|
UFL/COEL-97/016
SPECTRAL GROWTH OF HURRICANE GENERATED
SEAS
by
William Scott Finlayson
Thesis
1997
SPECTRAL GROWTH OF HURRICANE GENERATED SEAS
By
WILLIAM SCOTT FINLAYSON
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
1997
ACKNOWLEDGMENTS
I would like to thank Dr. Ochi for his considerable patience and instruction. I
would also like to thank Dr. Mehta and Dr. Shepherd for taking time out of their busy
schedules to serve on my committee. Particularly, I am grateful to Dr. Mehta for making
my graduate education a well rounded one. I am also grateful to the United States Navy
for giving me the opportunity to continue my education. Finally, I would like to thank my
mother for supporting me in all my endeavors and I would like to acknowledge the
considerable contribution my father made to my personal and professional development.
This thesis is dedicated to his memory.
TABLE OF CONTENTS
page
ACKNOW LEDGM ENTS ............................................................................................ ii
LIST OF FIGURES............................................... ............................ ..................... ............. iv
A B STR A C T ............................................................................................................ ... vii
1 INTRODUCTION ....................................... .............................. ........................... 1
2 LITERATURE SEARCH .............................................................. ......................6
3 SPECTRUM FORMULATION REPRESENTING HURRICANE
GENERATED SEA S ........................................................... ......................... ... 8
Verification of Modified JONSWAP Spectrum ............................ ..... ........... 8
Comparison with Wave Spectra obtained from Measured Data................................11
4 ESTIMATION AND PREDICTION OF HURRICANE WAVE SPECTRUM
G R O W TH ............................................................ ..................................................58
Estimation of significant wave height........................................... ...58
Prediction of m odal frequency ........................................................ .......................62
Estimation of hurricane wave spectrum growth ................................................ .... 64
5 CON CLU SION S........................................................................... ............................68
A appendix: D ata .......................................................... ..................................................69
LIST OF REFEREN CES ......................................................... ....................................73
BIOGRAPHICAL SKETCH ........................................ ...........................75
11mo
LIST OF FIGURES
Figure page
1 Representative data wave energy spectrum normalized by OOm and analyzed piecewise
for intervals of o/com of 0.05 .............................................................................................12
2 Modified JONSWAP spectrum versus significant wave height for o/0m=0.80.............13
3 Modified JONSWAP spectrum versus significant wave height for co/Om=0.85 .............14
4 Modified JONSWAP spectrum versus significant wave height for c)/M0=0.90.............15
5 Modified JONSWAP spectrum versus significant wave height for (/0m=0.95 .............16
6 Modified JONSWAP spectrum versus significant wave height for o/(0m=1.00.............17
7 Modified JONSWAP spectrum versus significant wave height for 0o/m=1.05.............18
8 Modified JONSWAP spectrum versus significant wave height for o/m=l1.10............19
9 Modified JONSWAP spectrum versus significant wave height for co/0m= 1.15.............20
10 Modified JONSWAP spectrum versus significant wave height for o/0m= 1.20...........21
11 Modified JONSWAP spectrum versus significant wave height for o0/0m=1.25 ..........22
12 Modified JONSWAP spectrum versus significant wave height for co/0m=1.30.........23
13 Modified JONSWAP spectrum versus significant wave height for O/0om=1.35 ..........24
14 Modified JONSWAP spectrum versus significant wave height for )/c0m=1.40...........25
15 Modified JONSWAP spectrum versus significant wave height for o/0om=1.45...........26
16 Modified JONSWAP spectrum versus significant wave height for )/o0m=1.50...........27
17 Modified JONSWAP spectrum versus significant wave height for co/Cm=l1.55...........28
18 Modified JONSWAP spectrum versus significant wave height for o/o)m=1.60...........29
19 Modified JONSWAP spectrum versus significant wave height for o/COm=1.65...........30
iv
20 Modified JONSWAP spectrum versus significant wave height for co/0m=1.70...........31
21 Modified JONSWAP spectrum versus significant wave height for (o/0m=1.75...........32
22 Modified JONSWAP spectrum versus significant wave height for co/Om=1.80...........33
23 Modified JONSWAP spectrum versus significant wave height for (o/COm=1.85 ...........34
24 Modified JONSWAP spectrum versus significant wave height for co/0m=1.90...........35
25 Modified JONSWAP spectrum versus significant wave height for (o/COm=1.95...........36
26 Modified JONSWAP spectrum versus significant wave height for 0o/Om=2.00...........37
27 Modified JONSWAP spectrum versus significant wave height for co/om=2.05...........38
28 Modified JONSWAP spectrum versus significant wave height for o/0m=2.10...........39
29 Modified JONSWAP spectrum versus significant wave height for Co/(0m=2.15 ...........40
30 Modified JONSWAP spectrum versus significant wave height for co/m=2.20...........41
31 Modified JONSWAP spectrum versus significant wave height for co/Om=2.25...........42
32 Modified JONSWAP spectrum versus significant wave height for /Om=2.30 ...........43
33 Modified JONSWAP spectrum versus significant wave height for c)/om=2.35...........44
34 Comparison between Modified JONSWAP and measured spectrum for Hurricane
B elle, H =3.2 m ................................................................................................... 47
35 Comparison between Modified JONSWAP and measured spectrum for Hurricane
B elle, H s=6.1 m ................................................................................................... 48
36 Comparison between Modified JONSWAP and measured spectrum for Hurricane
B elle, H s=7.1 m .........................................................................................................49
37 Comparison between Modified JONSWAP and measured spectrum for Hurricane
G loria, H =6.0 m ........................................................................................................50
38 Comparison between Modified JONSWAP and measured spectrum for Hurricane
G loria, H =8.1 m .................................................................................................51
39 Comparison between Modified JONSWAP and measured spectrum for Hurricane
Eloise, H =5.6 m .................................................................................................52
40 Comparison between Modified JONSWAP and measured spectrum for Hurricane
E lose, H =8.8 m ........................................................................................................53
v
41 Comparison between Modified JONSWAP and measured spectrum for Hurricane
Fredrick, H =4.5 m ............................................................................................... 54
42 Comparison between Modified JONSWAP and measured spectrum for Hurricane
Fredrick, H =5.5 m ............................................................................................... 55
43 Comparison between Modified JONSWAP and measured spectrum for Hurricane
Fredrick, H =8.5 m ............................................................................................... 56
44 Comparison between Modified JONSWAP and measured spectrum for Hurricane
K ate, H =10.7 m ................................................................................................... 57
45 Relationship between mean wind speed and significant wave height obtained in
various hurricanes (from Ochi, 1993) .....................................................................60
46 Relationship between mean wind speed and significant wave height, Tropical Cyclone
Gloria (solid) (Ochi, 1993) and a North Atlantic Storm (hollow) (Sneider and
Chakrabari, 1973..................................................................................................61
47 Relationship between mean wind speed and fetch................................................63
48 Dimensionless wind speed as a function of dimensionless fetch (based on Ross, 1980)
........................................... ................ ...................................................65
CHAPTER 1
INTRODUCTION
The purpose of this study is to evaluate the wave spectral energy growth of
hurricane associated seas. The best approach to achieve this goal is to analyze the shape
of wave spectra obtained from measured data at various stages of hurricane growth.
