Circulation and Vorticity

Ramesh Vellore
Dynamics of Atmosphere and Ocean
2021-22

Circular motion
Vorticity = 0
(no spin of the
Element)
Circular motion
Vorticity non-zero

The fluid element moving from A to B

on a circular path has no vorticity
while the fluid element moving from C
to D has vorticity.
 Vorticity and circulation
There are two complementary concepts that quantify the rotational property of a flow :
Vorticity and circulation.
For solid objects we do not speak of the vorticity of an object but instead we refer to its angular
velocity.
Vorticity is a local measure, whereas circulation is a bulk (integral) measure. Each can be defined
either in the Lagrangian sense or in the Eulerian sense.

Vorticity (ω) is defined as the curl of the velocity : Any velocity vector is intrinsically orthogonal to
 

ω    V    uˆi  vˆj  wkˆ  the corresponding vorticity vector

ˆi ˆj kˆ vorticity
    
ω  V    ˆi   ˆj   kˆ
x y z
u v w
 w v   u w   v u 
   ;     ;  =   
 y z   z x   x y 
   
 V  d l     V 
 ˆ
n dA  ω 
  nˆ dA
C A
 
A Circulation
 V  dl
 
kˆ    V    Lim 
A 0 A
Stokes's theorem: the circulation around a contour that contains a group of
vortices is just equal to the sum of the enclosed vortex strengths.
 Circulation and vorticity on sphere
Large-scale: the amount of (net) force that pushes Local scale: measure of the angular
along a closed boundary or path velocity of the fluid
One can compute the relative circulation   r  of a flow field on sphere:
Vorticity (vector)
 r  v ( ) a cos    v (   ) a  v (   ) a cos(   )   v ( ) a
r v ( ) a cos  v (   ) a v (   ) a cos(   ) v ( ) a
     Circulation (scalar)
    
r  v (   )  v ( )   v (   )cos(   )  v ( )cos  
 a    a 
      
For small  ,  a  radius of the Earth

r  v  
 a    v cos   
    
Recall that by Stokes' theorem,  r   r A [Circulation has the units of m 2s -1 ]
 r  vertical component of relative vorticity
where A  x x  (a cos   )(a  )  a 2 cos   Vorticity = Circulation
1  v   per unit area
 relsphere     v cos   constitutes
a cos      vorticity
Absolute vorticity = Relative vorticity + Planetary vorticity
Conservation of absolute vorticity :  a   r  f   r  2 sin   constant

sphere 1  v  
 abs     u cos   2 sin 
a cos      constitutes
circulation
  
Components Vc of a gradient flow Vg that contribute
to the circulation around the closed curve
  v u
On a Cartesian framework ˆ 
  k  V   
x y
 
Lim 
A 0
 V  dl  Circulation
  
A Area

Considering the circulation about a rectangular element of area  x y in the (x, y ) plane.
 
Evaluating V  d l for each side of the triangle gives the circulation
 
C   V  d l    udx  vdy 
The circulation around an infinitesimal fluid element ABCD:  C  C AB  CBC  CCD  CDA
 v   u   v u   v u 
 C  u x   v   x   y   u   y   x  v y      x y      A
 x   y   x y   x y 
Circulation  C  v u 
    
Area  A  x y 
 Vorticity and circulation in a rotating reference frame

Absolute vorticity (ωa )  vorticity as viewed in an inertial reference frame.
Relative vorticity  rel   vorticity as viewed in the rotating reference frame of the Earth.

Planetary vorticity (ω pla )  vorticity associated with the rotation of the Earth (f  2 sin  ).

 In Meteorology, by the term circulation, we mean the circulation of velocity vector V

Vorticity vector: Cyclonic

   v u
ω a    VInertial   rel  2 sin   0,  0    0
x y
Relative vorticity     Area  0
 Anticyclonic
 v u 
 rel     kˆ (Cartesian) v
 0,
u
0 0
 x y  x y
  1 v 1  (u cos  )  ˆ     Area  0
 rel     k (spherical)
 a  a cos  y 
Planetary vorticity (f ) =2 sin 

Low pressure systems  cyclones  :   0,   0, Anticlockwise flow.

