• TABLE OF CONTENTS
HIDE
 Front Cover
 Title Page
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 Abstract
 Introduction
 Background information and literature...
 Brief model descriptions
 Model calibration
 Model verification
 Model performance
 Summary and conclusions
 Measured and predicted profiles...
 Measured and predicted profiles...
 Measured and contour change for...
 Reference
 Biographical sketch














Title: Comparisons of erosion models for hurricanes Floyd and Irene at Ft. Pierce Beach, Florida
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00091421/00001
 Material Information
Title: Comparisons of erosion models for hurricanes Floyd and Irene at Ft. Pierce Beach, Florida
Series Title: Comparisons of erosion models for hurricanes Floyd and Irene at Ft. Pierce Beach, Florida
Physical Description: Book
Language: English
Creator: Heckman, Lisa Dawn
Publisher: Coastal & Oceanographic Engineering Dept. of Civil & Coastal Engineering, University of Florida
Place of Publication: Gainesville, Fla.
 Record Information
Bibliographic ID: UF00091421
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.

Table of Contents
    Front Cover
        Front Cover
    Title Page
        Page i
    Acknowledgement
        Page ii
    Table of Contents
        Page iii
        Page iv
        Page v
    List of Tables
        Page vi
    List of Figures
        Page vii
        Page viii
        Page ix
        Page x
    Abstract
        Page xi
        Page xii
    Introduction
        Page 1
        Page 2
    Background information and literature review
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
    Brief model descriptions
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
    Model calibration
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
    Model verification
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
    Model performance
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
    Summary and conclusions
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
    Measured and predicted profiles for Hurricane Floyd
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
    Measured and predicted profiles for Hurricane Irene
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
    Measured and contour change for Hurricanes Floyd and Irene
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
    Reference
        Page 133
        Page 134
    Biographical sketch
        Page 135
Full Text



UFL/COEL -2001/002


COMPARISONS OF EROSION MODELS FOR HURRICANES
FLOYD AND IRENE AT FT. PIERCE BEACH, FLORIDA







by




Lisa Dawn Heckman





Thesis


2001













COMPARISONS OF EROSION MODELS FOR HURRICANES FLOYD AND
IRENE AT FT. PIERCE BEACH, FLORIDA
















By

LISA DAWN HECKMAN


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2001













ACKNOWLEDGMENTS

I would first like to express my gratitude to Dr. Robert Dean, my advisor on

this thesis project. To be able to study under someone with his standing in the coastal

engineering field has been unforgettable. The insight, constructive advisement, and

support he has provided are invaluable. I would also like to thank Dr. Daniel Hanes,

Dr. Ashish Mehta, and Dr. Robert Thieke for serving on my committee and for their

guidance in the classroom.

Dr. Bruce Taylor gave me the opportunity to work full-time as a staff engineer

at Taylor Engineering, Inc. while I completed this thesis, and to him I owe my

deepest appreciation. I would also like to thank Dr. Mark Gosselin, Dr. Rajesh

Srinivas, and Mr. Ken Craig, co-workers at Taylor Engineering, who all patiently

took time out of their busy days to advise me on this thesis. All my other co-workers

deserve gratitude as well for having patience with me as I completed this thesis while

working full-time.

Finally, I would like to thank my family and friends, who have always

supported me.














TABLE OF CONTENTS

Page

ACKN OW LEDGM ENTS ..............................................- ---.............. ................ -ii

LIST OF TA BLES ................................................................... .................. vi

LIST OF FIGURES ............................ ....................... -- .......... ...... ......... vii

AB STRA CT .................................................... ........ ............................. xi

1 INTROD U CTION ........................ ............... ........-.. .... .... ................... I1

P u rpose ................................................................ ............................ 1
Report Organization ........................................................ 1

2 BACKGROUND INFORMATION AND LITERATURE REVIEW................. 3

The Ft. Pierce Shore Protection Project........................... .................... 3
Storm D escriptions.................................... .. ........... ... ... ......... 5
Hurricane Floyd..................................................... ... 5
Hurricane Irene....................................... .............. .... 6
Literature R eview .............................................. ........... ................... 6

3 BRIEF MODEL DESCRIPTIONS........................................... 10

The Equilibrium Beach Profile .................................................... 10
SBEACH M odel.................................. ... .......................... 11
EDUNE M odel....................... ...... .......... ..... .................... 15
C R O SS M odel.......................................................... ..................................... 16

4 MODEL CALIBRATION................................................... 18

Beach Profile Data and Profile Input ....................... .. ................ 18
Storm Input .................................................................. 19
W ave and W ater Level D ata ......... .... .......................................... ... 19
W ave Height......................................... ..... ... 21
W ave Period ............................................................ ... 22
W ave Direction. ........................... ...... ....... .. ................... 22
W ater Level..................... ....... .. .................................... 22
W ind D ata .................................. ....................... ... .......... ................... 23







Program Options .................................... ...... .................... 23
SBEA CH M odel................................. .................. ............................. 23
EDUNE and CROSS Models..................... ........................ 23
C alibration Procedure............................... .......................... ..................... 23
SBEACH Model Calibration........................... ................... 28
EDUNE M odel Calibration............................................... ..................... 30
CROSS Model Calibration................................. ............... ... 32

5 MODEL VERIFICATION............................................... 38

Profile Input ............. .......... ........ ....................... ... 38
Storm Input .................................... ................................... 38
W ave D ata............................... ............... .................... 39
W ave H eight.................................. ............................... 39
W ave Period ...................................................... ......................... 39
W ave D irection..................................................................... 39
W ater L evel ................................................................................................. 40
W ind D ata .................................................... ... ................................................ 4 1
V erification Procedure................ ................................................................. 41

6 MODEL PERFORMANCE............................ ......................... 49

Measures of Model Performance ...................................... 49
R esidual Param eter ..................................................................................-....... 49
E roded V olum e............ ........................... ... .................................... 50
Shoreline R etreat ......................................... .............. . .... ....... ............. 50
Contour Change Along Entire Profile ........................................ 52
Relative Model Performance .................... .......................... 53
SB E A C H M odel ......................................................... ................................... 53
EDUNE M odel................................... ...... .....................54
CR O SS M odel........................................ ......... . ............ ................. 55
Model Comparisons ........................................... 55
Mean Sediment Grain Size Sensitivity Analysis ............... ........ .. 59

7 SUMMARY AND CONCLUSIONS ................................................................. 70

Sum m ary .........................................................--.............. ..... .............................. 70
C onclu sio ns................. ................... .......................... ................................ 7 1

APPENDIX A MEASURED AND PREDICTED PROFILES FOR HURRICANE
FL O Y D ......................... ...... ........ ............-. .. ......... ......... .................................. 75

APPENDIX B MEASURED AND PREDICTED PROFILES FOR HURRICANE
IR E N E .................................................................................................................. 10 0

APPENDIX C MEASURED AND PREDICTED CONTOUR CHANGE FOR
HURRICANES FLOYD AND IRENE.......................................................... 119








LIST OF REFERENCES..................... ...... ........... ... 133

BIOGRAPHICAL SKETCH ......................................... .............. ..................... 135
























































V













LIST OF TABLES


Table Page

3.1 Summary of Recommended A Values (m1r ) (Dean, et al., 1994)..................... 11

4.1 Permissible calibration ranges for the SBEACH model parameters.................. 29

4.2 SBEACH Model Calibration for Hurricane Floyd.............................................. 30

4.3 Calibration ranges for the EDUNE model parameters....................................... 31

4.4 EDUNE Model Calibration for Hurricane Floyd................................................ 32

4.5 Calibration ranges for the CROSS model parameters........................................ 33

4.6 CROSS Model Calibration for Floyd.................................................... 34

6.1 Predicted Residuals, Measured and Predicted Eroded Volume Between +8 and
0 ft-NGVD and Shoreline Retreat at the 6-ft Contour for Hurricane
F loyd ......................................... ................... .... .................. 51

6.2 Predicted Residuals, Measured and Predicted Eroded Volume Between +8 and
0 ft-NGVD and Shoreline Retreat at the 6-ft Contour for Hurricane
Irene ............... ................... ...... .... ............................ 52

6.3 Summaries of Model Performances for Hurricane Floyd Model Calibration.... 56

6.4 Summaries of Model Performances for Hurricane Irene Model Verification ... 56

6.5 Mean Sediment Grain Size Sensitivity Analysis for Profile R-38................... 60

7.1 Summaries of Model Performances for Hurricanes Floyd and Irene ............... 72













LIST OF FIGURES


Figure Page

2.1 Ft. Pierce Shore Protection Project Location Map (Taylor Engineering,
2000) ............................................................. ............ .................. 4

3.1 4-Zone Transport Relationship for SBEACH (Larson, 1989)......................... 13

4.1 M measured Profile Data R-37 .............................................. ........................ 20

4.2 Hurricane Floyd Significant Wave Height and Period........................................ 24

4.3 Hurricane Floyd Wave Direction ............................................................. 25

4.4 Hurricane Floyd W ater Level........................................................................... 26

4.5 Hurricane Floyd Wind Speed.............................. ..................................... 27

4.6 Profile R-37 Measured and Predicted Post-Floyd from SBEACH Model ........... 35

4.7 Profile R-37 Measured and Predicted Post-Floyd from EDUNE Model ......... 36

4.8 Profile R-37 Measured and Predicted Post-Floyd from CROSS Model............ 37

5.1 Hurricane Irene Significant Wave Height and Period...................................... 42

5.2 Hurricane Irene Wind Speed and Direction.................................. ........... ... 43

5.3 Hurricane Irene Wave Direction ............................................................... 44

5.4 H hurricane Irene W ater Level........................................................................ .... 45

5.5 Profile R-37 Measured and Predicted Post-Irene from SBEACH Model............ 46

5.6 Profile R-37 Measured and Predicted Post-Irene from EDUNE Model .............. 47

5.7 Profile R-37 Measured and Predicted Post-Irene from CROSS Model ............ 48

6.1 Profile R-37 Contour Change for Hurricane Floyd.......................................... 61







6.2 Profile R-41 Post-fill, Post-Floyd, and Post-Floyd SBEACH............................. 62

6.3 Profile R-41 Post-fill, Post-Floyd, and Post-Floyd EDUNE ........................... 63

6.4 Profile R-38 Post-fill, Post-Floyd, and Post-Floyd CROSS ............................ 64

6.5 Profile R-37 Measured and Predicted Post-Floyd from Three Models ............. 65

6.6 Profile R-37 Measured and Predicted Post-Irene from Three Models................. 66

6.7 Average Contour Change for Hurricane Floyd................................................. 67

6.8 Average Contour Change for Hurricane Irene................................................ 68

6.9 Profile R-34 Post-fill, Post-Floyd, and Post-Floyd SBEACH............................. 69

A.1 Profile R-34 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH............. 76

A.2 Profile R-35 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH............. 77

A.3 Profile R-36 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH ............ 78