However, it is difficult in practice to derive a general conclusion by evaluating the
difference in the magnitude of energy for a specified wave frequency at different stages of
growth. One way to overcome this difficulty is to evaluate the growth of wave energy
through spectral formulation. It is then necessary to have a wave spectral formulation
which represents well the wave energy spectrum throughout the growing stage of
hurricanes. A good candidate is the Modified JONSWAP formulation developed by
Foster (1982) and Ochi (1993). However, the Modified JONSWAP formulation must be
verified to confirm it is valid over the entire frequency domain of the wave energy
spectrum for hurricane generated seas.
In order to use the Modified JONSWAP formula as the basis for evaluating the
growth of hurricane associated wave spectra, it is highly desirable to examine the validity
of the spectral formulation over the entire frequency domain. The subject is discussed in
detail in Chapter 3. A brief explanation of the Modified JONSWAP formulation for
wave spectra for hurricane generated seas is given. Most importantly, the Modified
JONSWAP formula is verified throughout the entire frequency domain. Also, wave
spectral data is compared to the Modified JONSWAP formulation and to existing spectral
formulation to show how well it agrees with observed hurricane data.
Since the Modified JONSWAP formula is a function of significant wave height
and modal frequency, and the design criterion usually specified is wind speed, a method
is prescribed in Chapter 4 to estimate modal frequency and significant wave height as
functions of wind speed. And therefore, growth of hurricane wave energy spectra can be
estimated through the Modified JONSWAP formula.
Before analyzing wave energy spectra for hurricanes, it is useful to understand the
atmospheric phenomenon which give rise to the wave energy spectra. Atlantic
hurricanes, or tropical cyclones, are most often formed from tropical low pressure
disturbances leaving the West Coast of Africa. The Atlantic Hurricane Season is from
May to November. Large convective energy from warm water feeds these tropical cells
as they are carried across the Mid-Atlantic. A characteristic Coriolis driven cyclonic
rotation develops to maintain the low-pressure disturbance as the storm becomes a
tropical depression. This rotation becomes more organized, the center's atmospheric
pressure drops, and wind speed increases. Strengthening, the tropical depression is
upgraded to a tropical storm and then to a hurricane. Based on maximum wind speeds
and, to a lesser extent, on the atmospheric pressure in the eye, these storms are classified
according to the Saffir-Simpson Scale given in Table 1.
Several favorable existing environmental conditions must be in place for a
tropical disturbance to grow to hurricane strength. Hurricanes require warm ocean waters
greater then 800 F and a large negative atmospheric temperature gradient. This
temperature gradient creates an unstable condition for moist convection and drives the
Table 1 Saffir-Simpson Hurricane Scale
Type Category Damage Pressure Winds
Hg (in) mph
Depression >35
Tropical storm 39-73
Hurricane 1 minimal >28.94 74-95
Hurricane 2 moderate 28.50-28.91 96-110
Hurricane 3 extensive 27.91-28.47 111-130
Hurricane 4 extreme 27.17-27.88 131-155
Hurricane 5 catastrophic <27.17 >155
necessary massive thunderstorms which releases thermal-energy contained in the ocean
water. Hurricanes cannot exist at distances any closer to the equator then 500 km, as a
non-negligible Coriolis force is necessary to offset the low pressure disturbance. A pre-
existing near-surface disturbance with sufficient vorticity and convergence is required.
Tropical cyclones cannot be generated spontaneously. To develop, they require a weakly
organized system with sizable rotation and low level inflow.
Although large storms can create sea-severity comparable to hurricanes, there are
characteristic atmospheric features of hurricanes which differentiate them from ordinary
large Atlantic storms. The low-pressure center, maintained by a Coriolis force, induces a
high rate of cyclonic rotation, causing extreme localized wind speeds. In contrast, this
low pressure center, known as the eye, has virtually no associated wind. The strongest
hurricanes have the lowest associated central atmospheric pressure and the smallest eye
diameter. This well defined eye can be used to track the location of the storm. Although
hurricanes can effect large areas, they are a local geostrophic phenomenon, driven by
greater atmospheric forces. Low and high-pressure fronts move hurricanes at speeds in
excess of 25 knots.
In contrast to hurricanes, ordinary storms are not as well organized and do not
have an easily defined center. Consequently, their location cannot be precisely tracked.
Ordinary storms are relatively stationary and have little rotation, so they are characterized
as having quasi-steady, non-localized winds blowing in a general direction over an
established fetch length.
To design a structure to operate in hurricane prone areas, an accurate prediction of
growth of hurricane generated wave spectra at a specific location with respect to the
storm is necessary. Because of differences in atmospheric characteristics, hurricanes
cause wave spectra to grow differently than for ordinary storms. When characterizing
hurricane associated events several factors are considered, in general, including
maximum sustained winds, radius of maximum winds, central barometric pressure,
distance to the storm, speed of the storm and others. However, it has been found that
shape and growth of wave spectra are generally dependent upon mean wind speed
measured at a known distance above the sea surface and a characteristic fetch length over
which the wind is acting. It is convenient to define the fetch length as the distance from
the eye to the design location.
Since hurricanes center are rapidly moving, wind speed at a fixed location and
fetch length change rapidly. This extreme increase in wind speed and change in fetch
length have the most profound effect on hurricane wave spectra shape and growth. In
5
contrast to normal wave spectra, hurricane wave spectra have a more pronounced peak at
the modal frequency, where more energy is contained. This is due to the rapid increase
of wind speed driving the growth. Wave energy is concentrated primarily in the
neighborhood of the peak frequency during hurricanes as contrasted with the energy
spread over a wide frequency range, including double peaks, for wave spectra obtained
during ordinary storms (Ochi, 1993). Sea-severity cannot develop in the high frequency
range as quickly as wind speed. Unlike for ordinary storms, where modal frequency
clearly migrates from higher to lower frequencies during spectral growth, hurricane wave
spectra grow predominately at lower modal frequencies. There is some leftward
migration of modal frequency with growth, but it quickly becomes concentrated at a low
frequency.
CHAPTER 2
LITERATURE SEARCH
Limited research has been done on finding a mathematical representation of
hurricane wave energy spectra. Cardone, Pierson and Ward (1976), Bretschneider and
Tamaye (1976), Young and Sobey (1981), Ross and Cardone (1978) and Young (1988)
have carried out studies on hurricane-generated seas, primarily through hindcasting and
forecasting approach. These techniques provide valuable insight, but they cannot be used
to predict growth. There have been many wave spectral formulation developed to
represent sea severity, none of which predict hurricane spectra well. The Pierson-
Moskowitz (1964), the two parameter, the three parameter, the six parameter (Ochi and
Hubble, 1976) and the JONSWAP (Hasselmenn et al., 1973) are all useful for
representing ordinary wind generated seas.