High pressure systems  anticyclones  :   0,   0, Clockwise flow.
  a   r  2Ae
 a   r  (2 sin  ) A

Area projected on
the equatorial
plane

The absolute circulation is related to the relative circulation by  a   r  2Ae

where Ae (  A sin  ) is the component of the area of the loop considered that is
perpendicular to the rotation axis of the Earth
circulation (2sin A) = planetary vorticity (2sin )  area (A)
 Vorticity and
Circulation is a scalar measure of fluid rotation.
circulation
If we consider a closed path c around the fluid
flow then circulation () is defined as the
line integral of the tangential velocity around c
   
   V  d l   V cos  dl
c c

By applying the Stoke’s theorem, circulation can be related to the

   
 
vorticity as;    V  d l     V  nˆ dS   ω  nˆ dS
c
S S

In conclusion, vorticity and circulation are two primary measures of

rotation in fluid flow. Circulation is a macroscopic scalar measure of
rotation for a given area of the fluid. However, vorticity being a vector
quantity is a microscopic measure of rotation for any point in the flow.
 Circulation and angular velocity
Considerations of angular momentum of fluid parcels is particularly
important in understanding atmospheric dynamics

 Circulation has the advantage over angular velocity

that no assumption of a solid body is required and so
it is suited to describing angular momentum ideas in a fluid.

Consider a circular ring of fluid of radius R


in solid-body rotation at angular velocity Ω
about the z -axis. The velocity field can be The fundamental definition of vorticity is (2Ω),
    that is, twice the local angular velocity
written as: U = Ω  R where R is the distance
from the axis of rotation to the ring of fluid Vorticity however has nothing to do with a
  2 path, it is defined at a point and would indicate
C   U  d l   R 2 d   2  R 2  the rotation in the flow field at that point.
0

C
2
 2  twice the angular speed of rotation of the ring
R
Unlike angular momentum or angular velocity, circulation can be computed without reference
to an axis of rotation; it can thus be used to characterize fluid rotation in situations where
"angular velocity" is not easily defined
 Circulation
Circulation is a scalar that measures the rotational property of a flow stemming
from the notions of vortex lines and vortex tubes
 A vortex line is defined to be a line to
which the vorticity vector is tangential
at every point of it
(analogous to streamline)
 The totality of all vortex lines passing
through  would make up the surface
of a tube is called vortex tube
 The fluid inside such a tube is called a
vortex filament
We now define a line integral of the velocity component tangential to this chosen
closed curve (). The sign convention is that this integration is performed in an

anticlockwise direction C   V  dl

Exact differentials
Exact differentials: which are differentials whose integral around a close path is zero.
If a function f of two variables x and y has a differential that is written as
M N
df  Mdx  Ndy  then it is an exact differential if    df  0
y x

 If a diffential is function of only a single variable and is of the form, df  M  x  dx,

then it is an exact differential as long as M is integrable  f is differentiable  .

Conservative vector fields  which are vector fields g (gravity vector) that can be

written in terms of the gradient of a scalar, such as "geopotential" g  
   
A conservative vector field has the property  g  d l     d l   (   )  dA  0
A

Remember the vector property : curl of gradient of a scalar function = 0      0

   u v v u
C   V  dl   (udx  vdy )  if V is conservative,    0 0
y x x y

If   V  0 everywhere, then the field is called "irrotational", thus curl free field is
conservative, A rotational vector is a vector field whose curl can never be zero.
 Circulation theorems
Circulation theorems deal with the change or evolution in circulation and its cause(s).