A.4 Profile R-37 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH .............. 79

A.5 Profile R-38 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH ............ 80

A.6 Profile R-39 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH ............ 81

A.7 Profile R-40 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH ............ 82

A.8 Profile R-41 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH ............ 83

A.9 Profile R-34 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE................ 84

A. 10 Profile R-35 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE ............ 85

A.11 Profile R-36 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE ............ 86

A. 12 Profile R-37 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE ............ 87

A. 13 Profile R-38 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE............. 88

A. 14 Profile R-39 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE............. 89

A. 15 Profile R-40 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE............. 90

A. 16 Profile R-41 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE.............. 91

A.17 Profile R-34 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd CROSS.............. 92







A. 18 Profile R-35 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd CROSS................. 93

A. 19 Profile R-36 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd CROSS................. 94

A.20 Profile R-37 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd CROSS............... 95

A.21 Profile R-38 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd CROSS................ 96

A.22 Profile R-39 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd CROSS.............. 97

A.23 Profile R-40 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd CROSS............... 98

A.24 Profile R-41 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd CROSS................ 99

B.1 Profile R-36 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
SB E A C H .................................................... ...................................... 10 1

B.2 Profile R-37 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
SB E A C H ............................................................................................. 102

B.3 Profile R-38 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
SB E A C H ............................................................................................. 103

B.4 Profile R-39 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
SB E A C H ........................................ ................................................... 104

B.5 Profile R-40 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
SB E A C H ............................................................................................. 105

B.6 Profile R-41 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
SB E A C H ............................................................................................. 106

B.7 Profile R-36 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
EDUNE ............................................................. .................... 107

B.8 Profile R-37 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
E D U N E ............................................................................................... 108

B.9 Profile R-38 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
ED U N E ..................................................... ............................... 109

B.10 Profile R-39 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
ED U N E ........................... .......................................................... 110

B. 11 Profile R-40 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
ED U N E ............................................................................ 111

B.12 Profile R-41 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
ED U N E ......................................... .. .... .. 112







B.13 Profile R-36 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
C R O SS ...................................................................... 113

B.14 Profile R-37 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
C R O SS ................................................. ........................ ........... 114

B.15 Profile R-38 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
C R O S S ................................................................................................ 115

B.16 Profile R-39 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
C R O S S .............................................................. ........... 116

B.17 Profile R-40 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
CR O SS ............................................................... 117

B. 18 Profile R-41 Pre-fill, Post-fill, Post-Floyd, Post-Irene, and Post-Irene
CROSS ......................................... 118

C.1 Profile R-34 Contour Change for Hurricane Floyd ........................................ 120

C.2 Profile R-35 Contour Change for Hurricane Floyd ........................................ 121

C.3 Profile R-36 Contour Change for Hurricane Floyd ................................... 122

C.4 Profile R-38 Contour Change for Hurricane Floyd ........................................ 123

C.5 Profile R-39 Contour Change for Hurricane Floyd ........................................ 124

C.6 Profile R-40 Contour Change for Hurricane Floyd ................................... 125

C.7 Profile R-41 Contour Change for Hurricane Floyd ........................................ 126

C.8 Profile R-36 Contour Change for Hurricane Irene ......................................... 127

C.9 Profile R-37 Contour Change for Hurricane Irene ....................................... 128

C. 10 Profile R-38 Contour Change for Hurricane Irene .................................... 129

C. 11 Profile R-39 Contour Change for Hurricane Irene ................................... 130

C. 12 Profile R-40 Contour Change for Hurricane Irene ....................................... 131

C. 13 Profile R-41 Contour Change for Hurricane Irene ........................................ 132













Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

COMPARISONS OF EROSION MODELS FOR HURRICANES FLOYD AND
IRENE AT FT. PIERCE BEACH, FLORIDA

By

Lisa Dawn Heckman

May 2001

Chairman: Dr. Robert G. Dean
Major Department: Civil and Coastal Engineering

This thesis presents a comparison of three cross-shore sediment transport

models applied to predict erosion at the Ft. Pierce Shore Protection Project area in

Florida during Hurricanes Floyd and Irene of 1999. Post-fill, post-Hurricane Floyd,

and post-Hurricane Irene survey data were available for Florida Department of

Environmental Protection reference monuments R-34 to R-41. The models were

calibrated with post-fill and post-Hurricane Floyd survey data, and Hurricane Floyd

storm data taken at a wave gage located offJensen Beach, Florida. The calibration

procedure was to vary the calibration parameters of each model, and determine which

set of parameters provided the least average root mean square error of measured to

predicted beach erosion for all profile lines. The models were then verified with post-

Hurricane Floyd and post-Hurricane Irene survey data, and Hurricane Irene storm

data taken from the NOAA buoy located offshore Cape Canaveral, Florida. The







calibrated parameters that were previously determined with Hurricane Floyd were

used in the verification procedure for the Hurricane Irene model runs.

The performances of the three models were evaluated and compared using

several methods including the residual parameter, eroded volume of sand, beach

retreat, and contour change. All three models provided reasonable predictions for

beach erosion for both Hurricanes Floyd and Irene. For Hurricane Floyd, the average

residual was between 0.47 and 0.51 for the three models, the average ratio of

predicted eroded volume to measured eroded volume was between 0.44 and 0.49, and

the average ratio of predicted retreat to measured retreat was between 0.46 and 0.52.

For Hurricane Irene, the average residual was between 0.47 and 0.74 for the three

models, the average ratio of predicted eroded volume to measured eroded volume was

between 0.67 and 2.50, and the average ratio of predicted retreat to measured retreat

was between 0.53 and 2.77. All three models underpredicted the average volume of

erosion during Hurricane Floyd, while for Hurricane Irene, the SBEACH and

EDUNE models overpredicted and the CROSS model underpredicted. The SBEACH

and EDUNE models incorporate dune overwash processes, but only the SBEACH

model predicts some very slight overwash during Hurricane Floyd. The SBEACH

model predicts the upper part of the beach face better than the EDUNE or CROSS

models, while the EDUNE and CROSS models provide better predictions of the

lower beach face. The SBEACH model allows non-monotonic profile input, while

the EDUNE and CROSS models allow only monotonic profile input at every contour.

The pre-storm profiles were therefore most accurately represented in the SBEACH

model, which led to a slightly better prediction of the overall post-storm profiles.












CHAPTER 1
INTRODUCTION


Purpose

Beaches are constantly evolving in response to waves, tides, sea level rise,

winds, and currents. While the time scale of a beach fill adjustment can take from

months to years, storm induced beach erosion can occur on time scales of only hours

to days and can have significant impact on coastal communities. The increasing use

of the coastal zone makes accurate predictions of beach profile evolution in response

to storm conditions very important in coastal engineering.

This thesis evaluates and compares the predictions of three cross-shore

sediment transport models with two sets of measured post-storm data at the Ft. Pierce

Shore Protection Project area. The three models are SBEACH (Larson and Kraus,

1989, Larson et al., 1989), EDUNE (Kriebel, 1986, 1989), and CROSS (Zheng, 1996,

Zheng and Dean, 1997).


Report Organization

This thesis is organized in the following manner. Chapter 2 provides

background information on the Ft. Pierce Shore Protection Project area and

Hurricanes Floyd and Irene. It also contains a brief review of a similar study

evaluating the predictions of the three models during two storm events at Ocean City,

Maryland. Brief descriptions of the equilibrium beach profile concept and the three

models are provided in Chapter 3. Chapter 4 discusses the model calibration




2

procedure and results. It includes detailed descriptions of the beach profile data,

profile input, storm input, and program options. The model verification procedure

and results are detailed in Chapter 5, which also includes detailed descriptions of the

program inputs. Chapter 6 evaluates and compares the performances of the calibrated

models using several methods and contains a mean grain size sensitivity analysis.

Finally, Chapter 7 presents a summary of this thesis as well as conclusions on the

performances of the three cross-shore sediment transport models.












CHAPTER 2
BACKGROUND INFORMATION AND LITERATURE REVIEW


This chapter presents background information on the Ft. Pierce Shore

Protection Project and brief descriptions of the storms that impacted the project area

shortly after construction. A review of the relevant literature is also presented.


The Ft. Pierce Shore Protection Project

The Ft. Pierce Shore Protection Project is located on the barrier island

shoreline of Hutchinson Island in St. Lucie County, Florida. The project area is

located immediately south of Ft. Pierce Inlet between Florida Department of

Environmental Protection (FDEP) reference monuments R-34 and R-41. Figure 2.1

provides a location map of the project area. The property adjacent to the project area

includes three St. Lucie County parks (South Jetty Park, South Beach Boardwalk, and

Surfside Beach State Park) and a number of private residential lots consisting of

condominiums and single-family homes (Taylor Engineering, June 1997).

The Ft. Pierce Shore Protection Project provides for nourishment of 1.3 miles

of shoreline between FDEP monuments R-34 and R-41. The Ft. Pierce Shore

Protection Project was completed in May 1999 with the placement of approximately

900,000 cubic yards of sand from Capron Shoal. Capron Shoal is located

approximately 3 miles offshore and 2.5 miles south of Ft. Pierce Inlet (Taylor

Engineering, June 1997). The mean sediment grain size of the native beach material





















































Figure 2.1 Ft. Pierce Shore Protection Project Location Map (Taylor Engineering,
2000)






was 0.30 mm (Taylor Engineering, 1997). The mean sediment grain size of the fill

material placed on the beach was 0.41 mm (Taylor Engineering, 2000).

The 1.3 mile project area contains extensive areas ofhardbottom. The

nearshore hardbottom occurs in nearshore parallel bands. The bands appear from

approximately 400 ft to 2,000 ft offshore, are denser in the northern project area, and

are less dense in the southern project area. Taylor Engineering, Inc. estimated that

the Ft. Pierce Shore Protection Project would cover approximately nine acres of the

hardbottom (Taylor Engineering, July 1997).

The construction profile template was one vertical to ten horizontal until

intersection with the existing bottom. The construction template was intended to

naturally evolve to a milder sloped equilibrium profile under the effects of waves and

water levels over a period of approximately one year (Taylor Engineering, June

1997).


Storm Descriptions

After construction of the Ft. Pierce Shore Protection Project in May 1999, two

hurricanes impacted the project area the following summer. Hurricanes Floyd and

Irene impacted the project area in mid-September and mid-October, respectively.

Hurricane Floyd

Hurricane Floyd originated from a tropical wave, which moved off the coast

of Africa and became a hurricane on September 11, 1999. Floyd became a strong

Category IV hurricane on the Saffir/Simpson scale on September 13, with winds of

155 miles per hour (mph), and it ravaged portions of the central and northwest

Bahamas on September 13-14. Floyd posed a serious threat of landfall to the East







Coast of Florida, but it took a turn to the northwest while slowly weakening.

Hurricane Floyd stayed approximately 100 miles off the coast of Florida as it moved

to the north as a Category 4 hurricane (Unisys, 2000). Hurricane Floyd impacted the

Ft. Pierce area during the period of September 12-17, 1999 as a Category 4 hurricane.

