Citation
The Development of a Scrap Tire Barrier as a Coastal Structure for Wave Damping Applications

Material Information

Title:
The Development of a Scrap Tire Barrier as a Coastal Structure for Wave Damping Applications
Creator:
Yousif, Ahmad
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (178 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Coastal and Oceanographic Engineering
Civil and Coastal Engineering
Committee Chair:
VALLE-LEVINSON,ARNOLDO
Committee Co-Chair:
SLINN,DONALD NICHOLAS
Committee Members:
OLABARRIETA LIZASO,MAITANE
ADAMS,PETER N
NEELAMANI,SUBRAMANIAN

Subjects

Subjects / Keywords:
coastal-structure -- floating-breakwater -- hydrodynamics -- inner-shelf -- scrap-tire -- wave-transmission
Civil and Coastal Engineering -- Dissertations, Academic -- UF
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Coastal and Oceanographic Engineering thesis, Ph.D.

Notes

Abstract:
The wave transmission, reflection, and dissipation of nine new configurations of scrap tire wave barriers have been investigated using physical models with regular and random waves. The study aim was to achieve the least wave transmission with a minimum number of scrap tires. Predictive relations which depend on relative water depth, d/L_p, and relative wave height, H_i/d, have been obtained to predict coefficient of transmission for nine newly proposed scrap tire configurations using random waves. After performing an oceanographic investigation, the optimal configurations of this study were recommended to be used for field application to protect the beaches of Kennedy Space Center (KSC), USA and Qaru Island, Kuwait from erosion. The tidal and subtidal hydrodynamics were analyzed over ridge-swale bathymetry in the inner shelf adjacent to Cape Canaveral, Florida using vessel-based data. Observations were compared to two analytical models that yield tidal and subtidal solutions. The north transect had relatively smoother bathymetry. Therefore, tidal and subtidal hydrodynamics were consistent with previous studies on flows passing over different bathymetries with less frictional influence. The south transect had relatively more complex bathymetry that defined a channel. Hence, tidal and subtidal hydrodynamics were consistent with previous observations on frictionally dominated flows. Results obtained with both tidal and subtidal analytical model solutions highlight the influence of ridge-swale bathymetry in inner shelves at those temporal scales and the applicability of open channel concepts to inner shelf dynamics. The subinertial hydrodynamics were also investigated using moored data. Wavelet coherence analysis techniques were used between subinertial flow and different forcings. Northward winds coincide with enhancement of the Florida current. As a result, positive across-shelf sea surface slope develops following geostrophy and the subinertial flow moves in the same direction of the western boundary current. On the other hand, southward winds coincide with weakening of the Florida current. Thus, a negative nearshore sea surface slope develops to maintain geostrophic balance and the nearshore flow moves southward. The across-shelf momentum balance showed that the subinertial flow was mainly geostrophic throughout the deployment, while the along-shelf momentum was mainly frictional. Therefore, the dynamics in the region are semi-geostrophic. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2017.
Local:
Adviser: VALLE-LEVINSON,ARNOLDO.
Local:
Co-adviser: SLINN,DONALD NICHOLAS.
Statement of Responsibility:
by Ahmad Yousif.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
LD1780 2017 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

THE DEVELOPMENT OF A SCRAP TIRE BARRIER AS A COASTAL STRUCTURE FOR WAVE DAMPING APPLICATIONS By AHMAD YOUSIF A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2017

PAGE 2

2017 Ahmad Yousif

PAGE 3

To my Parents

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my advisor, Arnoldo Valle Levinson, for all his guidance, support, and patience during my doctoral studies. I am very grateful to be part of his team. He has always been in my side whenever I needed him. I would also like to thank my committee members : Dr. Peter Adams, Dr. S. Neelamani Dr. Maitane Ol abarrieta, and Dr. Donald Slinn for all the knowledge provided and their feedback during my PhD studies. I gratefully acknowledge t he financial support of Kuwait Univer sity; the Bureau of Ocean Energy Management (B OEM) through the grant M13AS00010 ; and Kuwait Institute for Scientific Research (KISR) for the infrastructure facilities to carry out the physical modeling experiments which made this Ph.D. wo rk possible I thank my parents and family for their endless love and encouragement. Thanks go to my friends at University of Florida: Arma ndo Laurel Castillo, Mohammad Alk haldi, Gisselle Guerra, Dorukhan Ardag, Sangdon So, Juan Felipe Paniagua Arroyave, Zhendong Cao, Alessandro Aguiar, Christian Rojas, Braulio Juarez Huidi Liang and Fangjing Deng for the good memories, their help, and encouragement. Special thanks to my friends in KISR: Mr. J.M.S. Ashok, Mr. Khaled Elsyed Attaalla, Mr. G. Renganathan, E ng. Josko Ljubic, and Mr. George Joseph for their support while performing the physical modeling experiments.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 15 ABSTRACT ................................ ................................ ................................ ................... 16 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 18 This Dissertation ................................ ................................ ................................ ..... 19 Study Area ................................ ................................ ................................ .............. 19 Observations ................................ ................................ ................................ ........... 21 Vessel based Data ................................ ................................ ........................... 21 Mooring Da ta ................................ ................................ ................................ .... 21 Hydrographic Data ................................ ................................ ................................ .. 23 2 TIDAL AND SUBTIDAL HYDRODYNAMICS OVER RIDGE SWALE BATHYMETRY ................................ ................................ ................................ ....... 28 Background ................................ ................................ ................................ ............. 28 Data Analysis ................................ ................................ ................................ .......... 29 Matrix Arrangement and Compass Calibration ................................ ................. 29 Least Squares Fit to Observations ................................ ................................ ... 30 Tidal Analytical Model Solutions ................................ ................................ ....... 31 Subtidal Analytical Model Solutions ................................ ................................ .. 33 Results ................................ ................................ ................................ .................... 35 Tidal Model vs Observations ................................ ................................ ............ 35 Subtidal Model vs Observations ................................ ................................ ....... 37 Discussion ................................ ................................ ................................ .............. 38 Tidal Model ................................ ................................ ................................ ....... 38 Subtidal Model ................................ ................................ ................................ .. 40 Summary ................................ ................................ ................................ ................ 42 3 SUBINERTIAL HYDRODYNAMICS OVER RIDGE SWALE BATHYMETRY AROUND A CAPE ................................ ................................ ................................ .. 50 Background ................................ ................................ ................................ ............. 50 Data Analysis ................................ ................................ ................................ .......... 51 Filtering ................................ ................................ ................................ ............. 51

PAGE 6

6 Concatenated Hilbert Empirical Orthogonal Function (CHEOF) ....................... 51 Wavelet Analysis ................................ ................................ .............................. 51 Depth averaged Along Shelf Momentum Balance ................................ ........... 53 Depth averaged Across Shelf Momentum Balance ................................ .......... 55 Geostrophic Balance ................................ ................................ ............................... 56 Results ................................ ................................ ................................ .................... 57 Subinertial Parameters ................................ ................................ ..................... 57 CHEOF and Wavelet Coherence ................................ ................................ ..... 58 Along Shelf Momentum Balance ................................ ................................ ...... 59 Across Shelf Momentum Balance ................................ ................................ .... 60 Standard Deviation of the Momentum Terms ................................ ................... 61 Discussion ................................ ................................ ................................ .............. 62 Summary ................................ ................................ ................................ ................ 64 4 THE DEVELOPMENT OF A SCRAP TIRE BARRIER AS A COASTAL STRUCTURE FOR WAVE DAMPING APPLICATIONS ................................ ......... 73 Background ................................ ................................ ................................ ............. 73 Methodology ................................ ................................ ................................ ........... 76 Results and Discussion ................................ ................................ ........................... 79 Wave Transmission, Reflection, and Energy Dissipation Using Regular Waves ................................ ................................ ................................ ........... 79 Wave Transmission, Reflection, and Energy Dissipation Using Random Waves ................................ ................................ ................................ ........... 81 Comparison of Wave Transmission Characteristics for Different Scrap Tire Configurations Due to Random Waves ................................ ......................... 83 Multiple Regression Equations for the Prediction of the Coefficient of Transmission for Different Scrap Tire Configurations ................................ .... 85 Field Application ................................ ................................ ................................ ..... 86 Design Aspects of the Scrap Tire Configuration for Kennedy Space Center (KSC) at Cape Canaveral, Florida ................................ ................................ 86 Case 1: The Most Dominant Wave Condition ................................ ................... 87 Case 2: Longest and Most Frequent Wave Period ................................ ........... 89 Case 3: Largest Wave Height Condition ................................ ........................... 89 Time Series of Wave Heights and Periods with Design Recommendations ..... 90 Design Aspects of the Scrap Tire Configuration for Qaru Island, Kuwait ......... 91 Case 1: The Most Dominant Wave Condition ................................ ................... 92 Case 2: Longest and Most Frequent Wave Period ................................ ........... 93 Case 3: Largest Wave Height Condition ................................ ........................... 94 Design Recommendations for Qaru Island ................................ ....................... 94 Summary ................................ ................................ ................................ ................ 95 Future Work ................................ ................................ ................................ ............ 97 5 CONCLUSIONS ................................ ................................ ................................ ... 168 LIST OF REFREN CES ................................ ................................ ................................ 171

PAGE 7

7 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 178

PAGE 8

8 LIST OF TABLES Table page 1 1 Details of the ADCPs at the four mooring positions during spring season. ......... 27 4 1 Regular Wave Parameters ................................ ................................ ............... 163 4 2 Random Wave Parameters ................................ ................................ .............. 164 4 3 Details of the Various Scrap Tire Configurations ................................ .............. 164 4 4 Tire Measurements ................................ ................................ ........................... 165 4 5 Materials Used for Fabrication of Models ................................ ......................... 166 4 6 ................................ 166 4 7 Multiple Regression Equations and for the Prediction of for Different Scrap Tire Configurations ................................ ................................ ................. 167

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Bathymetric map of Cape Canaveral with the four locations of the ADCP moorings. ................................ ................................ ................................ ............ 25 1 2 Hydrographic data for the north and south transects during all tows performed between fall 2013 and summer 2016. ................................ ............... 26 2 1 Schematic representation of the model reference frame for the south transect bathymetry. The same reference frame is used for the north transect. .............. 43 2 2 A comparison between the tidal model and observation amplitudes for the north transect during fall 2013. ................................ ................................ ........... 44 2 3 A comparison between the tidal model and observation amplitudes for the south transect during spring 2015. ................................ ................................ ..... 44 2 4 A comparison between the tidal model and observation amplitudes for the south transect during summer 2015. ................................ ................................ .. 45 2 5 A comparison between the tidal model and observation amplitudes for the south transect during spring 2016. ................................ ................................ ..... 45 2 6 A comparison between the subtidal model and observation velocities for the north transect during fall 2013 ................................ ................................ ........... 46 2 7 A comparison between the subtidal model and observation velocities for the north transect during summer 2014. ................................ ................................ ... 46 2 8 A comparison between the subtidal model and observation velocities for the north transect during spring 2015. ................................ ................................ ...... 47 2 9 A comparison between the subtidal model and observation velocities for the south transect during summer 2013. ................................ ................................ .. 47 2 10 A comparison between the subtidal model and observation velocities for the south transect during winter 2014. ................................ ................................ ...... 48 2 11 A comparison between the subtidal model and observation velocities for the south transect during spring 2015. ................................ ................................ ..... 48 3 1 The subinertial parameters during spring season around Cape Canaveral a nd False Cape associated shoals ................................ ................................ .... 66 3 2 Result s of CHEOF during spring season ................................ ........................... 67

PAGE 10

10 3 3 Wavelet coherences results between CHEOF Mode 1 and different forcings during spring season. ................................ ................................ ......................... 68 3 4 Wavelet cohe rence results between the sea level at Trident Pier and the Florida current during spring season. ................................ ................................ 68 3 5 Wavelet coherence results between the along shelf wind and the Florida current during spring season. ................................ ................................ ............. 69 3 6 Time series of the along shelf momentum balance during spring season. ......... 69 3 7 Time series of the across shelf momentum balance during spring season. ....... 70 3 8 The standard deviation for each term of the momentum balance during spring season. ................................ ................................ ................................ .... 71 3 9 A diagram of the subinertial circulation in the inner shelf adjacent to Cape Canaveral, FL. ................................ ................................ ................................ .... 72 4 1 A sketch of the proposed Goodyear scrap tires as a floating breakwater via Candle and Fischer (1975). ................................ ................................ ................ 98 4 2 US patented design by Hibarger et al. (1979) for an interlocking array of tires to form a floating breakwater that is completely made of tire materials. ............. 99 4 3 A pipe tire floating breakwater that was investigated by Harms and Westerink (1980) using regular waves. ................................ ................................ ............... 99 4 4 US patented design by Walter (2000) for developing an artificial reef made of a concrete frame consisting of six concrete beams inserted through a number of tires. ................................ ................................ ................................ 100 4 5 US patented design by Cederlund (2009) for a wave attenuation system, which consists of a floating member above the waterline, interlocking tires below the waterline, and anchors to keep the system in position. .................... 101 4 6 Top and side views of model 1 configuration. It is completely slack and consists of 4 rows with a single layer of scrap tires. Measurements are not to scale. ................................ ................................ ................................ ................ 102 4 7 Top and side views of model 2 configuration. ................................ ................... 103 4 8 Top and side views of model 3 configuratio n. ................................ ................... 104 4 9 Top and side views of model 4 configuration. ................................ ................... 105 4 10 Top and side views of model 5 configuration.. ................................ .................. 106 4 11 Top and side views of model 6 configuration. ................................ ................... 107

PAGE 11

11 4 12 Top and side views of model 7 configuration. ................................ ................... 108 4 13 Top and side views of model 8 configuration. ................................ ................... 109 4 14 Top and side views of model 9 configuration.. ................................ .................. 110 4 15 Picture of model 1 configuration in the wave flume at the Hydraulic and Coastal Engineering Laboratory of the Kuwait Institute for Scientific Research.. ................................ ................................ ................................ ........ 111 4 16 Picture of model 2 configuration. ................................ ................................ ...... 112 4 17 Picture of model 3 configuration. ................................ ................................ ...... 113 4 18 Picture of model 4 configuration. ................................ ................................ ...... 114 4 19 Picture of model 6 configuration. ................................ ................................ ...... 115 4 20 Picture of model 7 configuration. ................................ ................................ ...... 116 4 21 Picture of model 8 configuration. ................................ ................................ ...... 117 4 22 Picture of model 9 configuration. ................................ ................................ ...... 118 4 23 The concrete wave flume at the Hydraulic and Coastal Engineering Laboratory of the Kuwait Institute for Scientific Research. ............................... 119 4 24 The flap type wavemaker for the concrete flume at the Hydraulic and Coastal Engineering Laboratory of the Kuwait Institute for Scientific Research. ........... 120 4 25 The beach for the concrete flume at the Hydraulic and Coastal Engineering Laboratory of the Kuwait Institute for Scientific Research. ............................... 121 4 26 Wave probes 2, 3, and 4 are used to estimate the reflected wave height. ....... 122 4 27 The work station by the concrete flume at the Hydraulic and Coastal Engineering Laboratory of the Kuwait Institute for Scientific Research. ........... 123 4 28 A wooden plank of 2.40 cm x 4 cm x 9 cm (LxWxH ). ................................ ....... 124 4 29 A stainless steel angle of 285 cm x 5 cm x 5 cm (LxWxH ). .............................. 124 4 30 Polyethylene rope of 6 mm thickness, used for tightening the scrap tire configurations. ................................ ................................ ................................ .. 125 4 31 A bowline knot was used while tightening the scrap tire configurations using a 6 mm polyethylene rope. ................................ ................................ ............... 126

PAGE 12

12 4 32 Polyethylene rope of 14 mm thickness, used for the scrap tires configurations as a mooring rope. ................................ ................................ ............................ 126 4 33 Polyf orm Buoy of 60 cm diameter. ................................ ................................ .... 127 4 34 Gravity anchors of 57 cm diameter, 18 cm thickness, and 100 kg weight. ....... 127 4 35 Effect of / on , and for model 1, where / equaling 0.041, 0.082, and 0.12 due to regular waves. ................................ ............................. 128 4 36 Effect of / on , and for model 2, where / equaling 0.041, 0.082, and 0.12 due to regular waves. ................................ ............................. 129 4 37 Effect of / on , and for model 3, where / equaling 0.041, 0.082, and 0.12 due to regular waves. ................................ ............................. 130 4 38 Effect of / on , and for model 4, where / equaling 0.041, 0.082, and 0.12 due to regular waves. ................................ ............................. 131 4 39 Effect of / on , and for model 5 where / equaling 0.041, 0.082, and 0.12 due to regular waves. ................................ ................................ ........ 132 4 40 Effect of / on , and for model 6, where / equaling 0.041, 0.082, and 0.12 due to regular waves. ................................ ............................. 133 4 41 Effect of / on , and for model 7, where / equaling 0.041, 0.082, and 0.12 due to regular waves. ................................ ............................. 134 4 42 Effect of / on , and for model 8, where / equaling 0.041, 0.082, and 0.12 due to regular waves. ................................ ............................. 135 4 43 Effect of / on , and for model 9, where / equaling 0.041, 0.082, and 0.12 due to regular waves. ................................ ............................. 136 4 44 Effect of / on , and for model 1, where / equaling 0.041 and 0.082 due to random waves. ................................ ................................ ............ 137 4 45 Effect of / on , and for model 2, where / equaling 0.041 and 0.082 due to random waves. ................................ ................................ ............ 138 4 46 Effect of / on , and for model 3, where / equaling 0.041 and 0.082 due to random waves. ................................ ................................ ............ 139 4 47 Effect of / on , and for model 4, where / equaling 0.041 and 0.082 due to random waves. ................................ ................................ ............ 140 4 48 Effect of / on , and for model 5, where / equaling 0.041 and 0.082 due to random waves. ................................ ................................ ............ 141

PAGE 13

13 4 49 Effect of / on , and for model 6, where / equaling 0.041 and 0.082 due to random waves. ................................ ................................ ............ 142 4 50 Effect of / on , and for model 7, where / equaling 0.041 and 0.082 due to random waves. ................................ ................................ ............ 143 4 51 Effect of / on , and for model 8, where / equaling 0.041 and 0.082 due to random waves. ................................ ................................ ............ 144 4 52 Effect of / on , and for model 9, where / equaling 0.041 and 0.082 due to random waves. ................................ ................................ ............ 145 4 53 Comparison of for different scrap tire floating barrier models for different / values and / = 0.041. ................................ ................................ ........ 146 4 54 Comparison of for different scrap tire floating barrier models for different / values and / = 0.082. ................................ ................................ ........ 147 4 55 Measured vs predicted values using the multiple regression equation for model 1. The blue solid line represents a slope of 45. The closer the values to the line, the more accurate the prediction of ................................ ........... 148 4 56 Measured vs predicted values using the multiple regression equation for model 2. ................................ ................................ ................................ ............ 148 4 57 Measured vs predicted values using the multiple regression equation for model 3. ................................ ................................ ................................ ............ 149 4 58 Measured vs predicted values using the multiple regression equation for model 4. ................................ ................................ ................................ ............ 149 4 59 Measured vs predicted values using the multiple regression equation for model 5. ................................ ................................ ................................ ............ 150 4 60 Measured vs predicted values using the multiple regression equation for model 6. ................................ ................................ ................................ ............ 150 4 61 Measured vs predicted values using the multiple regression equation for model 7. ................................ ................................ ................................ ............ 151 4 62 Measured vs predicted values using the multiple regression equation for model 8. ................................ ................................ ................................ ............ 151 4 63 Measured vs predicted values using the multiple regression equ ation for model 9. ................................ ................................ ................................ ............ 152

PAGE 14

14 4 64 Bathymetric map of Cape Canaveral. The inset map shows the location of the study area relative to Florida. The blue strip located north of False Cape represents the ~8 km of critically eroding beach at KSC. ................................ 153 4 65 The tidal r ange and wave seasonality at Cape Canaveral. ............................... 154 4 66 Seasonality of wave direction at Cape Canaveral. ................................ ........... 155 4 67 The total occurrence of significant wave height at Cape Canaveral. The total number of samples is 1756, which was collected seasonally. .......................... 156 4 68 Percent of occurrence of significant wave height at Cape Canaveral. The green boxes highlight the percentages around the most dominant wa ve condition. ................................ ................................ ................................ .......... 157 4 69 Time series of the seasonal max. orbital velocity at the bottom. The values above the dashed line represent the times of the year at which erosion is expected to happen at KSC. ................................ ................................ ............. 158 4 70 Side view of the adjusted cross section of model 7, which is proposed to be used for KSC. ................................ ................................ ................................ ... 158 4 71 Cumulative probability for and values at KSC using model 7. ............... 159 4 72 Google earth image of Kuwait showing the location of Qaru Island relative to the mainland ................................ ................................ ................................ .... 160 4 73 The total occurrences of significant wave height at Qaru Island. The total number of samples is 105178, which was collected via a model developed by the Kuwait Institute for Scientific Research. ................................ ..................... 161 4 74 Percent of occurrence of significant wave height at Qaru Island. The green boxes highlight the high percentages around the most dominant wave condition. ................................ ................................ ................................ .......... 162 4 75 Side view of the adjusted cross section of model 6, which is proposed to be used for Qaru Island. ................................ ................................ ........................ 163

PAGE 15

15 LIST OF ABBREVIATIONS ADCP Acoustic Doppler Current Profiler CTD Conductivity Temperature Depth DHI Danish Hydraulic Institute FBW Floating Breakwaters HEOF Hilbert Empirical Orthogonal Function KSC Kennedy Space Center SPSS Statistical Package for the Social Sciences

PAGE 16

16 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE DEVELOPMENT OF A SCRAP TIRE BARRIER AS A COASTAL STRUCTURE FOR WAVE DAMPING APPLICATIONS By Ahmad Yousif December 2017 Chair: Arnoldo Valle Levinson Major: Coastal and Oceanographic Engineering The wave transmission, reflection, and dissipation of nine new configurations of scrap tire wave barriers have been investigated using physical models with regular and random waves The study aim wa s to achieve the least wave transmission with a minimum number of scrap tires Predictive relations which depend on relative water depth, d/ and relative wave height, /d, h ave been obtained to predict coefficient of transmissi on for nine newly proposed scrap tire configurations using random waves. After performing an oceanographic investigation, t he optimal configurations of this study were recommended to be used for field application to protect the beaches of Kennedy Space Center (KSC), USA and Qaru Island, Kuwait from erosion The tidal and subtidal hydrodynamics were analyzed over ridge swale bathymetry in the inner shelf adjacent to Cape Canaveral, Florida using vessel based data Observations were compared to two analytical models that yield tidal and subtidal solutions. The north transect had relatively smoother bathymetry. Therefore, t idal and subtidal hydrodynamics were consistent with previous studies on flows passing over different bathym etries with less frictional influence The south transect had relatively more complex bathymetry that defined a channel. Hence, tidal and subtidal

PAGE 17

17 hydrodynamics were consistent with previous obser vations on frictional ly dominated flows. Results obtained wi th both tidal and subtidal analytical model solutions highlight the influence of ridge swale bathymetry in inner s helves at those temporal scales and the applicability of open channel concepts to inner shelf dynamics. The sub inertial hydrodynamics were als o investig ated using moored data. W avelet coherence analysis techniques were used between subinertial flow and different forcings N or thward winds coincide with enhancement of the Florida current As a result positive across shelf sea surface slope develops following geostrophy and the subinertial flow moves in the same direction of the western boundary current. On the other hand, southward winds coincide with weakening of the Florida current. Thus, a negative nearshore sea surfac e slope develops to maintain g eostrophic balance and the nearshore flow moves southward The across shelf momentum balance showed that the subinertial flow was mainly geostrophic throu ghout the deployment, while the along shelf momentu m was mainly frictional. Therefore, the dynamics in the region are semi geostrophic.