For an arbitrary vector field (B), the circulation theorem states that the time rate
 
of change of circulation of B is equal to the circulation of the time rate of change of B

d  dB Velocity vector on an inertial frame
Circulation theorem:   B  dl    dl    
dt dt Va  Vr  Ω  r

 Kelvin's circulation theorem, it is applied to the absolute velocity Va of fluid motion
   
dCa d  dVa  d dVa  Va  Va 
  Va  dl    dl   Va   dl     dl   d  
dt dt dt dt dt  2 
 as the line integral of an exact differential around a closed path = Zero
In Meteorology, circulation theorem
 simply states that the acceleration
d a Ca d a Va
   dl 
dt dt of circulation is equal to the circulation
of acceleration
 
Circulation theorems: Corollary d a Ca
 
d a Va
 dl
 dt dt
dV 1 
Equation for absolute motion is given by: a a   p  g
dt 

d a Va 1   1      p 
 dt  dl     p  dl   g  dl   
S   p  ˆ
n dS  S   2   nˆ dS

 g is a conservative force field. It is also known that work done by

a conservative force field around a closed path is zero  g  dl     dl =  d  0
gravity force would not be
d a Ca     p 
Therefore, for a frictionless flow,    2   nˆ dS involved in the generation
dt S   
of rotating motion
 in a barotropic atmosphere the density,    ( p )    p  0
d a Ca
Therefore, for a frictionless barotropic flow,  0  Circulation is conserved
dt
This is a direct corollary to the Kelvin’s theorem. Hence from Kelvin’s circulation
theorem it may be stated that for frictionless flow change in absolute circulation
is solely due to the baroclinicity of the atmosphere.

d a Ca dp Baroclinic fluid integrated effect of pressure-gradient force Bjerknes circulation

  0  
dt     p  0 can change the absolute circulation theorem
Baroclinicity term   
1
p  dl    p  dl 
1 Baroclinicity
 
  p  dl      p   dA
A

     p   d l        p   dA      p   dA
A A

dCa
      p   dA     p sin  dA
dt A A

0    clockwise
 
    2  anticlockwise
  Angle between the gradients of  and p
dCa
 In the given example 0
dt
For the atmosphere, which is an ideal gas, the solenoidal
term can be written in terms of the temperature and pressure
dCa
gradients as   R  T    ln p   dA
dt A
 Bjerknes circulation theorem - examples
dCabs where C is circulation, Ae is the area of the integral circuit projected
 0 in a baroclinic fluid
dt onto the equatorial plane. The term on the right-hand side represents
dCabs d dp
  Crel  2Ae     solenoidal or baroclinic generation of circulation and 2Ae term
dt dt 
represents the change in circulation due to rotation of the Earth.
dCabs
   RTd ln p  In a baroclinic fluid, circulation may be generated by the
dt
(along a vertical plane) f is negligible pressure-density solenoid term

Below we present an example of how Bjerknes' circulation theorem can be used

to predict the circulation induced by a horizontal temperature gradient such as
might be found along a mesoscale air mass boundary.i.e., we apply the circulation
theorem by integrating around a circuit in a vertical plane
dC
   RTd ln p
dt
dC p 
 R ln  b  Tw  Tc   0 Lines of constant density
dt  pt 
pt  900 hPa, pb  1000 hPa
Tw  300 K, Tc  290 K Warm
Cold air mass
dC
 302 m 2s 2  an increase of
dt
C of 1.09  106 m 2s 1 in 1 hour
Land and sea breeze
See Holton’s book for more details

h = 434 m

dCa dp
dt
  
  RTd ln p
dCa p 
 R ln  975  TLAND  TOCEAN   302 m 2s 2
dt  p925 
temperature difference (20o C) is inducing a clockwise
circulation
To estimate wind acceleration,
 
dC dV dV du dC dt
   dl    dl  
dt dt dt dt 2( L  h)

du R ln  p975 p925 
dt

2( L  h)
 TL  TO   7.4  103 m 2s 2

L = 20 km; h  434 m (using hypsometric equation)

 which could produce a wind speed
of 25 m s 1 in 1 hour in the absence of friction
 Land and sea breeze
For the atmosphere, which is an ideal gas, the solenoidal
term can be written in terms of the temperature and
dCa
pressure gradients as   R  T    ln p   dA
dt A
Solenoidal or baroclinicity