A PUV wave gage, located offshore of Jensen Beach, Florida (approximately 10 *

miles north of Ft. Pierce, Florida), in 26 ft of water, recorded a peak significant wave

height of 12.8 ft and a peak wave period of 14.2 s on September 15.

Hurricane Irene

Hurricane Irene originated from a broad area of low pressure in the southwest

Caribbean and became a tropical depression on October 13, 1999. Irene became a

Category 1 hurricane on October 14, and the center passed over Key West, Florida,

early on October 15. The storm moved offshore near Jupiter, Florida, later that day.

Irene retained her hurricane strength and moved northward parallel to the Florida

coast (Unisys, 2000). Hurricane Irene impacted the Ft. Pierce area during the period

of October 13-18, 1999, as a Category 1 hurricane. National Data Buoy Center

(NDBC) Buoy #41009, located east of Cape Canaveral, Florida, in 42 m of water,

recorded a peak significant wave height of 23.3 ft and a peak wave period of 8.0 s on

October 16.


Literature Review

A beach nourishment project can provide substantial storm protection and

erosion control for beach communities. Several researchers have documented the

responses of natural and nourished beaches to storms in order to improve beach fill

design, monitoring procedures, and cross-shore sediment transport models.







Leadon (1999) studied profile data collected along the Panhandle Coast of

Florida before and after Hurricane Opal in 1995. The extensive data set provided an

excellent opportunity to study beach and dune erosion and offshore sand transport

during a severe storm event. As expected with a hurricane, higher erosion occurred

near and to the right of Hurricane Opal's landfall, and decreasing erosion occurred

with increased distance from the point of landfall. The flattest post-storm beach

slopes also occurred nearest to Hurricane Opal's landfall, with increasing slopes seen

with increasing distance from the point of landfall. Survey data collected within two

to three years after Hurricane Opal indicate a very slow natural dune recovery and

very slow landward recovery of sand deposited in deep water during the hurricane.

Stauble (1993) studied three beach fill projects that were impacted by severe

storm events, including two major storms at the Ocean City, Maryland, project area.

He concluded that the storms eroded the beach fill from above the water line and

deposited a majority of the beach fill adjacent to the project between the water line

and the depth of closure. Stauble recommends analyzing pre- and post-storm profiles

at the project area, as well as updrift and downdrift of the project area to analyze the

alongshore sand movement and the variation in the beach responses of the nourished

beach versus the natural beach. The profiles should extend sufficiently far offshore in

order to estimate both the cross-shore and long-shore movement of sand. At the three

project sites, recovery of the sand back onto the subaerial beach took between 4 and

14 months, with between 68% and 100% of the pre-storm subaerial beach volume

returning.







Stauble and Grosskopf(1993) also analyzed the two storms impacting the

Ocean City, Maryland, project. A systematic monitoring program, with special

emphasis on response to major storms, provided an excellent set of profile, sediment,

and storm data. The authors found that the first storm, the 1991 Halloween storm,

had little effect on the constructed storm dune, but did remove sections of the

subaerial beach. Much of the beach fill did remain in the active nearshore area. The

second storm, in early January 1992, removed some sections of the storm dune and

most of the berm along the project area. As with the first storm, nearly all of the

eroded volume of sand remained in the nearshore area, as a low-tide terrace. As early

as the March 1992 survey, sand had been transported back onto the subaerial portions

of the beach at most survey locations.

Kraus and Wise (1993) made use of the extensive profile and storm data at the

Ocean City, Maryland, project area as well to test and improve the cross-shore

sediment transport model, SBEACH. With the first program simulation results, the

authors concluded that overwash occurred at the project area because the wave run-up

was exceeding the dune elevation, and they incorporated overwash into the SBEACH

model. The SBEACH model simulations indicated that the beach fill design had

functioned as predicted.

Zheng and Dean (1997) also made use of the Ocean City, Maryland, data set

to compare the predictions of their new cross-shore sediment transport model with

three existing models. The authors developed a modified non-linear cross-shore

sediment transport model, CROSS, based on equilibrium beach profile concepts and

scaling relationships. The CROSS model was calibrated and compared with the







linear transport relationship based on laboratory experiments. They concluded that

the nonlinear transport model yielded better overall predictions than the linear

transport equation. The models CROSS, EDUNE, and SBEACH were applied along

seven profile lines with the two storms at Ocean City, Maryland. The authors used a

non-dimensional mean square residual parameter to compare the predictions of the

models with the measured data, with smaller residual values indicating better model

performance. The authors found that while all three models underpredict the average

dune erosion, they all provide reasonable predictions for dune erosion and the entire

profile response. The predictions of the CROSS model were the best for an entire

profile and yielded the least average residual. In terms of errors of eroded volume of

sand and beach retreat, all three models provided comparable predictions. They

concluded that by incorporating dune overwash processes in the CROSS model,

which are included in the EDUNE and SBEACH models, the underprediction

associated with the CROSS model could be improved.











CHAPTER 3
BRIEF MODEL DESCRIPTIONS


The three models used in this thesis project are three of the most common

models used to predict cross-shore erosion in the United States. The models include

SBEACH, EDUNE, and CROSS, and all three are based on the equilibrium beach

profile concept discussed next.


The Equilibrium Beach Profile

A beach profile is considered to be in equilibrium when there is a balance

between the constructive and destructive forces. Dean (1977) proposed the following

equation for an equilibrium beach profile assuming that D., the wave energy

dissipation per unit water volume, is the dominant destructive force:

D.(d) (EC,) (3.1)
h dy

where h is the water depth at distance y offshore, E is the local wave energy density,

and Co is the local wave group velocity. Assuming shallow water and spilling wave

breaker assumptions, Equation 3.1 can be integrated to give

2
h(y)= A(d)y (3.2)

where A is defined as the sediment scale parameter in units of m13, which depends on

the sediment size, d, or equivalently, the sediment fall velocity (Dean, 1987). Table

3.1 presents A values for sand in 0.01 mm increments (Dean et al., 1994). Based on

equations 3.1 and 3.2, the sediment scale parameter is given by










A(d)


where K is the breaker index. Dean (1977) analyzed approximately 500 beach

profiles along the Atlantic and Gulf coasts that provide reasonable agreement with

this equilibrium form.




Table 3.1 Summary of Recommended A Values (m"3)

d
d 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
mm
0.1 0.063 0.0672 0.0714 0.0756 0.0798 0.084 0.0872 0.0904 0.0936 0.0968
0.2 0.100 0.103 0.106 0.109 0.112 0.115 0.117 0.119 0.121 0.123
0.3 0.125 0.127 0.129 0.131 0.133 0.135 0.137 0.139 0.141 0.143
0.4 0.145 0.1466 0.1482 0.1498 0.1514 0.153 0.1546 0.1562 0.1578 0.1594
0.5 0.161 0.1622 0.1634 0.1646 0.1658 0.167 0.1682 0.1694 0.1706 0.1718
0.6 0.173 0.1742 0.1754 0.1766 0.1778 0.179 0.1802 0.1814 0.1826 0.1838
0.7 0.185 0.1859 0.1868 0.1877 0.1886 0.1895 0.1904 0.1913 0.1922 0.1931
0.8 0.194 0.1948 0.1956 0.1964 0.1972 0.198 0.1988 0.1996 0.2004 0.2012
0.9 0.202 0.2028 0.2036 0.2044 0.2052 0.206 0.2068 0.2076 0.2084 0.2092
1.0 0.210 0.2108 0.2116 0.2124 0.2132 0.2140 0.2148 0.2156 0.2164 0.2172
(Dean et al., 1994)


SBEACH Model

The SBEACH model incorporates an extension of the breaking wave model

developed by Dally et al. (1985) to compute wave height and energy flux

distributions across the subaqueous beach profile. Four distinct zones with different

sediment transport expressions are identified across the active nearshore zone and are

as follows:

Zone 1: Pre-breaking zone

Zone 2: Transition zone


Zone 3: Broken wave zone


(3.3)






Zone 4: Swash zone

Figure 3.1 presents a summary of the transport relationship used in the four

zones. In all four zones, the direction of transport is determined by the following

criterion, specifying offshore transport when


H< 0.0007 (3.4)
Lo wT

and onshore transport when


Ho >_ 0.0007 HO (3.5)




where Ho is the deep water wave height, Lo is the deep water wave length, w is the

sediment fall velocity, and T is the wave period.

Zone 1, the pre-breaking zone, extends offshore from the break point, and the

sediment transport rate is assumed to decay exponentially with distance from the

break point as expressed by

Q=QBp exp[-A(y0-yB,)] (3.6)

where QBP is the sediment transport at the break point, %1 is the spatial decay

coefficient, and yeB is the break point location. The parameter X1 is expressed by

0.47
= 0.4 do (3.7)


where dso is the median grain size in mm, Hb is the breaking wave height in m, and 1i

is in m1.












































Bfeakina Wave Ihalah ht



pl Average Slope Over
One-Thlrd Local
Wave Length Seawrd
ol Break Point


flun tjlalhtg. ,z

IA7lt!.!Slope

[I = IitialSlope


Wavit Of akha aIA2"

ay- R-IECQ) -(E CoMj

1, aIIM, K'" 0.17
I'm 0.40


Figure 3.1 4-Zone Transport Relationship for SBEACH (Larson, 1989)







Zone 2, the transition zone, is the region between the break point and the

plunge point with a transport rate similar to that in Zone 1 given by

Q = Qpp exp[- (y- ypp)] (3.8)

where Qpp is the transport rate at the plunge point located at ypp. The decay rate, X2, is

expressed by

22 =0.21 (3.9)

Zone 3, the broken wave zone, extends from the plunge point to the seaward

limit of the swash zone, and the sediment transport in this region controls the

sediment transport of the other zones. The sediment transport rate is determined by


Q = +K D. + ,D > D. -- (3.10)


Q O,D< D.--- (3.11)
K

where E is an empirical constant (about 0.0006 m2/s) and K is the transport rate

coefficient.

Zone 4, the swash zone, extends from the seaward swash zone limit to the run-

up limit, where the transport is assumed to decay linearly up to the run-up limit such

as


Q=Q y"YRU (3.12)
YW -YRU

where the subscript RU denotes the run-up limit and WR the landward end of the surf


zone.






The continuity equation,

8h SQ
ah (3.13)
at ay

is combined with the transport relationships for Zones 1 through 4 to solve for the

beach profile evolution.


EDUNE Model

The cross-shore sediment transport rate, Q, in the surf zone is determined

directly from the degree of disequilibrium by

Q = K(D D.) (3.14)

where K is the transport rate coefficient, D is the actual value of the local wave

energy dissipation per unit volume, and D. is the equilibrium value of the wave

energy dissipation per unit volume. The parameter D is defined by


D = pg g h (3.15)
24 By

where K is taken to be 0.78. The parameter D is only dependent on the local water

depth and the bottom slope. During a storm, the water level increase causes the

degree of disequilibrium, and therefore the offshore transport to increase. The sand in

the profile will erode until the wave energy dissipation level decreases to that of the

equilibrium value of the wave energy dissipation.