PAGE 18

18 CHAPTER 1 INTRODUCTION In Kuwait, millions of unrecycled scrap tires have created one of the largest landfills in the w orld [Amusing planet, 2015] According to the U.S. Rubber Manufacturers Association, 275 million scrap tires were in stockpiles in the United States in 2003, and about 290 million new scrap tires were generate d [ Penoyer et. al., 2006 ] These piles pose severe p availability and low cost, they can be used for marine applications where the wave climate is moderate. Most of the year, Kuwait experiences wave heights ranging between 0.25 m and 1 m, with wave periods rangin g between 2 s and 6 s. Seasonal data measurements at Cape Canaveral, FL show a range of wave heights between 0.25 m and 2.5 m, with wave periods of 3 s to 12 s. The energy in such wave climates is moderate and hence suitable for using floating breakwaters (FBW) as wave barriers. Therefore, scrap tires assembled in a particular fashion can be used for such application. main launch center of human spaceflight, located on Cape Canavera l, Florida is facing serious erosion along ~8 km of the shoreline Erosion is increasing the risk of damaging the infrastructure. To determine the optimal scrap tire configuration to mitigate the erosion at Kennedy Space Center ( KSC ) a full understanding of the oceanographic aspects in the region must be carried out first Cape Canaveral, FL is located in an inner continental shelf region. It is the region that lies between the surf zone and the mid shelf. In this region, the water depth expands from a few meters to tens of meters [Lentz and Fewings, 2012] Also in this region, the surface and bottom boundary layers interact

PAGE 19

19 with each other [Lentz and Fewings, 2012; Lentz, 1995]. Observations have shown that across shelf currents are constrained by bathymet ry and therefore are weaker than the along shelf currents [Lentz and Fewings, 2012]. This Dissertation The objective of this dissertation was to achieve the least wave transmission with a minimum number of scrap tires. The ease of construction, installation, transport, and maintenance were considered while selectin g the optimal configuration Also vessel bas ed data were used to determine the influence of ridge swale bathymetry on the spatial structure of tidal and subtidal flows, and to assess whether hydrodynamics derived from open channel flow also apply to inner shelf region In addition, moored data were used to determine the subinertial hydrodynamics off a cape with a ridge swale bathymetry in an inner shelf This document is organized in f ive chapters. The remainder of Chapter 1 describes the study area and observations. Chapter 2 shows the tidal and subtidal hydrodynamics over ridge swale bathymetry. Chapter 3 describes the subinertial hydrodynamics over ridge swale bathymetry around a cap e After gaining a full understanding of the hydrodynamics dominating the region in Chapter 2 and 3, the development of a scrap tire barrier as a coastal structure for wave damping applications is presented in Chapter 4. Finally, Chapter 5 gives the conclu sion of this study. Study Area The study area is located in the inner shelf adjacent to Cape Canaveral, Florida (28.53 N, 80.45 W). It has a typical cape coastline configuration that is characterized by complex bathymetry and changes in coastline orien tation. It consists of a series of cape associated shoals and shore oblique ridges that vary several meters vertically

PAGE 20

20 over a horizontal scale of a few km. The southeast shoals consist of Canaveral Shoal and Canaveral II Shoal (Figure 1 1). Canaveral Shoal is attached to the tip of Cape Canaveral and extends offshore southeastward for ~7 km with a width of ~5 km. ~4 km and a width of 1.5 km. The northeast shoals con sist of Bull Shoal and Chester Shoal (Figure 1 1), which are located ~12 km from the tip of Cape Canaveral. Bull Shoal is 0.7 km wide and extends 7 km in the along shelf direction. Chester Shoal is ~4 km long and 0.7 km wide. It is part of multiple shore o blique ridges associated with a smaller cape located ~15 km northwest of Cape Canaveral known as False Cape (28.59 N, 80.58 W). Those ridges extend ~6.7 km offshore, southeastward of False Cape, with a width of ~1.5 km each. In this region, tides are pre dominantly semidiurnal with a mean range of ~2 m. Spring tides have a range between 1.3 and 1.8 m, while that for neap is 0.6 to 0.9 m. The significant wave height is ~1.2 m in winter, 1 m in spring, 0.5 m in summer, and 0.8 m in fall. The dominant wave pe riod is ~9 s in winter, ~8 s in spring, ~7 s in summer, and ~8 s in fall. The corresponding dominant wave direction is 71 T in winter, 81 T in spring, 107 T in summer, and 86 T in fall. Clearly, the wave direction is easterly. The instantaneous hourly wind speed reaches 16 m/s in winter, 12 m/s in spring, 13 m/s in summer, and 11 m/s in fall. T he subtidal wind speed attains 10 m/s in winter, 8 m/s in spring, 7 m/s in summer, and 6 m/s in fall. The corresponding wind direction is 142 T in winter (from t he southeast), 205 T in spring (from the southwest), 126 T in summer (from the southeast), and 208 T in fall (from the southwest).

PAGE 21

21 Observations Vessel based Data The study area included two sampling transects: the north transect (northeast of Cape Canav eral) and the south transect (southeast of Cape Canaveral; Figure 1 1). The north transect has a bottom slope of 0.002, while the south transect has a bottom slope of 0.004. In order to determine the influence of cape associated shoals on tidal and subtida l hydrodynamics, data were collected between September 24th, 2013 and June 29th, 2016. An Acoustic Doppler Current Profiler (ADCP) was attached to a catamaran, then towed back and forth along a distance of 4.5 km for about 12 hours. The ADCPs recorded the underway current velocity field, surface water temperature, and backscatter wit h a pinging rate of 2 Hz and average ensemble s of 20 profiles. The cruising speed was around 2 m/s and the bin size was 0.5 m. Ensembles had thus sampling intervals of ~ 10 s and spatial resolution of ~ 20 m. Fixed hydrographic profiles were collected from seasonal cruises between fall 2013 and summer 2016 using Castaway and Sea Bird 19 Plus conductivity temperature depth (CTD) profilers depending on their availability Along the t ransects, three CTD casts were performed at the start, middle, and end of each seaward transect repetition. Data from both were comparable. Mooring Data This study focused on sampling at four sites to determine the subinertial hydrodynamics at the study swale bathymetry. Two sites were at each side of Chester Shoal (named Chester swale west and Chester swale east) and the other two at each side of Canaveral II Shoal (named Canaveral swale west and Canaveral swale east; Figure 1 1 ). Data were collected during the spring season, between May 6 th

PAGE 22

22 and June 6 th of 2014, using Acoustic Doppler Current Profilers (ADCPs). The ADCPs recorded water bottom temperature, pressure, current profiles, and wave heights and direction. More details on the locatio ns of the ADCP moorings and the data sampling schemes are provided in Table 1 1 Hydrographic data were collected from seasonal cruises between fall 2013 and summer 2016 using Castaway and Sea Bird 19 Plus conductivity temperature depth (CTD) profilers. Se nsors were compared to each other to ensure consistency. Three CTD casts were taken along two transects. The location of these transects relevant to the ADCP moorings is also shown in Figure 1 1 Hourly winds were compiled for the period of the deployment from the National Oceanic and Atmospheric 28.52 N and 80.18 W ( http://www.ndbc.noaa.gov/ ). The station is approximately 25 km east of Bull Shoal over a water depth of ~45 m. Hourly sea level records at Trident Pier ( https://opendap.co ops.nos.noaa.gov/ ) at station TRDF1 (ID# 8721604), located at Canaveral Port inlet (28.42 N, 80.59 W). The daily mean transport of the Florida ( www.aoml.noaa.gov/phod/floridacurrent/ ) through a transect spanning from West Palm Beach, Florida to Eight Mile Rock, Bahamas. The transect is located ~200 km southeast of Cape Canaveral, Florida. Bathymetric data for the region were obtained from NOA National Geophysical Data Center (NGDC) using the United States Coastal Relief Model ( https://ngdc.noaa.gov/mgg/coastal/crm.html ).

PAGE 23

23 Hydrographic Data The results of the hydrographic measurements are presented in Figure 1 2. The results show that temperature and salinity are mostly uniform with depth. One discrepancy is noted during summer 2015 when temperature decreased 2.8C with depth. Another discrepancy is noted during spring 2015, when sali nity decreased 0.43 g/kg with depth. Density is also noted to be uniform with depth during summer 2013, fall 2013, and winter 2014. In other seasons, a small change in density with depth is noted. In summer 2014, the change in density with depth was 0.24 k g/m 3 ; in summer 2015, at the south transect, the change was 0.18 kg/m 3 ; and in spring 2015, the change was 0.43 kg/m 3 One discrepancy is noted during summer 2015 at the north transect, where the change was about 0.86 kg/m 3 due to the 2.8C change in the t emperature with depth. Since density is mainly uniform or has a small change with depth, the buoyancy forcing in the region is assumed to be negligible over one month Hydrographic data also show a small change in density between the north and south transe cts. In summer 2014, the depth averaged density was 23.5 kg/m 3 in the north transect and 23.4 kg/m 3 in the south transect. In summer 2016, the depth averaged density was 22.25 kg/m 3 in the north transect and 22.3 kg/m 3 in the south transect. In winter 2014, the depth averaged density was 24.1 kg/m 3 in the north transect and 23.9 kg/m 3 in the south transect. In spring 2015, the depth averaged density was 23.1 kg/m 3 in the north transect and 22.7 kg/m 3 in the south transe ct. An exception is noted in June 2015, between the depth average density at the north transect (23.5 kg/m 3 ) and the depth average density at the south transect (22.8 kg/m 3 ). This discrepancy can be related to the 2.8C change in temperature with depth at the north transect, as it was mostly uniform at the south transect. In September 2013, temperature and salinity (28C and

PAGE 24

24 35.7 g/kg respectively) were larger than temperature and salinity (27C and 34.7 g/kg respectively) in October 2013. Therefore, a dist inction is noted between the depth average density in September 2013 (22.8 kg/m 3 ) and October 2013 (22.4 kg/m 3 ). Water temperature is lowest between December and April, and highest between June and October. Although precipitation increases between June and October, enhanced evaporation causes an increase in salinity. The position of the Gulf Stream relative to the coast may also influence salinity. As the Gulf Stream gets closer to the coast, salinity in the region is expected to increase. On the other hand as the Gulf stream moves farther from the coast, salinity in the region is expected to decrease. Also, the Gulf Stream transport is influenced by seasonality. According to Geosat altimetry results, the Gulf stream transport is maximum in fall and minimum in spring, which is in phase with the north/south shifts from its mean position [Kelly and Gille, 1990; Zlotnicki, 1991]. However, satellite imagery of the sea surface temperature of Florida coast show s that the strength of the transport does not necessar ily have a direct influence on salinity near the coast, as the position of the Gulf Stream may shift offshore with increased transport. The T/S diagram shows that for all CTD casts made, salinity ranges between 33.4 kg/m 3 and 35.8 kg/m 3 while temperature ranges between 17.3 C and 30 C. 3 ) has more influence on density than the change in temperature (change of > 12 C). The T/S diagram also shows that water masses during the colder months are similar to one another. The water temperature during these colder months ranges between 18 C and 21 C, while salinity ranges between 33.4 kg/m 3 and 34 kg/m 3 Likewise, the water

PAGE 25

25 masses during the warmer months are similar to each other. The water temperature during these warmer months ranges between 25.7 C and 29 C, while salinity ranges between 35.4 kg/m 3 and 35.7 kg/m 3 Since October is a transitionary month, the corresponding water mass lies between warmer and colder seasons. Figure 1 1 Bathymetric map of Cape Canaveral with the four locations of the ADCP moorings. The green dot represents Chester swale west; the blue, Chester swale east; the cyan blue, Canaveral II swale west; the red, Can averal II swale east; the grey, the NOAA buoy; and the magenta, the sea level at Trident Pier. The black dashed lines show the location of the two transects (north and south) along which hydrographic data were collected. Plots A) and B) show the bathymetry of the north and south transects, respectively. The red solid line in the inset map of Florida shows the location of the Florida current transport.

PAGE 26

26 Figure 1 2 Hydrographic data for the north and south transects d uring all tows performed between fall 2013 and summer 2016. The blue line represents fall 2013 in the north transect and summer 2013 in the south transect; red, summer 2014; magenta, winter 2014; black, spring 2015; green, summer 2015; cyan blue, spring 20 16, and temperature (C) profiles at the two locations; the 2nd shows salinity (g/kg); the 3rd, density (kg/m3); and the last, the temperature/sali nity (T/S) diagram

PAGE 27

27 Table 1 1 Details of the ADCPs at the four mooring p ositions during spring s eason Shoal Chester Canaveral II Location Swale West Swale East Swale West Swale East Latitude, 28N 33.01' 32.64' 23.99' 23.64' Longitude, 80W 28.95' 27.80' 26.33' 25.42' Instrument RDI Workhorse RDI Workhorse Nortek Aquadopp Nortek AWAC Depth (m) 9 12 14 13 Start date (GMT) 6 May 14 6 May 14 6 May 14 6 May 14 Start time 15:30:00 14:10:00 20:10:00 18:42:00 End date 6 Jun 14 6 Jun 14 6 Jun 14 6 Jun 14 End time 15:10:00 14:00:00 19:40:00 18:51:00 Total time span (days) 31 31 31 31 Waves Burst interval (minutes) 120 60 60 Samples per burst 2400 2400 2400 Sampling rate (Hz) 2 2 2 Currents Blanking distance (m) 1.1 1.1 0.2 0.4 Interval (minutes) 10 10 30 24 Number of cells (bins) 30 25 40 40 Cell size (m) 0.5 0.5 0.5 0.5

PAGE 28

28 CHAPTER 2 TIDAL AND SUBTIDAL HYDRODYNAMICS OVER RIDGE SWALE BATHYMETRY Background The inner continental shelf is the region between surf zone and mid shelf, where water depth expands from a few meters to tens of meters [Lentz and Fewings, 2012]. In such region, the surface and bottom boundary layers interact with one another [Lentz, 1995; Lentz and Fewings, 2012]. Obser vations have shown that across shelf currents are constrained by bathymetry and therefore are weaker than the along shelf currents [Lentz and Fewings, 2012]. This clear influence of bathymetry on the flow, particularly in inner shelves, indicates the need to further investigate its spatial influence on tidal and subtidal hydrodynamics. The dynamics of flow have been investigated over rugged bathymetry such as sills [Farmer and Armi, 1986, Eriksen, 1991; Stenstrom, 2003], bathymetric depressions also known a Levinson and Guo, 2009; Salas Monreal and Valle Levinson, 2009], channels [Winters and Seim, 2000 ], estuaries [Valle Levinson et al., 2003; Wong, 1994], and inlets [ Valle Levinson et al., 2015]. These studies have demonstrated that frict ionally influenced flow has the maximum val ues over the deepest part of a cross section and that velocity contours are parallel to the bathymetry throughout the section. On the other hand, non frictionally influenced flow would be dominated by inertial eff ects and follow s Bernoulli type dynamics, where it accelerates over shallower depths and decelerates over deeper depths. It was also noted in cases of weak frictional and advective effects that the flow may tilt to the right due to the Coriolis effect in t he northern hemisphere [Valle Levinson et al., 2000; Valle Levinson et al., 2003].