Antani (1981) compared the shape of nearly 400 hurricane wave spectra to various
mathematical formulations, such as the one parameter (Pierson and Moskowitz, 1964),
the two parameter, the three parameter, the six parameter (Ochi and Hubble, 1976) and
the JONSWAP spectra (Hasselmann et al., 1973). His results showed that the six
parameter and JONSWAP most closely represent hurricane wave spectra shape. The
JONSWAP is a better candidate in that it produces a single spectrum and was developed
based on fetch limited data. It has also been suggested as the formulation of choice for
hurricanes by Lee (1980), Ross(1976), and Whalen and Ochi (1978).
7
Foster (1982) and Ochi (1993) developed a Modified JONSWAP formula. They
analyzed data from several hurricanes and represented the original JONSWAP
coefficients in terms of significant wave height and modal frequency. Ochi (1993)
compared the Modified JONSWAP formula to wave spectra data from several hurricanes
and showed that it was a good representation of the hurricane associated wave spectrum.
CHAPTER 3
SPECTRUM FORMULATION REPRESENTING HURRICANE GENERATED SEAS
As stated in Chapter 1, the Modified JONSWAP spectral formula will be used as
the basis for evaluating wave spectral growth for hurricane generated seas is to have a
wave spectral formulation which represents hurricane generated seas at various stages of
growth. The Modified JONSWAP formulation was derived based on peak energy
hurricane data and seems to represent hurricane wave spectra well. It is important to
validate the Modified JONSWAP formula over the entire frequency domain at various
stages of hurricane growth in order to determine how well it represents hurricane
generated seas.
Verification of Modified JONSWAP Spectrum
Ochi and Foster have done considerable work in determining the coefficients for
the JONSWAP spectral formulation for hurricane generated seas. A limited explanation
is provided here to outline the process used and to illustrate assumptions made during the
derivation. For a complete understanding of the determination of these parameters the
reader is directed to Foster (1982) and Ochi (1993).
The original JONSWAP formulation is given by
2 f _1)2/ 2 ,2(1 )
S(f) = a (2 exp -1.25 '[ 2
where. y = peak shape parameter, 3.30 as an average
a = 0.076 (x)-022
a = 0.07 for ooxom
fm = 3.5 (gU)(x)-033
x = dimensionless fetch = gr/U2
r = fetch length
U = mean wind speed
Foster analyzed the values of the parameters in the original JONSWAP
formulation, Equation (1), and determined functional relationships for a and the peak
shape parameter, y. The parameter a is a function of dimensionless fetch length and, due
to the rapidly changing wind speeds, is difficult to evaluate for hurricanes. However,
analysis showed a can be presented as the function of significant wave height, Hs, and
modal frequency, fm, given in Equation (2) (Foster, 1982). The constant, a, carries the
units of (sec2/meter)2.
a = 4.5H f4 (2)
The peak energy, S(fm), of the hurricane data was determined by analyzing S(fm)
as a function of Hs and found to be represented by
S(f,) = 0.75H.34. (3)
The peak energy, at f=fm, of the JONSWAP spectrum is determined from
Equation (1) and is given as
S(f) = a(2) exp(-125). (4)
From Equations (2) through (4) the peak shape parameter, y, can be shown to be a
function of significant wave height, Hs, and modal frequency, fm, as
y = 9.5H"4f, (5)
It is noted that y was determined using peak spectral energy, which is a key feature
of this derivation. Using these relationships Equation (1) can be written as a function of
co as
S(W) =( (Hg) exp{- 1.25 } H "m 22 (6)
where a = 0.07 for cocm.
Equation (6) will be referred to as the Modified JONSWAP formulation. It was
developed by determining y from data at the modal frequency. Consequently it is highly
desirable to confirm the formulation is valid over the entire frequency domain of the
spectrum through comparison with measured data.
In order to compare spectra to each other having various modal frequencies, the
frequency range for each energy density spectrum used for the current study was
normalized by dividing by modal frequency, cm. Each spectrum was then broken into
sections from om/Om=0.80 to o)/om=2.35 at intervals of 0.05. Figure 1 demonstrates the
procedure. The result gives values of spectral energy, S(co), known for each spectrum at
80% to 235% of the modal frequency, at 5% intervals. The corresponding values of
spectral energy for each co/o, are plotted versus significant wave height in Figures 2
through 34.
On each figure is a plot of the Modified JONSWAP formula at corresponding
normalized modal frequencies. The figures illustrate how well the Modified JONSWAP
formula represents hurricane data taken during the growing stage at various sea severities
(significant wave heights) throughout the frequency domain. The figures demonstrate the
utility of the Modified JONSWAP formula. Even though derived by analyzing peak
spectral energy (o)/om=1.00, Figure 6), it agrees reasonably well with values of spectra
data throughout the normalized frequency domain. Figure 27 shows even at twice the
modal frequency, the Modified JONSWAP formula adequately agrees with data. At
0/0om>2.00 the data begins to scatter but the magnitude of S(o) is substantially reduced in
comparison with that at the peak frequency.
Comparison with Wave Spectra obtained from Measured Data
Although some examples of comparison between the wave spectra obtained from wave
data during hurricanes and the Modified JONSWAP spectral formulation are given in the
references (Ochi 1993, for example) it may be of considerable interest to show
comparisons at various stages of hurricane intensities. Included in the comparisons are
the Pierson-Moskowitz and the two parameter formulations. Since these formulations
were developed for ordinary wind generated sea it may be of interest to examine how well
they represent hurricane generated sea conditions. Figures 34 through 44 show
comparison between selected data to the Modified JONSWAP formulas.
12 - -
10
6 -___- - - - - - - - --
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4
Figure 1 Representative data wave energy spectrum normalized by om and analyzed
piecewise for intervals of 0/Om of 0.05
~ U
100.00
10.00
E
U'
0.