dC 
      p   dA;   1 
 dt A

 T dC 
  R  T  (ln p )  dA
dt A
dC
0 dC
dt     p sin dA
dt A

dC
   T  ln p sin dA
dt A

If the atmosphere is barotropic

 kˆ   p     0

T  Pressure and density surfaces lie

on each other (barotropic atmosphere)

dC
0 If the atmosphere is baroclinic
dt
kˆ   p     0  cyclonic
In terms of temperature gradients,
kˆ   p     0  anticyclonic
Low level winds flow in the direction of T  0
 Vorticity and circulation in a rotating reference frame
Kelvin’s circulation theorem which states that if the fluid is barotropic on the material curve C and the
frictional force on C is zero then absolute circulation is conserved following the motion of the fluid
   

In a rotating fluid, the velocity vector (on an inertial frame) Vabs  Vrel  Ω  r frame

so that the vorticity associated with the velocity in an inertial frame is related to the velocity in a rotating
        

frame by ω abs  ω rel    Ω  r   ω abs  ω rel  2Ω where ω    V
frame

 so the vorticity in the inertial frame is equal to the vorticity seen in the rotating frame called the relative
vorticity plus the vorticity of the velocity due to the frame’s rotation which is just twice the rotation rate of
the frame
  
To examine the form C abs  C  c Ω r d r The circulation observed in a rotating frame
of Kelvin’s theorem  dCabs dC dAe   p
  2    nˆ dA
for a rotating frame Cabs  C   2Ω  nˆ dA  C  2Ae dt dt dt A  2
A

Consider the situation where viscosity can be neglected and where the fluid is barotropic    p  0  .

which just tells us that the absolute circulation is conserved

dC dAe
 2  0   In other words there will be a trade off between the relative
dt dt
and planetary vorticities
 C2  C1  2  A2 sin 2  A1 sin 1 

Ae  A sin 
 Vorticity and circulation in a rotating reference frame
  
To examine the form Cabs  C  
c

Ω  r  dr The circulation observed in a rotating frame
of Kelvin’s theorem  dCabs dC dAe   p
for a rotating frame Cabs  C   2Ω  nˆ dA  C  2Ae dt  dt  2 dt    2  nˆ dA
A
A

Consider the barotropic situation where viscosity can be neglected

which just tells us that the absolute
dC dA circulation is conserved  In other
 2 e  0 
dt dt words there will be a trade off between
the relative and planetary vorticities
 C2  C1  2  A2 sin 2  A1 sin 1 
 A material curve that is shifted to higher latitude. As the curve
An air parcel at 30o N moves northward conserving absolute
moves to higher latitude the area normal to the Earth’s rotation
vorticity. If its initial relative vorticity is 5  105 s 1 , what is
axis will increase and so the circulation associated with the
its relative vorticity upon reaching 90o N?
planetary vorticity increases (2Ae )   f 30    f 90  (5  105  )30  ( 2  2)90
 In order to conserve the absolute circulation the relative  2  2.3  105 s 1
 an anticyclonic circulation is 
circulation must decrease  
 induced 
Implication : conservation of absolute vorticity (CAV)
a negative absolute circulation in the Northern
Hemisphere can develop only if a closed chain of fluid particles
Ae  A sin 
is advected poleward  Rossby wave  CAV trajectory
 Vorticity and circulation in a rotating reference frame
The rate of change of the relative circulation
d  dC   p 
 V  dl   
  2Ω  V  dl    dl   Ffriction  dl
dt c dt c c  c

where Ffriction is the frictional force per unit mass. There are
therefore three terms that can act to alter the circulation.
 
 
 2Ω  V  dl
c

Consider the circulation around the curve C in a divergent flow.