PAGE 29

29 Capes and their associated shoals are characterized by complex bathymetry, as they represent a sudden change in the coastline orientatio n [Kumar at el., 2013] and water depth. Some studies have been targeted on the interaction between flow around these complex features and sediment transport [McNinch and Luettich, 2000; Sanay et al., 2007], plus the interaction between flow and shoaling de ep water waves [Kline et al., 2012]. However, the hydrodynamics of the flow over ridge swale bathymetry caused by such complex bathymetries remain unexplored. The objective of this study is to determine the spatial structure of tidal and subtidal hydrodyna mics off a cape with a complex bathymetry in an inner shelf. The objective is tackled with comparison of observations to two analytical models: a) a tidal model influenced by friction, local acceleration, and pressure gradient, and 2) a subtidal model influenced by friction, Coriolis, and pressure gradient. These analytical models have been developed to approximate the dynamics in estuarine environments. An additional objective is to assess whether the channelized dynamics over ridge sw ale bathymetry can be represented as in those of channelized environments [Gill, 1982; Armani, 1986; Wong, 1994; Valle Levinson et al., 2000; Valle Levinson et al., 2003; Valle Levinson and Guo, 2009; Salas Monreal and V alle Levinson, 2009; Valle Levinson et al., 2015]. Data Analysis Matrix Arrangement and Compass Calibration First, the data collected are trimmed with the following criteria: % good > 70 to 80%, |error| < 10 cm/s, discharge < 100 m3/s, and ship speed or bottom track speed > 0.15 m/s. Then, the method of Joyce (1989), similar to the method of Pollard & Reid (1989), is u sed to calibrate the ADCP compass. Correction coefficients ( and ) are

PAGE 30

30 used to correct the velocity components by considering both the bottom track velocity and navigation velocity as follow: ( 2 1) ( 2 2) ( 2 3) ( 2 4) where is the east component of the bottom track velocity, is the north comp onent of the bottom track velocity, is the east component of the navigation velocity (from GPS), is the north component of the navigation velocity, is the east component of the current velocity measured by the ADCP, is the north componen t of the current velocity measured by the ADCP, is the corrected east component of velocity, is the corrected north component of velocity, and < > indicates average throughout one transect repetition. Then, a regular matrix is generated for , and corresponding to each transect repetition in terms of depth and distance. Each transect repetition is identified according to the time of beginning and end of each repetition. The origin of the regular matrix (zero distance) is arbitrary. The distance from that origin to the location of each profile is calculated in order to generate the regular grid. The end result is a group of regular grids, where is the number of transect repetitions. Least Squares Fit to O bservations Since the semidiurnal tides are dominant in the region, a least squares fit with a semidiurnal (12.42hr) harmonic was applied to the flow. The amplitude and phase were therefore calculated as:

PAGE 31

31 (2 5) where is the along shelf velocity measured from the field, is the along shelf residual velocity that represents the subtidal flow, is the along shelf semidiurnal amplitude that represents the amplitude of the tidal flow, is the semidiurnal angular tidal frequency, is the time, and is the semidiurnal phase. Since the subtidal velocity can be positive or negative, and its direction influences the interpretation of the driving hydrodynamics, the subtidal flow was rotated in the direction of the principal axis. Tidal Analytical Model Solutions A tidally driven analytical model was used to solve tidal amplitude and phase for any arbitrary bathymetry [Huijts et al., 2006; Huijts et al., 2009; H uijts et al., 2011]. A schematic representation of the model reference frame for one of the locations is provided in Figure 2 1 Scaling analysis of the governing equations yields the model momentum balance as [Huijts et al., 2006; Huijts et al., 2009; Hui jts et al., 2011] : (2 6a) (2 6b) where represents local acceleration, represents Coriolis acceleration, is the model along shelf tidal flow (in the principal axis direction), is the Coriolis parameter, and represent the pressure gradient along and across shelf, is the acceleration due to gravity, is the free surface el evation, and represent the stress divergence (friction) along and across shelf, is the vertical eddy viscosity,

PAGE 32

32 and the subzero terms ( and ) represent the first order or dominant order tidal solutions. The analy tical model assumes that the motion in the along shelf is given by simplified momentum balance between friction, local acceleration, and the pressure gradient due to the free surface elevation. Only barotropic effects are considered. The model drawbacks ar e that no along shelf variation in bathymetry is allowed and no coastline curvature is considered [Valle Levinson et al., 2015]. It also explores whether the tidal flow can be explained as a simple damped wave [Valle Levinson et al., 2015]. The domina nt order solution of the model simplified momentum balance is obtained in terms of [Huijts et al., 2006; Huijts et al., 2009; Huijts et al., 2011 ]: (2 7) which is a depth scale parameter, and (2 8) which is the surface slope along the shelf. The tidal flow is: (2 9) The modulus of the complex analytical represents the model along shelf tidal amplitudes Equation 2 9 indicate s that is a function of the vertical direction the transverse distribution of water depth the gradient of the free surface elevation, and the vertical eddy viscosity As the tidal period increases, both and decrease. As increases, decreases. Values of represent the influence of friction in the water column. Small means high friction or frictional effects occupying most or all the water column. In equation 2 8 is the sectional average tidal amplitude, is the cross secti onal area, and is the cross sectional width. and are the the two free

PAGE 33

33 parameters for the solution of the analytical model Therefore, these parameters can be varied throughout possible ranges to find the optimal values to reduce the difference b etween model and observation. For more details on the analytical model solutions, please see Huijts et al. (2006). Subtidal Analytical Model Solutions In 2003, Valle Levinson et al. developed a density driven analytical model that k in 2000 to better explain exchange flows over various bathymetries. All equations presented in this section have been introduced by Valle Levinson et al. (2003) and Valle Levinson (2008). The same reference frame presented previously in Figure 2 1 will be used here. This model was used in our study to determine the subtidal hydrodynamics over the ridge swale bathymetry. It describes the momentum balance as follows: (2 10a) (2 10b) where and represents Coriolis acceleration along and across shelf, is the Coriolis parameter (7 x 10 5 s 1 ), is a reference seawater density, and represent the barotropic contribution to the pressure gradient force per unit mass, and represent the baroclinic contribution, while and represent friction. This analytical solution assumes that the mot ion is produced by pressure gradients and modified by Coriolis and frictional influences. The pressure gradient includes the influence of free surface elevation and the density gradient. Therefore, both barotropic and baroclinic contributions are considere d in this analytical model. An

PAGE 34

34 important restrictive assumption in this analytical solution is that the vertical eddy viscosity is constant. It also assumes uniform along shelf bathymetry and does not include the influence of curvature effects [Valle Levin son et al., 2015]. The Ekman number is a non dimensional number, which represents the ratio between friction and Coriolis as : (2 11 ) where is the maximum water depth of the cross section. Dynamically deeper cross sections would have a smaller Ekman number, which indicates that the effect of friction is restricted to the very deep part of the water column. Dynamically shallower cross sections would have a larger Ekman number, indicating that the frictional effects will exten d over the entire water column. The parameter is calculated as: (2 1 2 ) where stands for the Ekman layer depth. The density gradient arises from making it dynamically consistent with the surface slope (e.g. Valle Levinson et al., 2003): (2 1 3 ) where is the cross sectional average speed (m/s) of the net flux in the principal axis direction (along shelf) or in other words the net flux per unit area. The surface slope ( ) is prescribed as:

PAGE 35

35 (2 1 4 ) is a slope at the coast that decays exponentially. The parameter is the rate of exponential decay that can be related to the i nternal radius of deformation ; The barotropic and Baroclinic contributions of the flow are given as : and (2 1 5 ) Therefore, the subtidal model solution has contributions from and : (2 1 6 ) where stands for model. The subtidal flow in the principal axis direction (along shelf) is the real part of the complex subtidal flow presented in equation 2 1 6 The free parameters in this analytical model are , and Optimal values of the free parameters were obtained by varying them throughout possible ranges to reduce the difference between model and observa tion. For more details beyond this section on the analytical solution, please see Valle Levinson et al. (2003) and Valle Levinson (2008). Results First, the results of the tidal model vs observations are described. The n, the results of the subtidal model vs observations are explained. Tidal Model vs Observation s In order to obtain the model tidal amplitudes, the free parameters ( and ) in the analytical model were varied. The values of the parameters, which provided the best match between the model and observation were then used to create a contour plot of

PAGE 36

36 well as appreciate the capabilities of this model. The model is unable to reproduce the observation exactly bec ause of those limitations. However, physical insights can be drawn from the and values. A comparison between the tidal model and observation amplitudes is provided in Figures 2 2 to 2 5 The comparison is in terms of the contour plots of semidiurn al amplitudes and the corresponding and values. In the north transect the maximum tidal flow (0.08 m/s) during fall 2013, is located over the shoals (Figure 2 2 ). The best match between the tidal model and observations was achieved with = 0.060 m /s and = 1x10 3 In the south transect, the maximum tidal flow during spring 2015, summer 2015, and spring 2016 (0.15 m/s, 0.3 m/s, and 0.3 m/s respectively) is located in the channel, which is the deepest part of the cross section ( x/B = ~0.3 to 0.7; Figures 2 3 to 2 5 ). The corresponding values of and that achieved the best match between the tidal model and observations are: = 0.084 m/s and = 5x10 3 during spring 2015; = 0.176 m/s and = 3.1x10 3 during summer 2015; and = 0.175 m/ s and = 3x10 3 during spring 2016. Also, the tidal model contours of the flow in the south transect are following the bathymetry throughout the section. In order for the tidal model to emulate observations, the vertical eddy viscosity ( ) had to be decreased in the north transect with gentler bathymetry, while increased (three to five times) in the south transect with steeper bathymetry. Increasing or decreasing affects friction ( ). Therefore, the flow in the north transect is mor e influenced by local acceleration such that inertial effects cause the flow to accelerate over shallower depths. On the other hand, the flow in the south transect is dominated by

PAGE 37

37 friction such that the maximum flow is located in the channel. These results indicate that bathymetry and friction have influence on the tidal flow. The relationship between the tidal flow and bathymetry will be explored in the discussion. Subtidal Model vs Observation s To obtain the model subtidal velocities, the free parameters ( , and ) in the analytical model were varied based on trial and error until the best match between the model and observation was achieved. As with the tidal model, it is important to acknowledge the limitations as well as appreciat e the capabilities of the model. A comparison between the subtidal model and observation velocities is provided in Figures 2 6 to 2 12 plots and the corresponding values of , and In the north transect the maximum subtidal flow during fall 2013, summer 2014, and spring 2015 (0.3 m/s, 0.15 m/s, and 0.3 m/s respectively) is located over the shoals ( Figures 2 6 to 2 8 ). During these three seasons, the value of remained constant (0.6), while values ranged between 1.5x10 4 and 2.3x10 4 The absolute value s of ranged between 2.7x10 3 m/s and 5.5x10 3 m/s, and they ranged between 1.2x10 6 and 2.4x10 6 for In the south transect, the maximum flow during summer 2013, winter 2014, spring 2015, and summer 2016 (0.25 m/s, 0.3 m/s, 0.2 m/s, and 0.35 m/s respectively) is located over the deepest part of the cross section (x/B= ~0.3 to 0.7; Figures 2 9 to 2 12 ). The corresponding values of ranged be tween 6x10 2 and 7x10 2 The absolute value s of ranged between 0.74x10 5 m/s and 1.4x10 5 m/s, and t hey ranged between 0.8x10 6 and 1.8x10 6 for The value of continued to be constant (0.1) during these

PAGE 38

38 seasons. The modeled contours o f the subtidal flow in the south transect are also following the bathymetry throughout the section as noted with the tidal model. In order to use the subtidal model to represent the observations, Ekman number ( ), which indicates the effect of friction on the water column, had to be decreased in the north transect, while increased in the south transect by two orders of magnitude. This indicates that the frictional effects in the south transect with steeper bathymetry extend over the entire water column a nd the flow is dominated by friction. On the other hand, the effect of friction in the north transect with gentler bathymetry is restricted to the very deep part of the water column. Therefore, the subtidal flow in the north transect is more influenced by Coriolis during fall 2013 and spring 2015 ( Figures 2 6 and 2 8 ) such that the maximum current of subtidal flow moving southward is tilted to the right. During summer 2014, the maximum current of subtidal flow moving northward is ti lted to the le ft ( Figure 2 7 ). Since the subtidal model considers the influence of friction, Coriolis, and pressure gradient, we can only suggest when the subtidal flow is following Bernoulli type dynamics (pressure gradient balanced by advection). Similar to the tidal model, resu lts obtained with the subtidal model also indicate the influence of bathymetry and friction on the subtidal flow. The relationship between the subtidal flow and bathymetry will also be explored in the discussion. Discussion Tidal Model In general, the vert ical eddy viscosity ( ) had a larger influence on the shape of the contours than the sectional average tidal amplitude ( ). As increases, friction ( ) increases, and the maximum flow tends to be located over the deepest part of

PAGE 39

39 the c ross section ( x/B = ~0.3 to 0.7; Figures 2 3 to 2 5 ). The model tidal amplitude contours also tend to be parallel to the bathymetry throughout the section. This behavior has been seen in estuaries [Won g, 1994; Valle Levinson et al., 2003] and inlets [Valle Levinson et al., 2015]. On the other hand, as gets smaller, friction decreases and local acceleration becomes more important. Therefore, inertial effects cause the flow to accelerate over shallower depths (over the shoals; Figure 2 2 ) [Valle Levinson et al., 2015]. The south transect was dominated by friction and had larger values (3x10 3 m 2 /s to 5x10 3 m 2 /s). This is expected in inner shelves where both the surface and bottom boundary layers overlap [Lentz, 1995; Lentz and Fewings, 2012]. In cont rast, the north transect, which had a more influential local acceleration, had three to five times smaller value (1x10 3 m 2 /s). These results can be directly linked to the bathymetry at each location. The north transect tends to have gentler bathymetr y with a slope of 0.002. The effect of friction will be restricted to the very deep part of the water column section has a smaller and local acceleration has more influenc e. The south transect, however, has a steeper bathymetry with a slope of 0.004, twice the slope of the north transect, forming the shape of a channel. Therefore, frictional effects will extend over the entire water column because of larger In the nor th transect, where the tidal flow tends to be more influenced by local acceleration and less by friction, the corresponding along shelf momentum can be approximated with : (2 1 7 )

PAGE 40

40 which describes the tidal propagation as a wave like phenomenon [Valle Levinson, 2010]. In the south transect, where the tidal flow tends to be more influenced by friction, neglecting the local acceleration term leads to the following momentum equation in the along shelf : (2 1 8 ) which describes the tidal propagation as a diffusive phenomenon [LeBlond 1978; Friedrichs and Madsen 1992; Waterhouse and Valle Levinson, 2010]. The discrepancies between the model and observed tidal amplitudes in Figure 2 2 could be attributed to the restrictive model assumptions. It may also indicate that there are more forces influencing the north transect other than local acceleration. The tidal model used for this study was originally developed for channelized flow, while our observations are in t he open inner shelf. Despite that, the model was still very useful in representing the essence of the flow at each location and helped to determine the dominate hydrodynamics. This indicates that the tidal flows are heavily influenced by bathymetry. Subtid al Model In general, Ekman number ( ) and the cross sectional average speed of the net flux ( ) had larger influence on the shape of the contours than the surface slope ( ) and the rate of exponential decay ( ). Since is a function of and (see eqn. 9), smoother cross sections such as the north transect would have a smaller Ekman number (1.5x10 4 to 2.3x10 4 ). This indicates that the effect of friction is restricted to the hand, more complex bathymetry with a steeper cross section such as the south transect would have a larger Ekman number (6x10 2 to 7x10 2 ). This indicates that the frictional

PAGE 41

41 effects will extend over the entir e water column, which is expected in inner shelves where both the surface and bottom Ekman layers overlap [Li and Weisberg, 1999b]. Therefore, the south transect was dominated by friction and the strongest subtidal currents were located over the deepest pa rt of the cross section (x/B= ~0.3 to 0.7; Figures 2 9 to 2 12 ) [Wong, 1994; Valle subtidal velocity contours were parallel to the bathymetry throughout the section emulatin g open channel flow [Lwiza et al., 1991; Valle Levinson et al., 2015]. This is consistent with moored data that were collected in the same area of study during spring 2014 ( see chapter 3 ) Results showed that the along shelf momentum balance around the ridges is mainly frictional (dominated by pr essure gradient, bottom stress and the gradient of ). On the other hand, the north transect had smaller values by two sectional area, when the flow moved southward and the maximum curren t tilted to the right (looking northward), suggests the influence of Coriolis (Figures 2 6 2 8 and 2 9 ) [Valle Levinson and Levinson et al., 2000; Valle Levinson et al., 2003]. Furthermore sectional area, when the flow was northward and the maximum current tilted to the left (Figure 2 7 ), suggests that the flow was not influenced by Coriolis, but rather followed Bernoulli type dynamics. When the flow is following Bernoulli type dynamics, it accelerates over shallower depths and decelerates over deeper depths. Since the subtidal model considers only the contribution of friction, Coriolis, and pressure gradient, the exact hydrodynamics causing such behavio r cannot be verified It was also noted that the values of the cross sectional average speed of the net flux ( ) at the north transect (2.7x10 3 m/s to 5.5x10 3 m/s) were larger than those

PAGE 42

42 at the south transect by two orders of magnitude (0.74x10 5 m/s to 1.4x10 5 m/s). This can be related to the higher frictional influence of the complex bathymetry at the south transect, which decelerates. The discrepancies between the model and observed subtidal currents in Figures 2 6 to 2 8 could be attributed to the restrictive model assumptions such as constant or influential forces that are not considered in the model such as advection. Similar to the tidal model, the subtidal model used for this study was originally developed for channelized flow, while our obse rvations are in the open shelf. This was attributed to the marked bathymetric control on the inner shelf flow. The model was still very useful in representing the flow at each location and helped to determine the dominant hydrodynamics. As found w ith tidal flows, results indicate that the subtidal flows will display a structure that is consistent with frictional effects governed by bathymetry. Summary This study determined the influence of the ridge swale bathymetry on tidal and subtidal hydrodynam ics at two transects. The north transect has a smoother bathymetry than the south transect. At the north transect, the effect of friction was restricted to the 0.3 fr om the bottom). Therefore, in the north transect, tidal hydrodynamics were more influenced by local acceleration, while subtidal hydrodynamics were either influenced by Coriolis or followed Bernoulli type dynamics. This was consistent with previous studies on flows passing over different bathymetries with less frictional influence. Since the subtidal model considers the contribution of friction, Coriolis, and pressure gradient, we can only suggest when the flow follows Bernoulli type dynamics. At the south south transect, frictional effects could extend over the entire water column, which is expected

PAGE 43

43 in inner shelves where both the surface and bottom boundary layers overlap. As a result, the south transect was dominated by friction, and the maximum tidal and subtidal flows were located over the deepest part of the cross section (x/B= ~0.3 to 0.7). Also, the modeled contours of principal axis flow were parallel to the bathymetry throughout the section. This behavior followed open channel flow. The tidal and su btidal analytical model solutions support each other and highlight the strong influence of ridge swale bathymetry on tidal and subtidal hydrodynamics as also observed in channelized systems Figure 2 1 Schematic representation of the model reference frame for the south transect bathymetry. The same reference frame is used for the north transect.

PAGE 44

44 Figure 2 2 A comparison between the tidal model and observation amplitudes for the north transect during fall 2013. Figure 2 3 A comparison between the tidal model and observation amplitudes for the south transect during spring 2015.

PAGE 45

45 Figure 2 4 A comparison between the tidal model and observation amplitudes for the south transect during summer 2015. Figure 2 5 A comparison between the tidal model and observation amplitudes for the south transect during spring 2016.

PAGE 46

46 Figure 2 6 A comparison between the subtidal model and observation velocities for the north transect during fall 2013. The black line at the bottom represents bathymetry, while the black contour represents the location of the strongest currents. F igure 2 7 A comparison between the subtidal model and observation velocities for the north transect during summer 2014.

PAGE 47

47 Figure 2 8 A comparison between the subtidal model and observation velociti es for the north transect during spring 2015. Figure 2 9 A comparison between the subtidal model and observation velocities for the south transect during summer 2013.

PAGE 48

48 Figure 2 10 A comparison between the subtidal model and observation velocities for the south transect during winter 2014. Figure 2 11 A comparison between the subtidal model and observation velocities for the south transect during spring 2015.

PAGE 49

49 Figure 2 12 A comparison between the subtidal model and observation velocities for the south transect during summer 2016.

PAGE 50

50 CHAPTER 3 SUBINERTIAL HYDRODYNAMICS OVER RIDGE SWALE BATHYMETRY AROUND A CAPE Background The inner continental shelf lies between the surf zone and the mid shelf. In this region, the water depth expands from a few meters to tens of meters. Also in this region, the surface and bottom boundary layers interact with each other [Lentz and Fewings, 2012; Len tz, 1995]. Observations indicate that cross shelf currents are constrained by bathymetry and therefore are several times smaller than the along shelf currents [Lentz and Fewings, 2012]. In previous studies performed on inner, mid, and outer shelves, the ac ross shelf momentum balance is predominantly geostrophic [Fewings and Lentz 2010; Liu and Weisberg, 2005; Shearman and Lentz 2003; Lentz et al., 1999; Li and Weisberg, 1999a, b; Lee et al., 1984, 1989; Brown et al., 1985, 1987; Thompson and Pugh, 1986; Nob el and Butman, 1983; Allen and Kundu, 1987]. The along shelf momentum is expected to be frictional: pressure gradient mainly balanced by surface and bottom stresses [Scott and Csanady, 1976; Pettigrew, 1981; Lentz and Winant, 1986; Masse, 1988; Lee et al., 1989; Lentz, 1994; Lentz et al., 1999]. Inner shelf morphology and hydrodynamics can be modified by capes and points. Capes and their associated shoals are characterized by complex bathymetry, as they represent a sudden change in the coastline orientation [Kumar at el., 2013] and water depth. The effect of such bathymetry on subinertial circulation remains unclear [Kumar at el., 2013]. The objective of this study is to determine the subinertial hydrodynamics off a cape with a ridge swale bathymetry in an i nner shelf. This objective is addressed by: a) using statistical techniques to assess the response of the subinertial flow to different forcings, and b) calculating the momentum balance along and across the shelf.