CO
1.00
0.10
m,
____ --U--
-I-
10
Significant Wave Height, H, (
Significant Wave Height, H, (m)
1 Data Modified JONSWAP Trendline
Figure 2 Modified JONSWAP spectrum versus significant wave height for o/(Om=0.80
/
'
'"
14
100.00
10.00 -1
1.00
5/
w
0.10
110 100
01 - -Da --- - ______ - -
Significant Wave Height, H, (m)
SData Modified JONSWAP Trendline
Figure 3 Modified JONSWAP spectrum versus significant wave height for o/0m=0.85
100.00
1 10 100
Significant Wave Height, H, (m)
Data Modified JONSWAP Trendline
Figure 4 Modified JONSWAP spectrum versus significant wave height for 0/c0m=0.90
100.00 ~
S S10.00eg
10.001~ ------- -- - ~- -------- --- -- -
1--.00 ---
1 10
1.00 -- ---- -4 -- - -------- -- _
0.10- -
Significant Wave Height, H, (m)
N Data Modified JONSWAP Trendline
igure 5 Modified JONSWAP spectrum versus significant wave height for o/om=0.95
100.00
10.00
N0
E
-i
8
w
0)
Cl)
to
1.00
0.10
Figure 6 Modified JONSWAP spectrum versus significant wave height for o/B0m=1.00
-U
_ _- U----
10 10
Significant Wave Height, H, (m)
Data Modified JONSWAP Trendline
'E
100.00
10.00
) __ ------ -- -- -
0
w
1.00
0.10
1 10 100
Significant Wave Height, H. (m)
[ Data Modified JONSWAP Trendline
Figure 7 Modified JONSWAP spectrum versus significant wave height for mo/0m=1.05
19
100.00
10.0010
E
0.10
1.00 ---- -- - - ** - -
0.10 ---- -- -------- --- ----------- - - --
1 10 100
Significant Wave Height, H, (m)
Data Modified JONSWAP Trendline
Figure 8 Modified JONSWAP spectrum versus significant wave height for co/cm=1.10
1 10 100
Significant Wave Height, H, (m)
E Data Modified JONSWAP Trendline
Figure 9 Modified JONSWAP spectrum versus significant wave height for 0/0m=1.15
100.00
10.00---
IRJ I
1.00
l-5
0.10
1 10 100
Significant Wave Height, H, (m)
SData Modified JONSWAP Trendline
Figure 10 Modified JONSWAP spectrum versus significant wave height for c/om=1.20
100.00
10.00 -
5at -- *- -- -- ---JNWA Tedln
1.00
SData Modified JONSWAP Trendline
0.10Figure Modified JONSWAP spectrum versus significant wave height for =1.25
1 10 100
Significant Wave Height, H8 (m)
Data Modified JONSWAP Trendline
Figure 11 Modified JONSWAP spectrum versus significant wave height for /(0m=1.25
100.00
10.00
NIN
C4
E
C) -
C-O
1.00 --
0.10
1 10 100
Significant Wave Height, H, (m)
| Data Modified JONSWAP Trendline
Figure 12 Modified JONSWAP spectrum versus significant wave height for O/0m=1.30
100.00
10.00 -
1.00-1
1 10 100
Significant Wave Height, H, (m)
0.10 ----- ------ ---- U
Data Modified JONSWAP Trendline
Figure 13 Modified JONSWAP spectrum versus significant wave height for o/cm=1.35
100.00
10.00
SI/
C
w
1.00 --
0.10
1 10 100
Significant Wave Height, H, (m)
I Data Modified JONSWAP Trendline
Figure 14 Modified JONSWAP spectrum versus significant wave height for 0/Am=1.40
100.00
10.00
E
C
w
t5
0
v,
uJ
1.00
U?
_ _ -U_____ _
---U--- -- -
*
/ 1
i/
=- !
10 100
Significant Wave Height, H, (m)
| Data Modified JONSWAP Trendline
Figure 15 Modified JONSWAP spectrum versus significant wave height for (o/cm=1.45
100.00
10.00-
.. I
i *
0.10
1 10 1 00
Significant Wave Height, H, (m)
Co-
1.00 -- i-
-U---
0.10
10 100
Significant Wave Height, H, (m)
SData -- Modified JONSWAP Trendline
Figure 16 Modified JONSWAP spectrum versus significant wave height for co/cm=1.50
28
100.00
10.00 -
v,
I E
(0I
MI -
1.00
01 ---- -- ---- ---------I--i-I-I
0.10
1 10 100
Significant Wave Height, H, (m)
Data Modified JONSWAP Trendline
Figure 17 Modified JONSWAP spectrum versus significant wave height for o/Om=1.55
100.00
10.00 I--
Es ----- ---__ ---- -- _
w
t5
im
1.00 -
0.10
1 10 100
Significant Wave Height, H, (m)
a Data Modified JONSWAP Trendline
Figure 18 Modified JONSWAP spectrum versus significant wave height for Co/COm=1.60
100.00
10.00
- -- - -
E
1.00
0.10
10 100
Significant Wave Height, H, (m)
Data --Modified JONSWAP Trendline
Figure 19 Modified JONSWAP spectrum versus significant wave height for O/(Om=1.65
2 ~ ~ ~ - -- -- --- -- --- -- __-- -
0.10 ---- ----- ---- ---- -- - -
Figure 19 Modified JONSWAP spectrum versus significant wave height for o)/I0m=1.65
31
10.00 -
7'
--*/--- -
1.00=
U
0.10
1 10 100
Significant Wave Height, H, (m)
0 Data Modified JONSWAP Trendline
Figure 20 Modified JONSWAP spectrum versus significant wave height for o/cm=1.70
Ug
Figure 20 Modified JONSWAP spectrum versus significant wave height for C0/Cm=1 .70
32
10.00
S11 N11
NI
0.10
u5.5
^ 1 0 __- _ ^ _-------- -
1 10 100
Significant Wave Height, H, (m)
SData Modified JONSWAPTrendline
Figure 21 Modified JONSWAP spectrum versus significant wave height for (o/om=1.75
10.00
0.10
II
110 100
SData -ModifiedJONSWAPTrendline
u -II
C" ,/
I-
1 10 140
Significant Wave Height, H8(i)
,, ---- -- L c _---- _
uj "____ _____"_
a~~~~~ |z- ~T"~. -
0.10 ---- -----------
Figure 22 Modified JONSWAP spectrum versus significant wave height for o/Cm=1.80
10.00
'8
w
vi,
0.10
0.10
F-U
"II
----- ---__ _-_- ---- ---- - ---_
----- -- -J- -- --- ---- --- ---
"/
10 10
Significant Wave Height, H, (m)
Data Modified JONSWAP Trendline
Figure 23 Modified JONSWAP spectrum versus significant wave height for o/cm=1.85
N'
v1
1.oo
w
I,
uL
o
1 10 100
Significant Wave Height, Hs (m)
S Data Modified JONSWAP Trendline
Figure 24 Modified JONSWAP spectrum versus significant wave height for o/cOm=1.90
10.00
> I q
ii '*
1.00 ---
C,
0.10--
1 10 100
Significant Wave Height, H, (m)
Data -- Modified JONSWAP Trendline
Figure 25 Modified JONSWAP spectrum versus significant wave height for a0/0m=1.95
10.00
-I
CI
iil
S1.00 --
& -----U-------
0.10
1 10 100
Significant Wave Height, H, (m)
Data Modified JONSWAP Trendline
Figure 26 Modified JONSWAP spectrum versus significant wave height for oC/cOm=2.00
10.00
N I
o 1.00 --L
4)1
LLI
-s -- Utt- -- ---
Data -Modified JONSWAP Trendline-
0.10Figue 27 M d JP s m v s st w e h t fr
1 10 100
Significant Wave Height, H. (m)
Data Modified JONSWAP Trendline
Figure 27 Modified JONSWAP spectrum versus significant wave height for G)/am=2.05
10.00
U)
p 1.00 ----l
w
0.10
1 10 100
Significant Wave Height, H, (m)
SData ---Modified JONSWAP Trendline
Figure 28 Modified JONSWAP spectrum versus significant wave height for co/o0m=2.10
10.00
1 n
~ 1.o00o
SH LI
0.10
1 10 100
Significant Wave Height, H. (m)
I Data -- Modified JONSWAP Trendline
Figure 29 Modified JONSWAP spectrum versus significant wave height for o/0m=2.15
10.00
E
E/ *
) ___I_______
o1 .oo0
0..