As in the SBEACH model, the continuity equation, given by

Ox O
(3.16)
at Sh







is combined with the transport rate equation to solve for the beach profile evolution.

The wave break point offshore and the wave run-up limit onshore bound the profile

change.


CROSS Model

The equilibrium concept applied in the EDUNE model is also applied in the

CROSS model. The cross-shore sediment transport rate is determined directly from

the degree of disequilibrium, or the deviation of the actual wave energy dissipation

from the equilibrium wave energy dissipation. However, the transport equation in the

CROSS model is given by

Q =K(D-D.)' (3.17)

where the exponent in the transport relationship is set to three, not one as in the

SBEACH and EDUNE models.

The explanation for the exponent in the transport relationship being set to

three is as follows. The scale relationship of the term (D-D.) is length /2. If the

Froude relationship is applied, the scale relationship is given by

L'2 =
Q- = L, (3.18)
-r

where L, is the length scale and Tr is the time scale, the square root of the length

scale. Therefore, by scaling the transport relationship according to the Froude model,

the exponent of the transport relationship is three.

Once again, the continuity equation, given by

ax (3.19)
at ah





17

is combined with the transport rate equation to solve for the beach profile evolution.












CHAPTER 4
MODEL CALIBRATION


The three models SBEACH, EDUNE, and CROSS were run and calibrated

with data from Hurricane Floyd. The following sections describe the beach profile

data and profile input, storm input, program options specific to each model, and the

calibration procedure and results.


Beach Profile Data and Profile Input

Four sets of beach profile data were used for model calibration, model

verification, and model comparison purposes. The beach profile data sets represent

the project area at pre-fill, post-fill, post Hurricane Floyd (post-Floyd), and post

Hurricane Irene (post-Irene). The pre-fill survey, taken between February 19 and

March 20, 1999, by the Jacksonville District U.S. Army Corps of Engineers

(USACE), represents the project area before the beach fill was in place and is used for

comparison with the post-fill survey. The post-fill survey, taken between June 25 and

July 1,1999, by the USACE, represents the project area immediately after the beach

fill was in place and before the hurricanes influenced the area. For both the pre-fill

and post-fill surveys, profile lines R-34 to R-41 were surveyed to depths of

approximately -25 ft-NGVD.

On October 3, 1999, immediately following Hurricane Floyd, lidar data were

collected along Hutchinson Island by the Department of Environmental Protection,

including along profile lines R-34 to R-41 in the project area. These data were







filtered for buildings and vegetation, and are only accurate to the water line

(approximately 0 ft-NGVD). In November 1999, following Hurricane Irene, Morgan

and Eklund, Inc. performed wading depth surveys for St. Lucie County for profile

lines R-34 to R-41. The depths of these surveys are between 0 and -8 ft-NGVD.

Figure 4.1 illustrates the pre-fill, post-fill, post-Floyd, and post-Irene conditions for

profile line R-37.

The SBEACH model allows non-monotonic profile input, and therefore the

profile data were input exactly from the post-fill survey, including the dune features

and offshore features. The models EDUNE and CROSS allow only monotonic

profile data input at every contour. The post-fill profile data were interpolated at

every contour from the top of the dune to the offshore for input into the EDUNE and

CROSS models.


Storm Input

The three models require storm inputs of some or all of the following: wave

height, wave period, wave direction, water level, and wind speed.

Wave and Water Level Data

A PUV wave gage offshore Jensen Beach, Martin County, Florida, provided

significant wave height, peak wave period, peak wave direction, and water level data

during Hurricane Floyd. The wave gage is located offshore of the Jensen Beach

Holiday Inn at an average water depth of approximately 26 ft. The wave gage

collects wave data for the first 17 minutes of every hour and collects tide data every 7

minutes. After examining the wave data, it was determined that Hurricane Floyd
















- *- Pre-fill - - Post-fill --- Post-Floyd - Post-Irene


Distance from Monument (ft)

Figure 4.1 Measured Profile Data R-37







affected the Ft. Pierce area for 60 hours from September 13 at 4 pm until September

16 at 4 am.

Wave Height

The SBEACH model allows the input of a significant wave height time series.

The significant wave height time series used for the SBEACH model is taken directly

from the wave gage data and is shown in Figure 4.2.

The EDUNE and CROSS models allow the input of a root-mean-square

(RMS) breaking wave height time series. The significant wave height time series

shown in Figure 4.2 was converted to an RMS breaking wave height time series using

the following procedure. The significant wave heights were first changed to RMS

wave heights according to


H,. (4.1)
1.416

where HFIs is the RMS wave height and Hsig is the significant wave height (Dean and

Dalrymple, 1991). The RMS wave height was then changed to a RMS breaking wave

height using linear wave theory. The deep water wave heights were shoaled into

shallow water wave heights using Snell's Law and the Conservation of Energy.

Snell's Law states that

sin O sin b (4.2)
(4.2)
CO Cb

where 0o is the deep water wave direction, Ob is the breaking wave direction, and c is

the wave velocity. The Conservation of Energy is given by

Ho C o coso = Hb2CGb cosOb (4.3)







for straight and parallel contours where Ho is the deep water wave height, Hb is the

breaking wave height, Co is the group velocity, 60 is the deep water wave direction,

and Ob is the breaking wave direction (Dean and Dalrymple, 1991).

McCowan (1984) defines the breaking wave height, Hb, as

Hb = .78x hb (4.4)

where hb is the breaking water depth.

Wave Period

The SBEACH model is the only model to allow the input of a wave period

time series. The wave period time series used for the SBEACH model is taken

directly from the wave gage data and is shown in Figure 4.2.

Wave Direction

The SBEACH model is the only model to allow the input of a wave direction

time series. The wave direction time series used for the SBEACH model is taken

directly from the wave gage data and is shown in Figure 4.3.

Water Level

All three models, SBEACH, EDUNE, and CROSS, allow the input of a water

level time series. The water level time series taken by the wave gage does not include

the effects of barometric pressure, which is added back into the water level according

to

P P = pgAh (4.5)

where P2 is the ambient pressure, P1 is local barometric pressure, p is the density of

water, g is gravity, and h describes the increase or decrease of the water level due to

the pressure difference. The adjusted water level time series is shown in Figure 4.4.







Wind Data

A National Climatic Data Center (NCDC) weather station located at the Ft.

Pierce St. Lucie County Airport, just west of the Ft. Pierce Inlet, collected wind speed

and barometric pressure data during Hurricane Floyd. Data were taken at

approximate one-hour intervals.

The SBEACH model is the only model to allow the input of wind speed time

series. The wind speed time series used for the SBEACH model is taken directly

from the weather station and is shown in Figure 4.5.


Program Options

The following paragraphs describe the options specific to each model.

SBEACH Model

The input wave water depth is set to 26 ft, the depth at which the wave gage

collected the data. The option of irregular waves, which requires an input of a

significant wave height time series, is chosen. The option that all the sand remains on

the grid is chosen, as the post-fill profiles extend sufficiently offshore. The effective

grain size is entered at 0.41 mm. The USACE took post-fill beach samples in

September 1999 and determined the mean grain size diameter was 0.41 mm.

EDUNE and CROSS Models

The sediment scale parameter, A, is entered as 0.1466 mI', corresponding to

the mean grain size of 0.41 mm.


Calibration Procedure

Each model was run with several combinations of the calibration parameters

specific to that model for Hurricane Floyd. The predicted post-Floyd profile was














o-+-Wave Height - Wave Period

14 16


12 14

]12

S10

S* V

10
*m p -. -o


"1 . r M m 8
6 C
I-~



4 4


2 2



0 0
0 10 20 30 40 50 60
Hour


Figure 4.2 Hurricane Floyd Wave Height and Period















50


40


E
L- 30 -
0
O

S20

E




Hoo
0
( 0


0
e -20






-30


-40
0 10 20 30 40 50 60
Hour


Figure 4.3 Hurricane Floyd Wave Direction
















3.5


3


2.5


2


(9 1.5
z



-J,
7@

3 0.5


0


-0.5


-1 -


-1.5 -
0 10 20 30 40 50 60
Hour


Figure 4.4 Hurricane Floyd Water Level












40


35


30


' 25


0 20
C

15


10


5


0


!0 30 4
Hour

Figure 4.5 Hurricane Floyd Wind Speed


60


M /V


A V,




V







compared with the measured post-Floyd profile by using a mean square error

analysis. The mean square error across the entire length of the measured profile, L, is

defined as


C = (hm -hp)2d (4.6)
L L

where hm and hp are the measured and predicted elevations at cross-shore distance x.

The cross-shore distances are taken between consecutively measured post-storm data

points and are relatively evenly spaced. The sum of the cross-shore distances equals

the entire length of the profile. The cross-shore distances were taken at each post-fill

survey data point. A root mean square error was calculated for each profile line for

each run. The average of the root mean square errors for all profiles lines for each

run was used to estimate that run's goodness of fit with the measured data. The

calibrated model was determined by the run with the least average root mean square

error.

SBEACH Model Calibration

The SBEACH model allows for the calibration of four parameters. The

transport rate parameter, K, determines the magnitude of sediment transport and

influences the response time of the profile. Larger values of K result in faster

response times to equilibrium, whereas smaller values of K result in slower response

times. The slope-related transport parameter, s, influences bar formation, with

smaller values resulting in more enhanced bars and larger values resulting in more

subdued bars. The spatial decay factor, X, influences the rate of decay of transport

seaward of the break point, with smaller values of X resulting in slower rates of decay

and wider bars. The angle of advent of avalanching, (, influences the maximum







slope that the eroded profile is allowed to achieve, with larger values resulting in

steeper profiles. The values of the calibration parameters can only be specified within

a particular range of values, detailed in Table 4.1.




Table 4.1 Permissible calibration ranges for the SBEACH model parameters

Parameter Range
K 0.5 x 10 2.5 x 10 m4/N
s 0.001 0.003 m2/s
X 0.1 0.5 m-
150 450

The model was run twelve times, with varying combinations of the calibration

parameters. Table 4.2 details the calibration parameters and average root mean

square error for each run of the SBEACH model. The model was first run with the

default values for the calibration parameters: a K value of 1.75x10.6 m4/N, e value of

0.002 m2/s, X value of 0.5 m"1, and 4 value of 450, producing a root mean square error

of 1.49 ft. A K value of0.50x10-6 m4/N, value of 0.003 m2/s, value of 0.1 m1,

and 4 value of 150 produced the least average root mean square error of 1.42 ft. The

calibrated SBEACH model for this study consisted of these parameter values as

highlighted in Table 4.2. The average residual for each model run is included in the

table for comparison purposes only. The average residual will be discussed further in

Chapter 6, Model Performance.