PAGE 51

51 Data Analysis Filtering Low frequency time series of velocity profiles, wind, and sea level were obtained by filtering measured variables with a 30 hour low pass Lanczos filter to remove the oscillations associated with tidal motion and inertial oscillations of ~25 hr. Time series at differen t depths therefore only describe the variance at subtidal and subinertial periods. Concatenated Hilbert Empirical Orthogonal Function (CHEOF) CHEOF was implemented to the subtidal flow profiles at the four locations to obtain their dominant modes of variab ility. The along and across shelf velocity components at the four locations were concatenated into one matrix to build the covariance matrix related to interactions with each other. The real part of the complex time series is the original observation and the imaginary part is phase shifted by / 2. The principal modes of coverability are determined by finding the eigenvalues and eigenvectors of the complex co variance matrix [Emery and Thomson, 1998]. CHEOF was used instead of the Concatenated Real vector Empirical Orthogonal Function (CREOF) because CHEOF has a better phase resolution The reason is that CREOF yields varying time series that decrease or increase in magnitude whereas the spatial structure remains the same [Hannachi et al., 2007]. Concatenat ed Complex Empirical Orthogonal Function (CCEOF) does not preserve the vector nature and therefore may have direction ambiguity in the eigenstructures [Khaihatu and Shay, 1998]. Therefore, CHEOF was used instead. Wavelet Analysis Unlike spectral analysis, wavelet analysis is applicable to nonstationary time series in which the amplitudes and phases of a signal may be changing in time or space

PAGE 52

52 [Emery and Thomson, 1998]. Wavelet transform can determine the periods and time at which a time series has high powe r. The scaled and normalized wavelet is calculated as [Grinsted et al., 2004]: (3 1) where, is a uniform time step, is a scale, is a time series, is a Morlet wavelet, and is a time index. Wavelet coherence can determine the periods and time at which two time series (X and Y) have high common power, whether they are coherent at those periods, and their phase [Grinsted et al., 2004]. The scaled and normalized wavelet coherence, is calculated as [Torrence and Webster, 1999; Grinsted et al., 2004]: (3 2) where, is a smoothing operator in time and frequency, and are the wavelets for each time series, is the cross wavelet, the terms , and represent the power, and is used to convert to energy density. Two time series are considered highly coherent as their coherency amplitude gets closer to 1. esents the phase. Arrows pointing to the right mean that X and Y are in phase. Arrows pointing to the left mean that X and Y are out of phase. Arrows pointing downward mean X is leading Y by 90, while arrows pointing upward mean Y is leading X by 90. Usi ng wavelet analysis techniques to compare the time series of the most dominant mode of CHEOF with different forcings will help to understanding their relationship and how they interact with one another.

PAGE 53

53 Depth averaged Along Shelf Momentum Balance The depth averaged along shelf momentum balance is calculated to determine the dominant forces driving the along shelf subinertial currents around the nearshore bathymetric features. The equation is as follows [Lentz and Fewings, 2012], (3 3) w here is the along shelf current, is the cross shelf current, are the along shelf nonlinear advection terms, is the Coriolis frequency, is the is the latitude ( ), is the cross shelf Stokes drift velocity, g is the acceleration due to gravity, is the sea level, is the along shelf barotropi c pressure gradient, is the along shelf wind stress, is the along shelf bottom stress, is a reference seawater density, h is the undisturbed water depth, and are radiation stresses due to incident waves, is the alo ng shelf bottom stress due to waves, and the overbar represents depth averages. Seasonal cruises to collect hydrographic data at the north transect (northeast of Cape Canaveral) and the south transect (southeast of Cape Canaveral; Figure 1 1), showed chang es in density that are between 0.05 kg/m 3 and 0.4 kg/m 3 within the area of study. Therefore, only the barotropic pressure gradient was considered in the momentum balance. The cross shelf Stokes drift velocity was estimated using linear wave theory [Lentz a nd Fewings, 2012; Fewings and Lentz, 2010; Lentz et al., 2008; Fewings et al., 2008, Xu and Bowen, 1994; Stokes, 1847; Mei, 1983],

PAGE 54

54 (3 4) where is the significant wave height, is the angular wave frequency, is the wavenumber, is the vertical coordinate with at the mean water surface, and is the direction the waves are propagating, measured counterclockwise from the +x direction, so for waves propagating o nshore [Lentz and Fewings, 2012; Longuet Higgins and Stewart, 1964; Mei, 1983]. The along shelf surface stress was estimated as in Smith (1988), (3 5) where is the density of air (1.23 kg/m 3 according to the International Standard Atmosphere [ISA]), is the wind velocity, is the along shelf wind velocity component, and is the air water drag coefficient considered to be when < 11 m/s and calculated as for m/s [Large and Pond, 1981 ]. Bottom stress was calculated using the quadratic drag formula [Fewings and Lentz, 2010; Liu and Weisberg, 2005; Geyer et al., 2000; Lee et al., 1984], (3 6) where is the bottom velocity vector and is the nondimensional bottom drag coefficient generally for sandy bottoms [Valle Levinson et al., 2015; Salas Monreal and Valle Levinson, 2009; Valle Levinson et al., 2003; Liu and Weisberg, 2005]. The radi ation stresses were estimated using linear wave theory as follows [Lentz and Fewings, 2012; Longuet Higgins and Stewart, 1964],

PAGE 55

55 (3 7) (3 8) (3 9) where is the wave energy, is the root mean square wave height, is the wave group velocity, and c is the phase speed of the waves. The gradients were estimated using a finite difference approxim ation between lo cations. The along shelf bottom stress due to the waves was estimated as follows [Lentz et al., 2008; Xu and Bowen, 1994; Longuet Higgins, 1953], (3 10) where is the vertical eddy viscosity used to represent the turbulent processes, and is the height above the bottom. Depth averaged Across Shelf Momentum Balance The depth averaged across shelf momentum balance is [Lentz and Fewings, 2012], (3 11) where are the across shelf nonlinear advection terms, is the along shelf Stokes drift velocity, is the across shelf pressure gradient, is the across shelf surface stress due to the winds, is the across shelf bottom stress, and is the

PAGE 56

56 across shelf bottom stress due to the wav es. The along shelf Stokes drift velocity was estimated using linear wave theory [Lentz and Fewings, 2012; Fewings and Lentz, 2010; Lentz et al., 2008; Fewings et al., 2008, Xu and Bowen, 1994; Stokes, 1847; Mei, 1983]: (3 12) The across shelf surface stress was estimated as [Smith, 1988]: (3 13) where is the across shelf velocity component. The across shelf bottom stress was calculated using the quadratic drag law [Fewings and Lentz, 2010; Liu and Weisberg, 2005; Geyer et al., 2000; Lee et al., 1984]: ( 3 14) The across shelf near bottom stress generated by waves was estimated as follows [Xu and Bowen, 1994; Lentz et al., 2008]: ( 3 15) Geostrophic Balance The influence of the Florida current on sea level can be quantified using geostrophic balance: (3 16) where is the Florida current transport ( Sv ), is the difference in sea level across shelf, is the change in location across shelf, is the sea level at the location of the

PAGE 57

57 Florida current and is the coastal sea level obtained from the tidal gauge located at Trident Pier Equation 3 16 shows a direct correlation between the Florida current and sea level slope. As the Florida current gets stronger, the Coriolis force becomes stronger as well. Therefore, the sea surface slope has to increase in order to keep a balance of forces. In other words, the sea level at the location of the Florida current has to increase, while the sea level at the coast has to decrease to maintain a g eostrophic balance. Results Subinertial Parameters As expected in inner shelves, the cross shelf currents are more than 5 times smaller than the along shelf currents because of bathymetric constraints (Figure 3 1 ). The northward along shelf wind coincides with enhancement of the Florida current (days 135, 142, and 148), while the southward along shelf wind coincides with weakening of the Florida current (days 139, 145, and 152 in Figure 2). As Florida current increas es, the sea level at the coast decreases (days 135, 142, and 148) and vice versa. This can be associated with their linkage through a geostrophic balance. As a result, an inverse relationship is noted between the along shelf current and the sea level at th e coast (days 135, 139, 142, 145, 148, and 152). When the sea level at the coast decreases following geostrophy, the flow moves northward. In contrast, as the sea level at the coast increases the flow moves southward. There is also a clear link between the depth averaged subinertial along shelf currents and the along shelf wind (Figure 3 1 ). The flow is in phase with the along shelf wind (days 130, 135, 139, 14 2 145, 148, and 15 2 ), which enhances th e motion of the flow

PAGE 58

58 CHEOF and Wavelet Coherence CHEOF was used to determine the temporal and spatial modes of variability of the subinertial flow measured in the four locations of the study area. The CHEOF results were mainly associated with the along shore current. In Figure 3 2 the temporal variability of CHEOF Mode 1 accounted for 95.25% of the total variance. The spatial variability was unidirectional at the four locations. Wavelet analysis techniques were used between the principal mode of the subinertial flow (CHEOF Mode 1) and the subintertial forcings to determine the periods and phase at which they were coherent. The results of the wavelet coherence analysis (Figures 3 3 to 3 5 ) showed coherency larger than 0.8 between CHEOF Mode 1 and a) the along shelf wind; b) the sea level at the coast; and c) th e Florida current. Strong coherency between the along shelf wind and the subinertial flow have been documented before by Lee et al. (1984; 1989) when studying the South Atlantic Bight (SAB) and the South Carolina shelf during winter conditions. In 1982, Mi tchum and Sturges also found strong coherency among the currents, sea level, and along shelf wind while studying the west Florida shelf in winter In Figure 3 3 CHEOF Mode 1 has a coherency larger than 0.8 and 0 phase with the Florida current. CHEOF Mode 1 is also coherent with the along shelf wind and sea level at the coast, but has 180 phase with the latter. On day 141, both the Florida current and along shelf wind were moving in the same direction (Figure 3 5 ). The Florida current transport was enhanc ed, the sea level at the coast decreased, and the subinertial flow moved northward. On day 145, the Florida current and along shelf wind were moving in opposite directions (Figure 3 5 ). The Florida current transport weakened, the sea level at the coast inc reased, and the subinertial flow moved southward. On both days (141 and 145), at the same periods of > 0.9 coherency and 0

PAGE 59

59 phase between the Florida current and CHEOF Mode 1 (at periods of 6 8 days; Figure 3 3 ), > 0.9 coherency with 180 phase is observed between the Florida current and the sea level at the coast (Figure 3 4 ). Similarly, the Florida current has > 0.9 coherency and 0 phase with the along shelf wind (Figure 3 5) at the same periods it has > 0.9 coherency and 180 phase with the sea level at the coast (at periods 6 8 days; Figure 3 4). Also, the along shelf wind has > 0.9 coherency and 0 phase with CHEOF Mode 1 (Figure 3 3) at the same periods it has > 0.9 coherency and 0 phase with the Florida current (at periods of 6 8 days; Figure 3 5 ). These relationships suggest a tight coupling of sea level at the coast with shelf currents and the Florida current. They also indicate that the wind affects shelf currents and seems to have influence on the Florida current. These relationships will be expl ored in the discussion Along Shelf Momentum Balance Time series of along shelf momentum balance during the spring season are provided in Figure 3 6 Pressure gradient, bottom stress, and the gradient of were on the same order of magnitude (ranging between x10 7 m/s2 and x10 5 m/s2). Generally, they had larger values than other terms in the along shelf momentum balance. However, in some occasions, surface stress (days 130, 134, 136, and 152), local acceleration (days 135, 138, and 141), advective ac celeration (days 130, 136, and 142), Coriolis acceleration (days 136, 145, and 151), and the gradient of (days 136, 147, and 153) had orders of magnitude ranging between x10 6 m/s2 and x10 5 m/s2. Therefore, the along shelf momentum is not simply ba lanced by pressure gradient, surface, and bottom stresses as found in previous studies on inner shelves [Scott and Csanady, 1976; Pettigrew, 1981; Lentz and Winant, 1986; Masse, 1988; Lee et al.,

PAGE 60

60 1989; Lentz, 1994; Lentz et al., 1999]. However, other terms can still be influential to the along shelf hydrodynamics. On day 134, surface, bottom, and the gradient of had the same order of magnitude (x10 6 m/s2), which exceeded other terms. Therefore, the pressure gradient signal was mainly influenced by t hese three forces. On day 154, the gradient of at Canaveral II swale east was one or more orders of magnitude (x10 5 m/s2) larger than other terms. As a result, the pressure gradient signal on day 154 was mainly influenced by this term. Similarly, o n day 152, the bottom stress at Chester swale west was about 1.7x10 5 m/s2, which exceeded other terms and hence highly influenced the pressure gradient signal. On day 130, advective acceleration (1.1x10 5 m/s2) competed with bottom stress (1.3x10 5 m/s2) at Canaveral II shoal. However, the relatively larger bottom stress had more influence on the pressure gradient signal. Across Shelf Momentum Balance Time series of the across shelf momentum balance during the spring season is provided in Figure 3 7 Results shows that the across shelf momentum balance is mainly geostrophic (pressure gradient balanced by Coriolis). The pressure gradient and Coriolis acceleration had the same order of magnitude (x10 5), which exceeded all other terms. As Coriolis accele ration increases, pressure gradient increases according to geostrophic dynamics This is clear o n days 130, 135, 139, 142, 145, 14 8 15 2 and 155 (Figure 3 7 ) O n d ay 153, the surface stress (0.5 x10 5 m/s 2 ), the gradient of (1.1x10 5 m/s 2 ), and the gradient of (1.6x10 5 m/s 2 ) had the same order of magnitude as Coriolis acceleration (1.64 x 10 5 m/s 2 ). Therefore, these terms had some influence on the pressure gradient signal along with the Cor iolis acceleration. Similarly, on day 134, the gradient of (1x10 5 m/s2) had the same order of magnitude as

PAGE 61

61 Coriolis acceleration (1.9x10 5 m/s2). Therefore, it had some influence on the pressure gradient signal. However, Coriolis acceleration had the greatest influenc e. Since the across shelf momentum balance is clearly geostrophic, while the along shelf momentum balance is mainly frictional, this indicates that the dynamics in the region are semi geostrophic. Standard Deviation of the Momentum Terms The standard devia tion of the dynamical terms can be used to compare the size of the fluctuation of each term [Lentz and Fewings, 2012; Fewings and Lentz, 2010; Liu and Weisberg, 2005; Lee et al., 1984]. A larger standard deviation means larger variability and therefore a r elevant contributing force. A summary of the results for the depth averaged momentum balance (along and across shelf) using the standard deviation is provided in Figure 3 8 In the along shelf momentum balance, at the four locations, the standard deviati on of pressure gradient ranged between 4x10 6 m/s2 and 7x10 6 m/s2. This standard deviation of the pressure gradient was not only balanced by the standard deviation of bottom stress (3.7x10 6 m/s2 to 7x10 6 m/s2) and gradient of (reached 4.8x10 6 m/ s2). Other terms in the momentum balance such as surface stress, local acceleration, advective acceleration, Coriolis, and the gradient of had relatively lower standard deviations, but the same order of magnitude (reaching 1.9x10 6 m/s2 to 2.6 x10 6 m/s2). Therefore, their influence on the along shelf hydrodynamics is not negligible. In the across shelf momentum balance, the standard deviation of pressure gradient ranged between 1.2x10 5 m/s2 and 1.4x10 5 m/s2, which was mainly balanced by the standa rd deviation of Coriolis acceleration (1.2x10 5 m/s2 to 1.3 x10 5 m/s2).

PAGE 62

62 These results reinforce the findings obtained with along and across shelf momentum balances that the dynamics in the region are semi geostrophic during this deployment. Discussion Th e influence of cape associated shoals on the subinertial currents is evident in Figures 3 1 and 3 2 The nearshore bathymetry is forcing the flow to move primarily along shelf. Also, the along shelf wind has a clear influence on the subinertial flow by enh ancing the motion in the same direction (Figure 3 1 and 3 3 ). The Florida current seems also to be influenced by the along shelf wind (Figure 3 5 ). When both the Florida current and along shelf wind move in the same direction, it coincides with an increase in the Florida current strength and vice versa (Figure 3 9 ). As the Florida current increases, the Coriolis acceleration increases as well as it is proportional to the flow. As a result, the offshore sea surface slope increases to keep a geostrophic bala nce. The result, is a positive across shelf sea surface slope that drives the flow northward (Figure 3 9 ). The northward along shelf wind enhances this northward motion of the subinertial flow. On the other hand, as the Florida current decreases, the Corio lis influence decreases and the sea surface slope offshore relaxes to keep a geostrophic balance. At the same time, a negative nearshore sea surface slope develops to maintain this balance, which drives the nearshore flow southward. Similarly, the southwar d along shelf wind enhances this southward motion of the subinertial flow. The result, is a negative nearshore sea surface slope that drives the nearshore flow southward and a positive but more relaxed offshore sea surface slope that drives the offshore fl ow northward (Figure 3 9 ). Two exceptions to the process explained previously have been noted on days 133 and 138 (Figure 3 3 ). Comparing day 133 with 145, both had southward along shelf

PAGE 63

63 wind of ~3.5 m/s. However, the southward along shelf wind coincided with a reduction in the Florida current transport of ~0.6 Sv on day 133 and a reduction in Florida current transport of ~3 Sv on day 145. Consequently, the sea level at the coast increased to 0.01 m on day 133, while it increased to 0.04 m on day 145. Ther efore, the corresponding nearshore sea surface slope on day 133 caused a reduction in the northward subinertial flow strength, but did not change the flow direction. On the other hand, the corresponding nearshore sea surface slope on day 145 caused the sub inertial flow to move southward. On day 138, a decrease in the strength of the southward along shelf wind coincided with an increase in the Florida current transport. As a result, the sea level at the coast decreased and the corresponding nearshore sea sur face slope caused a reduction in the southward subinertial flow strength, but did not change the flow direction. The results from the across shelf momentum balance in Figures 3 7 and 3 8 ratify the conclusion derived from the wavelet coherence analysis on the dynamics being geostrophic. This result is consistent with previous studies performed in inner, mid, and outer shelves [Fewings and Lentz 2010; Liu and Weisberg, 2005; Shearman and Lentz 2003; Lentz et al., 1999; Li and Weisberg, 1999a, b; Lee et al., 1984, 1989; Brown et al., 1985, 1987; Thompson and Pugh, 1986; Nobel and Butman, 1983; Allen and Kundu, 1987]. The across shelf momentum balance was dominated by Coriolis and the pressure gradient, which is very clear on days 130, 135, 139, 142, 145, 148, 152, and 155 (Figure 3 7 ) Both terms had the same order of magnitude (x10 5 m/s 2 ), which exceeded all other terms in the momentum balance Similarly, in Figure 3 8 the standard deviation of the pressure gradient (1.2 x10 5 m/s 2 to 1.4 x10 5 m/s 2 ) and Cori olis

PAGE 64

64 acceleration (1.2 x10 5 m/s 2 to 1.3 m/s 2 ) exceeded the standard deviation of all the other terms in the momentum balance (along and across shelf). In the along shelf momentum ( Figures 3 6 and 3 8 ) pressure gradient, bottom stress and the gradient of were generally larger than other terms (ranging between x10 7 m/s2 and x10 5 m/s2), which suggests that the hydrodynamics are mainly frictional. This is consistent with vessel based data that were collected in the same area of study during summer 20 13, winter 2014, spring 2015, and summer 2016 (see chapter 2) Using a subtidal density driven model [Valle Levinson et al., 2003; Valle Levinson, 2008] influenced by friction, Coriolis, and pressure gradient, results showed that the spatial structure of s ubtidal flow in the along shelf was dominated by friction. Therefore, the maximum flow was located over the deepest parts of the cross section. Other terms in the along shelf momentum balance (i.e. surface stress, local acceleration, advective acceleration Coriolis acceleration, and the gradient of ) were occasionally of the same order of magnitude (x10 6 m/s2 to x10 5 m/s2) with relatively lower values. Therefore, they can still have contribution on the along shelf hydrodynamics. Similar pattern was noted with standard deviation results (Figure 3 8). These results suggest that the dynamics in the region are semi geostrophic. Summary The bathymetry in the inner shelf adjacent to Cape Canaveral is considered complex due to its series of shoals and shor e oblique ridges. The flow over this bathymetry mainly moves along shelf. As the along shelf wind and the Florida current move in the same direction, it coincides with an increase in the Florida current transport. As a result, a positive across shelf sea s urface slope develops and the inner shelf water

PAGE 65

65 moves in the same direction of the western boundary current. This motion suggests that an across shelf geostrophic balance is driving the subinertial flow. On the other hand, as the along shelf wind and the F lorida current are in opposite direction, it coincides with a decrease in the Florida current transport. Consequently, the positive offshore sea surface slope relaxes to keep geostrophic balance and the outer shelf flow moves in the same direction as the F lorida current. At the same time, a negative nearshore sea surface slope develops to maintain the geostrophic balance and the nearshore flow moves southward. The along shelf wind in both cases enhances the subinertial flow motion (northward or southward). The across shelf momentum balance shows that the subinertial flow is dominated by the pressure gradient and Coriolis (geostrophic balance). This was consistent with the result obtained from the wavelet coherence analysis. In the along shelf momentum balanc e, dominant terms were pressure gradient, bottom stress, and the gradient of but other estimated terms were occasionally on the same order of magnitude with relatively smaller values. Thus, can still be influential.

PAGE 66

66 Figure 3 1 The subinertial parameters during spring season around Cape Canaveral and False Cape associated shoals. Having all these parameters next to each other makes it easier to observe how they interact with one another. A) Subinertial sea level at the Trident Pier (magenta) IS on the left y axis, while the e stimated transport rate of the Florida current (dashed purple) is on the right y axis. B) Along shelf wind component in the direction of propagation (grey). The depth averaged subinertial currents at: C) Chester swale west, D) Chester swale east, E) Canave ral II swale west and F) Canaveral II swale east.