0.10 ---
1 10 100
Significant Wave Height, H. (m)
SData Modified JONSWAP Trendline
Figure 30 Modified JONSWAP spectrum versus significant wave height for co/m=2.20
42
10.00
(D
t 1.00
wm
i/
u i
0.10 -
1 10 100
Significant Wave Height, H, (m)
Data Modified JONSWAP Trendline
Figure 31 Modified JONSWAP spectrum versus significant wave height for o/Om=2.25
10.00
E
i m
S1.00 I -
c ------- U .--
0.10-
1 10 100
Significant Wave Height, H, (m)
S Data -Modified JONSWAP Trendline
Figure 32 Modified JONSWAP spectrum versus significant wave height for co/tom=2.30
UI
II
UI
0.10 H II _____ -
1 10 10
SiniicntWae eiht H1(n
I aa-MdiidJNWA rnln
Fiue3 odfe OSWPsetumvru igiiatwaehih1fr0023
44
10.00
E
CO
l-oo ---
I I
7E
L. ___---__s f -----_
0.10
1 10 100
Significant Wave Height, H, (m)
a Data --Modified JONSWAP Trendline
Figure 33 Modified JONSWAP spectrum versus significant wave height for co/(m=2.35
The Pierson-Moskowitz and the two parameter spectra have been included to show how
well they represent hurricane generated wave spectra.
The two parameter and Modified JONSWAP both require a given modal
frequency and significant wave height. For the plots in Figure 34 through 44 the
corresponding modal frequency from the data presented in the figure is used. For this
reason the two parameter and Modified JONSWAP have the same modal frequency as the
data. On the other hand, the Pierson-Moskowitz formula requires only significant wave
height. Modal frequency is then a function of significant wave height.
Wave spectra data for hurricane Belle (Figures 34 through 36) and Gloria
(Figures 37 and 38) are similar in that both storms grew after several days of moderate
sea severity. For Belle at a significant wave height of 3.2m (Figure 34), a second peak
can be seen at 1 rad/sec. This represents a significant preexisting sea severity before the
hurricane. As the hurricane grows to a significant wave height of 6.1m (Figure 35) and
then to 7.1m (Figure 36) this second peak becomes small in comparison to the energy
contained at the modal frequency. As Gloria grows from a significant wave height of
6.0m (Figure 37) to 8.1m (Figure 38) a similar second peak is present and then
disappears because of energy associated with the hurricane dominates the energy pre-
existing in the spectrum.
Gloria at a significant wave height of 8.1m shows a good comparison between the
Pierson-Moskowitz, the two parameter and the Modified JONSWAP. Before Gloria
passed the NOAA buoy, a continues wind of 7-11 m/s blew for 10 days so the sea had a
sufficient amount of time to be fully developed (Pierson et al., 1958). Consequently, at
this stage in the growth of Gloria, the Pierson-Moskowitz formula predicts the same
modal frequency as the data, but slightly under predicts the shape of the spectra and
maximum energy. The two parameter formula gives similar results. Belle and Gloria
both grow significantly at their modal frequencies, which are concentrated in the lower
end of the frequency domain.
The wave spectral growth of Eloise from a significant wave height of 5.6m
(Figure 39) to 8.8m (Figure 40) shows a similar trend as that of Fredrick from 4.5m
(Figure 41) to 5.5m (Figure 42) and then to 8.5m (Figure 43). They exhibit growth
predominantly at the modal frequency, again located at lower frequencies. In all stages of
growth for all storms, the data shows spectra with a sharp peak at the modal frequency
which is represented well by the Modified JONSWAP formula. This peak becomes more
pronounced at higher significant wave heights which can be seen in Kate at a significant
wave height of 10.7m (Figure 44).
Plotting individual spectra shows the obvious advantage of using the Modified
JONSWAP formula to represent wave spectra for hurricane generated seas, as it
represents maximum energy and shape reasonably well at various hurricane intensities.
The figures show that the Pierson-Moskowitz and two parameter formulas both under
predict maximum values and give an extended shape at the modal frequency. The
Pierson-Moskowitz and two parameter formulations do not, in general, represent
hurricane generated sea state.
2.0
Belle
8/9/76
0:00 Modified
1.8 Hs=3.2m JONSWAP
Parameter
1.6
1.4
N
E 1.2 Pierson-
SMoskowitz
CO
2 1.0
I 0.8
0.6 o-
0.4-
0.2
0.0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
Frequency, o (rad/sec)
Figure 34 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Belle, H,=3.2 m
7.0-
6.0-
E
CO
S5.0
E
2
I5 4.0
2)
-
3.0-
2.0
1.0
0.0
0.3
0.9 1.0 1.1
Figure 35 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Belle, Hs=6.1 m
0.4 0.5 0.6 0.7 0.8
Frequency, co (rad/sec)
49
14.0
Belle
8/9/76 o
6:00
Hs=7.1m
12.0
Modified
JONSWAP
10.0
o kowitz
Two Parameter
8.0
4.0
2.0
0.0-
I-
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
Frequency, eo (rad/sec)
Figure 36 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Belle, Hs=7.1 m
8.0
Gloria o Modified
9/26/85 /\JONSWAP
7:00
Hs=6.0l Parameter
7.0
Pierson-
oskowitz
6.0
5.0
N
2 4.0
2.0
oo
Lu 3.0 o-
0
0
2.0 O
1.0
0.0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
Frequency, w (rad/sec)
Figure 37 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Gloria, H,=6.0 m
18.0
Modified
16.0
Gloria
9/26/85
14:00 o
Hs=8.lm
14.0 Parameter
Pierson-
Moskowitz
12.0
E
3 10.0
t 8.0
CO
2.0
0.0-
LU6.0------------P----------
00 00
2.0 ---- 0---------------------
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Frequency, co (rad/sec)
Figure 38 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Gloria, H,=8.1 m
7.0
Eloise Modified
9/22/ JONSWAP
22:00
Hs=5.6m
6.0-
erson-
skowitz
5.0
5. Two arameter
N
4.0
3.0
a-
2.0
1.0
0.0
0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
Frequency, o (rad/sec)
Figure 39 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Eloise, Hs=5.6 m
25.0
Eloise
9/23/75
1:00 Modified
Hs=8.8m JONSWAP
20.0
Piers n-
os
S15.0
T Parameter
E
S10.0
5.0
0.0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3
Frequency, c (rad/sec)
Figure 40 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Eloise, H,=8.8 m
4.5
Modified
JONSWAP
4.0
Fredrick
9/11/79
12:00
Hs= 4.5m
3.5
iersor& oskowitz
3.0
T Parameter
j2.5
w 2.0
1.5
1.0
0.5
0.0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3
Frequency, o (rad/sec)
Figure 41 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Fredrick, Hs=4.5 m
7.0
Fredrick
Modified
9/11/79
5:0 AJONSWAP
15:00
Hs=5.5m a
6.0
person
skowi
5.0
SParameter
N
4.0
2
> 3.0
2.0
1.0
0.0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3
Frequency, o (rad/sec)
Figure 42 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Fredrick, Hs=5.5 m
20.0
18.0 Modified
Fredrick JONSWAP
9/11/79
22:00
16.0 Hs=8.5m
16.0\
Pie son
14.0
T Parameter
S) 8.0 -"-- t -- \ \ A--------------
i--v--
4.0 -----------N -------
12.0
10.0
4.0
0.0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3
Frequency, a (rad/sec)
Figure 43 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Fredrick, Hs=8.5 m
Hurricane Fredrick, H=8.5 mn
35.0
Modified
ONSWAP
30.0
o\ Kate
S 11/20/85
Piers n- 117:00
osk tl Hs=10.7m
25.0
ST Parameter
20.0
15.0
10.0
10.0
5.0
0.0 1
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
Frequency, co (rad/sec)
Figure 44 Comparison between Modified JONSWAP and measured spectrum for
Hurricane Kate, H,=10.7 m
CHAPTER 4
ESTIMATION AND PREDICTION OF HURRICANE WAVE SPECTRUM GROWTH
This chapter discusses a method for estimating the growth of sea severity
(significant wave height) and modal frequency when a hurricane is approaching a specific
location from knowledge of wind speed at that location and distance to the storm. The
method will provide significant information for design of offshore structures. As
presented in the previous chapter, hurricane-generated are represented by the Modified
JONSWAP formula given as a function of significant wave height and modal frequency.