Table 4.2 SBEACH Model Calibration for Hurricane Floyd

Run Calibration Parameters Average Root Mean Average
K (m4/N) (m2/s) X (m) ) Square Error (ft) Residual
1 1.75 x 10-6 0.002 0.5 45 1.4906 0.5373
2 2.50 x 10-6 0.002 0.5 30 1.6608 0.6541
3 1.75 x 10-6 0.002 0.5 30 1.5017 0.5487
4 1.00 x 10-6 0.002 0.5 30 1.4596 0.5145
5 0.50 x 10-6 0.002 0.5 30 1.4595 0.5030
6 0.50 x 10-6 0.002 0.1 30 1.4557 0.5022
7 0.50 x 10-6 0.002 0.3 30 1.4585 0.5024
8 0.50 x 10-6 0.002 0.2 30 1.4584 0.5036
9 0.50 x 10-6 0.001 0.1 30 1.4615 0.5063
10 0.50 x 10-6 0.003 0.1 30 1.4508 0.4982
~ rX;IiO"4 001 O-T, )A LA J- jj V13j I
12 0.50 x 10-6 0.003 0.1 23 1.4498 0.4977

Figure 4.6 illustrates the post-fill profile, measured post-Floyd profile, and the

SBEACH model's predicted post-Floyd profile at R-37 in the project area. Figures

A. 1 through A.8 in Appendix A present the pre-fill profiles, post-fill profiles,

measured post-Floyd profiles, and the predicted post-Floyd profiles for the SBEACH

model for profiles R-34 through R-41.

EDUNE Model Calibration

The EDUNE model allows for the calibration of four parameters. The

transport rate parameter, K, determines the magnitude of sediment transport. The

dune slope, d, determines the equilibrium eroded profile slope above the run-up limit,

where the larger the value, the steeper the slope. The shoreline slope, b, determines

the equilibrium eroded profile slope below the run-up limit, where the larger the

value, the steeper the slope. The run-up limit, R, controls the location of the dune

scarp above the peak still water flood level, with higher values indicating a higher







run-up limit along the profile. Table 4.3 details the range of values used for the

calibration parameters.




Table 4.3 Calibration ranges for the EDUNE model parameters

Parameter Range
K 0.0001 0.0045 ft4/lb
d 1-5
b 0.05 -0.20
R 1-4ft

The model was run fourteen times, with varying combinations of the

calibration parameters. Table 4.4 details the calibration parameters and average root

mean square error for each run of the EDUNE model. The model was first run with

the default values for the calibration parameters: a K value of 0.0045 ft4/lb, d value of

1.0, b value of 0.05, and R value of 1.0 ft, producing a root mean square error of 2.75

ft. A K value of 0.0003 ft4/lb, d value of 5.0, b value of 0.1, and R value of 3.0 ft

produced the least average root mean square error of 1.46 ft. The calibrated EDUNE

model for this study consisted of these parameter values and as highlighted in Table

4.4. The average residual for each model run is included in the table for comparison

purposes only. The average residual will be discussed further in Chapter 6, Model

Performance.







Table 4.4 EDUNE Model Calibration for Hurricane Floyd

Run Calibration Parameters Average Root Mean Average
K (ft4/lb) d B R (ft) Square Error (ft) Residual
1 0.0045 1 0.05 1 2.7541 1.6070
2 0.0045 2 0.10 2 2.4473 1.2679
3 0.0020 2 0.10 2 2.2628 1.0164
4 0.0010 2 0.10 2 1.9773 0.7338
5 0.0005 2 0.10 2 1.5730 0.5237
6 0.0003 2 0.10 2 1.4819 0.5021
7 0.0001 2 0.10 2 1.8320 0.7021
8 0.0002 2 0.10 2 1.5875 0.5464
9 0.0003 2 0.10 4 1.4798 0.4963
10 0.0003 2 0.10 3 1.4646 0.4925
I:M I L.464t 042
12 0.0003 4 0.10 3 1.4642 0.4923
13 0.0003 5 0.0625 3 1.4714 0.4924
14 0.0003 5 0.20 3 1.5172 0.5140

Figure 4.7 illustrates the post-fill profile, measured post-Floyd profile, and the

EDUNE model's predicted post-Floyd profile at R-37 in the project area. Figures

A.9 through A. 16 in Appendix A present the pre-fill profiles, post-fill profiles,

measured post-Floyd profiles, and the predicted post-Floyd profiles for the EDUNE

model for profiles R-34 through R-41.

CROSS Model Calibration

The CROSS model allows for the calibration of four parameters. The

transport parameter, K, determines the magnitude of sediment transport. The dune

slope is the maximum slope that the profile is allowed to achieve. The shoreline

slope controls the profile evolution between the shoreline and the run-up limit. The

offshore slope controls the slope at the end of the deposited volume of sand offshore.

Table 4.5 details the range of values used for the calibration parameters.







Table 4.5 Calibration ranges for the CROSS model parameters


Parameter Range
K 8.00 xl104 8.00x10-7 ftgs2/lb3
Offshore slope 0.15 0.50
Dune slope 0.5 3.0
Shoreline Slope 0.05 0.20


The model was run twelve times, with varying combinations of the calibration

parameters. Table 4.6 details the calibration parameters and average root mean

square error for each run of the CROSS model. The model was first run with the

default values for the calibration parameters: a K value of 8.0x104 ft8s2/1b3, offshore

slope of 0.15, dune slope of 1.0, and shoreline slope of 0.05, producing a root mean

square error of 3.45 ft. AK value of 7.0x 107 ft8s2/lb3, offshore slope of 0.5, dune

slope of 1.5, and shoreline slope of 0.13 produced the least average root mean square

error of 1.56 ft. The calibrated CROSS model for this study consisted of these

parameter values and as highlighted in Table 4.6. The average residual for each

model run is included in the table for comparison purposes only. The average

residual will be discussed further in Chapter 6, Model Performance.






Table 4.6 CROSS Model Calibration for Floyd


Run Calibration Parameters Average Root Mean Average
K offshore dune shoreline Square Error (ft) Residual
(ft8s2/lb3) slope slope slope
1 8.00x104 0.15 1.0 0.05 3.4524 2.4091
2 3.68x10'7 0.5 3.0 0.2 1.5893 0.5264
3 4.41x10'7 0.5 3.0 0.2 1.5770 0.5181
4 6.00x10'7 0.5 3.0 0.2 1.5649 0.5100
5 8.00x10-7 0.5 3.0 0.2 1.5666 0.5100
6 7.00x10'7 0.5 3.0 0.2 1.5610 0.5074
7 7.00x10-7 0.15 3.0 0.2 1.5679 0.5117
8 7.00x10'7 0.3 3.0 0.2 1.5625 0.5083
9 7.00x107 0.5 0.5 0.2 1.5620 0.5081
10 7.00x10'7 0.5 1.5 0.2 1.5609 0.5073

12 7.00x10.7 0.5 1.5 0.05 1.5598 0.5065

Figure 4.8 illustrates the post-fill profile, measured post-Floyd profile, and the

CROSS model's predicted post-Floyd profile at R-37 in the project area. Figures

A. 17 through A.24 in Appendix A present the pre-fill profiles, post-fill profiles,

measured post-Floyd profiles, and the predicted post-Floyd profiles for the CROSS


model for profiles R-34 through R-41.


















- Post-fill -A- Post-Floyd *- Post-Floyd SBEACH


15






5


0
Z 0
S100 200 300 40 1 500 600 700 800 910
0
=C -5

w \
-10



-15



-20
Distance from Monument (ft)

Figure 4.6 Profile R-37 Measured and Predicted Post-Floyd from SBEACH Model















- - Post-fill --- Post-Floyd - Post-Floyd EDUNE


Distance from Monument (ft)

Figure 4.7 Profile R-37 Measured and Predicted Post-Floyd from EDUNE Model



















- - Post-fill --- Post-Floyd - Post-Floyd CROSS


15


10


5 M





Z 0 100 200 300 400% 500 600 700 800 90
10
5 X,
C)
Z 0







-10



-15



-20

Distance from Monument (ft)


Figure 4.8 Profile R-37 Measured and Predicted Post-Floyd from CROSS Model












CHAPTER 5
MODEL VERIFICATION


The three calibrated models were next run for verification purposes with

Hurricane Irene. The following sections describe the profile input, storm input, and

model verification procedure and results.


Profile Input

The post-Floyd profile data were used as the profile input data for the

Hurricane Irene model runs. The actual post-Floyd data are only available to the

water line, and therefore the profile data are a combination of the actual post-Floyd

data and the predicted post-Floyd data. For each profile in each of the three models,

the actual post-Floyd data to approximately 0 ft-NGVD were spliced with the

predicted post-Floyd data below 0 ft-NGVD. For example, the R-36 profile input in

the EDUNE model consisted of the actual R-36 post-Floyd data down to 0 ft-NGVD

and the predicted (from EDUNE) R-34 post-Floyd data below 0-ft NGVD. Only

profiles R-36 through R-41 were used for the Hurricane Irene analysis because the

spliced profiles for R-34 and R-35 appeared to be anomalous.


Storm Input

As stated in the model calibration chapter, the three models require storm

inputs of some or all of the following: wave height, wave period, wave direction,

water level, and wind speed.







Wave Data

NDBC Buoy #41009, located 20 nautical miles east of Cape Canaveral,

Florida, in an approximate water depth of 42 m, recorded significant wave height and

dominant wave period during Hurricane Irene. The buoy collects data every hour.

After examining the wave data, it was determined that Hurricane Irene affected the

Ft. Pierce area for 42 hours from October 15 at 7 pm until October 17 at 1 pm.

Wave Height

The SBEACH model allows the input of a significant wave height time series.

The significant wave height time series used for the SBEACH model is taken directly

from the buoy data and is shown in Figure 5.1. The significant wave height time

series was converted to an RMS breaking wave height time series using the same

procedure outlined in Chapter 4 for input into the EDUNE and CROSS models.

Wave Period

The SBEACH model is the only model to allow the input of a wave period

time series. The wave period time series used for the SBEACH model is taken

directly from the buoy data and is shown in Figure 5.1.

Wave Direction

The SBEACH model is the only model to allow the input of a wave direction

time series. No wave direction data were available during Hurricane Irene in the Ft.

Pierce area. Therefore, the following procedure was used to produce a wave direction

time series. The wave direction during Hurricane Floyd shows a linear trend (see

Figure 4.3), with the wave direction coming from approximately -200 at the

beginning of the storm, from shore normal at the peak of the storm, and from

approximately + 200 towards the end of the storm. This is consistent with the storm







passing the Ft. Pierce area offshore from south to north. Hurricane Irene took a

similar offshore south to north track as it passed the Ft. Pierce area. The wind

direction time series shown in Figure 5.2 for Hurricane Irene illustrates the south to

north shift of the hurricane, with wind coming off the water as the storm approached

and off the land as it passed by. Therefore, the wave direction time series shown in

Figure 5.3 was approximated using the wind direction and then a linear trend

continuation of the wind direction. The wave direction for the first half of the storm

was taken as the corresponding wind direction. A linear trend was then fit to

correspond to a zero wave direction at the peak of the storm, and at the end of the

storm the wave direction was the opposite of what it was at the beginning of the

storm.