PAGE 67

67 Figure 3 2 Results of CHEOF during spring season The spatial variability of Mode 1 across shelf (solid line) and along shelf (dashed line) for each location is as follows: A) Chester swale west, B) Chester swale east, C) Canaveral II swale west, and D) Canaveral II swale east. E) The temporal variability of CHEOF Mode 1 (solid black). The spatial structure of CHEOF Mode 1 is positive and unidirectional throughout the deploymen t. Therefore, positive temporal variability of CHEOF Mode 1 means that the flow is traveling northward, while negative temporal variability of CHEOF Mode 1 means that the flow is traveling southward.

PAGE 68

68 Figure 3 3 Wavelet coherences results between CHEOF Mode 1 and different forcings during spring season A) Along shelf wind, B) the Florida current, and C) Sea level at Trident Pier. The red and blue contours represent high and low coherency, respectively. The bold black contour represents a 95% confidence level. The enclosed areas within the cone of influence represent significant coherence, while shaded areas outside the cone of influence are less reliable. Figure 3 4 Wavelet cohere nce results between the sea level at Trident Pier and the Florida current during spring season

PAGE 69

69 Figure 3 5 Wavelet coherence results be tween the along shelf wind and the Florida current during spring season Figure 3 6 Time series of the along shelf momentum balance during spring season The green line represents the results for Chester swale west; the blue, Chester swale east; the cyan blue, Canaveral II swale west; and the red, Canaveral II swale east. A) Local acceleration: B) Advective acceleration: C) Coriolis acceleration: D) Stokes Coriolis: E) Sum of left hand side, F) Surface stress: G) Bottom stress: H) Gradient of the radiation stress of x component in the y direction: I) Gradient of the radiation stress of y component in the y direction: J) Bottom stress due to waves: K) Pressure gradient: and L) Sum of the right hand side.

PAGE 70

70 Figure 3 7 Time series of the across shelf momentum balance during spring season The green line represents the results for Chester swale west; the blue, Chester swale east; the cyan blue, Canaveral II swale west; and the red, Canaveral II swale east. A) Local acceleration: B) Advective acceleration: C) Coriolis acceleration: D) Stokes Coriolis: E) Sum of left hand side, F) Surf ace stress: G) Bottom stress: H) Gradient of the radiation stress of x component in the y direction: I) Gradient of the radiation stress of x component in the x direction: J) Bottom stress due to waves: K) Pressure gradient: and L) Sum of the right hand side.

PAGE 71

71 Figure 3 8 The standard deviation for each term of the momentum balance during spring season The gree n dot represents the results for Chester swale west; the blue, Chester swale east; the cyan blue, Canaveral II swale west; and the red, Canaveral II swale east. stands for Local Acceleration, stands for Advective Acceleration, stands for Corio lis, stands for Stokes Coriolis, stands for Surface Stress, stands for Bottom Stress, stands for Gradient of Radiation Stress stands for Gradient of Radiation Stress stands for Gradient of Radi ation Stress and stands for Bottom Stress Due to Waves.

PAGE 72

72 Figure 3 9 A diagram of the subinertial circulation in the inner shelf adjacent to Cape Canaveral, FL. stands for along shelf wind, Stands for along shelf current, stands for the change in sea surface elevation, the cross means that the direction of propagation is northward, and the dot means that the direction of propagation is southward. The diagram is not to scale, but for ill ustration purposes.

PAGE 73

73 CHAPTER 4 THE DEVELOPMENT OF A SCRAP TIRE BARRIER AS A COASTAL STRUCTURE FOR WAVE DAMPING APPLICATIONS Background The goal of any wave barrier is to intercept the incoming incident wave, transmitting and reflecting part of the incident wave energy while dissipating the remaining energy, so that the sum of all these three energies is equal to the incident wave energy. The research element in a wave barrier is how to reduce the transmission and increase the reflection and dissipation by i nvesting a minimum cost for the construction of the barrier. Using scrap tires as wave barriers has some advantages such as the ease of fabrication and installation. Scrap tires can be assembled into smaller segments and then attached together to form the required configuration. The raw material is available at no cost or at competitive prices compared with other materials. Also, this raw material is easy to mobilize, is independent of sea bed soil conditions, causes little or no damage to coral reefs, and provides better circulation of sea water than bottom resting structures on the lee side after its field installation. One potential drawback of tires is the environmental impact of leaching. From the available literature [ Col lins et al. 2002 and Evans 1997 ] it is found that the leaching rate from scrap tires is acceptable and decreases with time. Before identifying the main objectives of this study, it is important to review the existing literature on scrap tires for marine applications. Noble (1969 and 19 76) proposed the Wave Maze floating breakwater, which consists of used truck tires filled with floating material bolted to a center sandwich of vertical tires arranged in a triangular pattern. The width of the breakwater has to be approximately the length of the wave to effect attenuation. Candle and Fischer (1975) proposed the use of scrap tires to form a

PAGE 74

74 large floating breakwater for shore protection from erosion. The idea consists of assembling smaller segments and then connecting them with high strengt h cable. The tires are stacked flat but vertically, and the four corner tires are rotated 100 ( Figure 4 1 ). Floating materials were placed at each tire to keep the structure floating. McGregor (1978) constructed a floating breakwater using scrap tire particularly for the application of fish farm protection. Models were tested in a hydrodynamics laboratory tank, where the wave attenuation characteristics were obtained using computerized data collection apply to the environmental requirements of fish farms. Hibarger et al. (1979) secured a US patent for developing an interlocking array of tires to form a floating breakwater completely made of tire materials ( Figure 4 2 ). Cables were only used to keep the structure in position. Harms and Westerink (1980) investigated the wave transmission and mooring force characteristics of a pipe tire floating breakwater using regular waves ( Figure 4 3 ). The results were compared with earlier experiments on the Goodyear f loating breakwater. A buoyancy test was performed to determine the flotation requirement. Day et al. (1993) performed a laboratory study to determine the toxicity of the leachate from tires. Results showed that for short term installation (a few days to a month), the leachate remained relatively toxic for one species (rainbow trout). However, for tires with longer term installation (10 years) there was no release of chemicals toxic to any tested species. In 2000, Walter secured a US patent for developing an artificial reef made of a concrete frame, which consisted of six concrete beams inserted through a number of tires ( Figure 4 4 ). Gu (2005) discussed many aspects related to the reuse of scrap tires such as identifying the problems associated with scrap ti res, summarizing the physical and chemical

PAGE 75

75 properties of tires, and discussing the hydraulic engineering applications of scrap tires. significant effects on the growth of phyt paper also introduced studies on the impacts of leachates on the surface and groundwater. These studies concluded that the leachate from scrap tires had little or no effect on either the surface or groundwater. In 2009, Cederlund secured a US patent on a wave attenuation system consisting of a floating member above the waterline, interlocking tires below the waterline, and anchors to keep the system in position ( Figure 4 5 ). From the literature review, it is cle ar that researchers have tried to understand the wave transmission characteristics of certain scrap tire configurations. There is a range of possibilities to investigate regarding new types of configurations and the associated wave transmission characteris tics. One of the most important aspects is to search for a configuration with better wave t ransmission characteristics and a minimum number of scrap tires. The present investigation is focused on this particular point. It is also important to consider the ease of construction, installation, transport, and maintenance while selecting the optimal configuration. To achieve this result, the research objectives are: To assess the wave transmission, reflection, and energy dissipation characteristics of different innovative configurations of scrap tire barriers. To investigate regular waves to increase the fundamental understanding of wave interaction with scrap tires for different configurations. To carry out the research with random waves to better predict the fi eld performance and optimal design of using scrap tires as a floating breakwater.

PAGE 76

76 To obtain predictive relations of the coefficient of transmission for the different scrap tire configurations covering a wide range of relative water depth and wave height in random wave fields. To carry out a design case study for erosion protection of beaches in Kennedy Space Centre, Florida, USA and in Qaru Island in Kuwait for the prevailing environmental conditions. Methodology Theoretical modeling of wave interaction wit h different scrap tire configurations is challenging because the problem involves turbulence and wave energy dissipation. Hence, physical modeling is an adequate alternative to solve this problem. Physical model tests are performed in the concrete wave flu me of 2.5 m wide, 2.0 m deep and 40 m long, fixed with a bottom hinged flap type wave maker at the Kuwait Institute for Scientific Research (KISR; Figure 4 23 to 3 27 ), which has an active wave absorption capability for reflected waves. Regular and random waves of a wide range of wave results are used for field applications and design purposes. Incident, reflected, and transmitted wave heights are measured using wave probes (WPs) located before and after the scrap tire floating structure. The data collected using WP1 are used to estimate incident wave heights; data collected using WP5 are used t o estimate transmitted wave heights; and data collected using WP2, 3, and 4 are used to estimate the reflected wave heights. Three wave probes are recommended for reflection analysis to avoid singularity problems that may occur when using two wave probes o nly. The distances between all wave probes and the wavemaker are provided in table 6. The normalized inputs are: relative incident wave ( / ), relative water depth ( / ), relative wave barrier width ( / ), and relative draft ( / ), where stands for incident wave height, stands for wave

PAGE 77

77 length, stands for water depth, stands for the width of the scrap tire wave barrier, and stands for the thickness of the tire assembly. The normalized outputs are the coefficient of transmission ( = / ), coefficient of reflection ( = / ), and coefficient of dissipation ( = / ie. = [1 ] ). stands for transmitted wave height; stands for reflected wave height; and stands for di ssipated wave height. is estimated using the law of conservation of wave energy. According to this law, the incident wave energy must be equal to the sum of the transmitted wave energy, reflected wave energy, and wave energy loss. The Danish Hydraulic wave synthesizer analytical software is used for wave transmission and reflection analysis. The software has a reflection analysis package that provides the coefficient of reflection ( ). For regular wave tests, the estimated inciden t ( ) and transmitted wave heights ( ) are obtained using the time domain analysis. Although the output data is collected for 60 s, only a 20 s window of repeating data is used for the analysis. These 20 s have a clear repetition of the wave cycle wi thout any reflection effect from the flume beach. Also, the estimated incident ( ) and transmitted waves ( ) are the average of the 20 s duration of the selected data window. For random wave tests, the estimated incident significant wave ( ) and transmitted wave heights ( ) is obtained using the frequency domain analysis. The output data are collected for 5 minutes (300 s) with a 40 Hz data acquisition rate. The estimated significant incident wave height is calculated as = 4.0 In this equation, stands for the zero th moment of the incident wave spectrum, which is estimated as Here, is the spectral density value of the incident waves, and is the frequency bin.

PAGE 78

78 A total of 9 different scrap tire config urations is used for the physical model investigation. For each configuration, 26 regular wave runs and 9 random wave runs are used with different combinations of wave heights and wave periods. A total of 315 runs are conducted for the 9 different wave bar rier models. All experiments are performed with a water depth of 122 cm. Table 4 parameters, while Table 4 eters For regular wave experiments, the wavemaker of the concrete wave flume cannot generate the flap type wavemaker. Similarly, for random wave experiments, the wavemaker cannot generate incident waves larger than 10 cm for the selected peak wave periods Table 4 3 provides details of the various scrap tire configurations. Table 4 4 provides det ails of the tire m easurements; Table 4 5 provides details of the different materials used fo fabrication; and Table 4 the wavemaker. Figures 4 6 to 4 14 provide sketches of the 9 different scrap t ire configurations, while Figure 4 1 5 to 3 22 show pictures of 8 out of the 9 configurations placed inside the concrete wave flume at KISR. Figures 4 2 8 to 4 3 4 show pictures of Model 1 to 4 needs th e same number of scrap tire units. Model 1 will be the cheapest, since there is no need for strong mooring arrangement, but needs additional buoy to keep it floating. Model 2 and 3 need to be moored at the seabed at one end, hence needs spending for moorin g arrangements, but needs only 50% of buoys to keep the other end floating. Model 4 needs mooring on both sides and also needs more

PAGE 79

79 buoys to keep the other parts floating. Hence, overall, model 1 will be the easiest to install and cheapest and 4 will be mo st difficult to install and costliest. Model 5 to 9 needs double the number of scrap tires, required for model 1 to 4. If similar analysis is done, then model 6 and 7 will be the easiest to install and cheapest and model 5, 8 and 9 needs additional investm ent for mooring one end on the sea bed. This information needs to be considered while discussing the wave transmission, reflection, and dissipation of individual models and for final selection of a suitable model for field conditions. The specific weight o f the scrap tire is slightly more than the specific weight of fresh water. Hence, the configuration needs buoys to be attached to make sure the tires are floating during all the experimental investigations. It is difficult to keep the scrap tires floating horizontally by using only a few units of buoys. Hence, when the scrap tire configuration is in the water in floating mode, the cross sectional shape of the units is measured and is provided in Figures 4 6 to 4 14 Results and Discussion First, the results of the study with regular waves are described. Then the results with random waves are explained Wave T ransmission, R eflection and Energy Dissipation Using R egular W aves The effect of wave period is studied by using the normalized parameter, d/ and the wave height by / The effect of relative water depth ( / ) on the coefficient of transmission, reflection, and energy dissipation using regular waves is shown in Figures 4 35 to 4 4 3 for the 9 different scrap tire configurations. The res ults are for small, medium, and high wave energy climates with a relative incident wave height ( / ) of 0.041, 0.082, and 0.12. The range of / is between 0.1 and 0.5. This means the result

PAGE 80

80 covers a wide range of intermediate water depth conditions. For all configurations, the wavemaker of the concrete wave flume could not generate regular waves of / and / important to remember that while performing regular w ave experiments, the incident under the influence of a particular wave height and perio d. In reality, waves are random and only the results with random waves can be used for field applications. The results for the 9 different scrap tire configurations using regular waves can be summarized in the following points: ons, as / increases, the coefficient of transmission becomes smaller. This means that the relatively shorter wave lengths are effectively interacting with the scrap tire wave barrier with a wave transmission coefficient of 0. 1 to 0.8 and a wave dissipa tion coefficient of 0.6 to 0.97 Therefore, they had less wave transmission and more energy dissipation. On the other hand, the relatively longer wave lengths are passing through the model with less interaction. Therefore, they had a relatively higher wave transmission of 0. 6 to 1.0 and less energy dissipation of 0. 1 to 0. 7 Model 5 (8 rows of scrap tire with a fixed front at the flume bed) and model 6 (8 rows of completely slack scrap tires), achieved the least coefficient of transmission among all configu rations. Increasing the number of rows from 4 to 8 helped to increase the structure interaction with the incident waves and helped to allow fewer waves to transmit to the other side of the structure. This is because when the number of rows of scrap tires i s higher, then each row contributes to the wave dissipation and finally the transmission is expected to be smaller. Models 1, 5, and 6 seemed to perform better with larger / values (0.082 and 0.012). Models 1 and 6 consisted of a completely slack sin gle layer of scrap tires. This condition allowed the structure to better adjust to the size of the incident waves and therefore perform more effectively with the relatively larger waves and with less wave transmission. Although model 5 is fixed at the fron t, its relatively longer (8 rows of scrap tires) and completely slack end allowed for better adjustment to the size of the incident waves as in models 1 and 6. Therefore, it had a better interaction with larger waves and had less wave transmission.

PAGE 81

81 Model 4 (4 rows with fixed front and back) and model 8 (4 rows of double layer with fixed front) had better interaction with shorter waves and smaller / values. Model 4 had a relatively high stiffness as a result of being fixed on both ends and hence did no t provide the structure with enough flexibility to adjust itself to the different /d values. Therefore, it had better interaction with smaller / values. On the other hand, model 8 had a relatively high mass due to the double layers of scrap tires fixed front, and relatively short completely slack end of 4 rows. Therefore, it was not able to adjust itself to the different / values and therefore performed better with smaller waves. Also, the fixed front facing the incoming waves with a relati vely shorter end (4 rows) allowed larger waves to push the structure downward as it was passing and therefore overtopped the structure. layer of scrap tires to model 1, forming model 7, helped to increase reflection and therefore reduce wave transmission. The completely slack structure showed a better ability to reflect waves than other configurations. When the structure is completely slack, less tension is applied on th e structure. Also, the structure has less stiffness, allowing each tire to interact individually with the incident waves. This individual movement for each tire helps to increase the efficiency of the structure in blocking and reflecting waves. For the sam e reasons mentioned in point 3, models 1, 5, and 6 seemed to dissipate more wave energy with larger / values. Also, models 4 and 8 seem to dissipate more wave energy for smaller / for the same reasons mentioned in point 4. In model 1, the coeff icient of dissipation could not be calculated for smaller /L values (longer wave lengths) due to the resonance condition. This model is completely slack with a relatively small mass (a single layer of 4 rows of scrap tires). Therefore, at longer wave lengths the structure began moving with the incoming wave witho ut any loss in wave energy. To avoid this condition, the natural frequency of the wave barrier has to be increased or decreased. Increasing the wave barrier stiffness by increasing the mooring force (fixing one or both ends of the structure) or increasing the wave barrier mass by increasing the number of rows or layers, can help to achieve that. Adding a second layer of scrap tires to model 1 to form model 7 helped to delay the occurrence of resonance to longer wave lengths. Adding a third layer will increa se the structure mass even more and therefore shift the natural period of oscillation. Wave Transmission, Reflection, and Energy Dissipation Using Random Waves The effect of relative water depth ( / ; where stands for wave length corresponding t o peak wave period) on the coefficient of transmission, reflection, and

PAGE 82

82 energy dissipation using random waves is shown in Figures 4 44 to 4 52 for the 9 different scrap tire configurations. The results are for small and medium/high wave climates with a rel ative incident wave height ( / ) of 0.041 and 0.082. The range of / is between 0.1 and 0.5. For all configurations, the wavemaker of the concrete wave flume could not generate random waves of / of the flap type wavemakers. Also, the wavemaker could not generate random waves of / < 0.2 and / evaluate and select the optimal design of scrap tire configurations. The random waves a re generated using the Pierson Moskowitz spectrum. The results from the 9 different scrap tire configurations using random waves can be summarized in the following points: Similar to the results obtained with regular waves, the relatively shorter wave leng ths were effectively interacting with the scrap tire wave barrier and hence a smaller wave transmission coefficient such as 0.0 2 to 0. 21 is achieved. Therefore, the shorter wave lengths had less wave transmission and more energy dissipation. On the other h and, the relatively longer wave lengths were passing through the model with less interaction. Therefore, they had a relatively higher wave transmission of 0.1 to 0.35 and less energy dissipation. / val ues increased the coefficient of transmission, meaning that larger incident waves would transmit more to the other side of the structure than smaller incident waves. Model 5 (a single layer of 8 rows of scrap tire with a fixed front at the flume bed), mode l 6 (a single layer of 8 rows of completely slack scrap tires), and model 7 (a double layer of 4 rows of completely slack scrap tires) achieved the least coefficient of transmission among all configurations. Increasing the number of rows or layers helped t o increase the structure interaction with the incident waves and helped to allow lesser waves to transmit to the other side of the structure. Model 5 had a slightly smaller coefficient of transmission than models 6 and 7 due to fixing the front of the stru cture, in addition to increasing the number of rows. It is found that the coefficient of reflection consistently increases with increase in /

PAGE 83

83 Models 5 and 7 had the largest coefficient of reflection among all configurations / (0.082). Model 7 had also the largest coefficient of / (0.041). Therefore, increasing the number of layers is effective in reflecting both larger and smaller incident waves, while increasing the number of rows in addition to fixing one end proves to be more effective with larger waves. No resonance condition has been noted with any model configuration with random wave runs. The reason is that, as in reality, the wave frequency of each wave changes. Therefore, the f requency of the incident wave that equals the natural frequency of the structure will last for a few seconds only. Therefore, if resonance happens, it will not last for long, and therefore will not be noted. Comparison of Wave Transmission Characteristics for Different Scrap Tire Configurations Due to Random Waves When it comes to choosing the best floating scrap tire configuration for a particular field condition, three main factors must be considered: the value of the coefficient of transmission, the rela tive depth ( / ) and the relative wave height ( / ). The other normalized outputs ( and ) are required for understanding the general performance of different models. However, the coefficient of transmission ( ) is the main factor in evaluating the hydrodynamic performance of different configurations when it comes to field application (especially for beach protection from erosion). The reason is that this coefficient will estimate the size of the waves that will actually reach the coast and whether that wave will still cause beach erosion. The relative depth consists of the depth at which the model configuration will be placed and the most dominant wave length in the study area. As mentioned earlier, for all conf igurations, the wavemaker for the concrete wave flume could not generate random waves of / 0.12, since it is beyond the capacity of the flap type wavemakers. Also, the wavemaker could not generate random waves of / < 0.2 or /

