More on Tropical Waves
¾ Steady forced motions
¾ Large-scale response of the tropical atmosphere to transient convection
¾ A theory for midlatitude forcing of tropical motions during winter monsoons
Topics
¾ Consider a homogeneous ocean layer of mean depth H forced by a surface wind stress X = (X,Y)per unit area.
¾ Assume that wind stress is distributed uniformly with depth as a body force X/(ρH)per unit mass.
¾ Suppose that there is also a drag per unit mass acting on the water, modelled by the linear friction law - ruper unit mass.
¾ See e.g. A. E. Gill(1982): Atmosphere – Ocean Dynamics.
f
H(1 + η) H
Response to steady forcing
X/(ρH) X
Equations
x y
2 2
x y
yv gH X /( H) ru yu gH Y /( H) rv c ( u v) gE / c r
−β = − ∂ η + ρ − β = − ∂ η + ρ −
∂ + ∂ = − ρ − η
( )
( )
2 2
r r2 f2 v r v v v
2 2 2 x
c x y
1 r E Y X
rY fX r fE
H c2 y x x x
∂ ∂ ∂
+ − + − β =
∂ ∂ ∂
− + ∂ − ∂ ∂ −∂ +
ρ ∂ ∂ ∂ ∂
Evaporation/heating Friction
Eliminate uand η
( )
( )
2 2
r r2 f2 v r v v v
2 2 2 x
c x y
1 r E Y X
rY fX r fE
H c2 y x x x
∂ ∂ ∂
+ − + − β =
∂ ∂ ∂
− + ∂ − ∂ ∂ −∂ +
ρ ∂ ∂ ∂ ∂
Case of small friction: r →0
β∂
∂ ρ
∂
∂
∂
∂
∂
∂ v
x H x
Y x
X x fE
= − +
1
β ρ
∂
∂
∂ v ∂
H Y
x X
x fE
= − +
1
Integrate w. r. t. x
⇐when E = 0⇒ Sverdrup’s formula
( )
( )
2 2
r r2 f2 v r v v v
2 2 2 x
c x y
1 r E Y X
rY fX r fE
H c2 y x x x
∂ ∂ ∂
+ − + − β =
∂ ∂ ∂
− + ∂ − ∂ ∂ −∂ +
ρ ∂ ∂ ∂ ∂
Zonally-independent flows: ∂/∂x = 0
(
2 2)
2( )
2 2 2
r f v 1 rY fX E
c v y H c y
+ −∂∂ − =ρ − +∂∂ Valid on an f-plane or when f = βy
The linear solution due to a line source of heat/evaporation.
pressure distribution
meridional circulation heat source
Equatorial Rossby radius √(c/2β)
y z
y
Heating confined to the range of longitudes|x| < 2LR
Heating confined to the range of longitudes|x| < 2LR
Zonal flow ⇒
Meridional flow ⇒
Meridionally-averaged Walker circulation
Large-scale response of the tropical atmosphere to transient convection
Silva Dias, Schubert
& DeMaria, 1983
¾ Large west-east asymmetries in the tropical circulations are associated with the equatorial continental areas of South America, Africa and the Maritime continent.
¾ SDSDshow average 200and 700mb flow over South America for four winters (JJA) and four summers (DJF) between 1996and 1970.
¾ A prominent feature is the Bolivian high above 200mb.
