Coastal Engineering 56 (2009) 479–483

Contents lists available at ScienceDirect

Coastal Engineering
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c o a s t a l e n g

Characteristics of developing waves as a function of atmospheric conditions, water

properties, fetch and duration
J.P. Le Roux ⁎
Departamento de Geología, Facultad de Ciencias Físicas y Matemáticas, Universidad de Chile, Casilla 13518, Correo 21, Santiago, Chile

a r t i c l e i n f o a b s t r a c t

Article history: For any specific wind speed, waves grow in period, height and length as a function of the wind duration and
Received 18 December 2007 fetch until maximum values are reached, at which point the waves are considered to be fully developed.
Received in revised form 1 September 2008 Although equations and nomograms exist to predict the parameters of developing waves for shorter fetch or
Accepted 28 October 2008
duration conditions at different wind speeds, these either do not incorporate important variables such as the
Available online 2 December 2008
air and water temperature, or do not consider the combined effect of fetch and duration. Here, the wind
Keywords:
conditions required for a fully developed sea are calculated from maximum wave heights as determined from
Developing waves the wind speed, together with a published growth law based on the friction velocity. This allows the
Wave period parameters of developing waves to be estimated for any combination of wind velocity, fetch and duration,
Wave height while also taking account of atmospheric conditions and water properties.
Wavelength © 2008 Elsevier B.V. All rights reserved.
Fetch
Wind duration

1. Introduction 2. Atmospheric conditions and water properties

Estimation of long-term wave conditions is of vital importance in The characteristics of gravity waves are not only a function of the wind
ocean and coastal engineering, whereas the forecast of short-term conditions, but also depend on the physical properties of the water and air,
conditions is critical for maritime activities. While fully developed sea in particular the difference between the water and air temperature
conditions (FDS) are often assumed in long-term forecasts, such a state is (Geernaert et al., 1986). This difference (Δ°C) has a direct influence on the
not necessarily reached in enclosed or semi-enclosed water bodies such drag force generated by wind friction on the water surface, so that it is
as lakes or bays, where a short fetch will limit wave development. This is necessary to determine the drag for different atmospheric conditions. The
particularly true of storms, because higher wind speeds require a longer wind friction velocity Ua⁎ (the subscripts a and w referring to air and water,
fetch to produce FDS waves. For short-term forecasts, the wind duration respectively) is estimated by Demirbilek et al. (1993) as follows:
also plays an important role both on the open sea and in smaller water
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
bodies. Developing waves may be more hazardous than in their fully Ua4 = Cda Ua2 ð1Þ
developed state, because in spite of being lower they are normally much
steeper, which poses a danger especially for smaller vessels.
where the dimensionless wind drag coefficient Cda is given by
Although much work has been done on wave growth as a function
of wind speed, fetch or duration (e.g. Inoue, 1967; Barnett, 1968; Cda = 0:001ð1:1 + 0:035Ua Þ: ð2Þ
Bunting, 1970; Hasselmann et al., 1973; Toba, 1978; Forristall et al.,
1978; Resio and Vincent, 1979; Resio, 1981, 1987, 1988; Kahma, 1981; Eq. (2) gives an approximation of the drag coefficient for “normal”
Kitaigorodskii, 1983; Hasselmann et al., 1985; Resio and Perrie, 1989; weather conditions, but does not take the air–water temperature
Cardone, 1992; Van Vledder and Holthuisjen, 1993; Demirbilek et al., difference (Δ°C = °Ca − °Cw) into account. A more rigorous solution is
1993), a simple model taking account of all three variables at the same given by a 3-D graph in Resio et al. (2003, Fig. II-2-5) that plots Cda as a
time has been lacking. In this paper, previous work on fully developed series of curves against Ua for different values of Δ°C (Geernaert et al.,
waves (Le Roux, 2007a,b, in press) is used together with a growth law 1986; Smith, 1988). This graph can be recast into the single equation:
based on the JONSWAP (Joint North Sea Wave Project) spectrum  
(Hasselmann et al., 1973; Demirbilek et al., 1993; Resio et al., 2003) to Cda = −1:7×10−8 ΔoC 3 −1:4×10−6 ΔoC 2 −3×10−5 ΔoC + 0:001
develop such a method. h  i ð3Þ
exp Ua −1:6×10−6 ΔoC 3 + 2×10−5 ΔoC 2 + 0:001ΔoC + 0:0324 :

⁎ Tel.: +56 2 9784123. At a wind speed of 10 m s− 1, an atmospheric pressure of 1010 mb, a

E-mail address: [email protected]. relative humidity of 80%, and air and water temperatures of 20 and

0378-3839/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.coastaleng.2008.10.007
480 J.P. Le Roux / Coastal Engineering 56 (2009) 479–483

23 °C, respectively, the values from Eqs. (2) and (3) inserted into Eq. (1) energy increases with the wave height and length as well as the water
coincide at 0.3808 m s− 1. For practical reasons, this is taken here as the density. It follows that, in the case of wind energy being transferred to
“normal condition” or NC. the waves, more energy would be required to reach the same
Another factor probably playing a role in wave formation is the wavelength and height in water with a higher density than the
water and air density, as discussed below. The density of seawater other way round. Considering Eq. (7), the wave height should
depends on its temperature and salinity. For a normal seawater therefore be directly proportional to the ratio ρaU2a /ρwg (the gravity
salinity of 35‰ acceleration being added to maintain the dimensional correctness of
  this ratio).
ρw = 1000 + −0:0051oC 2w −0:064oC w + 28:109 kg m−3 ð4Þ Le Roux (in press), based on Demirbilek et al. (1993) and Resio et al.
(2003), showed that the fully developed deepwater wave height is
which yields 1023.9391 kg m− 3 for sea water at 23 °C. given by
Air density ρa, on the other hand, is a function of its temperature
2
and relative humidity Γ, as well as the isobaric pressure Pa, vapor Ho = gTw =18π2 ð9Þ
pressure Pv and saturated vapor pressure Pvs in millibar (mb).
where Tw is the fully developed wave period obtained from
ρa = 1000fPa =½2870:5ð273:15 + Ca Þ−Pv =½4614:95ð273:15 + C a Þg kg m−3 ð5Þ
o o

Tw = 2πUa =g: ð10Þ

where
Setting C1ρaU2a / ρwg = gT w
2
/ 18π2 (where C1 is a dimensionless propor-
Pv = ðC=100ÞPvs and Pvs = 6:1078×107:5 Ca =ð237:3 + Ca Þ
o o

: ð6Þ
tionality constant) and replacing Tw by 2πUa/g, this yields
For the NC, ρa works out at 1.18643 kg m− 3.
Ho = 2C1 ρa Ua2 =9gρw : ð11Þ
3. Wave height as related to wind speed and the air/water density
ratio For the “normal condition” or NC, C1 has a value of 863.042.
Rearranged, C1(ρa / ρw) = 9gHo/2U2a = 1, where the second term is an
Air density probably plays a role in wave height in that the wind inverted wave Froude number.
energy (Ea) would increase with a higher ρa given the same wind Table 1 compares the values of Ho and Tw as obtained from
speed Ua, because the total energy of any moving fluid according to Eqs. (11) and (10) with the nomograms of Resio et al. (2003) and an
the Bernoulli equation is given by equation of Demirbilek et al. (1993). Tw as calculated here corresponds
very well with the average between the nomogram values of Resio
Ea = 0:5ρa Ua2 + ρa gh + Pa ð7Þ et al. (2003) and the equation of Demirbilek et al. (1993), whereas Ho
agrees exactly with the equation of Demirbilek et al. (1993), but
where the first and second terms in Eq. (7) refer to the kinetic and somewhat underestimates the nomogram values of Resio et al. (2003).
potential energies, respectively, and h is a length term. Both the
kinematic and potential energies of the wind must therefore increase 4. Fetch and duration as a function of the fully developed wave
with the air density, whereas Eq. (5) shows that Pa is also directly height
proportional to ρa.
The total wave energy Ew per unit length of wave crest, on the Waves grow in height and length not only in relation to the
other hand, is the sum of the kinetic energy and potential energy given velocity Ua of the wind, but also to its duration Ta and fetch F, the latter
by (Demirbilek and Vincent, 2002) being defined as the distance that the wind blows over open water
without a significant change in direction (b15°) or sustained speed
Ew = ρw gHo2 Lo =16 + ρw gHo2 Lo =16 = ρw gHo2 Lo =8 ð8Þ
(b2.5 m s− 1).
Demirbilek et al. (1993) proposed an equation modeling wave
where H is the wave height and L the wavelength, the subscript o
growth with fetch based on the JONSWAP data:
referring to the deepwater condition. The potential energy results
from that part of the fluid mass being above the still water level (SWL),  
2 2 1=2
thus being directly related to the wave height, whereas the kinetic gHFL =Ua4 = 0:0413 gF=Ua4 ð12Þ
energy is due to water particle velocities associated with wave motion
(Demirbilek and Vincent, 2002). Eq. (8) shows that the total wave where HFL is a fetch-limited, energy-based significant wave height.