PAGE 84

84 The effect of nine different floating scrap tire models on wave transmission for five different / ( / =0.1, 0.2, 0.3, 0.4 and 0.5) and for / =0. 0 41 is provided in Figure 4 53 The following are the main findings: Within the parameters studied, the range of values is from 0.0 2 to 0.2, which are encouragingly smaller values for the field applications. If the beach soil is fine sand, then it is better to select a model, which allows much smaller value (say, less th an 0.1). If the beach soil is coarse sand, then the designer can select the model that offers larger value (say closer to 0.2). Among the models selected, the range of value for model 4 (moored on both ends at the seabed) is smaller (0.09 to 0.14). It means, th is model is not sensitive for variation of wave period. Among the models selected, the range of value for model 5 and 6 are wider (0.02 to 0.15), meaning that these models are sensitive for changing wave periods. If scrap tire is available in the market for cheaper cost and the design is needed for minimum value, then model 6 and 7 are good options, since it is under slack mooring condition. If scrap tire is costlier, then model 1 will become the cheapest option. For any model, value is the smallest for the highest / value. Hence one can select deeper water for placing the floating breakwater. However, it will result / value used for the study ( Tabl e 4 3 ). The length of the breakwater also will become higher. Hence selecting / value as small as possible, is desirable. A similar plot for / = 0.082 is provided in Figure 4 5 4 The range of values is from 0.0 36 to 0.35, which is wider, when compared to Figure 4 53 for low energy wave field condition. The suitable scrap tire wave barrier configuration for a typical site should be selected based on the following criteria: Overall cost of initial installation (Model 1 may be the cheapest. Model 6 and 7 also will be cheaper, if the scrap tire unit cost is low) Better performance, i.e. a minimum number of scrap tires but higher efficiency or lesser wave transmission

PAGE 85

85 Ease of assembling and installation, e.g. slack arrangement is better than mo oring arrangement Easy reorientation and less maintenance during its field performance From these points, it is advisable to avoid models that need strong mooring lines to fix one end to the sea bed. The preferred configurations are scrap tires, floating f reely with slack mooring. Therefore, the ease of construction and installation for models 6 and 7, in addition to their high performance, give them an advantage over all the other models. Multiple Regression Equations for the Prediction of the Coefficient of Transmission for Different Scrap Tire Configurations Multiple regression analysis to correlate between the dependent variable and independent variable is necessary for a better understanding and application of the present study. Relative wave height and relative water depths are the most important independent variable. Wave transmission, reflection and dissipation coefficients are the dependent variables. For field application, wave transmission is the most important parameter and hence the multiple regr ession analysis is carried out for that parameter. The Statistical Package for the Social Sciences (SPSS) software was used to predict the multiple regression equations and the coefficient of determination ( ) for the of different scrap tire configurations. The software uses the range of values of / (0.041 and 0.081) and / (0.1 to 0.5) that were used while performing the experiments to calculate these equations. If / and / values in the field lie between or close to the range of values used while performing the experiments, then the certainty of predicting increases. As / and / values in the field get farther from the range of values used while performing the experiments, th e certainty of

PAGE 86

86 predicting decreases. Table 4 7 shows a summary of all multiple regression equations to predict for the different scrap tire configurations. In addition to monitoring values to predict the accuracy of the multiple regression eq uations, both predicted and measured values are plotted against one another to get a better visualization of experimental results and predicted values. The closer the values to a 45 slope, the more accurate the prediction of Figures 4 55 to 4 6 3 show a comparison between the measured and predicted values of The results show that the multiple regression equations are reliable for predicting values for the different scrap tire configurations, which is proved by having a higher value (> 0.80). Field Application Scrap tires as a wave barrier can be used for numerous applications such as a floating breakwater for a marina, coastal erosion protection for mainland and islands, and protecting an offshore oil & gas loading/unloading termina ls from direct wave attack. They can also be used for facilitating open sea construction activity for wider durations during a year by reducing wave activity and as artificial reef to attract fish community to populate around the scrap tire. In this study, the proposed scrap tire as a Space Center (KSC), located on a Florida cape, and Qaru Island in Kuwait. Design Aspects of the Scrap Tire Configuration for Kennedy Space Cente r (KSC) at Cape Canaveral, Florida According to Florida D epartment of Environmental Protection (2016), main launch center of human spaceflight, located on Cape Canaveral, Florida is facing serious erosion along ~8 km of the shoreline. Erosion is increasing the risk of damaging the infrastructure. To determine the optimal scrap tire configuration to m itigate the

PAGE 87

87 erosion at KSC, a full understanding of the oceanographic aspects in the region have been carried out in Chapter 2 and Chapter 3. Looking at the seasonal wave direction in Figure 4 66 the medium and large size waves of the fall, winter, and sp ring seasons arrive at the beach from the northeast. In this direction, the bathymetry is relatively smoother and deeper ( Figure 4 64 ). This allows waves to travel faster and act on the beach with higher energy than other locations which are naturally prot ected with series of cape associated shoals and shore oblique ridges. Therefore, more beach erosion is expected during these seasons. On the other hand, the small waves of the summer season with relatively low energy arrive at the beach dominantly from the east. These waves encounter more bathymetric obstacles before hitting the beach of KSC. Plus, its small size and relatively lower energy will not cause any major erosion to the beach. The seasonal wave heights and periods, which are provided in Figure 4 6 5 can be reorganized as in Figure 4 67 and 3 68 to determine their total and percentage of occurrences. This way of presenting the wave data will help in determining the most dominant wave conditions, longest and most frequent wave period, and the largest wave height. These three cases are important in the design and selection of the optimal scrap tire configuration. Usually, the design is made to protect the beach from the most dominant wave conditions. The cost of the design increases if it is intended t o be used to protect the beach from extreme conditions (long wave periods or large wave heights ) Case 1: The Most Dominant Wave Condition Referring to Figure 4 67, which has a total number of seasonally collected samples equal to 1756, the most dominant c ondition has a peak wave period ( ) of 9 s,

PAGE 88

88 significant wave height ( ) of 0.375 m, and a percent of occurrence equal to 7.6% (Figure 4 68). It is also important to mention that the percentage of occurrences that are dominant condition and produce a total percent of occurrence equal to ~64% (including the percentage of the most dominant wave condition of 7.6%). Locating the proposed design at a water depth of 4 m, the field becomes 0.094 and equals 0. 033, where Using volume 2 of the shore protection manual (1984), for of 0.033, the corresponding value of is 0.0759 and turns to be 52.7 m. Using the empirical equation for Model 7 (equation 4 7), presented in Table 4 7, and are calculated to be 0.221 and 0.083 m (8.3 cm) respectively. The max. orbital velocity of the water particle of this wave at the sea bed is calculated as = 6.23 cm/s; where g stands for gravity acceleration and stands for wave number. Using the plot of minimum velocities for sediment erosion and deposition, which was developed by Vincent in 1975 [Herbich, 1981]: erosion starts to happen at of 15 cm/s. T his value is associated with the median sediments size ( environmental assessment for KSC shoreline protection project (2015). The calculated of 6.23 cm/s is less than 15 cm/s, thus no erosion is expected to happen. The scale ( ) is calculated to be ~3.3 and the design dimensions will be as follows: Width of the structure in the field ( ) = = 2.23 m x 3.3 = 7.4 m Depth of tires (double layers; model 7) in the field ( ) = = 0.382 m x 3.3 = 1.26 m

PAGE 89

89 Length of the structure in the field ( ) = = 2.4 m x 3. 3 = ~8 m. However, selection of length is based on the budget available and length of coastal protection from erosion. Case 2: Longest and Most Frequent Wave Period Referring to Figure 4 67, the longest and most frequent wave period has a of 15 s with an of 1.25 m, and an occurrence equal to 1.3% (Figure 4 68). Locating the proposed design at a water depth of 6 m, the field becomes 0.28 and equals 0.017. Using volume 2 of the shore protection manual (1984), for of 0.017, the corresponding value of is 0.0529 and turns to be 113.3 m. Using the empirical equation for model 7 (equation 4 7), and are calculated to be 0.468 and 0.527 m (52.7 cm) respectively. The calculated of 32.4 cm/s is larger than 15 cm/s. Therefore, erosion is expected to happen. The scale is calculated to be ~5 and the design dimensions will be as follows: Width of the structure in the field ( ) = = 2.23 m x 5 = 11 m Depth of tires (double layers; model 7) in the field ( ) = = 0.382 m x 5 = 1.87 m Case 3: Largest Wave Height Condition Referring to Figure 4 67, the largest wave height has a of 8 s with an of 2.375 m, and a perce nt of occurrence equal to 0.2% (Figure 4 68). Locating the proposed design at a water depth of 6 m, the field becomes 0.40 and equals 0.0625. Using volume 2 of the shore protection manual (1984), for of 0.0625, the corresponding valu e of is 0.1072 and turns to be 56 m. Using the empirical equation for model 7 (equation 4 7), and are calculated to be 0.598 and 1.42 m

PAGE 90

90 (142 cm) respectively. The calculated of 80.56 cm/s is larger than 15 cm/s. T herefore, erosion is expected to happen. Similar to case 2, the scale is calculated to be ~5 and the design dimensions will be as follows: Width of the structure in the field ( ) = 11 m Depth of tires (double layers; model 7) in the field ( ) = 1.87 m Time S eries of Wave Heights and Periods with Design Recommendations Referring to the time series of significant wave heights and periods provided in Figure 4 65 the max. orbital velocity at the bottom was calculated accordingly ( Figure 4 69 ). This way we can predict the time at which erosion will take place and therefore adjust the design of the structure accordingly. According to the results obtained in Figure 4 69 the following design modifications are recommended: If the design propos ed for model 7 in cases 2 and 3 (extreme conditions) is used ( Figure 4 70 ), KSC beach will be protected from erosion as long as the max. cases 2 and 3 by 2 will protect the beach from erosion as long as the max. orbital 2 and 3 by 3 will protect the beach from erosion as long as the max. orbital velocity at the he structure size will increase cost. However, it will increase the beach protection as well. It will be up to NASA to decide which design they would like to go with according to their budget and need. A combination between beach nourishment and an increa se in the structural cross section can be very helpful. Using the plot of minimum velocities for sediment erosion and deposition, which wa s developed by Vincent in 1975 [Herbich, 1981] for median sediments size ( ) of 2 mm, the erosion will start to hap pen when = 30 cm/s. Using the dimensions proposed in cases 2 and 3 will protect the beach from erosion as long as the max. orbital velocity at If there is any environmental concern associated will beach nourish ment, since KSC beach is a m ajor nesting beach for turtle hatchlings in Florida with more than 5000 turtles nest [Herridge, 2014]. T hen option 1 can be used instead

PAGE 91

91 The design appearance of the scrap tire barrier for coastal protection on KSC beach is sh own in Figure 4 70 Using the empirical equation of for model 7 (equation 4 7) and the time series of incident sign ificant wave heights ( ) provided in Figure 4 65 the corresponding values were calculated Then the cumulative probability was calculated for each and as shown in Figure 4 7 1 Results show the following: 50% of the time, the incident wave height does not exceed 0.82 m while the transmitted wave height using model 7 does not exceed 0.18 m 70% of the time, the inci dent wave height does not exceed 1.05 m while the transmitted wave height using model 7 does not exceed 0.30 m 90% of the time, the incident wave height does not exceed 1.45 m while the transmitted wave height using model 7 does not exceed 0.57 m These findings prove the efficiency of the proposed configuration for KSC beach (model 7) in reducing the transmitted wave height and thus provide better protection from erosion. Design Aspects of the Scrap Tire Configuration for Qaru Island, Kuwait Qaru Island is among the nine islands that belong to Kuwait ( Figure 4 7 2 ). It is the furthest island offshore and smallest among all nine islands with a length of 275 m and a wi dth of 175 m (an approximate area of 0.035 km 2) It has historical significance to all Kuwaiti people since it was the first Kuwaiti land liberated from Iraq during the Gulf War. Also, the island has a rich biological life and is famous for its coral reefs, which make it an important recreational attra ction for locals especially during warm seasons. Due to its relatively small size and offshore location from the mainland, the island is subject to wave actions and therefore erosion. To determine the optimal scrap tire configuration to mitigate erosion at Qaru Island, the oceanographic aspects of the

PAGE 92

92 region must be investigated. For this section, the outputs of a model developed by the Kuwait Institute for Scientific Research will be used to determine the wave climate in the region. The model outputs are p rovided in Figures 4 7 3 and 4 7 4 to determine the total number of occurrences and percent of occurrences, respectively. The most dominant wave conditions, longest and most frequent wave period, and the largest wave height can therefore be determined. As me ntioned earlier, the configuration is usually designed to protect the beach from the most dominant wave conditions. The cost of the design will increase accordingly if the structure is intended to protect the beach from extreme conditions (long wave period s or large wave heights). Case 1: The Most Dominant Wave Condition Referring to Figure 4 73, which has a total number of 105178 samples collected via a model developed by the Kuwait Institute for Scientific Research, the most dominant condition has a peak wave period ( ) of 2.5 s, a significant wave height ( ) of 0.125 m, and a percent of occurrence equal to 25.6% (Figure 4 74). It is also around the most dominant cond ition and produce a total percent of occurrence equal to ~74% (including the percentage of the most dominant wave condition of 25.6%). Locating the proposed design at a water depth of 3 m, the field becomes 0.041 and equals 0.3. Using volum e 2 of the shore protection manual (1984), for of 0.3, the corresponding value of is 0.33 and turns to be 9.1 m. Since and values are available in the figures produced by the experimental results, the value can be obt ained without using the empirical equations in Table 4 7. Therefore, at of 0.3 and

PAGE 93

93 of 0.041, is 0.064 for model 6 (Figure 4 49) and becomes 0.008 m (0.8 cm). The max. orbital velocity of the water particle of this wave at the sea bed is calculated as 0.267 cm/s. Using the plot of minimum velocities for sediment erosion and deposition, which was developed by Vincent in 1975 [Herbich, 1981]: erosion starts to happen at of 20 cm/s. This value is associated with th e median sediments size ( ) of 1.4 mm in Qaru Island [Neelamani et al., 2009]. The calculated of 0.267 cm/s is less than 20 cm/s, thus no erosion is expected to happen. The scale is calculated to be ~2.5 and the design dimensions wi ll be as follows: Width of the structure in the field ( ) = = 4.46 m x 2.5 = 11.15 m ~ 11 m Depth of tires (single layer; model 6) in the field ( ) = = 0.191 m x 2.5 = 0.4775 m = 47.75 cm ~ 48 cm Length of the structure in the field ( ) = = 2.4 m x 2.5 = 6 m. This can be selected as needed to protect the whole island. Case 2: Longest and Most Frequent Wave Period Referring to Figure 4 73, the longest and most frequent wave period has a of 6 s with an of 1.625 m, and a percent of occurrence equal to 0. 03 % ( Figure 4 74). Locating the proposed design at a water depth of 6 m, the field becomes 0.272 and equals 0.095. Using volume 2 of the shore protection manual (1984), for of 0.095, the corresponding value of is 0.137 and turns to be 43.9 m. Using the empirical equation for model 7 (equation 4 7), and are calculated to be 0.43 and 0.70 m (70 cm) respectively. The calculated of 36.6 cm/s is larger than 20 cm/s. Therefore, erosion is expected to happen. The scale is calculated to be ~5 and the design dimensions will be as follows:

PAGE 94

94 Width of the structure in the field ( ) = 11 m Depth of tires (double layers; model 7) in the field ( ) = 1.87 m Case 3: Largest Wave Height Condition Referring to Figure 4 73, the largest wave height has a of 6 s with an of 3.375 m, and a percent of occurrence equal to 0.003 % (Figure 4 74). Locating the proposed design at a water depth of 6 m, the field becomes 0.56 and equals 0.0945. Using volume 2 of the shore protection manual (1984), for of 0.0945, the correspond ing value of is 0.1364 and turns to be ~ 44 m. Using the empirical equation for model 7 (equation 4 7), and are calculated to be 0.803 and 2.71 m (271 cm) respectively. The calculated of 141.3 cm/s is larger than 20 cm/s. Therefore, erosion is expected to happen. Similar to case 2; the scale is calculated to be ~5 and the design dimensions will be as follows: Width of the structure in the field ( ) = 11 m Depth of tires (double layers; model 7) in the fi eld ( ) = 1.87 m Design Recommendations for Qaru Island Referring to the percentage of occurrences provided in Figure 4 7 4 the most dominant wave condition has a percentage of occurrence equal to 25.6%. The conditions surrounding the most dominant condition have a total percent of occurrence equal to ~74% (including the percentage of the most dominant wave condition of 25.6%). Therefore, most of the year the wave climate is mild/moderate and can be protected with the dimensions proposed in case 1 ( Figure 4 7 5 ). The extreme conditions (longest wave length and largest wave height) have a very rare percentage of

PAGE 95

95 occurrence (0.03% and 0.003%, respectively). Therefore, there is no need to make any changes to the design proposed in case 1 for mode l 6. The productive performance of the structure throughout the year will accommodate the erosion due to very rare conditions. Summary In this study, the wave transmission, reflection, and dissipation of nine new configurations of scrap tire wave barriers has been investigated. The goal was focused on achieving the minimum wave transmission characteristics with a minimum number of scrap tires in the configuration. The ease of construction, installation, transport, and maintenance was also considered while s electing the optimal configuration. Regular waves were used to increase the fundamental understanding of wave throughout the duration of the experiment to better configurations, the relatively shorter wave lengths were effectively interacting with the scrap tire wave barrier with a wave transmission co efficient of 0. 1 to 0.8. Therefore, the shorter wave lengths had less wave transmission and more energy dissipation. On the other hand, the relatively longer wave lengths were passing through the model with less interaction. Therefore, they had a relativel y higher wave transmission of 0. 6 to 1.0 and less energy dissipation. Increasing the number of rows from 4 to 8 (i.e. models 5 and 6) helped to reduce wave transmission significantly. Each row contributed to wave dissipation and finally the transmission wa s smaller. Completely slack configurations with a single layer (i.e. models 1 and 5) were lighter, more flexible, and had better adjustment to the size of the incident waves. Therefore, they performed more effectively

PAGE 96

96 with relatively larger wave heights by allowing less wave transmission and more energy dissipation. On the other hand, completely fixed configurations (i.e. model 4) or a fixed configuration at one end with double layers (i.e. model 8), had less ability to adjust to the size of the incident wa ves. Therefore, they performed more effectively with relatively smaller wave heights. Increasing the number of layers (i.e. model 7) helped to increase the efficiency of the structure in blocking and reflecting waves. Hence, the coefficient of the reflecti configurations (i.e. models 1 and 7). This condition could be avoided by changing the natural frequency of the structure by increasing the structure stiffness (i.e. increasing the moo ring force) or mass (i.e. increasing the number of rows or layers). Random waves were used to better predict the field performance and optimal design of scrap tire as a floating breakwater. Similar to the results obtained with regular waves, for all model atively shorter wave lengths effectively interacted with the scrap tire wave barrier and hence a smaller wave transmission coefficient (0.0 2 to 0.2 1 ) was achieved. Therefore, the shorter wave lengths had less wave transmission and more energy dissipation. On the other hand, the re latively longer wave lengths passed through the model with less interaction. Therefore, they had relatively higher wave transmissions (0.1 to 0.35) and less energy dissipation. Increasing the number of row s or layers (i.e. models 6 or 7) helped to increase the structure interaction with the incident waves and therefore allow fewer waves to transmit configurations, larger i ncident wave heights had more transmission to the other side of the structure than smaller incident wave heights. Hence, floating scrap tire is expected

PAGE 97

97 to perform well for locations dominated by higher relative water depth and lower wave energy. Similar t o regular waves, increasing the number of layers (i.e. model 7) is effective in reflecting both larger and smaller incident wave heights. It is also found that the coefficient of reflection consistently increases with increase in / No resonance cond ition has been noted with any model configuration. As in reality, the frequency of the incident wave that equals the natural frequency of the structure last for a few seconds only. Therefore, if resonance happens, it will not last for a long time, and henc e will not be noted. To add more value to the findings using random waves, multiple regression equations, which depend on relative water depth and relative wave height, have been obtained to predict the coefficient of transmission for the nine newly propo sed scrap tire configurations. Considering the wave transmission results of the nine configurations along with the ease of construction, installation, transport, and maintenance, models 6 and 7 were selected as the optimal configurations. They were also re commended to be used to protect the beaches of KSC, USA and Qaru Island, Kuwait from erosion. Future Work Due to the limitations of the flap type wavemakers in concrete wave flumes, they cannot generate regular incident waves of / / 0.100, random incident waves of / / / < 0.2. Therefore, using smaller size models of tires produced by the manufacturers (instead of real scrap tires) inside a glass wave flume will increase the study input range of the experiments. Such an approach will allow the coverage of a wider range of / and d/ values to better understand the performance of the structure under the influence of

PAGE 98

98 a wider range of marine conditions. The certainty of predicting the coefficient of transmission using the multiple regression equations will also increa se. Moreover, it will be very important to monitor the performance of the structure in the field for at least one year. In case of any adverse influence on the beach, the structure must be adjusted accordingly. Since the design is relatively simple and the structure is floating, it will be relatively easy to mobilize. During this year, any physical change to the assembly must be identified. The structure may attract marine life, which will affect the buoyancy force and may therefore cause the structure to s ink over time. While assembling the structure, the buoyancy force should accommodate this possibility. Also, different types of tying materials should be tested to determine the most appropriate one. Figure 4 1 A sketch of th e proposed Goodyear scrap tires as a floating breakwater via Candle and Fischer (1975).

