Fla. , Dec. 1969.
TRASK, PARKER D„, "Movement of Sand around Southern California Promon-

tories," TM-76, U.S. Army, Corps of Engineers, Beach Erosion Board,
Washington, D.C., June 1955.
4-176
TUCKER, M.J., "Recent Measurement and Analysis Techniques Developed at
The National Institute of Oceanography," National Institute of
Oceanography 3 Collected Reprints, V. 11, No. 465, 1963. Reprinted
from Ocean Wave Spectra; Proceedings of Conference on Ocean Wave
Spectra, 1961, pp. 219-226.
URBAN, H.D., and CALVIN, C.J., "Pipe Profile Data and Wave Observations
from the CERC Beach Evaluation Program, January - March, 1968,"
Center, Washington, D.C., Sept. 1969.
U.S. ARMY, CORPS OF ENGINEERS, "Relation Between Sand Size and Slope of
the Foreshore," Interim Report, Beach Erosion Board, Washington, D.C.
1933.
U.S. ARMY, CORPS OF ENGINEERS, "Beach Erosion Report on Cooperative Study

at Palm Beach, Florida," Beach Erosion Board, Washington, D.C, 1947,
(Unpublished)
U.S. ARMY, CORPS OF ENGINEERS, "Shoreline Effects, Harbor at Playa Del

Rey, Calif.," Enclosure 20, Engineer District, Los Angeles, Calif.,
1948a, (Unpublished).
U.So ARMY, CORPS OF ENGINEERS, "Survey of Ocean City Harbor and Inlet and
Sinepuxent Bay, Maryland," Engineer District, Baltimore, Md. , 1948b,
(Unpublished)
U.S. ARMY, CORPS OF ENGINEERS, "Illinois Shore of Lake Michigan - Beach

Erosion Control Study," House Document 28, 83rd Congress, 1953a.
U.S. ARMY, CORPS OF ENGINEERS, "Ohio Shoreline of Lake Erie, (Sandusky

to Vermillieni) Beach Erosion Control Study," House Document 32, 83rd
Congress, 1953b.
U.S. ARMY, CORPS OF ENGINEERS, "Interim Report on Harbor Entrance Improve-

Camp Pendleton, California," Engineer District, Los Angeles, Calif.,
1953c, (Unpublished).
U.S. ARMY, CORPS OF ENGINEERS, "Preliminary Analysis of Cooperative Beach

Erosion Study, City of Kenosha, Wisconsin, Engineer District, Milwaukee,
Wis., 1953d, (Unpublished).
U.S. ARMY, CORPS OF ENGINEERS, "Beach Erosion Control Report, Cooperative

Study of Perdido Pass, Alabama," Engineer District, Mobile, Ala., 1954a,
(Unpublished)
U.S. ARMY, CORPS OF ENGINEERS, "Atlantic Coast of New Jersey, Sandy Hook
to Bamegat Inlet, Beach Erosion Control Report on Cooperative Study,"
Engineer District, New York, N.Y., 1954b, (Unpublished).
U.S. ARMY, CORPS OF ENGINEERS, "Study of Monrovia Harbor, Liberia, and

Adjoining Shore Line," Beach Erosion Board, Washington, D.C, 1955,
(Unpublished)
4-177
U.S. ARMY, CORPS OF ENGINEERS, "Cliff Walk, Newport, Rhode Island, Beach
Erosion Control Study," Letter from Secretary of the Army, U.S. Govern-
ment Printing Office, Washington, 1965.
U.S. ARMY, CORPS OF ENGINEERS, "Shore Protection, Planning, and Design,"

TR No. 4, Coastal Engineering Research Center, Washington, D.C., 1966.
U.S. ARMY, CORPS OF ENGINEERS, "Littoral Environment Observations, Florida

Panhandle, 1969-1970," Coastal Engineering Research Center, Washington,
D.Co
U.S. ARMY, CORPS OF ENGINEERS, "A Study of Methods to Preserve Gay Head
Cliffs, Martha's Vineyard, Massachusetts," Report prepared by Woodard-
Moorhouse and Associates, Inc., for the Engineer, New England Division,
Oct. 1970.
U.S. ARMY, CORPS OF ENGINEERS, "Investigation of Erosion, Carolina Beach,

N.C.," Rpt. 1-69, Engineer District, Wilmington, N.C, 1970.
U.S. ARMY, CORPS OF ENGINEERS, "National Shoreline Study," Great Lakes

Region- Inventory Report, Aug. 1971.
U.S. CONGRESS, "Beach Erosion Study, Lake Michigan Shore Line of Milwaukee
County, Wisconsin," House Document 526, 79th Congress, 2nd Session,
p. 16, 1946.
U.S. CONGRESS, "North Carolina Shore Line, Beach Erosion Study," House
Document 763, 80th Congress, 2nd Session, 1948.
U.S. CONGRESS, "Ocean City, New Jersey, Beach Erosion Control Study,"
House Document 184, 83d Congress, 1st Session, 1953a.
U.S. CONGRESS, "Cold Spring Inlet (Cape May Harbor), New Jersey," House
Document 206, 83d Congress, 1st Session, 1953b.
UoS. CONGRESS, "Gulf Shore of Galveston Island, Texas, Beach Erosion