For design use, these parameters will be presented in terms of wind speed.
Estimation of significant wave height
Ordinary storms are relatively stationary and associated wind speeds change at a
much slower rate and blow over an established quasi-steady fetch length. Wind-
generated seas in hurricanes differ from ordinary storms in that the hurricane is moving
from 5 to 12 knots with respect to a specific location. This causes a high rate of change
of wind speed blowing over a rapidly changing fetch. Ochi (1993) addressed this topic
and developed wind speed-sea severity functional relationships for two cases; the
growing stage of hurricane-generated seas in which the wind speed is increasing at an
extremely high rate but the sea severity was comparatively moderate because the sea
condition prior to the hurricane was very mild, the second is the sea condition resulting
from continuous winds of mild severity blowing for one week or longer then followed by
a storm, usually a tropical storm which has a wind speed much less then a hurricane.
For an extremely high rate of increase in wind speed following a very mild sea,
analysis shows an almost linear increase in sea severity with increase in wind speed. The
significant wave height during the growing stage of the hurricane is a function of mean
wind speed at a 10-meter level given as Equation (7) and presented in Figure 45.
H. = 0235U10 (7)
For a tropical storm following continues mild winds blowing for one week or
longer, sea severity increases rapidly with increase in wind speed. The significant wave
height-wind speed relationship seems to be approximated by Equation (8) and is
presented in Figure 46. Equation (8) is derived from the Pierson-Moskowitz formulation
corrected for a wind speed at a 10-meter height and making a narrow-band assumption.
H, = 0.237(U'0/g) (8)
Note that the Pierson-Moskowitz spectral formulation is only applicable to fully
developed seas. For the sea to become fully-developed at high wind speeds and
corresponding large significant wave heights, a significantly long duration is needed
(Pierson-Moskowitz, 1964). For example, 42 hours is required for a Uio of 20 m/s.
Hurricanes do not behave in this manner. Therefore even though the significant wave
height-wind speed relationship for the tropical storm close to that given in Equations (8).
The shape of the spectrum is different from that of the Pierson-Moskowitz spectrum. For
large wind speeds associated with hurricanes, U1o>33 m/s, Equation (8) would predict
unrealistically high significant wave heights which observation does not support.
60
60
SELOISE /
A FREDERIC
v ANITA
BELLE (100 KTS)-
50- o KATE -,
a CARMEN
CAMILLE
( ]SEVEREST
WIND SPEED
40 /*_"ZZ ------ -- ^ -----
0 (75 KTS) --H
| 3 -----------^ -------- -----
W E9
o ] Ol- HURRICANE
0) 0e o
o *0
20 -
%" % FULLY DEVELOPED SEA
10 ---
0 2 4 6 8 10 12 14
SIGNIFICANT WAVE HEIGHT IN M
Figure 45 Relationship between mean wind speed and significant wave height obtained in
various hurricanes (from Ochi, 1993)
20
w FULLY DEVELOPED SEA
cL
0
1 o
//
0 4 8 12 16
SIGNIFICANT WAVE HEIGHT IN M
Figure 46 Relationship between mean wind speed and significant wave height, Tropical
Cyclone Gloria (solid) (Ochi, 1993) and a North Atlantic Storm (hollow) (Sneider and
Chakrabari, 1973
In summary, Equation (7) should be used to estimate significant wave height from
knowledge of wind speed at a specified location. Equation (8) applies to specific cases
involving tropical storms following several days of steady mild wind. However, it
predicts unrealistically high significant wave heights at wind speeds associated with
hurricanes.
Prediction of modal frequency
The method given here to predict a trend in change in modal frequency with
respect to wind speed is useful in representing growth of hurricane spectra. Unlike for
fully developed conditions, where modal frequency is directly related to significant wave
height, a modal frequency-significant wave height relationship is difficult to establish for
hurricane associated wave spectra. However, a relationship can be found by first
estimating the relationship between fetch length and wind speed. Then modal frequency
can be estimated through obtaining the relationship between wind speed and modal
frequency.
Figure 47 is a plot of fetch length versus wind speed for the various hurricanes
considered in this study. It shows a general fetch length wind speed relationship trend
given by Equation (9).
r = 5.4x107U;U (9)
At small fetch lengths there is a small but finite wind speed. As fetch length
decreases wind speed increases. At large wind speeds fetch length approaches 10 km, a
good approximations of the eye diameter, where wind speed is a maximum. Specifying
wind speed yields a theoretical fetch length.
500,000
0
450,000
400,000
-01-
0
350,000
0
-250,000 --' ----------- ---
300,000
0
150,000 : : :
100,000 A o
50,000 5
S. I i i r
100,000 --0---
50,000 ----------B 4^,------r=5E
10 20 30 40
Wind Speed, U10 (m/s)
o Belle u Eloise A Fredrick o Gloria a Kate
Figure 47 Relationship between mean wind speed and fetch
Next, the relationship between fetch length and modal frequency of hurricane
associated wave spectra is discussed. Ross (1976) developed a relationship between the
dimensionless modal frequency, u, and the dimensionless fetch, 6, from the analysis of
three storms. This is given in Equation (10). Data considered in the present study
compares well to this relationship. Figure 48 is a plot of data and Equation (10), which
shows good agreement.
o = 0.97.2'1 (10)
where
wUo
1-
27rg
rg
U210
U'10
Thus, from Equation (9) and (10) it is possible to estimate the modal frequency of
hurricane associated wave spectra from knowledge of wind speed.
Estimation of hurricane wave spectrum growth
Hurricane wave energy spectra growth for various wind speeds is now estimated.