PAGE 99

99 Figure 4 2 US patented design by Hibarger et al. (1979) for an interlocking array of tires to form a floating breakwater that is complete ly made of tire materials. Figure 4 3 A pipe tire floating breakwater that was investigated by Harms and Westerink (1980) using regular waves.

PAGE 100

100 Figure 4 4 US patented design by Walter (2000) for developing an artificial reef made of a concrete frame consisting of six concrete beams inserted through a number of tires.

PAGE 101

101 Figure 4 5 US patented design by Cederlund (2009) for a wave attenuation system, which consists of a floating member above the waterline, interlocking tires below the waterline, and anchors to keep the system in position.

PAGE 102

102 Figure 4 6 Top and side views of model 1 configuration. It is completely slack and consists of 4 rows with a single layer of scrap tires. Measurements are not to scale.

PAGE 103

103 Figure 4 7 Top and side views of model 2 configuration. The front is fixed at the flume bed using a gravity anchor. The back is completely slack. The model consists of 4 rows, a single layer of scrap tires, and is inclined 20.2 . Measurements are not to scale.

PAGE 104

104 Figure 4 8 Top and side views of model 3 configuration. The back is fixed at the flume bed using a gravity anchor. The front is completely slack. The model consists of 4 rows, a single layer of scrap tires, and is inclined 20.2. Measurements are not to scale.

PAGE 105

105 Figure 4 9 Top and side views of model 4 configuration. Th e front and back are fixed at the flume bed using gravity anchors. The model consists of 4 rows and a single layer of scrap tires. The front is inclined 59, while the back is inclined 90. Measurements are not to scale.

PAGE 106

106 Figure 4 10 Top and side views of model 5 configuration. The front is fixed at the flume bed using gravity anchors and is inclined 20.2. The back is completely slack. The model consists of 8 rows and a single layer of scrap tires. Meas urements are not to scale.

PAGE 107

107 Figure 4 11 Top and side views of model 6 configuration. It is completely slack and consists of 8 rows with a single layer of scrap tires. Measurements are not to scale.

PAGE 108

108 Figure 4 12 Top and side views of model 7 configuration. It is completely slack and consists of 4 rows with a double layer of scrap tires. Measurements are not to scale.

PAGE 109

109 Figure 4 13 Top and side views of model 8 configuration. The front is fixed at the flume bed using a gravity anchor. The back is completely slack. The model consists of 4 rows, double layers of scrap tires, and is inclined 16.5. Measurements are not to scale.

PAGE 110

110 Figure 4 14 Top and side views of model 9 configuration. The back is fixed at the flume bed using a gravity anchor. The front is completely slack. The model consists of 4 rows, double layers of scrap tires, and is inclined 16.5. Measurements are no t to scale.

PAGE 111

111 Figure 4 15 Picture of model 1 configuration in the wave flume at the Hydraulic and Coastal Engineering Laboratory of the Kuwait Institute for Scientific Research. It is completely slack and consists of 4 rows w ith a single layer of scrap tires. The white arrow represents the wave direction (Photo courtesy of author).

PAGE 112

112 Figure 4 16 Picture of model 2 configuration. The front is fixed at the flume bed using a gravity anchor. The back is completely slack. The model consists of 4 rows, a single layer of scrap tires, and is inclined 20.2 . The white arrow represents the wave direction (Photo courtesy of author).

PAGE 113

113 Figure 4 17 Picture of model 3 configu ration. The back is fixed at the flume bed using a gravity anchor. The front is completely slack. The model consists of 4 rows, a single layer of scrap tires, and is inclined 20.2. The white arrow represents the wave direction (Photo courtesy of author).

PAGE 114

114 Figure 4 18 Picture of model 4 configuration. The front and back are fixed at the flume bed using gravity anchors. The model consists of 4 rows and a single layer of scrap tires. The front is inclined 59, while the back is inclined 90. The white arrow represents the wave direction (Photo courtesy of author).

PAGE 115

115 Figure 4 19 Picture of model 6 configuration. It is completely slack and consists of 8 rows with a single layer of scrap tires. The white arrow represents the wave direction (Pho to courtesy of author).

PAGE 116

116 Figure 4 20 Picture of model 7 configuration. It is completely slack and consists of 4 rows with a double layer of scrap tires. The white arrow represents the wave direction (Photo courtesy of author).

PAGE 117

117 Figure 4 21 Picture of model 8 configuration. The front is fixed at the flume bed using a gravity anchor. The back is completely slack. The model consists of 4 rows, double layers of scrap tires, and is inclined 1 6.5. The white arrow represents the wave direction (Photo courtesy of author).

PAGE 118

118 Figure 4 22 Picture of model 9 configuration. The back is fixed at the flume bed using a gravity anchor. The front is completely slack. The mo del consists of 4 rows, double layers of scrap tires, and is inclined 16.5. The white arrow represents the wave direction (Photo courtesy of author).

PAGE 119

119 Figure 4 23 The concrete wave flume at the Hydraulic and Coastal Engineering Laboratory of the Kuwait Institute for Scientific Research (Photo courtesy of author)

PAGE 120

120 Figure 4 24 The flap type wavemaker for the concrete flume at the Hydraulic and Coastal Engineering Laboratory of the Kuwait Institute for Scientific Research (Photo courtesy of author)

PAGE 121

121 Figure 4 25 The beach for the concrete flume at the Hydraulic and Coastal Engineering Laboratory of the Kuwait Institute for Scientific Research (Photo courtesy of author)

PAGE 122

122 Figure 4 26 Wave probes 2, 3, and 4 are used to estimate the reflected wave height (Photo courtesy of author) WP2 WP3 WP4

PAGE 123

123 Figure 4 27 The work station by the concrete flume at the Hydraulic and Coastal Engineering Laboratory of the Kuwait Institute for Scientific Research. It consists of an amplifier and a PC. The DHI wave synthesizer analytical software was used for wave transmission an d reflection analysis (Photo courtesy of author)

PAGE 124

124 Figure 4 28 A wooden plank of 2.40 cm x 4 cm x 9 cm (LxWxH; Photo courtesy of author). Figure 4 29 A stainless steel angle of 285 cm x 5 c m x 5 cm (LxWxH; Photo courtesy of author).

PAGE 125

125 Figure 4 30 Polyethylene rope of 6 mm thickness, used for tightening the scrap tire configurations (Photo courtesy of author)

PAGE 126

126 Figure 4 31 A bowline knot was used while tightening the scrap tire configurations using a 6 mm polyethylene rope (Photo courtesy of author) Figure 4 32 Polyethylene rope of 14 mm thickness, used for the scrap tires configurations as a mooring rope (Photo courtesy of author)

PAGE 127

127 Figure 4 33 Polyform Buoy of 60 cm diameter (Photo courtesy of author) Figure 4 34 Gravity anchors of 57 cm diameter, 18 cm thickness, and 100 kg weight (Photo courtesy of author)

PAGE 128

128 Figure 4 35 Effect of / on , and for model 1, where / equaling 0.041, 0.082, and 0.12 due to regular waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12

PAGE 129

129 Figure 4 36 Effect of / on , and for model 2, where / equaling 0.041, 0.082, and 0.12 due to regular waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0 0.2 0.4 0.6 0.8 1 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12

PAGE 130

130 Figure 4 37 Effect of / on , and for model 3, where / equaling 0.041, 0.082, and 0.12 due to regular waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12

PAGE 131

131 Figure 4 38 Effect of / on , and for model 4, where / equaling 0.041, 0.082, and 0.12 due to regular waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.012

PAGE 132

132 Figure 4 39 Effect of / on , and for model 5 where / equaling 0.041, 0.082, and 0.12 due to regular waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.012

PAGE 133

133 Figure 4 40 Effect of / on , and for model 6, where / equaling 0.041, 0.082, and 0.12 due to regular waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12

PAGE 134

134 Figure 4 41 Effect of / on , and for model 7, where / equaling 0.041, 0.082, and 0.12 due to regular waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12

PAGE 135

135 Figure 4 42 Effect of / on , and for model 8, where / equaling 0.041, 0.082, and 0.12 due to regular waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12

PAGE 136

136 Figure 4 43 Effect of / on , and for model 9, where / equaling 0.041, 0.082, and 0.12 due to regular waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L Hi/d= 0.041 Hi/d= 0.082 Hi/d= 0.12

PAGE 137

137 Figure 4 44 Effect of / on , and for model 1, where / equaling 0.041 and 0.082 due to random waves. 0 0.2 0.4 0.6 0.8 1 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/Lp His/d= 0.041 His/d= 0.082 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L p His/d= 0.041 His/d= 0.082 0.800 0.850 0.900 0.950 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L p His/d= 0.041 His/d= 0.082

PAGE 138

138 Figure 4 45 Effect of / on , and for model 2, where / equaling 0.041 and 0.082 due to random waves. 0 0.2 0.4 0.6 0.8 1 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L p His/d= 0.041 His/d= 0.082 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L p His/d= 0.041 His/d= 0.082 0.800 0.850 0.900 0.950 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L p His/d= 0.041 His/d= 0.082

PAGE 139

139 Figure 4 46 Effect of / on , and for model 3, where / equaling 0.041 and 0.082 due to random waves. 0 0.2 0.4 0.6 0.8 1 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L p His/d= 0.041 His/d= 0.082 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L p His/d= 0.041 His/d= 0.082 0.800 0.850 0.900 0.950 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L p His/d= 0.041 His/d= 0.082

PAGE 140

140 Figure 4 47 Effect of / on , and for model 4, where / equaling 0.041 and 0.082 due to random waves. 0 0.2 0.4 0.6 0.8 1 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L p His/d= 0.041 His/d= 0.082 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L p His/d= 0.041 His/d= 0.082 0.800 0.850 0.900 0.950 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L p His/d= 0.041 His/d= 0.082

PAGE 141

141 Figure 4 48 Effect of / on , and for model 5, where / equaling 0.041 and 0.082 due to random waves. 0 0.2 0.4 0.6 0.8 1 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L p His/d= 0.041 His/d= 0.082 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L p His/d= 0.041 His/d= 0.082 0.800 0.850 0.900 0.950 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L p His/d= 0.041 His/d= 0.082

PAGE 142

142 Figure 4 49 Effect of / on , and for model 6, where / equaling 0.041 and 0.082 due to random waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L p His/d= 0.041 His/d= 0.082 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L p His/d= 0.041 His/d= 0.082 0.800 0.850 0.900 0.950 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L p His/d= 0.041 His/d= 0.082

PAGE 143

143 Figure 4 50 Effect of / on , and for model 7, where / equaling 0.041 and 0.082 due to random waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L p His/d= 0.041 His/d= 0.082 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L p His/d= 0.041 His/d= 0.082 0.800 0.850 0.900 0.950 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L p His/d= 0.041 His/d= 0.082

PAGE 144

144 Figure 4 51 Effect of / on , and for model 8, where / equaling 0.041 and 0.082 due to random waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L p His/d= 0.041 His/d= 0.082 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L p His/d= 0.041 His/d= 0.082 0.800 0.850 0.900 0.950 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L p His/d= 0.041 His/d= 0.082

PAGE 145

145 Figure 4 52 Effect of / on , and for model 9, where / equaling 0.041 and 0.082 due to random waves. 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Transmission, K t d/L p His/d= 0.041 His/d= 0.082 0.000 0.200 0.400 0.600 0.800 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Reflection, Kr d/L p His/d= 0.041 His/d= 0.082 0.800 0.850 0.900 0.950 1.000 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 Coefficient of Dissipation, K l d/L p His/d= 0.041 His/d= 0.082

PAGE 146

146 Figure 4 53 Comparison of for different scrap tire floating barrier models for different / values and / = 0.041. 0.00 0.05 0.10 0.15 0.20 0.25 0.30 1 2 3 4 5 6 7 8 9 Coefficient of Transmission, Kt Models H is /d= 0.041 d/Lp= 0.1 d/Lp= 0.2 d/Lp= 0.3 d/Lp= 0.4 d/Lp= 0.5

PAGE 147

147 Figure 4 54 Comparison of for different scrap tire floating barrier models for different / values and / = 0.082. 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 1 2 3 4 5 6 7 8 9 Coefficient of Transmission, Kt Models H is /d= 0.082 d/Lp= 0.2 d/Lp= 0.3 d/Lp= 0.4 d/Lp= 0.5

PAGE 148

148 Figure 4 55 Measured vs predicted values using the multiple regression equation for model 1. The blue solid line represents a slope of 45. The closer the values to the line, the more accurate the prediction of Figure 4 56 Measured vs predicted values using the multiple regression equation for mo del 2. 0.0 0.1 0.2 0.3 0.4 0 0.1 0.2 0.3 0.4 Predicted Measured Measured vs Predicted t 0.0 0.1 0.2 0.3 0.4 0 0.1 0.2 0.3 0.4 Predicted Measured Measured vs. Predicted t

PAGE 149

149 Figure 4 57 Measured vs predicted values using the multiple regression equation for model 3. Figure 4 58 Measure d vs predicted values using the multiple regression equation fo r model 4. 0.0 0.1 0.2 0.3 0.4 0 0.1 0.2 0.3 0.4 Predicted Measured Measured vs Predicted t 0.0 0.1 0.2 0.3 0 0.1 0.2 0.3 Predicted Measured Measured vs Predicted t

PAGE 150

150 Figure 4 59 Measured vs predicted values using the multiple regression equation for model 5. Figure 4 60 Measured vs predicted values using the multiple regression equation for model 6. 0.0 0.1 0.2 0 0.1 0.2 Predicted Measured Measured vs Predicted t 0.0 0.1 0.2 0.3 0.0 0.1 0.2 0.3 Predicted Measured Measured vs Predicted t

PAGE 151

151 Figure 4 61 Measured vs predicted values using the multiple regression equation for model 7. Figure 4 62 Measured vs predicted values using the multiple regres sion equation for model 8. 0.0 0.1 0.2 0.3 0.0 0.1 0.2 0.3 Predicted Measured Measured vs Predicted t 0.0 0.1 0.2 0.3 0.0 0.1 0.2 0.3 Predicted Measured Measured vs Predicted t

PAGE 152

152 Figure 4 63 Measured vs predicted values using the multiple regression equation for model 9. 0.0 0.1 0.2 0.3 0.0 0.1 0.2 0.3 Predicted Measured Measured vs Predicted t

PAGE 153

153 Figure 4 64 Bathymetric map of Cape Canaveral. The inset map shows the location of the study area relative to Florida. The blue strip located north of False Cape represents the ~8 km of critically eroding beach at KSC.

PAGE 154

154 Figure 4 65 The tidal range and wave seasonality at Cape Canaver al.

PAGE 155

155 Figure 4 66 Seasonality of wave direction at Cape Canaveral.

PAGE 156

156 Figure 4 67 The total occurrence of significant wave height at Cape Canaveral. The total number of samples is 1756, which was collected seasonally. The three blue boxes highlight three cases: the most dominant wave condition, the longest and most frequent wave length, and the largest wave height.

PAGE 157

157 Figure 4 68 Percent of occurrence of significant wave height at Cape Canaveral. The green boxes highlight the percentages around the most dominant wave condition.

PAGE 158

158 Figure 4 69 Time series of the seasonal max. orbital velocity at the bottom. The values above the dashed line represent the times of the year at which erosion is expected to happen at KSC. Figure 4 70 Side view of the adjusted cross section of model 7, which is proposed to be used for KSC.

PAGE 159

159 Figure 4 71 Cumulative probability for and values at KSC using model 7.

PAGE 160

160 Figure 4 7 2 Google earth image of Kuwait showing the location of Qaru Island relative to the mainland. The inset image shows a top view of the island surrounded by coral reefs (photo courtesy of http://www.shalay.com/style/images/resource/island8.jpg ).

PAGE 161

161 Figure 4 73 The total occurrences of significant wave height at Qaru Island. The total number of samples is 105178, which was collected via a model developed by the Kuwait Institute for Scientific Research.

PAGE 162

162 Figure 4 74 Percent of occurrence of significant wave height at Qaru Island. The green boxes highlight the high percentages around the most dominant wave condition.

PAGE 163

163 Figure 4 75 Side view of the adjusted cross section of model 6, which is proposed to be used for Qaru Island. Table 4 1 Regular Wave Parameters Wave Period, T (s) Wave Height, H i (cm) 1.253 5, 10, 15 1.323 5, 10, 15 1.408 5, 10, 15 1.513 5, 10, 15 1.652 5, 10, 15 1.847 5, 10, 15 2.145 5, 10, 15 2.660 5, 10, 15 3.747 5, 10

PAGE 164

164 Table 4 2 Random Wave Parameters Peak Wave Period, T p (s) Significant Wave Height, H is (cm) 1.253 5, 10 1.408 5, 10 1.652 5, 10 2.145 5, 10 3.747 5 Table 4 3 Details of the Various Scrap Tire Configurations Model Condition No. of Layers No. of Rows No. of Buoys B/d Fig. No. 1 Completely slack 1 4 4 1.83 4 6 2 Front is fixed at sea bed 1 4 2 1.83 4 7 3 Back is fixed at sea bed 1 4 2 1.83 4 8 4 Front and Back are fixed at sea bed 1 4 4 1.50 4 9 5 Front is fixed at sea bed 1 8 8 3.41 4 10 6 Completely slack 1 8 8 3.66 4 11 7 Completely slack 2 4 8 1.83 4 12 8 Front is fixed at sea bed 2 4 4 1.83 4 13 9 Back is fixed at sea bed 2 4 4 1.83 4 14

PAGE 165

165 Table 4 4 Tire Measurements Tire No. Weight (kg) Outer Diameter (cm) Inner Diameter (cm) Width (cm) 1 7.5 59 37 19 2 8 63.2 35 17.6 3 7.07 58 32 15.7 4 7.08 59.5 37 19 5 8.04 62 37 20.6 6 7.23 59 37 19.4 7 7.62 60 37 19.8 8 6.94 58.6 37 18.4 9 7.88 60.5 37 21 10 6.8 58.2 37 19.2 11 7.72 60 37 19.2 12 8.54 63 37.4 21 13 7.67 62.6 37.2 20.6 14 7.72 62.3 37.1 20.2 15 10.13 65 34.5 16 16 6.81 57.2 40 19 17 18.08 63 37.3 19.8 18 8.19 63 37 19.6 19 6.76 59.5 37.4 19.3 20 6.8 59 37 19 21 7.3 57.5 40 19.3 22 7.88 61.2 37.5 17.5 23 6.28 57.2 32.5 17 24 5.47 56.2 32.4 15 25 7.86 63.4 39.2 20 26 9.4 61.5 39.8 21 27 8.15 62.5 37.5 19.5 28 10.16 65.5 42 22.2 Mean 8.039 60.629 36.993 19.104 Std 2.219 2.497 2.222 1.681

PAGE 166

166 Table 4 5 Materials Used for Fabrication of Models Material Measurement Figure No. Wooden Plank 2.40 cm x 4 cm x 9 cm (LxWxH) 4 28 Stainless Steel Angle 285 cm x 5 cm x 5 cm (LxWxH) 4 29 Tightening Ropes 6 mm (Thickness) 4 30 and 4 31 Mooring Ropes 14 mm (Thickness) 4 32 Polyform Buoy Diameter = 60 cm 4 33 Gravity Anchor Weight = 100 kg 4 34 Diameter = 57 cm; Height = 18 cm Table 4 6 Sensor No. Calibration Factor (cm/volt) Distance from the wavemaker (m) 1 1.593 5.000 2 1.466 9.900 3 1.427 10.505 4 1.464 10.860 5 1.454 4 row structure: 22.23 8 row structure: 24.46

PAGE 167

167 Table 4 7 Multiple Regression Equations and for the Prediction of for Different Scrap Tire Configurations Model Equation Fig. No. Eqn. No. 1 0.87 4 55 4 1 2 0.93 4 56 4 2 3 0.91 4 57 4 3 4 0.93 4 58 4 4 5 0.84 4 59 4 5 6 0.85 4 60 4 6 7 0.85 4 61 4 7 8 0.89 4 62 4 8 9 0.93 4 63 4 9

PAGE 168

168 CHAPTER 5 CONCLUSIONS The wave transmission, reflection, and dissipation of nine new configurations of scrap tire wave barriers were investigated using physical models with regular and random waves. Regular waves were used to increase the fundamental understanding of wave inter action with scrap tires, while random waves were used to predict the field aim wa s to achieve the least wave transmission with a minimum number of scrap tires. The ease of c onstruction, installation, transport, and maintenance were also considered shorter wave lengths yielded smaller wave transmission coefficient (regular waves: 0.2 to 0 .8; random waves: 0.036 to 0.2). On the other hand, the waves with relatively longer wave lengths easily passed through the scrap tire model and yielded relatively higher wave transmissions (regular waves: 0.7 to 1.0; random wave: 0.13 to 0.35). Increasing the number of rows from 4 to 8 helped to increase the wave structure interaction and therefore allowed fewer waves to transmit and allowed more energy dissipation. Increasing the number of layers from 1 to 2 was also found to be effective in increasing wa ve reflection and hence reducing the transmission. Predictive relations which depend on relative water depth, d/ and relative wave height, /d, were obtained to predict the coefficient of transmission for the nine newly proposed scrap tire configur ations using random waves. The optimal configurations of this study were recommended to be used for field application to protect the beaches of Kennedy Space Center (KSC), USA and Qaru Island, Kuwait from erosion.