It is clear that the coriolis force acting on the flow field acts to
induce a circulation around the curve C
Consider friction to be a linear drag on the velocity with some
timescale 
friction acts to spin-down
dC  1  C
  Ffriction  dl    V  dl    the circulation
dt c  c 
spin-down = decay
 Helmholtz theorem

Theorem: Every vector field V whose divergence and rotation possess potentials
can be written as the sum of a divergence-free vector field plus another vector field
that is irrotational.

     non-divergent (divergence-free; V )
ˆ
V = Vψ + Vχ  k ×  +   
divergent (irrotational/curl-free; V )
where ψ is the stream function and χ is the velocity potential. The relative
vorticity ( ) and divergence ( ) can be expressed as follows:
v u
 1  v     
ˆ
  k    Vψ    u cos     2
 x y

a cos     
 u v
 1  u    
    V    v cos     2
 x y

a cos     
  
 
u   ; v 
1  ˆ 1  ˆ 1  ˆ 1  ˆ y x
V   i j; V  i j
a  a cos   a cos   a   
  u  ; v 
Typically, V >> V x y
Helmholtz theorem   
ˆ
 
V = Vψ + Vχ  k ×  + 

Irrotational vector V

In vector calculus an irrotational vector field is a vector V with curl zero at all points
 Any motion in which the curl of the velocity vector is
in the field   V  0 
zero is said to be "irrotational"  potential flow

In a vector calculus a solenoidal vector field is a vector field V with zero divergence
Any motion in which the divergence

at all points in the field   V  0  of the velocity vector is zero is said
to be "solenoidal"  can produce vorticity
     
  V    Vψ    V   V    Vψ    V
    
 
  V    kˆ ×      
ˆ 
  V    k ×      
 2

  Vψ leads to       V leads to  2   
 
rotational part of V  vorticity irrotational part of V  divergence
(Measure of the non-divergent part of the wind) (measure of the divergent part of the wind)
Stream function and velocity potential
Helmholtz decomposition: On a line of constant velocity potential  ( x, y ) is constant,
  
The horizontal velocity field VH  V  V so that d  =0

In terms of scalar functions, we can write VH  kˆ     d     dsˆ  0
On a streamline,  ( x, y ) is constant, so that the  
d  dx  dy  0 (3)
total differential d =0 x y

d    dsˆ  0 Irrotational part of the winds: V  u ˆi  v ˆj
  ˆ   ˆ    ˆ 
d   ˆi j  ˆ 
  idx  jdy  0 V     i ˆj 

 x y   x y 
   
d = dx  dy  0 (1)  u ; v (4)
x y x y

Rotational part of the winds: V  u ˆi  v ˆj  dy  u
Substituting (4) in (3), udx  vdy  0    
   ˆ  ˆ   dx   v
V  kˆ      i j
 y x  This is the equation of the velocity potential.
  Since the slopes of a streamline and of a line of constant
 u ; v (2)
y x velocity potential are negative reciprocals of one another,
 dy  v streamlines and velocity potential lines must be mutually
Substituting (2) in (1), vdx  udy  0    
 dx  u perpendicular.
This is the equation of the streamline.

Lines of constant ψ (streamlines) are perpendicular

to lines of constant χ (velocity potential lines).
 Composite monsoon depression
depicted by the 850-mb
streamline field superimposed
with stream function departures
(shaded areas) from their
summer (JJA) mean values in
the short-wave regime, Δψs(850
mb), (b) composite 850-mb
velocity potential and divergent
wind departures in the short-
wave regime superimposed with
precipitation, [Δ(χs, VD) (850
mb), P]

streamlines and velocity

potential lines are mutually
perpendicular to each other.

Velocity potential
Stream function
  is irrotational, but has divergence  2  . Because of this last property,
examination of the velocity potential is especially useful as a diagnostic
tool for isolating the divergent circulation

Distribution of the upper tropospheric (200 hPa) mean seasonal velocity

potential (solid lines) and arrows indicating the divergent part of the mean seasonal
wind which is proportional to  2  . (Adapted from Krishnamurti et al. 1973).
 In x - y system  Laplacian operator  "curvature of function or field"
 v u  i 1, j  2 i , j   i 1, j  i , j 1  2 i , j   i , j 1
ˆ 
k  V    
x y
2
 
x 2
+
y 2
 u v 2
 i 1, j  2  i , j   i 1, j  i , j 1  2  i , j   i , j 1
  VH        2
+
x y x y 2
 It tells you how much the value of the field differs from its average value
taken over the surrounding points.
2 min
What is the physical significance of the Laplacian ?
2