Silva Dias et al., Fig. 1
JJA DJF
Winter Summer
Rainfall in mm
JJA DJF
Winter Summer
700 mb streamlines
Silva Dias et al., Fig. 1
JJA DJF
Winter Summer
200 mb streamlines
(Linearized) Equations of motion
u u u u
u v w yv F
t x y z x
v v v v
u v w yu G
t x y z y
z RT 0
u v w
x y z w 0
∂ + ∂ + ∂ + ∂ − β = −∂φ +
∂ ∂ ∂ ∂ ∂
∂ + ∂ + ∂ + ∂ + β = −∂φ +
∂ ∂ ∂ ∂ ∂
∂φ− =
∂
∂ + ∂ + ∂ − =
∂ ∂ ∂
z ln(p / p)= 0
Momentum sources
Heat source
z z
z z
u yv u
t x t
v v
t yu y t
e u v e
e e
t z R z x y t z R z
(z) dT T dz
− −
∂ − β + ∂φ = ∂
∂ ∂ ∂
∂ + β + ∂φ= ∂
∂ ∂ ∂
∂ ∂ ∂φ−∂ + ∂ = ∂ ∂ ∂φ
∂ ∂ Γ ∂ ∂ ∂ ∂ ∂ Γ ∂
Γ = + κ
Forcing terms:
p
u v RQ
F , G ,
t t c t z
∂ ∂ ∂ ∂φ
= ∂ = ∂ = ∂ ∂
Basic state static stability:
n n n
n
n n n
n
n 2 n n n
n
u u
t yv x t
v v
t yu y t
u v
t c x y t
∂ − β + ∂φ = ∂
∂ ∂ ∂
∂ + β + ∂φ = ∂
∂ ∂ ∂
∂φ∂ + ∂∂ +∂∂ = ∂φ∂
Inverse:
T z / 2
n n
z
f (x, y, t)=
∫
0 f (x, y,z)Ψ (z)e− dz Stürm-Liouville Transform:Transformed equations
z / 2
n n
n 0
f (x, y,z, t) f (x, y,z) (z)e
=
=
∑
∞ Ψn n n
f (x, y,z, t) u, v, , u , v ,= φ φ
1/ 2
n n
c =(gh )
Time (hours) Dimensionless time
3 2 t
1 t e 2
α −α
1 2 3
32
8 16 24
α= 4/3 α= 4
0.5 1.0
3
Total fieldlvin W.
Wind and geopotential
16 h 32 h
MRG waveRossby W.
Wind and geopotential
16 h 32 h
Total fieldKelvin W.
Wind and geopotential
48 h 64 h
48 h 64 h
MRG waveRossby W.
Wind and geopotential
Wind and geopotential
48 h 64 h
16 h 32 h
16 h 32 h
16 h 32 h
Wind and geopotential
DisturbanceAfter NMI
Summary - 1
¾ The amount of Kelvin wave energy generated is greater when the forcing is rapid and when the heat source is close to the equator.
¾ The convection that occurs over the South American continent and also over the Indonesian region has large variations on short time scales.
¾ Since both of these regions are close to the equator, there should be a large Kelvin wave response to this convective forcing.
Summary - 2
¾ At upper levels, convergence occurs on the eastern side of the Kelvin wave group with divergence on the western side.
¾ This convergence-divergence pattern and associated vertical motion field may be important for modulating convection in areas far to the east of the forcing.
¾ There is some observational evidence for this modulation of convection by the Kelvin waves.
Summary - 3
¾ The generation of Kelvin waves by transient convection has some implications for the initialization of tropical forecast models.
¾ The basic idea of many nonlinear normal mode
initialization procedures is to obtain the slow mode part of an initial field from observations, and then to diagnose the fast mode part in such a way that the initial tendency of the fast modes is zero.
¾ When this type of procedure is applied, the magnitude of die nonlinear interaction of the slow modes and the
Summary - 4
¾ For the case of transient convection, the amplitude of the Kelvin modes will be fairly large after the amplitude of the forcing has decreased.
¾ Since the Kelvin waves are considered to be fast modes (except for very long waves), it may not be possible to diagnose these after the magnitude of the forcing has decreased and the waves have propagated away from the source region.
A theory for midlatitude forcing of tropical motions during winter monsoons
Lim. H, C.-P. Chang, 1981
¾ In the mid1atitudes, a surge in the northeast trades arrives with a steep rise of surface pressure, a sharp drop of temperature and a strengthening of northerly winds.
¾ The cold front leading the surge sometimes brings stratus and rain but a strong surge is generally associated with subsiding motions which leads to clearing of weather (Danielsen and Ho, 19692; Ramage, 1971).
¾ Although the front associated with a surge cannot normally be followed southward of about 250N.
¾ The surge propagates equatorwards in dramatic fashion, ss a vigorous surge reaches the South China coast, northerly winds freshen almost simultaneously several hundred kilometres to the south, far beyond the region where the winds could have pushed the front. (Ramage, 1971; Chang et al., 1979).
¾ A belt of strong northeasterly winds from within 24 h off the South China and Vietnam coasts.
Northeasterly surges
Northeasterly surge
22 Dec 78 25 Dec 78
Wind surges
A shallow water model calculation for surges
Based on:
On the dynamics of midlatitude-tropical interactions and thew winter monsoon, by Hoch Lim and C.-P. Chang.
In Monsoon Meteorology, Ed. C.-P. Chang and T. N.