Water Level

As stated earlier, all three models allow the input of a water level time series.

No water level data were available during Hurricane Irene in the Ft. Pierce area.

Therefore, a predictive model was run to produce a time series for water level during

the storm. The model is based on the theory ofa bathystrophic storm tide (Dean,

1991). Longshore currents generated by wind during a hurricane are influenced by

the Coriolis forces. The Coriolis forces require balancing hydrostatic gradients that

will either add to or reduce the surface gradient induced by the wind. A hurricane's

circular wind patterns induce longshore currents in both directions.

The predictive model requires beach profile and storm data for the area and

storm of interest as input. The beach profile data include the range and elevation data

for each of the profile lines R-34 to R-41. The storm data include time series of

location nothingg and eatingg, center (latitude) location, forward velocity, maximum







wind velocity, radius to the maximum wind, and central pressure for Hurricane Irene.

This information was taken from the unisys website (www.weather.unisys.com). The

time series for water level produced by the model was then increased by 60% to

account for wave set-up. The predicted water level time series is shown in Figure 5.4

for Hurricane Irene at the Ft. Pierce Shore Protection Project.

Wind Data

The NCDC weather station located at the Ft. Pierce St. Lucie County Airport

provided wind speed and wind direction data during Hurricane Irene. Data were

taken at approximate one-hour intervals.

The SBEACH model is the only model to allow the input of wind speed time

series. The wind speed time series used for the SBEACH model is taken directly

from the weather station and is shown in Figure 5.2. The wind direction time series

used to estimate the wave direction time series is also shown in Figure 5.2.


Verification Procedure

Each of three calibrated models was run with the profile and storm input

described above. Figures 5.5 through 5.7 illustrate the post-fill profile, measured

post-Floyd profile, measured post-Irene profile, and the model's predicted post-Irene

profile at R-37 in the project area for the SBEACH, EDUNE, and CROSS models.

Figures B.1 through B.6, B.7 through B. 12, and B. 13 through B18 in Appendix B

illustrate the pre-fill profiles, post-fill profiles, measured post-Floyd profiles,

measured post-Irene profiles, and the predicted post-Irene profiles for the SBEACH,

EDUNE, and CROSS models, respectively for R-34 through R-41.














-- Wave Height - -Wave Period

25


8


20 ---------. 7


. 6

S15






10 3V
> 2
S 101




5





0 i 0
0 6 12 18 24 30 36 42
Hour


Figure 5.1 Hurricane Irene Wave Height and Period














-I- Wind Speed - -Wind Direction


ir *i i



*~~~ ------f-----------------------* *-'
V*-* "












---m-m- - 1n n. n--n-nm- --" ----








0 6 12 18 24 30 36 42


150



0

0
100 I
In


50 .2
0)
50 S)


0 )



-50 .|


-100


Hour

Figure 5.2 Hurricane Irene Wind Speed and Direction





















20

E
0
C 10

0
a-

0
E
0


-10



S-20



S-30
TO


40



-50
-50


0 6 12 18 24 30 36 42
Hour


Figure 5.3 Hurricane Irene Wave Direction
















2


1.8


1.6


1.4



C11
>. 1 2

C) L I U"
J.
S0.8


0.6


0.4


0.2


0
0 6 12 18 24 30 36 42
Hour


Figure 5.4 Hurricane Irene Water Level















- - Post-fill -A- Post-Floyd -x- Post-Irene .-- Post-Irene EDUNE


Distance from Monument (ft)

Figure 5.6 Profile R-37 Measured and Predicted Post-Irene from EDUNE Model




















- Post-fill Post-Floyd -x- Post-Irene - Post-Irene SBEACH


15








5-------^----------
0 \
5



z 0
b, 100 200 300 500 600 700 800 9
C





-10



-15



-20

Distance from Monument (ft)


Figure 5.5 Profile R-37 Measured and Predicted Post-Irene from SBEACH Model
















- -- Post-fill --- Post-Floyd -x-- Post-Irene -- Post-Irene CROSS


Distance from Monument (ft)

Figure 5.7 Profile R-37 Measured and Predicted Post-Irene from CROSS Model












CHAPTER 6
MODEL PERFORMANCE


This chapter evaluates and compares the performances of calibrated

SBEACH, EDUNE, and CROSS models using several methods. These methods

include the residual parameter, eroded volume of sand, beach retreat, and contour

change. The models are evaluated individually as well as in comparison to each

other.


Measures of Model Performance

Several methods were used to compare the predicted post-storm profiles with

the measured post-storm profiles. The residual parameter, eroded beach volume,

shoreline retreat at the 6-ft contour, and contour change along the entire profile were

used to evaluate the accuracy of the model predictions.

Residual Parameter

The residual parameter, Res, provides a meaningful comparison of the

measured and predicted profile change. The Res is defined in non-dimensional form

as:

n
S(hp,, hma, )
Res= '=- (6.1)
C(h, h,,)2


where h is the elevation from 0-ft NGVD, the subscripts p and m describe predicted

and measured, the subscripts b and a describe before and after the storm, and the sum






of all the ith locations describes the entire profile. In a perfect model simulation, the

residual parameter would equal zero. The residual parameters are shown in Tables

6.1 and 6.2 for all predicted post-Floyd and post-Irene profiles.

Eroded Volume

The eroded volume of sand in the profile between +8 and 0 ft-NGVD also

provide a valid comparison of the measured and predicted profiles. The upper limit

for the erosion calculations was chosen because the berm height of the post-fill

profile is approximately +8-ft NGVD. Although the overwash experienced from

Hurricane Floyd extends well above +8 ft-NGVD, SBEACH predicts only a very

slight overwash on a few profiles, while EDUNE and CROSS do not predict any

overwash. There will not be an accurate comparison of eroded volume if this

accreted volume of sand is used to compare the measured to predicted eroded

volumes of sand because the net erosion will be quite small. The lower limit for the

erosion calculation was chosen because the post-Floyd survey data only extend to

approximately 0 ft-NGVD. Eroded volumes of sand are shown in Tables 6.1 and 6.2

for all measured and predicted post-Floyd and post-Irene profiles.

Shoreline Retreat

Shoreline retreat at the 6-ft contour was also used to compare the measured

and predicted profiles. The retreat at the 6-ft contour is representative of the erosion

along the beach face between the berm and 0-ft NGVD. The shoreline retreats at the

6-ft contour are shown in Tables 6.1 and 6.2 for all measured and predicted post-

Floyd and post-Irene profiles.







Table 6.1 Predicted Residuals, Measured and Predicted Eroded Volume Between +8
and 0 ft-NGVD and Shoreline Retreat at the 6-ft Contour for Hurricane Floyd

Profile Model Residual Eroded Volume Retreat at 6-ft
Between +8 and 0 ft- Contour (ft)
NGVD (ft2)
R-34 Measured -1150.62 -127.05
SBEACH 1.001 -42.58 -2.77
EDUNE 0.858 -20.43 -3.06
CROSS 0.858 -26.64 -10.32
R-35 Measured -642.86 -71.93
SBEACH 0.653 -395.59 -32.70
EDUNE 0.880 -371.71 -27.12
CROSS 0.878 -356.98 -27.86
R-36 Measured -343.79 -44.20
SBEACH 0.121 -249.06 -36.20
EDUNE 0.075 -245.25 -29.38
CROSS 0.109 -215.69 -30.49
R-37 Measured -179.68 -22.58
SBEACH 0.673 -235.51 -31.99
EDUNE 0.641 -195.87 -23.86
CROSS 0.619 -179.07 -28.50
R-38 Measured -243.88 -28.53
SBEACH 0.335 -197.36 -21.43
EDUNE 0.460 -183.79 -20.21
CROSS 0.464 -169.77 -21.76
R-39 Measured -230.54 -27.48
SBEACH 0.342 -189.03 -30.13
EDUNE 0.394 -203.45 -28.82
CROSS 0.418 -189.44 -29.88
R-40 Measured -347.80 -48.47
SBEACH 0.295 -185.07 -27.38
EDUNE 0.232 -205.51 -28.81
CROSS 0.240 -189.10 -29.71
R-41 Measured -282.46 -35.02
SBEACH 0.371 -194.12 -28.34
EDUNE 0.398 -197.18 -26.63
CROSS 0.457 -177.07 -27.55
Average of Measured -427.70 -50.66
All Profiles SBEACH 0.474 -211.04 -26.37
R-34 to R- EDUNE 0.492 -197.79 -23.49
41 CROSS 0.505 -187.97 -25.76







Table 6.2 Predicted Residuals, Measured and Predicted Eroded Volume Between +8
and 0 ft-NGVD and Shoreline Retreat at the 6-ft Contour for Hurricane Irene

Profile Model Residual Eroded Volume Retreat at 6-ft
Between +8 and 0 ft- Contour (ft)
NGVD (ft2)
R-36 Measured -128.27 -13.33
SBEACH 0.274 -190.55 -21.11
EDUNE 0.312 -117.05 -19.89
CROSS 0.293 -44.70 -1.39
R-37 Measured -72.09 -13.44
SBEACH 0.596 -162.38 -14.62
EDUNE 0.406 -127.12 -23.65
CROSS 0.249 -60.21 -5.88
R-38 Measured -14.52 -2.67
SBEACH 0.568 -194.90 -23.83
EDUNE 0.423 -77.22 -15.65
CROSS 0.305 -38.37 0.11
R-39 Measured -245.88 -17.97
SBEACH 0.612 -117.57 -11.62
EDUNE 0.399 -62.95 -2.12
CROSS 0.566 -47.22 -7.31
R-40 Measured 5.33 6.55
SBEACH 0.774 -163.93 -20.26
EDUNE 0.567 -30.36 -8.43
CROSS 0.449 -23.21 0.46
R-41 Measured 83.58 5.68
SBEACH 1.601 -100.99 -6.08
EDUNE 1.145 -81.07 -16.32
CROSS 0.974 -33.91 -4.52
Average of Measured -61.98 -5.86
All Profiles SBEACH 0.738 -155.05 -16.25
R-36 to R- EDUNE 0.542 -82.63 -14.34
41 CROSS 0.473 -41.27 -3.09


Contour Change Along Entire Profile


Another method of comparison between the measured and predicted profiles

is a contour change analysis. The pre-storm to post-storm contour change is

calculated at every contour along the entire profile. Graphs of the contour change as

a function of the contour illustrate well where sand is eroded from and accreted to

along the profile as well as compare the accuracy of the predicted profiles to the







measured profiles. Figure 6.1 illustrates the contour change as a function of the

contour for the measured and predicted profile data at R-37 for Hurricane Floyd.