The following procedure may be used to estimate Hs and com needed for the Modified
JONSWAP formula; 1) Specify Uo1, 2) calculate H, using Equation (7), 3) calculate r
using Equation (9), and 4) knowing Ulo and r, calculated om using Equation (10).
Figure 49 is a plot of hurricane energy spectrum, S(co), for wind speeds from
20m/s to 60m/s and is useful in showing a trend in spectral growth. The plot shows the
characteristics of hurricane spectral growth. Modal frequency is concentrated at lower
values throughout growth and most growth takes place at the modal frequency.
1.00 I I
L ----- : I ,
u=.974-.21
: i P ! i :
0 1 ; ; : : .. 0 .1 . o
S 43
0)
0.10
:o0P
fo
0-
i f
0.01
10 100 1,000 10,000 100,000
r,=rg/Uo2
o Belle 0 Eloise A Fredrick o Gloria a Kate
Figure 48 Dimensionless wind speed as a function of dimensionless fetch (based on Ross,
1980)
60 -- =60 m/s
H=14.1 m
50
0=55 m/s
s=12.9 m
40
10=50 m/
s=11.8 m
30 -I =45 m/
S30
S//10.6
2
20
s 8.
10
0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1
Frequency, w (rps)
Figure 49 Predicted hurricane spectrum growth showing mean wind speed and
corresponding significant wave height
To show the growth of the wave spectra with respect to significant wave height,
the Modified JONSWAP formula is differentiated with respect to significant wave height
(2)4H, exp 1.25 H. 4m
x 2 + 0.34exp (CL- if 2&}
showing growth is proportional to significant wave height at m=mm as,
SCO H1.34
and at co>> m as,
oc H.
Clearly, hurricane spectral growth occurs most predominantly at the modal
frequency.
CHAPTER 5
CONCLUSIONS
This study dealt with evaluating wave spectra energy growth for hurricane
generated seas at various stages of intensities to develop a mathematical formulation
representing a trend in the growth of the wave energy spectrum. Since it is difficult to
derive a general conclusion by evaluating the difference in the magnitude of energy for a
specified wave frequency at different intensities from analysis of spectra obtained from
field data, a mathematical formula is needed which represents hurricane wave spectra at
various stages of growth. The Modified JONSWAP formula is a good candidate, but
needed to be verified over the entire frequency domain. In Chapter 3, it was verified
throughout the frequency domain by comparison to hurricane wave spectra from
measured data and was shown to represent the wave spectrum associated with hurricane
generated seas for various stages of intensity.
The Modified JONSWAP formula is a function of significant wave height and
modal frequency. To use it to present a trend in the growth of the wave spectrum,
mathematical relationships were presented in Chapter 4 to present significant wave height
and modal frequency as functions of wind speed. Using these relationships a plot of the
trend in growth of the wave energy spectrum was given for wind speeds of 20 to 60 m/s.
The plot shows the characteristics of hurricane wave spectrum growth. Modal frequency
is located at lower values throughout and is where most growth takes place.
APPENDIX
DATA
General
The data used for this study were taken from National Oceanographic and
Atmospheric Administration's (NOAA) Data Buoys. Atmospheric and oceanographic
data has been collected by NOAA Data Buoys since the early 1970's. They are generally
located in deep water locations and collect a wide variety of atmospheric and general
oceanographic data in addition to wave data. Steele and Johnson (1977) describe the
payload and operation of the buoys. There is considerable information at the NOAA
world wide web homepage at http://www.nhc.noaa.gov. The data used consists of an
hourly energy density spectrum, significant wave height determined by integration, and
wind speed for five Atlantic and Gulf of Mexico hurricanes. The purpose of this study is
spectral growth, so the data was generally reduced to reflect increasing significant wave
height with increasing wind speed. Hurricane path information was obtained at
http://www.nhc.noaa.gov/tracks.html Following are narratives taken from yearly
NOAA Mariners Weather Logs for each storm considered.
Hurricane Belle (August 9, 1976) (Mariners Weather Log, 1977)
Belle reached hurricane strength at 1800 on 7 August, 1976 as it moved across the
Atlantic in the trade wind belt just east of the Northern Bahamas. Belle's maximum
strength was attained on August 9 with sustained winds of 105 knots and a minimum sea
level pressure of 957 mb. While moving northward, parallel to the U.S. East Coast, Belle
passed almost directly over NOAA buoy EB 15 (32N, 75.20W). The largest significant
wave height measured was 7.1 m with a wind speed of 30.7 m/s. Belle made landfall on
1000 on August 10 on southern Long Island with sustained winds of 65 knots and
minimum pressure of 980 mb. Belle caused tides three feet above normal and four to five
inches of rainfall into the mountains of New England. Five fatalities were associated
with Hurricane Belle.
Hurricane Eloise (September 22-23, 1975) (Mariners Weather Log .1976)
The disturbance from which Eloise formed left the African coast on September 6.
Winds reached tropical strength early on September 16. Eloise intensified rapidly and
reached minimal hurricane strength before it struck the northeastern coast of the
Dominican Republic. Fifty-nine deaths occurred in the area, with 34 in Puerto Rico; and
damage was estimated at $60 million. Rainfall amounts of 10-20 inches were common
over eastern and southeastern Puerto Rico. Eloise weakened as it tracked westward
across the mountains of Hispanola and eastern Cuba. By 19 September Eloise was barely
a tropical storm as it passed over the Yucatan.
Eloise began a steady strengthening north of the Yucatan Peninsula, regaining
hurricane force in the central Gulf of Mexico about 350 miles south of New Orleans on
the morning of 22 September. At 0300 on 23 September the eye passed within 10 miles
of NOAA buoy EB10 (27.50N, 88.00W). The largest significant wave height measured
was 8.8 m with a wind speed of 35.1 m/s. It continued to strengthen as it reached landfall
midway between Fort Walden Beach and Panama City, Florida, shortly after 1200 on 23
September. A sustained wind near 80 knots with a gust to 135 knots was measured on a
98 foot tower 13 miles offshore. Hurricane tides of 12 to 16 feet were measured. Four
deaths in Florida were indirectly attributed to Eloise. The hurricane caused $500 million
in loss of property and crops in Florida, Alabama and over the northeastern U.S. due to
flooding.
Hurricane Fredrick (September 11-12, 1979) (Mariners Weather Log, 1980)
Fredrick was the first hurricane to strike Mobile, Alabama directly since 1926.
The central pressure of 946 mb and maximum sustained winds of 115 knots made
Fredrick the most intense hurricane to affect Mobile this century. The highest wind
reported in the U.S. was a gust to 126 knots on Dauphin Island bridge in Alabama. The
peak storm surge of 12 feet over Gulf Shores, Alabama destroyed most of the Island. An
11 foot surge destroyed the Dauphin Island causeway. Five deaths were attributed to
Fredrick. The estimated damage totaled $2.3 billion. Data were obtained from NOAA
buoy 42003 (26.00N, 86.00W). The largest significant wave height measured was 8.9 m
with a wind speed of 37.0 m/s.