Krishnamurti.
u h
yv g 0
t x
v yu g h 0
t y
h u v
t H x y
∂ − β + ∂ =
∂ ∂
∂ + β + ∂ =
∂ ∂
∂∂ + ∂∂ + ∂∂ = Φ
Linearized equations on an equatorial β-plane
Motions forced by a mass source term
Variation of forcing amplitude with time
Fast
Slow
Slow forcing
day 3 day 2
Lim Hoch Fig 4b
day 4
day 5day 5 day 4
Slow forcing
Lim Hoch Fig 4c
day 6
day 9
Slow forcing
Kelvin wave group (n = −1)
Mixed Rossby-gravity wave group (n = 0)
Rossby-gravity wave group (n = 1 − 2)
Rossby-gravity wave group (n = 3 − 4)
Combined Rossby-gravity wave group
n = 0 Hermite modes excited by the fast forcing
from day 4 to day 9
Trajectories of the centre (area of maximum amplitude) of the mixed
Rossby-gravity wave group
Trajectories of the centre (area of maximum amplitude) of the inertia-
gravity wave group
Day 1
Day 2
Fast forcing
Day 3 - 4
Day 3
Day 4
Day 5 - 6
Day 5
Day 6
Day 7 - 8
Day 7
Day 8
Day 9
Day 9
Realistic forcing
day 2
day 7
Summary - 1
¾ Midlatitude forcing with large time and space scales excite mainly Rossby and, to a much smaller amplitude, Kelvin waves.
¾ As either of these scales decreases, the amplitudes of mixed Rossby-gravity and inertia-gravity waves increase.
¾ Immediately following a pressure forcing, the strong wind surge is more northerly with a gravity wave character.
¾ The strong cross-isobaric divergent flow occurs without the presence of surface friction.
¾ It becomes more easterly as the Rossby wave group is
Summary - 2
¾ The development of the main tropical flow pattern after a midlatitude pressure forcing, including the strong northeast wind belt southeast of the main anticyclone and a cyclonic shear zone further southeast of it, is represented by the slowly moving Rossby wave group response.
¾ This Rossby wave group develops a pronounced northeast- southwest tilt which is caused by the differential westward movement of wave modes, with the lower modes having relatively faster group velocities.
¾ The equatorial cyclonic shear flow southeast of the main anticyclone is an inherent property of the Rossby wave group response even in the absence of orography.
Summary - 3
¾ lt is consistent with the mean winter condition of a northeast-southwest equatorial trough along Borneo- Philippines, and with an enhanced cyclonic circulation over Borneo following surges.
¾ The surge forcing also gives rise to eastward moving wave groups of the Kelvin, mixed Rossby-gravity, and inertia- gravity (mainly n = 0) modes.
¾ This result suggests a possible interpretation for the eastward moving cloud patterns observed recently during winter.
Summary - 4
¾ The Kelvin waves forced by midlatitude forcing. as is the case for the thermally forced Kelvin waves, have more energy in the larger wavelengths for a white noise random forcing.
¾ It appears that the diverse events of the monsoon surge might be given a coherent interpretation in terms of simple, inviscid, barotropic β-plane dynamics.
¾ However, many important questions remain …!
¾ Lim, H., C.-P. Chang, 1981: A theory for midlatitude forcing of tropical motions during winter monsoons. J.
Atmos. Sci., 38, 2377-2392 .
¾ Lim, H., C.-P. Chang, 1983: Dynamics of teleconnections and Walker circulations forced by equatorial heating. J.
Atmos. Sci., 38, 2377-2392 .
¾ Silva Dias, P. L., W. H. Schubertand M. DeMaria, 1983:
Large-scale response of the tropical atmosphere to transient convection. J. Atmos. Sci., 40, 2689-2707.
References
OLR and 700 mb winds 21 – 25 April 1986
OLR and 700 mb winds 26 – 30 April 1986
OLR and 700 mb winds 01 – 05 May 1986
OLR and 700 mb winds 06 – 10 May 1986
OLR and 700 mb winds 11 – 15 May 1986
OLR and 700 mb winds 21 – 25 April 1986
A shallow water model calculation
A shallow water model calculation
Summary
1. These calculations indicate the complexity of the response of the tropics to forcing.
2. The response is in the form of various wave types that can exist on the equatorial wave-guide.
3. The interaction of these waves with convection is still not well understood.
A relevant paper
http://www.meteo.physik.uni-
muenchen.de/~roger/Tropical_Meteorology/MatWheeler2001.pdf