Appendix C contains the contour change as a function of the contour for the measured

and predicted post-Floyd and post-Irene profile lines. Figures C. 1 through C.7

compare the measured and predicted post-Floyd contour changes and Figures C.8

through C. 13 compare the measured and predicted post-Irene contour changes. The

measured and predicted post-Floyd to post-Irene contour changes only extend to 0 ft-

NGVD because the measured post-Floyd data only extend to that depth.


Relative Model Performance

The predictions of the three models, SBEACH, EDUNE, and CROSS are

discussed next.

SBEACH Model

On average, the SBEACH model underpredicts the volume of erosion

between +8 and 0 ft-NGVD experienced during Hurricane Floyd, but overpredicts the

erosion during Hurricane Irene. The model performs better, on average, for

Hurricane Floyd than Irene. The average residual for Hurricane Floyd is 0.474, and

for Hurricane Irene is 0.738. Several of the profiles experienced significant overwash

from Hurricane Floyd. The SBEACH model does predict some slight overwash for

profile R-41 from Hurricane Floyd, and predicts the upper beach face better than the

lower beach face for several of the profiles. In several of the post-storm profiles for

both Hurricanes Floyd and Irene, the SBEACH model predicts an offshore bar at

approximately -5 ft to -10 ft-NGVD. Figure 6.2 illustrates a good example of the







slight overwash, the agreement in the upper beach face, and the offshore bar predicted

by the SBEACH model at profile R-41.

The SBEACH model predicts retreat at the 6-ft contour quite well for

Hurricane Floyd. The average retreat for profiles R-36 through R-41 is -34.38 ft, and

the average retreat predicted by the SBEACH model for these profiles is -26.37 ft.

The prediction is not quite as good for Hurricane Irene. The average retreat for

profiles R-36 through R-41 is -5.86 ft, and the average retreat predicted by the

SBEACH model is -16.25 ft.

EDUNE Model

On average, the EDUNE model underpredicts the volume of erosion between

+8 and 0 ft-NGVD experienced during Hurricane Floyd, but overpredicts the erosion

during Hurricane Irene. The model performs slightly better for Hurricane Floyd than

Irene. The average residual for Hurricane Floyd is 0.492, and for Hurricane Irene is

0.542. Although the EDUNE incorporates dune overwash processes, none of the

predicted profiles show any overwash. One explanation for this may be that the dune

is located so far back on the post-fill profile that the predicted erosion did not extend

that far back, and thus there was not overwash of the dune. The EDUNE model

predicts the lower beach face better than the upper beach face and does not predict

any sort of offshore bar feature. Figure 6.3 presents the measured and predicted post-

Floyd profiles at R-41, and provides a representative example.

The EDUNE model predicts retreat at the 6-ft contour reasonably well for

both Hurricanes Floyd and Irene. For Hurricane Floyd, the average retreat for

profiles R-36 through R-41 is -34.38 ft, and the average retreat predicted by the

EDUNE model for these profiles is -23.49 ft. For Hurricane Irene, the average






retreat for profiles R-36 through R-41 is -5.86 ft, and the average retreat predicted by

the EDUNE model is -14.34 ft.

CROSS Model

On average, the CROSS model underpredicts the volume of erosion between

+8 and 0 ft-NGVD experienced during both Hurricanes Floyd and Irene. For

Hurricane Irene, the model predicts very close to the measured eroded volume of

sand. The model performs slightly better for Hurricane Irene than Floyd. The

average residual for Hurricane Floyd is 0.505, and for Hurricane Irene is 0.473. The

CROSS model does not incorporate dune overwash processes, and therefore does not

predict any of the overwash experienced during either hurricane. The CROSS model

predicts the lower beach face better than the upper beach face and does not predict

any sort of offshore bar feature. Figure 6.4 provides a good example of these

predictions at profile R-38.

The CROSS model predicts retreat at the 6-ft contour reasonably well for

Hurricane Floyd and very well for Irene. For Hurricane Floyd, the average retreat for

profiles R-36 through R-41 is -34.38 ft, and the average retreat predicted by the

CROSS model for these profiles is -25.76 ft. For Hurricane Irene, the average retreat

for profiles R-36 through R-41 is -5.86 ft, and the average retreat predicted by the

EDUNE model is -3.09 ft.


Model Comparisons

Tables 6.3 and 6.4 provide summaries of the performances of the three models

for Hurricanes Floyd and Irene. The tables detail the average residuals, the ratio of

the average predicted eroded volume of sand to measured eroded volume of sand







between +8 and 0 ft-NGVD, and the ratio of the average predicted beach retreat to

measured beach retreat at the 6-ft contour. Figure 6.5 illustrates the post-fill profile,

measured post-Floyd profile, and each of the three model's predicted post-Floyd

profiles at R-37 in the project area. Figure 6.6 illustrates the measured post-Floyd

profile, measured post-Irene profile, and each of the three model's predicted post-

Irene profiles at R-37 in the project area.


Table 6.3 Summaries of Model Performances for Hurricane Floyd Model
Calibration

Average Model Residual Ratio of Predicted Eroded Ratio of Predicted
of All Volume to Measured Retreat to
Profiles Eroded Volume Measured Retreat
R-34 to R- SBEACH 0.474 0.493 0.521
41 EDUNE 0.492 0.463 0.464
_CROSS 0.505 0.440 0.509





Table 6.4 Summaries of Model Performances for Hurricane Irene Model
Verification

Average Model Residual Ratio of Predicted Eroded Ratio of Predicted
of All Volume to Measured Retreat to
Profiles Eroded Volume Measured Retreat
R-36 to R- SBEACH 0.738 2.502 2.773
41 EDUNE 0.542 1.333 2.447
CROSS 0.473 0.666 0.527


For Hurricane Floyd, the SBEACH model provided the smallest average

residual, the closest prediction for the average eroded volume of sand, and the closest

prediction for the average beach retreat. From Table 6.1, the SBEACH model's

residual was the smallest for four of the profiles, the EDUNE model's for two of the







profiles, and the CROSS model's for one of the profiles. For eroded volume of sand,

the SBEACH model predicted four of the profiles the closest, the EDUNE model

three of the profiles the closest, and the CROSS model one of the profiles the closest.

The CROSS model predicted the beach retreat the closest for four of the profiles,

while the SBEACH and EDUNE models each predicted the beach retreat the closest

for two of the profiles.

For Hurricane Irene, the CROSS model provided the smallest average

residual, the EDUNE and CROSS models provided the closest predictions for the

average eroded volume of sand, and the CROSS model provided the closest

prediction for the average beach retreat. From Table 6.2, the CROSS model's

residual was the smallest for four of the profiles, while each of the SBEACH and

EDUNE model's residual was the smallest for one of the profiles. For eroded volume

of sand, the CROSS model predicted four of the profiles the closest, while the

SBEACH and EDUNE models each predicted one of the profiles the closest. The

CROSS model predicted the beach retreat the closest for three of the profiles, the

SBEACH model for two of the profiles, and the EDUNE model for one of the

profiles.

The contour change as a function of the contour for the measured and

predicted profile data illustrate well how the models perform along the entire profile.

Figures 6.7 and 6.8 illustrate the average contour changes for Hurricanes Floyd and

Irene. Figure 6.7 illustrates that the models do not predict the overwash experienced

during Hurricane Floyd. Figures 6.7 and 6.8, as well as the individual profile figures,

also illustrate that the SBEACH model predicts the upper part of the beach face better







than the EDUNE or CROSS models, while the EDUNE and CROSS better predict the

lower beach face. The SBEACH model is the only one of the three models to allow

non-monotonic profile input with as many input data points as desired. The EDUNE

and CROSS models allow only monotonic profile input at every contour. Therefore,

the pre-storm profiles were most accurately represented in the SBEACH model. The

profile input for the SBEACH model could describe the irregular shape of the berm,

such as the overwash from Hurricane Floyd. This led to a better prediction of the

overall post-storm profile.

All three of the models fail to predict any erosion for profile R-34, and only

slight erosion for profile R-35 for Hurricane Floyd. Figure 6.9 illustrates the

measured and predicted post-Floyd conditions at R-34 for the SBEACH model, which

is very similar to that predicted in both the EDUNE and CROSS models. These two

profile lines most likely experienced the greatest amount of erosion in the form of

longshore sediment transport due to their proximity to the inlet and jetty (Figure 2.1

in Chapter 2). The only noticeable difference in the post-fill profiles at R-34 and R-

35 compared to the profiles at R-36 though R-41 is their gentler slopes.

Both the SBEACH and EDUNE models overpredict the erosion experienced

during Hurricane Irene. As seen from the measured data, the eroded volumes of sand

from the post-Irene profiles are quite inconsistent. The hurricane caused varied

amounts of erosion at four profiles and accretion at two profiles. At profiles R-37 and

R-38 there was minor erosion, at profiles R-36 and R-39 there was significant

erosion, and at profiles R-40 and R-41 the beach actually saw accretion from

Hurricane Irene. Therefore, on average, Hurricane Irene did not cause a significant






amount of erosion along the project area. The CROSS model consistently predicted

the least amount of erosion of the three models, and therefore, performed the best

overall in terms of beach retreat. It also performed better than the SBEACH model in

terms of eroded volume of sand. The average difference in eroded volume of sand

was comparable for the CROSS and EDUNE models, however, the CROSS model

underpredicted the amount of erosion while the EDUNE model overpredicted.


Mean Sediment Grain Size Sensitivity Analysis

The post-fill mean sediment grain size was 0.41 mm and was used for input

into all three models. The mean sediment grain size is a quantity that is readily

measured in the field. Although the mean sediment grain size of a beach is

representative of the sand on that beach, the beach is actually composed of a range of

sediment grain sizes. In order to test the sensitivity of each model with the grain size,

the following procedure was used. Each of the three calibrated models was run with

varying sediment grain sizes between 0.21 and 0.61 mm at profile R-38 for Hurricane

Floyd.

Table 6.5 provides a listing of the model and the grain size used for each run,

and the corresponding root mean square error and residual for each run. For the

SBEACH model, the post-fill sediment grain size of 0.41 mm provided the least root

mean square and smallest residual in comparison to the finer and coarser sediment

sizes. For the EDUNE and CROSS models, the coarser sediment size of 0.51 mm

provided smaller root mean square errors and residuals than the post-fill sediment

grain size of 0.41 mm. Although a coarser sediment size provided slightly smaller

root mean square errors and residuals in the EDUNE and CROSS models, the







reduction did not justify altering the input sediment size from its measured mean

value.