PAGE 169

169 In order to be able to achieve the optimal configuration for KSC, an oceanographic investigation was first completed to have full understanding of the hydrodynamics controlling the region. The tidal and subtidal hydrodynamics were analyzed over ridge swale bathymetry in the inner shelf adjacent to Cape Canaveral, Florida using vessel based data. The main objectives of this study were to determine the influence of cape associated shoals on the spatial structure of tidal and subtidal flows, and to assess whet her hydrodynamics derived from open channel flow also apply to this inner shelf region. To accomplish the objectives, results from vessel based ADCP measurements were compared to two analytical models that yield tidal and subtidal solutions. The region was sampled at two locations: the north transect and the south transect. The north transect had relatively smoother bathymetry with a bottom slope of 0.002. Tidal hydrodynamics were more influenced by local acceleration than frictional effects, while subtidal hydrodynamics were influenced either by Coriolis or followed Bernoulli type dynamics. The south transect had relatively more complex bathymetry, with a steeper bottom slope of 0.004, that defined a channel. Hence, frictional effects were dominant and the maximum flow was located over the deepest part of the cross section. The results obtained with both the tidal and subtidal analytical model solutions highlight the influence of ridge swale bathymetry in inner shelves at those temporal scales. The subinerti al hydrodynamics were also investigated using moored data. This study focused on two ridges located northeast and southeast of Cape Canaveral in central Florida. Four Acoustic Doppler Current Profilers (ADCPs) were moored at each side of the two ridges. Th e bathymetry in the region steered the flow to move mainly

PAGE 170

170 along shelf. As a result, most of the variance accounted for by Concatenated Hilbert Empirical Orthogonal Function (CHEOF) Mode 1 (95.25%) was due to the along shelf current. The vertical structure of the flow was unidirectional at the 4 locations. Wavelet coherence analysis techniques were used between the CHEOF Mode 1 and different forcings. When the along shelf wind and the Florida current move in the same direction, that coincide with an increas e in Florida current transport and vice versa. The along shelf wind in both cases enhances the subinertial flow motion. The across shelf momentum showed that the subinertial flow was in geostrophic balance throughout the deployment. The along shelf momentu m was dominated by pressure gradient, bottom stress, and the gradient of However, other terms can still be influential. Enhancement of the Florida current is associated with increased offshore sea surface slope following geostrophy. This causes the subinertial flow on the shelf to move in the same direction as the western boundary current. On the other hand, southward winds coincide with weakening of the Florida current. As a result, a negative nearshore sea surface slope develops to maintain offsho re geostrophy and the nearshore flow moves southward.

PAGE 171

171 LIST OF REFRENCES Allen, J. S., and P. K. Kundu (1978), On the momentum, vorticity, and mass balance on the Oregon shelf, J. Phys. Oceanogr., 8, 13 27. Amusing Planet (2015), World biggest tire gravey ard in Sulabiya, Kuwait. Web. 31 May 2017, < http://www.amusingplanet.com/2015/01/worlds biggest tire graveyard in.html > Armi, L. (1986), The hydraulics of two flowing layers with different densities, J. Fluid Mech., 163, 27 58. Brown, W. S., N. R. Pettigrew, and J. D. Irish (1985), The Nantucket Shoals Flux Experiment (NSFE79). Part II: The structure and variability of aacross shelf pressure gradients, J. Phys. Oceanogr., 15, 749 771. Brown, W. S., J. D. Irish, and C. D. Winant (1987), A description of subtidal pressure field observations on the northern California continental shelf during the Coastal Ocean Dynamics Experiment, J. Geophys. Res., 92, 1605 1636. C andle, R. and W. Fischer (1975), Scrap tire shore protection structures, Goodyear Publication, Hydraulic Engineering Reports, uuid:26f882f2 04a7 4c72 853e abf4471fcbfe. Cederlund, J. W. (2009). Wave attenuation system. US Patent No. 7575396. Day, K.E., K.E Holtze, J.L. Metcalfe Smith, C.T. Bishop, and B.J. Dutka (1993). Toxicity of leachate from automobile tires to aquatic biota. Chemosphere. 27 (4), pp.665 675. Emery, W. J. and R. E. Thomson (1998), Data Analysis Methods in Physical Oceanography, 1st ed., Elsevier, New York. Eriksen, C. C. (1991), Observations of amplified flows atop a large seamount, J. Geophys. Res., 96(C8), 15,227 15,236. Farmer, D. M., and L. Armi (1986), Maximal two layer exchange over a sill and through the combination of a sill and a contraction with barotropic flow, J. Fluid Mech., 164, 53 76. Fewings, M. R. and Lentz, S. J. 2010, Momentum balances on the inner continental shelf at Martha's Vineyard Coastal Observatory, J. Geophys. Res., 115(C12), doi: 10.1029/2009JC005578. Fewings M. R., Lentz S. J., Fredericks J. 2008. Observations of across shelf flow driven by across shelf winds on the inner continental shelf. J. Phys. Oceanogr. 38:2358 78.

PAGE 172

172 Florida Department of Environmental Protection (2016). Critically Eroded Beaches In Florida. Divisio n of Water Resource Management. Friedrichs, C. T., & Madsen, O. S. (1992). Nonlinear diffusion of the tidal signal in frictionally dominated embayments. Journal of Geophysical Research, 97(C4), 5637. doi:10.1029/92jc00354 Geyer, W. R., Trowbridge, J. H. and Bowen, M. M. 2000. The Dynamics of a Partially Mixed Estuary, J. Phys. Oceanogr. 2035 48, doi: 10.1175/1520 0485(2000)030<2035:TDOAPM>2.0.CO;2. Gill, A. E. (1982), Atmosphere Ocean Dynamics, vol. 30, 662 pp., Elsevier, New York. Gu, R.R., (2005). Beneficial Reuses of Scrap Tires in Hydraulic Engineering. The Handbook of Environmental Chemistry, 2005, Vol. 5F, pp.183 215. Grinsted, A., J. C. Moore, and S. Jevrejeva (2004), Application of the across wavelet transform and wavelet cohere nce to geophysical time series, Nonlinear processes in geophysics, 11(5/6), 561 566. Hannachi, A., Jolliffe, I.T., Stephenson, D.B., 2007. Empirical Orthogonal Functions and related techniques in atmospheric science: A review. International Journal of Clim atology 27, 11191152. Harms, V. W. and J. J. Westerink (1980), Wave transmission and mooring force characteristics of pipe tire floating breakwaters. doi:10.2172/1108708 Herbich, J. B. (1981). Offshore pipeline design elements. New York: M. Dekker Herrid ge, L. (2014 ). Dark Skies Essential for Sea Turtle Nesting Season. Retrieved August 30, 2017 < https://www.nasa.gov/content/kennedy space center keeps dark skies for sea turtle nesting >. Hibarger, G.E., G. G. Hibarger and D. W. Daniel. (1979). Breakwater System. US Patent No.4150909. Huijts, K. M. H., H. M. Schuttelaars, H. E. de Swart, and A. Valle Levinson (2006), Lateral entrapment of sediment in tida l estuaries: An idealized model study, J. Geophys. Res., 111, C12016, doi:10.1029/2006JC003615. Huijts, K., H. Schuttelaars, H. D. Swart, and C. Friedrichs (2009), Analytical study of the transverse distribution of along channel and transverse residual flo ws in tidal estuaries. Continental Shelf Research, 29(1), 89 100. doi:10.1016/j.csr.2007.09.007. Huijts, K. M., H. E. Swart, G. P. Schramkowski, and H. M. Schuttelaars (2011), Transverse structure of tidal and residual flow and sediment concentration in es tuaries. Ocean Dynamics, 61(8), 1067 1091. doi:10.1007/s10236 011 0414 7

PAGE 173

173 Atmospheric and Oceanic Technology, 6(1), 169 172. doi:10.1175/1520 0426(1989)006<0169:oisosa>2.0.co;2 K aihatu, J. M., R. A. Handler, G. O. Marmorino, and L. K. Shay (1998). Empirical Orthogonal Function Analysis of Ocean Surface Currents Using Complex and Real Vector Methods*. Journal of Atmospheric and Oceanic Technology, 15(4), 927 941. doi:10.1175/1520 0 426(1998)015<0927:eofaoo>2.0.co;2 16 961 16 969. Kelly, K.A., and S.T. Gille, 1990: Gulf Stream surface transport and statistics at 69W from the Geosat altimeter. Journal of Geophysical Research, 95, 3149 3161. Kline, S. W., P. N. Adams, N. G. Plant, R. A. MacKenzie, and J. M. Jaeger (2012), Effect of a shore oblique storm ridge on nearshore wave energy dissipation and beach/bar morphology at Kennedy Space Center, Cape Canaveral, Florida, AGU Fall Meeting 2012. Kumar, N., G. Voulgaris, J. H. List, and J. C. Warner (2013), Alongshore momentum balance analysis on a cuspate foreland, J Geophys. Res. Oceans, 118, 5280 5295, doi:10.1002/jgrc.20358. Large, W. and S. Pond (1981), Open ocean momentum flux measurements in moderate to strong winds, J. Phys. Oceanogr., 11(3), 324 336, doi: 10.1175/1520 0485(1981)011<0324:OOMFMI>2.0.CO;2. LeBl ond, P. H., 1978: On tidal propagation in shallow rivers. J. Geophys. Res., 83 (C9), 4717 4721. Lee, T. N., W. J. Ho, V. Kourafalou, and J. D. Wang (1984), Circulation on the continental shelf of the southeastern United States. Part I: Subtidal response to wind and Gulf Stream forcing during winter, J. Phys.Oceanogr.,14, 1001 1012. Lee, T. N., E. Williams, R. E. J. Wang, and L. Atkinson (1989), Response of South Carolina continental shelf waters to wind and Gulf Stream forcing during winter of 1986, J. Geoph ys. Res., 94, 10,715 10,754. Lentz, S. J., and C. D. Winant (1986), Subinertial currents on the southern California shelf, J. Phys. Oceanogr., 16, 1737 1750. Lentz, S. J. ( 1995 ) Sensitivity of the inner shelf circulation to the eddy viscosity profile. J. Phys. Oceanogr. 25:19 28.

PAGE 174

174 Lentz, S. J., R. T. Guza, S. Elgar, F. Feddersen, and T. H. C. Herbers (1999), Momentum balances on the North Carolina inner shelf, J. Geophys. Res., 104, 18,205 18,226. Lentz, S. J., M. Fewings, P. Howd, J. Fredericks, and K. Hat haway (2008), Observations and a model of undertow over the inner continental shelf, J. Phys. Oceanogr., 38(11), 2341 2357, doi: 10.1175/2008JPO3986.1. Lentz, S. J. and M. R. Fewings (2012), The wind and wave driven inner shelf circulation, Annual review o f marine science, 4, 317 343, doi: 10.1146/annurev marine 120709 142745. Li, Z., Weisberg, R.H., 1999a. West Florida Shelf response to upwelling favorable wind forcing: kinematics. Journal of Geophysical Research 104, 13507 13527. Li, Z., Weisberg, R.H., 1 999b. West Florida continental shelf response to upwelling favorable wind forcing, 2, Dynamics. Journal of Geophysical Research 104, 23427 23442.Liu, Y., and R. H. Weisberg (2005), Momentum balance diagnoses for the West Florida Shelf, Cont. Shelf Res., 25 2054 2074. Liu, Y., and R. H. Weisberg (2005), Momentum balance diagnoses for the West Florida Shelf, Cont. Shelf Res., 25, 2054 2074. Longuet Higgins, M. S. 1953. On wave set up in shoaling water with a rough sea bed. J. FluidMech. 527:217 34. Longuet H iggins, M. S. and R. W. Stewart (1964), Radiation stresses in water waves; a physical discussion, with applications, Deep Sea Research and Oceanographic Abstracts, 11(4), 529 562, doi: 10.1016/0011 7471(64)90001 4. Lwiza, K.M.M., Bowers, D.G., Simpson, J.H ., 1991. Residual and tidal flow at a tidal mixing front in the North Sea. Cont. Shelf Res. 11 (11), 1379e1395. Masse, A. (1988), Estuary shelf interaction: Delaware Bay and the inner shelf, Ph.D. thesis, Univ. of Del., Newark, Del. McGregor, R. C. and N. S. Miller (1978). Scrap Tyre Breakwaters in Coastal Engineering. Coastal Engineering 1978. doi:10.1061/9780872621909.134 McNinch, J. E., and R. A. Luettich Jr. (2000), Physical processes around a cuspate foreland headland: Implications to the evolution an d long term maintenance of a cape associated shoal, Cont. Shelf Res., 20(17), 2367 2389. Mei CC. 1983. The Applied Dynamics of Ocean Surface Waves. New York: McGraw Hill. Mitchum, G., and W. Sturges (1982), Wind driven currents on the west Florida shelf, J Phys. Oceanogr., 12, 1310 1317.

PAGE 175

175 Neelamani, S., K. Al Banaa, and A. Al Raqum (2009). Soil classification of Kuwaiti beach soils. Conference on land degradation in dry environments. 8 14 March 2009. National Aeronautics and Space Administration ( 2015), Env ironmental assessment for K SC shoreline protection project John F. Kennedy Space Center. Environmental Management Branch. Noble, H. M. (1969). Wave Maze Floating Breakwater. Proceedings Civil Engineering in the Oceans II, ASCE Conference, Miami Beach, FL. Dec. 10 12, 1969. Noble, H. M. (1976). Use of Wave Maze Flexible Floating Breakwater to Protect Offshore structures and Landings. Offshore Technology Conference. doi:10.4043/2542 ms Noble, M., and B. Butman (1983), On the longshelf structure and dynamics of subtidal currents on the eastern United States continental shelf, J. Phys. Oceanogr., 13, 2125 2146. Penoyer, D., L. Kennelly, and R. Dever (2006), Use of Tire Chips in Landfill Gas Gas Symposium. Pettigrew, N. R. (1981), The dynamics and kinematics of the coastal boundary layer off Long Island, Ph.D. thesis, Woods Hole Oceanogr. Inst., Woods Hole, Mass. Pollard, D. (1989). Asymptotics via Empirical Processes. Statistical Science, 4(4 ), 341 354. doi:10.1214/ss/1177012394 Russel, R. (2007). Thermohaline Ocean Circulation. Retrieved July 07, 2017, < https://www.windows2universe.o rg/?page=%2Fearth%2FWater%2Fthermohalin e_ocean_circulation.html > Salas Monreal, D., and A. Valle Levinson (2009), Continuously stratified flow dynamics over a hollow, J. Geophys. Res., 114, C03021, doi:10.1029/2007JC004648. Sanay R., G. Voulgaris, and J. C. Warner (2007), Influence of tidal asymmetry and residual circulation on sediment transport in linear sandbanks: A processes oriented numerical study, J. Geophys. Res., 112, C12015, doi: 10.1029/2007JC004101. Scott, J. T., and G. T. Csanady (1976), Nears hore currents off Long Island, J. Geophys. Res., 81, 5401 5402. Shearman, K., and S. J. Lentz (2003), Dynamics of mean and subtidal flow on the New England shelf, J. Geophys. Res., 108(C8), 3281, doi:10.1029/ 2002JC001417.

PAGE 176

176 Shore Protection Manual (2nd ed., Vol. 2). (1984). Vicksburg, MS: Dept. of the Army, Waterways Experiment Station, Corps of Engineers, Coastal Engineering Research Center. Stenstrom, P. (2003), Mixing and recirculation in two layer exchange flows, J. Geophys. Res., 108(C8), 3256, doi:10.1 029/2002JC001696. Stokes GG. 1847. On the theory of oscillatory waves. Trans. Camb. Philos. Soc. 8:441 55. Stuart D. S. 1988. Coefficients for Sea Surface Wind Stress, Heat Flux, and Wind Profiles as a Function of Wind Speed and Temperature, J. Geophys. Re s., 93, 15467 15472, doi: 10.1029/JC093iC12p15467. Thompson, K. R., and D. T. Pugh (1986), The subtidal behavior of the Celtic Sea Part II. Currents, Cont. Shelf Res., 5, 321 346. Torrence, C. and G. P. Compo (1998), A practical guide to wavelet analysis, Bull. Am. Meteorol. Soc., 79(1), 61 78. Valle coastal plain estuary. Buoyancy Effects on Coastal and Estuarine Dynamics, D. G. Aubrey and C. T. Friedrichs, Eds., Coastal and Estuarine Studies, Vol. 53, Amer. Geophys. Union, 265 281 Valle Levinson, A., K. C. Wong, and K. M. M. Lwiza (2000), Fortnightly variability in the transverse dynamics of a coastal plain estuary, J. Geophys. Res., 105(C2), 3413 3424. Valle Levinson, A., C. A. Reyes, and R. Sanay (2003), Effects of bathymetry, friction, and rotation on estuary ocean exchange, J. Phys. Oceanogr., 33(11), 2375 2393. Valle Levinson, A. (2008), Density driven exchange flow in terms of the Kelvin and Ekman numbers, J. Geophys. Res.,113, C04001, doi:10.1029/2007JC004144 Valle Levinson, A., and X. Guo (2009), Asymmetries in tidal flow over a Seto Inland Sea scour pit. Journal of Marine Research, 67(5), 619 635. doi:10.1357/002224009791218850 Valle Levinson, A. (Ed.). (2010). Con temporary Issues in Estuarine Physics. Cambridge: Cambridge University Press. doi:10.1017/CBO9780511676567 Valle Levinson, A., K. Huguenard, L. Ross, J. Branyon, J. Macmahan, and A. Reniers (2015), Tidal and nontidal exchange at a subtropical inlet: Destin Inlet, Northwest Florida. Estuarine, Coastal and Shelf Science, 155, 137 147. doi:10.1016/j.ecss.2015.01.020

PAGE 177

177 Walter, D. M. (2000). Concrete and tire artificial reef. US Patent No. 6042300. Waterhouse, A.F., A. Valle Levinson, and C.D. Winant, 2011: Tides in a System of Connected Estuaries. J. Phys. Oceanogr., 41, 946 959, < https://doi.org/10.1175/2010JPO4504.1 >. Wint ers, K. B., and H. E. Seim (2000 ), The role of dissipation and mixing in exchange flow through a contracting channel, J. Fluid Mech., 407, 265 290. Wong, K. C., 1994: On the nature of transverse variability in a coastal plain estuary. J. Geophys. Res., 99, 14 209 14 222. Xu, Z. and A. Bowen (1994), Wave and wind driven flow in water o f finite depth, J. Phys. Oceanogr., 24(9), 1850 1866, doi: 10.1175/15200485(1994)024<1850:WAWDFI>2.0.CO;2. Zlotnicki, V., 1991: Sea level differences across the Gulf Stream and Kuroshio Extension. Journal of Physical Oceanography, 21, 599 609.

PAGE 178

178 BIOGRAPHICAL SKETCH Ahmad Yous if earned a bachelor degree in civil e ngineering from Temple University in August 2010 within 3.5 years and a GPA of 3.93. It was the highest GPA in the Civil & Environmental Engineering D epartment at Temple University. His se nior design project was selected by the department faculty members to be the best senior design. In August 2012, he graduated from North Ca rolina State University with a master of science (M.S.) in civil e ngineering within 1.5 years. After ~2.5 years of wo rking for industry and Kuwait government, he earned a sc holarship to pursue a Ph.D. in coastal and o ceanogra phic e ngineering. In 2015, he started his doctoral studies at the University of Florida under the supervision of Prof. Arnoldo Valle Levinson. He re ceived his Ph.D. from the University of Florida in the fall of 2017.