Table 1
Comparison of wave heights and periods under fully developed sea conditions at different wind velocities as predicted Eqs. (9) and (11), nomograms in Resio et al. (2003) (DBT), and
Eqs. (12) of Demirbilek et al. (1993) (DBT) and (14) of Resio et al. (2003) (RBT)

Ua Ua⁎ Tws Tw s Tw s Ho m Ho m Ho m FFDS m TaFDS s

m s− 1 m s− 1 This paper Nomogram RBT Eq. (18) DBT This paper Eq. (12) DBT Nomogram RBT This paper This paper
2.5 0.0851 1.6 1.6 1.6 0.14 0.14 0.14 15,566 21,974
5 0.1768 3.2 3.3 3.1 0.57 0.57 0.60 59,780 42,233
7.5 0.2756 4.8 5.1 4.6 1.27 1.27 1.35 122,129 58,645
10 0.3808 6.4 6.6 6.1 2.27 2.27 2.45 204,375 74,216
12.5 0.4937 8.0 8.4 7.5 3.54 3.54 4.13 295,699 87,067
15 0.6148 9.6 11.1 8.9 5.10 5.10 6.14 395,769 98,286
17.5 0.7445 11.2 11.8 10.2 6.94 6.94 8.25 499,756 107,727
20 0.8832 12.8 13.0 11.5 9.06 9.06 11.51 605,212 115,619

Ua⁎ was calculated for both sets of equations using Eqs. (1), (3) and (4)–(6) with water and air temperatures of 23 and 20 °C, respectively, a water salinity of 35%, a relative air humidity
of 80% and a barometric pressure of 1010 mb.
 J.P. Le Roux / Coastal Engineering 56 (2009) 479–483 481

Eq. (12) can be rearranged to yield F for FDS conditions (FFDS) at any Table 2
particular wind speed, because HFL can be replaced by Ho, whereas the Comparison of wave heights at different wind velocities for fetch- or duration-limited
conditions as predicted by the equations proposed here, Eq. (12) of Demirbilek et al.
latter can be calculated by Eq. (11). This gives (1993) (DBT) and a nomogram in Resio et al. (2003, Fig. II-2-25) (RBT)

FFDS = gHo2 =1:70569×10−3 Ua4

2
ð13Þ Ua F HFL m HFL m Ua Ta HTL m HTL m
m s− 1 (km) This Eq. (12) DBT m s− 1 (h) This Nomogram
paper paper RBT
where Ua⁎ is obtained from Eqs. (1) and (3). For the NC, Ho = 2.27 m
and FFDS would be 204,375 m. 2.5 5 0.08 0.08 5 5 0.30 0.30
5 7 0.19 0.20 5 10 0.50 0.50
For any combination of wind speed and fetch, the time required for 7.5 3 0.20 0.20 7.5 9 0.82 0.80
waves to become fetch-limited (TaFL) is given in Resio et al. (2003) as 7.5 60 0.89 0.89 10 7 1.01 1.00
10 40 1.00 1.00 12.5 4 0.92 0.90
  2
2=3
TaFL = ð gF Þ= 0:00523Ua4 ðUa4 =g Þ: ð14Þ 12.5 6 0.50 0.50 15 8 2.03 2.00
12.5 100 2.06 2.06 15 20 4.04 4.00
15 10 0.81 0.81 17.5 6 2.08 2.00
Eq. (14) can be used to determine the limiting wind duration at FDS 17.5 5 0.69 0.69 17.5 10 3.05 3.00
conditions (TaFDS) by replacing F with FFDS as obtained from Eq. (13), 20 70 3.08 3.08 20.0 8 3.19 3.00
which would give 74,216 s or 20.6 h for the NC. Ua⁎ was calculated in the same way as for Table 1.
For a fetch or wind duration shorter than the FFDS or TaFDS
corresponding to any particular wind speed, the waves would still be
growing towards their FDS condition and the wave celerity Co would (1993), it turns out that if the wave height as obtained from Eq. (11) is
be less than Ua, so that there would be excess wind shear and the wind multiplied by the square root of the ratio between the actual fetch and
velocity at 10 m would be higher than at the SWL. It can also be the limiting fetch for the specific wind speed, the resulting wave heights
expected that at higher wind velocities, where an increasing correspond almost exactly to that given by Eq. (12) for any wind velocity
proportion of the wind energy is expended in breaking wave crests (Table 2). Therefore the fetch-limited wave height is given by
and spray production, the wind velocity would always exceed the
wave celerity. When this happens, Eqs. (13) and (14) probably become  
HFL = 863:042ðρa =ρw Þ 2Ua2 =9g ðF=FFDS Þ1=2 : ð15Þ
invalid and the maximum required fetch and wind duration for FDS
conditions may level off. Gross and Gross (1996) noted, for example, Similarly, evaluating the effect of wind duration on the wave height
that after about 30 h there is very little increase in wave growth for any against the nomogram in Resio et al. (2003, Fig. II-2-25) shows that the
wind condition regardless of fetch. duration-limited wave height can be obtained by
The condition at which wind energy becomes increasingly
 
expended in breaking wave crests and spray at the expense of wave HTL = 863:042ðρa =ρw Þ 2Ua2 =9g ðTa =TaFDS Þ3=4 : ð16Þ
growth, is approximated by observations and theoretical modeling.
According to the Beaufort wind scale (see, e.g., internet site http:// Table 2 compares the values of Eq. (12) and the nomogram with the
www.wikipedia.org/wiki/Beaufort_scale), foam begins to streak at results of Eqs. (15) and (16).
wind speeds exceeding 17.2 m s− 1, but considerable spray only The wave height for any combination of fetch and wind duration
develops at the transition from gales to strong gales at 20.8 m s− 1, can therefore be calculated by
where the wave crests also begin to roll over. This is supported by Bye
 2 
and Jenkins (2006), who theoretically modeled the drag coefficient for HFTL = 863:042ðρa =ρw Þ 2Ua10 =9g ðF=FFDS Þ1=2 ðTa =TaFDS Þ3=4 ð17Þ
momentum transfer (Cda) of an air–sea system with shear layers in
both fluids. They concluded that the maximum reached by Cda under where the subscript FTL indicates both fetch- and time-limited conditions.
any wind speed is 0.002. According to Eqs. (1) and (3) at the NC, this The advantage of Eq. (17), in comparison with Eqs. (12) and (14), is that
value coincides with a wind speed of 20.9 m s− 1, which is considered it represents a combination of fetch and duration, whereas the latter two
here to be the maximum wind speed for which the equations in this equations cannot be combined. For example, for the NC, a fetch of 100 km
paper are still valid. Inserted into Eqs. (8) and (9), respectively, and wind duration of 10 h, Eq. (12) would yield a wave height of 1.59 m,
together with the calculated friction velocity of 0.9347 m s− 1, this whereas Eq. (14) would indicate that 12.8 h are required to reach this
corresponds to a wave period of 13.39 s and wave height Ho of 9.90 m, fetch-limited height. However, it cannot compute the actual fetch- and
which used with Eqs. (13) and (14) yield a maximum fetch and duration-limited wave height (HFTL) after 10 h. Eq. (17) in this case yields a
duration of 645,201 m and 118,399 s (32.89 h), respectively. wave height of 0.92 m. A second advantage is that the maximum wave
height is limited by Eq. (17) to that given by Eqs. (9) or (11), whereas Eq.
5. Characteristics of developing waves (12) simply increases with the fetch. For a fetch of 1000 km it would give a
wave height of 5.02 m for the NC, for example, more than twice the
The considerations above are valid for FDS conditions where the maximum height of 2.27 m for FDS conditions. Demirbilek et al. (1993)
wind blows for an unlimited time over an unlimited fetch. However, calculate the maximum wave height for a FDS by Ho =211.5U 2a⁎ /g, which
there are many situations where the fetch is limited (denoted by the in this case would be 3.13 m. This still exceeds the value of 2.27 m given by
subscript FL), for example in the case of lakes or narrow sea straits. Eq. (11), for example, as well as the value shown by the nomogram (Fig. II-
Similarly, the wind may blow for only a few hours (time-limited 2-25) of Resio et al. (2003).
conditions, denoted by TL), in which case the resultant waves will not
reach their full height even with an unlimited fetch. 5.2. Period of developing waves

5.1. Height of developing waves Demirbilek et al. (1993) developed an equation to determine the
wave period with growing fetch, given by:
To estimate the height of developing waves, the actual fetch and  
2 1=3
duration can be added as ratios of the maximum fetch and duration as TwFL = 0:651 gF=Ua4 ðUa4 =g Þ: ð18Þ
calculated by Eqs. (13) and (14) for the particular wind speed, thus
maintaining the dimensional correctness of Eq. (11). Comparing this However, Eq. (18) does not correlate very well with Eq. (10). For the
approach with the wave heights given by Eq. (12) of Demirbilek et al. NC, it would indicate a wave period of 6.07 s for a FDS, with a
482 J.P. Le Roux / Coastal Engineering 56 (2009) 479–483

corresponding wave height of 2.03 m according to Eq. (9), which is Table 4

lower than the maximum FDS height of 2.27 m (a difference of 10.6%). Development of wave parameters with fetch under NC (see text for details)

An alternative method is to express TwFL as a function of the fully F (km) HFL (m) LFL (m) TwFL (s) CwFL (m s− 1) Steepness (HFL/LFL)
developed period Tw, the wave height Ho, as well as the fetch and 20 0.71 14.16 2.53 5.6 0.0501
duration ratios. This can be done by first rearranging Eq. (9): 40 1 22.19 3.34 6.64 0.0451
60 1.23 28.95 3.93 7.37 0.0425
 1=2 80 1.42 34.91 4.41 7.92 0.0407
Tw = 18π2 Ho =g : ð19Þ
100 1.59 40.38 4.82 8.38 0.0394
120 1.74 45.40 5.18 8.76 0.0383
Eq. (19) gives exactly the same values as Eq. (10) for FDS conditions 140 1.88 50.20 5.51 9.11 0.0374
and compares reasonably well with those given by Eq. (18) if the same 160 2.01 54.82 5.82 9.42 0.0367
fetch is used for wind velocities up to 20 m s− 1 (Table 1). At a wind speed 180 2.13 59.15 6.1 9.7 0.0360
200 2.25 63.39 6.36 9.97 0.0355
of 20.9 m s− 1, Eqs. (9) and (19) yield a wave height and period of 9.90 m 204,375 2.27 64.05 6.40 10 0.0354
and 13.39 s, respectively, which agree well with observed 5-year
maximum wave heights of 10.4 m and peak periods of 14.1 s in the Pacific
Ocean (Resio et al., 2003, Table I-2-5). Eq. (18) in this case would
underestimate the wave period somewhat at 12.0 s (FFDS = 645,201 m, Table 5
Ua⁎ = 0.9347 m s− 1). Development of wave parameters with wind duration under NC (see text for details)
The wave period can also be evaluated for both fetch- and Ta (h) HTL (m) LTL (m) TwTL (s) CwTL (m s− 1) Steepness (HTL/LTL)
duration-limited conditions by comparison with Figs. II-2-24 and II- 2 0.39 7.26 1.75 4.15 0.0537
2-26 of Resio et al. (2003). This yields: 4 0.66 13.93 2.58 5.41 0.0474
6 0.9 20.36 3.23 6.3 0.0446
 1=2
TwFTL = 18π2 Ho =g ðF=FFDS Þ2=5 ðTa =TaFDS Þ5=9 : ð20Þ 8 1.12 26.65 3.79 7.02 0.0419
10 1.32 32.75 4.29 7.63 0.0403
12 1.51 38.78 4.75 8.17 0.0389
Table 3 shows an excellent correlation, with the difference never 14 1.70 44.80 5.17 8.66 0.0379
exceeding 0.4 s. However, the advantage of Eq. (20) is again that a 16 1.88 50.74 5.57 9.10 0.0371
combination of both fetch and duration can be used, whereas the 18 2.05 56.6 5.95 9.51 0.0362
nomograms in Resio et al. (2003, Figs. II-2-24 and II-2-26) or Eq. (18) 20 2.22 62.37 6.3 9.90 0.0356
20.62 2.27 64.05 6.40 10 0.0354
can be used for either of the two, but not both.

5.3. Length of developing waves 8.4 h, the wave period and height according to Eqs. (20) and (17)
would be 5 s and 2.11 m, respectively, corresponding to a deepwater
The wavelength of fully developed deepwater waves is given by wavelength of 48.26 m, which is considerably longer than for fully
the standard Airy equation Lo = gTw2 / 2π. Le Roux (2007a) demonstrated developed 5 s waves. This is to be expected, because the higher wind
that the fully developed wavelength is also related to the fully speed would impart more energy to the waves, which is manifested in
developed wave height by Lo = 9πHo. Multiplying these two terms an increased wavelength and height. The wave steepness given by Ho/
expresses Lo in terms of the wave period and height: L2o = (gTw 2
/ 2π) Lo, however, is 0.0354 for a fully developed 5 s wave and 0.0437 for the
(9πHo). Because both the period and height of developing waves can developing 5 s wave. The wave steepness thus increases with a
be calculated by the equations proposed above, the wavelength should decrease in fetch and duration, until it reaches a maximum value
therefore be related to these parameters in the form when the wave will presumably break. This is in accordance with
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi observations at sea that developing waves are shorter and steeper
ffi
LFTL = 3 HFTL gTwFTL 2 =2 : ð21Þ than fully developed waves (Resio et al., 2003).
Tables 4 and 5 show the growth in NC wave parameters with fetch
Eq. (21) yields exactly the same wavelength in deepwater and time, respectively, whereas Table 6 illustrates wave development
conditions for fully developed waves as the standard Airy equation. under different combinations of fetch and time. In all cases, a potential
However, for developing waves, it gives longer wavelengths for the curve of the form y = axb can be fitted to the data. From the tables it is
same wave period. A wind speed of 7.8 m s− 1, for example, would clear that, for similar fetch and time increments, fetch-limited waves
correspond to a fully developed wave period of 5 s, with Ho = 1.38 m are higher during the initial stages of wave development than time-
and Lo = 39.03 m. For a 15 m s− 1 wind speed and a limited duration of limited waves and also have higher initial wavelengths and celerities.
Waves that are both fetch- and time-limited are significantly steeper
for the same wave wind speed than either fetch- or time-limited
Table 3
Comparison of wave periods at different wind velocities for fetch- or duration-limited
conditions as predicted by the equations proposed here and nomograms in Resio et al. Table 6
(2003, Figs. II-224 and II-2-26) (RBT) Development of wave parameters with wind fetch and duration under NC (see text for
details)
Ua F TwFL s TwFL s Ua Ta TwTL s TwTL s
m s− 1 (km) This paper Nomogram RBT m s− 1 (h) This paper Nomogram RBT F (km) Ta (h) HFL (m) LFTL (m) TwFTL (s) CwFTL (m s− 1) Steepness (HFTL/LFTL)
7.5 10 1.8 2.0 5 10 2.9 3.0 20 2 0.12 1.59 0.69 2.3 0.0755
7.5 80 4.0 4.0 7.5 3 1.9 2.0 40 4 0.29 4.83 1.35 3.58 0.06
10 60 3.9 4.0 7.5 7 3.0 3.0 60 6 0.49 9.20 1.98 4.64 0.0533
10 200 6.4 6.0 10 5 2.9 3.0 80 8 0.7 14.50 2.61 5.56 0.0483
12.5 20 2.7 3.0 10 9 4.0 4.0 100 10 0.92 20.58 3.23 6.37 0.0446
12.5 90 5.0 5.0 12.5 4 2.9 3.0 120 12 1.16 27.48 3.84 7.16 0.0421
15 200 7.3 7.0 12.5 7 4.0 4.0 140 14 1.41 35.12 4.45 7.89 0.0401
17.5 30 3.6 4.0 15 9 5.2 5.0 160 16 1.66 43.32 5.06 8.57 0.0383
17.5 60 4.8 5.0 17.5 6 4.6 5.0 180 18 1.93 52.25 5.66 9.23 0.0369
20 50 4.7 5.0 20.0 4 4.0 4.0 200 20 2.20 61.69 6.26 9.85 0.0357
204.375 20.62 2.27 64.05 6.40 10.00 0.0354
Ua⁎ was calculated in the same way as for Table 1.
 J.P. Le Roux / Coastal Engineering 56 (2009) 479–483 483

waves. For example, for a fetch of 50 km and wind duration of 12 h, NC References

waves have a steepness of 0.0455, decreasing to 0.039 over an
Barnett, T.P., 1968. On the generation, dissipation, and prediction of ocean wind waves.
unlimited fetch for the same wind duration.
J. Geophys. Res. 2, 531–534.
In Table 6, wave steepness values of 0.0755 and 0.06 are indicated Bunting, D.C., 1970. Evaluating forecasts of ocean-wave spectra. J. Geophys. Res. 75,
after 2 and 4 h on fetch distances of 20 and 40 km, respectively, but 4131–4143.
Bye, J.A.T., Jenkins, A.D., 2006. Drag coefficient reduction at very high wind speeds.
such waves will probably be breaking (whitecaps), so that the wave
Journal of Geophysical Research — Oceans 111 (C3), C03024 Art.
heights will be restricted to less than the calculated values. Camfield, F.E., 1977. Wind-wave propagation of flooded, vegetated land. Technical Paper,
No. 77-12, Coastal Engineering Research Center, U.S. Army Engineer Waterways
6. Conclusions Experiment Station, Vicksburg, MS.
Cardone, V.J., 1992. On the structure of the marine surface wind field. Extended Abstr.,
3rd Internat. Workshop of Wave Hindcasting and Forecasting, Montreal, Quebec,
The method developed here provides a convenient way to estimate pp. 54–66.
the wave height, length and period for any combination of wind Demirbilek, Z., Vincent, C.L., 2002. Water wave mechanics. Chapter II-1 Coastal
Engineering Manual (EM 1110-2-1100). US Army Corps of Engineers, Washington,
velocity, fetch, and duration. It incorporates atmospheric conditions D.C., p. 121 pp.
such as air temperature, pressure, relative humidity and density, as Demirbilek, Z., Bratos, S.M., Thompson, E.F., 1993. Wind products for use in coastal wave
well as the water temperature, salinity and density. and surge models. Misc. Paper, CERC-93-7, US Army Engineer Waterways
Experiment Station, Vicksburg, MS.
Applying the proposed equations to calculate the wave climate Forristall, G.Z., Ward, E.G., Cardone, V.J., Borgman, L.E., 1978. The directional spectra and
under different atmospheric conditions, leads to the following general kinematics of surface gravity waves in tropical storm Delta. J. Phys. Oceanogr. 8,
conclusions: 888–909.
Geernaert, G.L., Katsaros, K.B., Richter, K., 1986. Variation of the drag coefficient and its
1) Waves should be higher when the atmospheric pressure increases dependence on sea state. J. Geophys. Res. 91, 7667–7679.
Gross, M.G., Gross, E., 1996. Oceanography: a View of the Earth. Prentice Hall, New
because of an increase in the total energy (Eq. (7)). For example, for
Jersey. 472 pp.
a wind speed of 10 m s− 1 and an atmospheric pressure of 1000 mb, Hasselmann, K., Barnett, T.P., Bouws, E., Carlson, H., Cartwright, D.E., Enke, K., Weing, J.A.,
water and air temperatures both being 20 °C and the air humidity Gienapp, H., Hasselmann, D.E., Kruseman, P., Meerburg, A., Muller, P., Olbers, K.J.,
80%, waves would be 2.24 m high as compared to 2.31 m at Richter, K., Sell, W., Walden, W.H., 1973. Measurements of wind-wave growth and
swell decay during the Joint North Sea Wave Project (JONSWAP). Deutsche
1030 mb, a difference of 3.1%. This means that waves associated Hydrograph. Zeitschr., Erganzung-self Reihe, A 8 (12).
with anticyclones should be higher than those of cyclones, given Hasselmann, S., Hasselmann, K., Allender, J.H., Barnett, T.P., 1985. Computations and
the same wind speed. By the same reasoning, high-altitude water parameterizations of nonlinear energy transfer in a gravity-wave spectrum. Part II:
parameterization of non-linear transfer for application in wave models. J. Phys.
bodies such as Lake Titicaca in Bolivia should have lower waves Oceanogr. 15, 1378–1391.
because of the decrease in atmospheric pressure and air density, Inoue, T., 1967. On the growth of the spectrum of a wind-generated sea according to a
given the same temperature. modified Miles–Phillips mechanism and its application to wave forecasting.
Geophysical Sciences Laboratory Rep. No. TR-67-5. Department of Meteorology
2) On fresh water lakes, waves should be higher for the same and Oceanography. New York University.
duration- and fetch-limited conditions than on the ocean, because Kahma, K.K., 1981. A study of the growth of the wave spectrum with fetch. J. Phys.
of the lower water salinity and density. For the NC and a FDS, waves Oceanogr. 11, 1503–1515.
Kitaigorodskii, S.A., 1983. On the theory of the equilibrium range in the spectrum of
would be 2.27 m high in seawater and 2.33 m in fresh water, an wind-generated gravity waves. J. Phys. Oceanogr. 13, 816–827.
increase of 2.6%. The opposite effect will be caused by the presence Le Roux, J.P., 2007a. A simple method to determine breaker height and depth for
of a substantial amount of sediment in suspension, for example in different deepwater wave height/length ratios and sea floor slopes. Coastal Engr. 54,
271–277.
large effluent plumes opposite river mouths such as the Amazon,
Le Roux, J.P., 2007b. A function to determine wavelength from deep into shallow water
where wave heights should decrease because of an increase in based upon the length of the cnoidal wave at breaking. Coastal Engr. 54, 770–774.
water density. This is supported by the work of Camfield (1977) on Le Roux, J.P., in press. An extension of the Airy theory for linear waves into shallow
the large dissipation rate of wave energy in the presence of water. Coastal Engr.
Resio, D.T., 1981. The estimation of a wind wave spectrum in a discrete spectral model.
extreme amounts of sediment or plant material in the water J. Phys. Oceanogr. 11, 510–525.
column. Resio, D.T., 1987. Shallow water waves. Part II: Theory. J. Waterway, Port, Coastal and
3) Developing waves are shorter and steeper than fully developed Ocean Eng. 113, 264–281.
Resio, D.T., 1988. Shallow water waves. Part II: data comparisons. J. Waterway, Port,
waves for the same wind speed, which suggests that the most Coastal and Ocean Eng. 114, 50–65.
hazardous conditions for small vessels will be found on lakes or in Resio, D.T., Perrie, W., 1989. Implications d and f equilibrium range for wind-generated
narrow sea straits after the first hours of stormy conditions. The waves. J. Phys. Oceanogr. 19, 193–204.
Resio, D.T., Vincent, C.L., 1979. A comparison of various numerical wave prediction
procedure outlined above allows such conditions to be calculated, techniques. Proc. 11th Annual Offshore Technology Conf., Houston, TX, pp. 2471–2481.
so that certain maritime activities can be restricted during these Resio, D.T., Bratos, S.M., Thompson, E.F., 2003. Meteorology and Wave Climate, Chapter II-
periods. 2, Coastal Engineering Manual. US Army Corps of Engineers, Washington D.C. 72 pp.
Smith, S.D., 1988. Coefficients for sea surface wind stress, heat flux, and wind profiles as
a function of wind speed and temperature. J. Geophys. Res. 93, 467–470.
Acknowledgements Toba, Y., 1978. Stochastic form of the growth of wind waves in a single parameter
representation with physical implications. J. Phys. Oceanogr. 8, 494–507.
Van Vledder, G.P., Holthuisjen, L.H., 1993. The directional response of ocean waves to
This research was conducted as a Fellow of the Hanse Institute for
turning winds. J. Phys. Oceanogr. 23, 177–192.
Advanced Study in Delmenhorst, Germany. The logistic and financial
support of the Institute is gratefully acknowledged. I am grateful to an
anonymous reviewer who suggested many improvements to the
original text.