Table 6.5 Mean Sediment Grain Size Sensitivity Analysis for Profile R-38

Model Mean Sediment Grain Root Mean Square Residual
Size (mm) Error (ft)
SBEACH 0.21 1.4184 0.6160
0.31 1.0691 0.3848
0.41 0.9998 0.3346
0.51 1.3027 0.5057
0.61 1.5720 0.6873
EDUNE 0.21 0.9156 0.4909
0.31 0.8542 0.4613
0.41 0.8449 0.4602
0.51 0.8218 0.4585
0.61 0.8338 0.4623
CROSS 0.21 1.2279 0.4889
0.31 1.2035 0.4668
0.41 1.2039 0.4639
0.51 1.2004 0.4581
__0.61 1.2134 0.4656


















--- Post-fill to Post-Floyd
- -- Post-fill to Post-Floyd EDUNE


* Post-fill to Post-Floyd SBEACH
- -x- Post-fill to Post-Floyd CROSS


Contour Change (ft)


Figure 6.1 Profile R-37 Contour Change for Hurricane Floyd


I


-~---



















I - Post-fill --A- Post-Floyd -m- Post-Floyd SBEACH


15



10




->



5
__ 100 200 300 400 0O 600 700 800 9 0






-10



-15



-20
Distance from Monument (ft)


Figure 6.2 Profile R-41 Post-fill, Post-Floyd, and Post-Floyd SBEACH
















S-. - Post-fill --- Post-Floyd - Post-Floyd EDUNE


Distance from Monument (ft)

Figure 6.3 Profile R-41 Post-fill, Post-Floyd, and Post-Floyd EDUNE
















S - Post-fill --- Post-Floyd -x Post-Floyd CROSS


Distance from Monument (ft)

Figure 6.4 Profile R-38 Post-fill, Post-Floyd, and Post-Floyd CROSS















- -A Post-fill - - Post-Floyd -*--Post-Floyd SBEACH -- Post-Floyd EDUNE -. Post-Floyd CROSS


Distance from Monument (ft)

Figure 6.5 Profile R-37 Measured and Predicted Post-Floyd from Three Models
















-A Post-Floyd - - Post-Irene - Post-Irene SBEACH C- Post-Irene EDUNE - Post-Irene CROSS

15


10


5


Z0 -
a 20 250 300 350 450 500 550 61
:I -5---- *-








-20

Distance from Monument (ft)

Figure 6.6 Profile R-37 Measured and Predicted Post-Irene from Three Models













-+- Post-fill to Post-Floyd -* - Post-fill to Post-Floyd SBEACH
- -A Post-fill to Post-Floyd EDUNE -x- Post-fill to Post-Floyd CROSS


Contour (ft)


Figure 6.7 Average Contour Change for Hurricane Floyd




































I /
o I -i
o















-80 -60 -40 -20 0 20 40
Contour Change (ft)


Figure 6.8 Average Contour Change for Hurricane Irene
















S - - Post-fill -*--- Post-Floyd -x- Post-Floyd SBEACH


Distance from Monument (ft)

Figure 6.9 Profile R-34 Post-fill, Post-Floyd, and Post-Floyd SBEACH












CHAPTER 7
SUMMARY AND CONCLUSIONS


Summary

The predictions of three cross-shore sediment transport models, SBEACH,

EDUNE, and CROSS were compared with measured results at the Ft. Pierce Shore

Protection Project due to Hurricanes Floyd and Irene. All three models are based on

the equilibrium beach profile concept and brief descriptions of each model were

provided. The three models were run and calibrated with Hurricane Floyd data in the

project area along the profile lines of the FDEP reference monuments R-34 to R-41.

The profile input consisted of the post-fill survey data and the storm input consisted

of Hurricane Floyd data. The models were each run several times with varying

combinations of their calibration parameters. The calibrated model was established

by the run with the least average root mean square error of the predicted versus the

measured data across the profile. Next, the three models were run for verification

with Hurricane Irene in the project area. The calibrated models were run with the

post-Floyd survey data as profile input and the Hurricane Irene storm data. Finally,

the performances of the models were evaluated and compared using the residual

parameter, eroded volume of sand, beach retreat, and contour change. A grain size

sensitivity analysis was also performed.







Conclusions

The availability of relevant beach surveys as well as measured storm data at

the Ft. Pierce Shore Protection Project area provided a great opportunity to apply the

three most commonly used cross-shore sediment transport models. A limitation of

the survey profile data employed in this study was its shallow offshore extent. The

south jetty at Ft. Pierce Inlet, located just north of the project area, acts as a littoral

barrier. The limited extent of the offshore profile data did not allow determination as

to whether sand was conserved in the project area profiles from pre- to post-storm,

which is assumed in the three models.

The three models all provided reasonable beach erosion predictions during

Hurricanes Floyd and Irene at the project area. Table 7.1 provides a summary of the

performances of each of the three models for Hurricanes Floyd and Irene. For

Hurricane Floyd, the average residual was between 0.47 and 0.51 for the three

models, the average ratio of predicted eroded volume to measured eroded volume was

between 0.44 and 0.49, and the average ratio of predicted retreat to measured retreat

was between 0.46 and 0.52. For Hurricane Irene, the average residual was between

0.47 and 0.74 for the three models, the average ratio of predicted eroded volume to

measured eroded volume was between 0.67 and 2.50, and the average ratio of

predicted retreat to measured retreat was between 0.53 and 2.77.







Table 7.1 Summaries of Model Performances for Hurricanes Floyd and Irene

Hurricane Floyd Model Calibration
Average Model Residual Ratio of Predicted Eroded Ratio of Predicted
of All Volume to Measured Retreat to
Profiles Eroded Volume Measured Retreat
R-34 to R- SBEACH 0.474 0.493 0.521
41 EDUNE 0.492 0.463 0.464
CROSS 0.505 0.440 0.509
Hurricane Irene Model Verification
Average Model Residual Ratio of Predicted Eroded Ratio of Predicted
of All Volume to Measured Retreat to
Profiles Eroded Volume Measured Retreat
R-36 to R- SBEACH 0.738 2.502 2.773
41 EDUNE 0.542 1.333 2.447
CROSS 0.473 0.666 0.527


As indicated by Table 7.1, the average residual indicates that the SBEACH

model performed the best for Hurricane Floyd, while the CROSS model performed

the best for Hurricane Irene. In terms of eroded volume of sand, the SBEACH model

provided the closest prediction for Hurricane Floyd, while the EDUNE and CROSS

models provided the closest predictions for Hurricane Irene. The SBEACH model

provided the best prediction for beach retreat for Hurricane Floyd, while the CROSS

model provided the best prediction for Hurricane Irene.

Overall, the predictions for Hurricane Floyd were better than the predictions

for Hurricane Irene. One reason for this is that the models were calibrated with

Hurricane Floyd, not with Hurricane Irene, and models were considered calibrated

when the comparison of predicted to measured post-Floyd profiles was the best. The

calibrated models were then tested with Hurricane Irene. Had the models instead

been calibrated with Hurricane Irene, they might have performed better for Hurricane

Irene than for Hurricane Floyd. Another reason for the better predictions for







Hurricane Floyd is that for Hurricane Floyd, all the profile and storm input data were

measured, whereas the wave direction and water level for Hurricane Irene were

estimated, and the profile input was a combination of measured and predicted data. A

final reason for the models performing better for Hurricane Floyd than Hurricane

Irene is the inconsistency in behavior of the post-Irene profiles. Some areas

experienced significant erosion, while others experienced accretion, resulting in an

overall trend of relatively minor erosion. Overall, the models underpredicted the

average volume of erosion during Hurricane Floyd, while for Hurricane Irene the

SBEACH and EDUNE models overpredicted, and the CROSS model underpredicted

the volume of erosion.

The overwash experienced during both storms is not accurately predicted by

any of the models, even though the SBEACH and EDUNE models do incorporate

overwash processes. The predictions of the CROSS model would benefit with the

addition of overwash processes. A drawback of both the EDUNE and CROSS

models is that they allow only monotonic profile input at every contour. The

SBEACH model allows non-monotonic profile input that can capture features such as

overwash much better, and can therefore provide better post-storm predictions. For

this reason, SBEACH performed better than the EDUNE or CROSS models along the

upper beach face. However, the models EDUNE and CROSS performed better along

the lower beach face.

The grain size sensitivity analysis indicated that a coarser grain size than the

measured mean resulted in slightly smaller root mean square errors and residuals in

the EDUNE and CROSS models, but the reduction did not justify altering the input




74

sediment size from the measured mean value. In terms of ease of use, the SBEACH

model has a very user-friendly interface that simplifies model input, runs, and output.

The EDUNE and CROSS models are run in Fortran, and organizing the input files

can be somewhat time consuming.

It is beneficial to the coastal engineering community to apply cross-shore

sediment transport models whenever the necessary profile and storm data are

available. The more times the models can be evaluated under different

circumstances, the more improved the models can become.
























APPENDIX A
MEASURED AND PREDICTED PROFILES FOR HURRICANE FLOYD


















- -- Pre-fill - - Post-fill --- Post-Floyd - Post-Floyd SBEACH


Distance from Monument (ft)


Figure A. 1 Profile R-34 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH


C0
z


0
cc
CU
o\ )
w
















- - Pre-fill - - Post-fill --- Post-Floyd -w Post-Floyd SBEACH


Distance from Monument (ft)

Figure A.2 Profile R-35 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH
















- -- Pre-fill - - Post-fill --- Post-Floyd - Post-Floyd SBEACH


Distance from Monument (ft)

Figure A.3 Profile R-36 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH
















- - Pre-fill - - Post-fill Post-Floyd - Post-Floyd SBEACH


Distance from Monument (ft)

Figure A.4 Profile R-37 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH
















- -* Pre-fill - - Post-fill -*- Post-Floyd - Post-Floyd SBEACH


Distance from Monument (ft)

Figure A.5 Profile R-38 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH



















- 4- Pre-fill - - Post-fill --- Post-Floyd - Post-Floyd SBEACH


15



10



5
>

0 00

100 200 400 500 600 700 800 90

0
. -5


U1
-10



-15



-20
Distance from Monument (ft)


Figure A.6 Profile R-39 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH
















- -- Pre-fill - - Post-fill -A- Post-Floyd - Post-Floyd SBEACH


Distance from Monument (ft)

Figure A.7 Profile R-40 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH

















- -*- Pre-fill - - Post-fill -- Post-Floyd - Post-Floyd SBEACH


m


Distance from Monument (ft)

Figure A.8 Profile R-41 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd SBEACH
















- -*- Pre-fill - - Post-fill -*- Post-Floyd -x Post-Floyd EDUNE


Distance from Monument (ft)

Figure A.9 Profile R-34 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE
















- +- Pre-fill - -- Post-fill -A- Post-Floyd -x Post-Floyd EDUNE


Distance from Monument (ft)

Figure A. 10 Profile R-35 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE
















- +- Pre-fill - - Post-fill --- Post-Floyd -- Post-Floyd EDUNE


Distance from Monument (ft)

Figure A. 11 Profile R-36 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE















- --- Pre-fill - - Post-fill --- Post-Floyd -m- Post-Floyd EDUNE


Distance from Monument (ft)

Figure A. 12 Profile R-37 Pre-fill, Post-fill, Post-Floyd, and Post-Floyd EDUNE




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs