Lecture Notes on Fluid Dynamics: Transient Longshore Wind and Sinusoidal Wind Stress, Study notes of Fluid Mechanics

Lecture notes on fluid dynamics, specifically focusing on transient longshore wind and sinusoidal wind stress. The notes include equations and their solutions, boundary conditions, and discussions on the rossby radius of deformation. The document also includes exercises for students to solve.

Typology: Study notes

2010/2011

Uploaded on 10/11/2011

gwen
gwen 🇺🇸

5

(8)

283 documents

1 / 5

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
1
Lecture Notes on Fluid Dynamics
(1.63J/2.21J)
by Chiang C. Mei, 2006
7.6 Transient longshore wind
[Ref]: Chapter 14, p. 195 ff, Cushman-Roisin
Csanady: Circulation in the Coastal Ocean
Figure 7.6.1: Longshore wind
Let the xbe the coast, and the sea is in y>0, as seen in Figure 7.6.1.
In view of the last section, we ignore the bottom stress. Assume that the wind is uniform
in space but transient in time, so that ∂/∂x = 0, The flux equations are
∂η
∂t +∂V
∂y = 0 (7.6.1)
∂U
∂t fV =τS
x
ρ(7.6.2)
∂V
∂t +fU =gh ∂η
∂y.(7.6.3)
The boundary condition on the coast x=0: U=0.
7.6.1 Sudden long-shore wind
Let the wind stress be
τS
x=0,t0,
T, t > 0.(7.6.4)
the initial conditions are
η, U, V =0,t=0,y. (7.6.5)
This initial-boundary value problem can be solved by Laplace transform (Cr´epon, 1967).
The solution consists of two parts: one part is oscillatory and decays with time; the other
pf3
pf4
pf5

Partial preview of the text

Download Lecture Notes on Fluid Dynamics: Transient Longshore Wind and Sinusoidal Wind Stress and more Study notes Fluid Mechanics in PDF only on Docsity!

Lecture Notes on Fluid Dynamics (1.63J/2.21J) by Chiang C. Mei, 2006

7.6 Transient longshore wind

[Ref]: Chapter 14, p. 195 ff, Cushman-Roisin Csanady: Circulation in the Coastal Ocean

Figure 7.6.1: Longshore wind

Let the x be the coast, and the sea is in y > 0, as seen in Figure 7.6.1. In view of the last section, we ignore the bottom stress. Assume that the wind is uniform in space but transient in time, so that ∂/∂x = 0, The flux equations are

∂η ∂t

∂V

∂y

∂U

∂t

− f V =

τ (^) xS ρ

∂V

∂t

  • f U = −gh

∂η ∂y

The boundary condition on the coast x = 0 : U = 0.

7.6.1 Sudden long-shore wind

Let the wind stress be

τ (^) xS =

{ 0 , t ≤ 0 , T, t > 0. (7.6.4)

the initial conditions are η, U, V = 0, t = 0, ∀y. (7.6.5)

This initial-boundary value problem can be solved by Laplace transform (Cr´epon, 1967). The solution consists of two parts: one part is oscillatory and decays with time; the other

part increases monotonically with time. To avoid the complex mathematics we only examine the latter which is the dominant part for large time,

U = t U¯ (y), V = V¯ (y), η = tη¯(y) (7.6.6)

The oscillatory part is needed to ensure the initial condition on U. This solution clears fails for very large t, indicating the there is no steady- state solution in the linearized approxi- mation. For a realistic theory nonlinearity must be accounted for, or the wind duration has to be finite. It is easy to see from (7.6.1) to (7.6.3) that

η¯ +

d V¯ dy

f U¯ = −gh

dη¯ dy

U¯ − f V¯ = T /ρ (7.6.9)

These three equations can be combined into one :

d^2 V¯ dy^2

f 2 gh

V¯ = f T ρgh

The solution satisfies no flux on the coast is

V¯ = − T ρf

( 1 − e−x/Ro

) (7.6.11)

where

Ro =

gh f

is called the Rossby radius of deformation. If we take f = 10−^4 1/s in a shallow sea of h = 10 m the Rossby radius is about 10^5 m = 100 km. It is easy to find that η = tη¯ = t

T

ρgh

e−y/Ro^ (7.6.13)

and

U = t U¯ = t

T

ρf

e−y/Ro^ (7.6.14)

Clearly when y/Ro  1, the coast line has no influence. The flux is U = 0, V = −T /ρf , and is inclined to the right of the wind by 90 degrees, as predicted by the Ekman layer theory. The sea surface rises if T > 0 (coast is on the right of wind), and sinks near the coast if T < 0 (coast is on the left of wind).

Low frequency: ω < f

The solution to (7.6.22) bounded at infinity is

η 0 = A e−y/R^0 (7.6.23)

where

R 0 =

√ gh f 2 − ω^2

is the modified Rossby radius. Applying the B.C. on the coast : V 0 = 0, y = 0, we get from (7.6.20),

dη 0 dy

f ρgH

τ 0 ω

(7. 6 =.23) −A

R 0

and,

A = −

τ 0 ω

f ρgh

R 0 (7.6.25)

Hence

η 0 = −

f τ 0 ρωgh

R 0 e−y/Ro

and finally

η = −

f τ 0 ρωgh

R 0 e−y/R^0 e−iωt^ (7.6.26)

Its real part is the solution. Now from Eqn. (7.6.20), we get

U =

τoω ρ(f 2 − ω^2 )

( 1 −

f 2 ω^2

e−y/R^0

) e−iωt. (7.6.27)

From Eqn. (7.6.21)

V 0 = −

iτ 0 f ρ(f 2 − ω^2 )

( 1 − e−y/R^0

) (7.6.28)

(7.6.29)

Hence,

V = −

τ 0 f ρ (f 2 − ω^2 )

( 1 −

f 2 ω^2

e−x/R^0

) sin ωt. (7.6.30)

If τ 0 > 0 (wind blows eastward, along the coast), the sea level near the coast sinks at the start.

High frequency : ω > f

Of the two possible oscillatory solutions to (7.6.22), we must choose the one that represents outgoing waves at infinty (the radiation condition),

η 0 =^ Beikx,^ (7.6.31)

where the wavenumber is the inverse of the modified Rossby radius of deformation,

k =

√ ω^2 − f 2 gh

It is easy to show that from the no-flux confditon on the coast thata

B =

−iτ 0 f ρghωk

hence, in complex form,

η = −

iτ 0 f ρghωk

eikx−iωt^ (7.6.34)

or, in real form,

η = −

τ 0 f ρghωk

sin(kx − ωt), (7.6.35)

The final formulas for U, V are left as an exercise. Homework: For the oscillating-wind problem, find the flow in the inviscid interior and then the flow in the surface boundary layer.