Control Study," House Document 218, 8d Congress, 1st Session, 1953c.
U.S. CONGRESS, "Coast of California, Carpinteria to Point Mugu," House

Document 29, 83d Congress, 1st Session, 1953d.
U.S. CONGRESS, "Racine County, Wisconsin," House Document 88, 83d Congress,
1st Session, 1953e.
U.S. CONGRESS, "Illinois Shore of Lake Michigan," House Document 28, 83d
Congress, 1st Session 1953f.
U.S. CONGRESS, "Waikiki Beach, Island of Oahu, T.H. , Beach Erosion Study,"
House Document No. 227, 83d Congress, 1st Session, 1953g.
U.S. CONGRESS, "Pinellas County, Florida," House Document 380, 83d, Congress,
2d Session, 1954a.
4-178
U.S. CONGRESS, "Port Hueneme, California," House Docioment 362, 83d,
Congress, 2d Session, 1954b.
U.S. CONGRESS, "Anaheim Bay Harbor, California," House Document 349,

83d Congress, 2d Session, 1954c.
VALLIANOS, F., "Recent History of Erosion at Carolina Beach, N.C.,"

Pvooeedings of the 12th Conference on Coastal Eng-ineeving ^ ASCE,
Vol. 2, 1970.
VAN DORN, W.Go, "Wind Stress on an Artificial Pond," Journal of Marine

Research:, Vol. 12, 1953, pp. 249-276.
VOLLBRECHT, K., "The Relationship Between Wind Records, Energy of Long-

shore Drift, and the Energy Balance off the Coast of a Restricted
Water Body, as Applied to the Baltic," Marine Geology^ Vol. 4, No. 2,
Apr. 1966, pp. 119-148.
WALTON, T.L., Jr., "Littoral Drift Computations Along the Coast of

Florida by Use of Ship Wave Observations," Unpublished Thesis,
University of Florida, Gainesville, Fla. , 1972.
WALTON, T.L. and DEAN, R.G., "Application of Littoral Drift Roses to

Coastal Engineering Problems," Proceedings of the Conference on
Engineering Dynamics in the Coastal Zone, 1973.
WATSON, R.L., "Origin of Shell Beaches, Padre Island, Texas," Journal

of Sedimentary Petrology^ Vol. 41, No. 4, Dec. 1971.
WATSON, R.L., "Influence of Mean Grain Diameter on the Relationship

Between Littoral Drift Rate and the Alongshore Component of Wave
Energy Flux," EOS 53, No. 11, p. 1028, Nov. 1972.
WATTS, G.M., "Development and Field Tests of a Sampler for Suspended

Sediment in Wave Action," TM-34, U.S. Army, Corps of Engineers,
Beach Erosion Board, Washington, D.C., Mar. 1953a.
WATTS, G.M., "A Study of Sand Movement at South Lake Worth Inlet, Florida,"
D.C., Oct. 1953b.
WATTS, G.M., "Laboratory Study of the Effect of Varying Wave Periods on

Beach Profiles," TM-53, U.S. Army, Corps of Engineers, Beach Erosion
Board, Washington, D.C., Sept. 1954.
WATTS, G.M., "Behavior of Beach Fill at Ocean City, New Jersey," TM-77
Feb. 1956.
WATTS, G.M., and DEARDUFF, R.F., "Laboratory Study of Effect of Tidal

Action on Wave-Formed Beach Profiles," TM-52, U.S. Army, Corps of
Engineers, Beach Erosion Board, Washington, D.C. , Dec. 1954.
4-179
WIEGEL, R,L., Oaeanogrccphio Engineering^ Prentice-Hall, Englewood Cliffs,
N.J., 1964.
WOLMAN, G»M., and MILLER, J. P., "Magnitude and Frequency of Forces in

Geomorphic Processes," The Jowcnal of Geology y Vol. 68, No. 1, 1960,
pp. 54-74.
WOODARD, D.W., et al., "The Use of Grasses for Dune Stabilization Along
the Gulf Coast with Initial Emphasis on the Texas Coast," Report No.
114, Gulf Universities Research Corporation, Galveston, Tex., 1971.
WRIGHT, F., and COLEMAN, Jr., "River Delta Morphology: Wave Climate and
the Role of the Subaqueous Profile," Scienae, Vol. 176, 1972, pp. 282-
284.
YASSO, W.E., "Fluorescent Coatings on Coast Sediment: An Integrated

System," TR-16, Columbia University, Department of Geology, New York,
N.Y., 1962.
ZEIGLER, J.M., and GILL, B., "Tables and Graphs for the Settling Velocity
of Quartz in Water Above the Range of Stokes Law," Ref. No. 59-36,
Woods Hole Oceanographic Institution, Woods Hole, Mass., July 1959,
ZEIGLER, J. Mo, and TUTTLE, S.D., "Beach Changes Based on Daily Measure-
ments of Four Cape Cod Beaches," Journal of Geology, Vol. 69, No. 5,
1961, pp. 583-599.
ZEIGLER, J.Mo, et al., "Residence Time of Sand Composing the Beaches and
Bars of Outer Cape Cod," Proceedings of the Ninth Conference on Coastal
Engineering J ASCE, Vol. 26, 1964, pp. 403-416.
ZENKOVICH, V.P., "Applications of Luminescent Substances for Sand Drift

Investigation in the Nearshore Zones of the Sea," Die Ingenieur,
Vol. 13, 1967a, pp, 81-89,
ZENKOVICH, V.P., Processes of Coastal Develoipmenty Interscience Publishers,

New York, 1967b.
ZWAMBORN, J. A., FROMME, G.A.W., and FITZPATRICK, J.B., "Underwater Mound

for the Protection of Durban's Beaches," Proceedings of the 12th
Conference on Coastal Engineering, ASCE, 1970.
ZWAMBORN, J. A., et al., "Coastal Engineering Measurements," Proceedings of

the 13th International Conference on Coastal Engineering, Canada, 1972.
4-180
O U. S. GOVERNMENT PRINTING OFFICE : 1976 O - 508-951 (Vol. I)

SPM 1 - Coastal BW

Uploaded by

Copyright:

Available Formats

SPM 1 - Coastal BW

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SPM 1 - Coastal BW

Uploaded by

Copyright:

Available Formats

c?

DEPARTMENT OF THE ARMY

U.S. Army Coastal Engineering Research Center

SONOMA COAST, CALIFORNIA (GOAT ROCK) - 10 December 1958

U.S. ARMY COASTAL ENGINEERING RESEARCH CENTER

For sale by the Superintendent of Documents, U.S. Government Printing Office

editions thereof appeared in 1957, 1961, and 1966.

technical contributions to this manual are: R. H. Allen, B. R. Bodine, M. T. Czemiak, A. E. DeWall, D. B.

Fusch was responsible for the completion of the final manuscript.

7 STRUCTURAL DESIGN-PHYSICAL FACTORS

8 ENGINEERING ANALYSIS-CASE STUDY

CHAPTER 1 - INTRODUCTION TO COASTAL ENGINEERING

1.1 INTRODUCTION TO THE SHORE PROTECTION MANUAL 1-1

1.2 THE SHORE ZONE 1-2

1.3 THE SEA IN MOTION 1-5

1.4 THE BEHAVIOR OF BEACHES 1-9

1.5 EFFECTS OF MAN ON THE SHORE 1-13

1.6 CONSERVATION OF SAND 1-21

CHAPTER 2 - MECHANICS OF WAVE MOTION

2.1 INTRODUCTION 2-1

2.2 WAVE MECHANICS 2-1

2.22 WAVE FUNDAMENTALS AND CLASSIFICATION OF WAVES 2-3

2.3 WAVE REFRACTION 2-62

2.4 WAVE DIFFRACTION 2-79

2.5 WAVE REFLECTION 2-110

2.6 BREAKING WAVES 2-120

REFERENCES AND SELECTED BIBLIOGRAPHY 2-129

CHAPTER 3 - WAVE AND WATER LEVEL PREDICTIONS

3.1 INTRODUCTION 3-1

3.2 CHARACTERISTICS OF OCEAN WAVES 3-2

3.3 WAVE FIELD 3-15

3.4 WIND INFORMATION NEEDED FOR WAVE PREDICTION 3-20

3.5 SIMPLIFIED WAVE-PREDICTION MODELS 3-33

3.6 WAVE FORECASTING FOR SHALLOW WATER 3-42

3.7 HURRICANE WAVES 3-52

3.8 WATER LEVEL FLUCTUATIONS 3-69

REFERENCES AND SELECTED BIBLIOGRAPHY 3-145

CHAPTER 4 - LITTORAL PROCESSES

4.1 INTRODUCTION 4-1

4.2 LITTORAL MATERIALS 4-11

SIZE ANALYSES 4-26

LITTORAL WAVE CONDITIONS 4-27

NEARSHORE CURRENTS 4-39

LITTORAL TRANSPORT 4-50

4.6 ROLE OF FOREDUNES IN SHORE PROCESSES 4-111

4.7 SEDIMENT BUDGET 4-116

4.8 ENGINEERING STUDY OF LITTORAL PROCESSES 4-139

REFERENCES AND SELECTED BIBLIOGRAPHY 4-155

1-1 Beach Profile - Related Terms 1-3

2-1 Approximate Distribution of Ocean Surface Wave Energy

2-24 Refraction by a Submarine Ridge and Submarine Canyon . . . 2-77

2-58 Diffraction Diagram for a Gap of Two Wave Lengths and a

3-1 Sample Wave Records 3-3

3-24 Forecasting Curves for Shallow- Water Waves; Constant

3-57 Lake Surface Contours on Lake Okeechobee, Florida

4-29 Berra -Bar Criterion Based on Dimensionless Fall Time and

1-1 Shoreline Characteristics 1-3

2-1 Distribution of Wave Heights in a Short Train of Waves. . . 2-32

3-1 Correction for Sea-Air Temperature 3-27

4-1 Seasonal Profile Changes on Southern California Beaches . . 4-10