Hurricane Gloria (September 25-26, 1985) (Mariners Weather Log, 1986)
Gloria had its start near the Cape Verde Islands on 16 September. On 22
September, hurricane Gloria turned to the northwest as it approached the Leeward
Islands. The hurricane weakened as it passed Cape Hatteras after midnight on 27
September. Ten hours later, Gloria crossed western Long Island, New York and emerged
back over the open waters of the far North Atlantic. As Gloria again approached the East
Coast of the U.S. it passed within 60 miles of the NOAA buoy 41002 (32.30N, 75.30W) at
2000 on 26 September. The buoy measured a significant wave height of 14.3 m with a
wind speed of 25 m/s.
Hurricane Kate (November 20, 1985) (Mariners Weather Log, 1986)
Development of Kate began just northeast of the Virgin Islands when a weak
tropical wave began to interact with an upper air trough on 13-14 November. The
tropical storm began moving in a general westerly direction and reached hurricane
strength by 1800 on 16 November. Kate moved onto the north central Cuban coast on the
19th and emerged over the waters of the southeastern Gulf of Mexico. It passed the
NOAA buoy 42003 (26.00N, 86.0W) at 1700 on 20 November. The largest significant
wave height measured was 10.7 m with a wind speed of 47.3 m/s. Kate weakened
slowly before making landfall near Mexico Beach, Florida on 21 November.
LIST OF REFERENCES
Antani, J. K. (1981), "Mathematical Representation of Hurricane Associated Wave
Spectra," UFL/COEL-81/007, University of Florida, Gainesville.
Bretschneider, C. L., and Tamaye, E. (1976), "Hurricane Wind and Wave Forecasting
Techniques," Fifteenth Coastal Engineering Conference, American Society of
Civil Engineers, New York, pp. 202-237.
Cardone, V. J., Pierson, W. J., and Ward, E. G. (1976), "Hindcasting the Direction
Spectra of Hurricane-Generated Waves," Journal of Petroleum Technology, Vol.
28, 1976, pp. 385-394.
Foster, E. R. (1982), "JONSWAP Spectral Formulation Applied to Hurricane-Generated
Seas," UFL/COEL-81/004, University of Florida, Gainesville.
Hasselman, K., Barnett, T. P., Bouws, E., Carlson, H., Cartwright, D. E., Enke, K.,
Ewing, J. A., Gienapp, H., Hasselmen, D. F., Kruseman, P., Meerburg, A.,
Muller, P., Olbers, D. J., Richter, K., Sell, W., and Walden, H. (1973),
"Measurements of Wind-Wave Growth and Swell Decay during the Joint Sea
Wave Project (JONSWAP)," Deutsches Hydrographisches Institut, Hamburg.
Lee, Y. Keen (1980), "Hurricane Eloise Wave Spectra," Coastal Engineering, Vol. 4, pp.
151-156.
National Oceanographic and Atmospheric Administration, Mariners Weather Log (1976),
"Hurricane Eloise, September 13-24," Vol. 20, No. 2, p. 69.
National Oceanographic and Atmospheric Administration, Mariners Weather Log (1977),
"Hurricane Belle, August 6-10," Vol. 21, No. 2, p. 65.
National Oceanographic and Atmospheric Administration, Mariners Weather Log (1980),
"Hurricane Fredrick, August 29 September 14," Vol. 24, No. 2, p. 100.
National Oceanographic and Atmospheric Administration, Mariners Weather Log (1986),
"Hurricane Gloria, September 16 October 2," Vol. 30, No. 1, p. 9.
National Oceanographic and Atmospheric Administration, Mariners Weather Log (1986),
"Hurricane Kate, November 15-23," Vol. 30, No. 1, p. 11.
Ochi, M. K. (1993), "On Hurricane-Generated Seas," Proceedings Second Symposium
on Ocean Wave Measurement and Analysis, American Society of Civil Engineers,
pp. 374-387
Ochi, M. K. and Hubble, N. E. (1976), "Six-Parameter Wave Spectra," Proc. Fifteenth
Coastal Engineering Conference, American Society of Civil Engineers, Honolulu,
Hawaii, July 11-17.
Pierson, W. J. and Moskowitz, L. (1964), "A proposed Spectral Form for Fully
Developed Wind Seas Based on the Similarity Theory of S.A. Kitaigorodski,"
Journal of Geophysical Research, Vol. 69, No. 24, pp. 5181-5190.
Pierson, W. J., Neuman, G., and James, R. W. (1958), "Observing and Forecasting Ocean
Waves by Means of Wave Spectra and Statistics," U.S. Navy Hydrographic Office
Publication No. 603.
Ross, D. (1976), "Observing and Predicting Hurricane Wind and Wave Conditions,"
Seminar on Ocean Products and IGOSS Data Processing and Services System
(IDSS), Moscow, U.S.S.R, April 2-11.
Ross, D. and Cardone, V. J. (1978), "A comparison of Parametric and Spectral Hurricane
Wave Prediction Products," in Turbulent Fluxes through the Sea Surface, Wave
Dynamic, and Prediction, Plenum Press., New York.
Sneider, R. H. and Chakrabari, S. K. (1973), "High Wave Conditions Observed Over the
North Atlantic in March 1968," Journal of Geophysical Research, Vol. 78, No. 36,
pp. 8793-8807.
Steele, K. and Johnson, A. Jr. (1979), "Data Buoy Wave Measurements," in Ocean Wave
Climate, Plenum Press, New York, 1979.
Whalen, J. E. and Ochi, M. K. (1978), "Variability of Wave Spectra Shapes Associated
with Hurricanes," Offshore Technology Conference, Paper No. 3228, Houston,
Texas, May 6-8.
Young, I. R., (1988), "Parametric Hurricane Wave Prediction Model," Journal of
Waterway, Port, Coastal and Ocean Engineering, Vol. 114, No. 5, 1988, pp. 637-
652.
BIOGRAPHICAL SKETCH
William Scott Finlayson was born on December 27, 1969, in Alamogordo, New
Mexico and grew up in Arizona and New Mexico. He attended New Mexico State
University in Las Cruces and received a Bachelor of Science degree in mechanical
engineering in May 1991. He entered the Navy through the Navy Civil Engineer
Collegiate Program in 1989 and received his commission through Officer Candidate
School in Newport, Rhode Island, in November 1991.
After attending Civil Engineer Corps Officer School in Port Hueneme,
California, he was assigned to Marine Corps Air Station Iwakuni, Japan as the Facilities
Maintenance Officer and later as the Maintenance Control Officer, from April 1992
through April 1994. LT Finlayson reported to Naval Mobile Construction Battalion
SEVEN, Gulfport, Mississippi in April 1994 and served as Officer in Charge for Details
deployed to Sasebo, Japan, and El Salvador. He was assigned to the University of Florida
in September 1996 as part of the Civil Engineer Corps Ocean Facilities Program of the
United States Navy and will received a Master of Science degree in coastal and
oceanographic engineering in December 1997.
|
RSS
HIDE