    0   i, j  0
2  2         2 max
In 1-D    reduces to
x 2    0   i, j  0

 2  2
If  0   (x) is concave (convex if < 0)
x 2 x 2
 2  is less than the average of  in its y
 0 
x 2 surroundings
2
x x
   0  the slope of  increases in all directions,
 being less than the average of the local value. y
 2  0  the slope of  decreases in all directions,
 being locally greater than the average value.
 Velocity potential

When interpreting the  -fields, a note of caution is appropriate. Remember that

2
 
     VH and that |w| is proportional to   VH . Therefore centres of 
maximum or minimum do not coincide with centres of w maximum or minimum.
 The latter occur where  2  is a maximum or minimum.
Relationship between divergence and vertical motion
u v 
The continuity equation (in isobaric coordinates) is given by:   0
x y p

  u v  
       VH  
p  x y  H
 the velocity divergence on an isobaric surface is related to how vertical velocity
changes with height.
Let us now integrate this between two arbitrary isobaric surfaces pB and pT ,
 i.e., pB is found closer to the surface than is pT given that pressure 
where pB  pT  
 decreases with increasing altitude 
pT pT p
 T
    gw
p  dp   p p dp   p d   ( pB )   ( pT ) 0w0
B B B

 the difference in vertical motion  over some vertical layer bounded by two isobaric
levels pB and pT  where pB  pT  is equal to the vertically integrated divergence within
pT

that layer  ( pB )   ( pT )     dp
pB
 Dines compensation and level of non-divergence
Let us now consider a hypothetical atmosphere comprised of two layers :
one between the surface  psfc  and some mid-tropospheric isobaric level  pL  ,
and one between some mid-tropospheric isobaric level  pL  and the tropopause  ptrop  .
pL
 ( ptrop )  0
For the lower layer,   dp   ( p
psfc
sfc )   ( pL )

pL

The surface is a rigid bound on vertical motions, so,   dp   ( p

psfc
L )

ptrop
 ( psfc )  0
For the upper layer,
pL
  dp   ( p L )   ( ptrop )

Due to the large atmospheric stability found at the tropopause, the

tropopause itself is also treated as a rigid bound on vertical motion. therefore  ( ptrop )  0,
ptrop

thus simplifies to   dp   ( p ) L
pL

pL ptrop
    gw
Upon inspection, we can notice,    dp    dp
psfc pL 0w0
 Dines compensation and level of non-divergence
In other words, the vertically integrated divergence pL ptrop

in the lower layer is cancelled out by the vertically   dp    dp  0

psfc pL
integrated divergence in the upper layer.
Stated differently, the divergence within the lower
layer is equal in magnitude and opposite in sign to
the divergence in the upper layer. This implies that
the two are in balance with each other, such that
one compensates for the other. This important
principle is known as Dines’ compensation principle
Thus, an important corollary to Dines’ compensation principle states that there
must be at least one level at which the divergence is zero.

 where  0  max
p
 This level is the level of non-divergence. In the troposphere, we often find a
level of non-divergence in the middle troposphere, typically between 500  600 hPa.
  0  0

Horizontal divergence (solid red line) and vertical motion (dashed red line). LND = Level of non-divergence
The real atmosphere typically cannot be considered by two vertical layers. The vertical
motion at any isobaric level p is equal to the negative of the integrated divergence
p

between the surface and p   ( p )     dp

psfc

 0
In real atmosphere, there would be  0
multiple LND’s wherever there is a  0
shift from divergence to convergence,
or ascent is maximized.
Vertical distribution of vertical motion and
horizontal divergence in an extra- tropical cyclone,
an anticyclone and in a more complex synoptic-
scale disturbance, after (Sutcliffe, 1947).
Convergence/divergence and vertical motion

w  u v 
      
z  x y  z
w
 0 where w  wmax and   0
z
This level is level of
non-divergence. It is found on
average near 550-600 mb.
 Vertical motion   u v 
Typical profiles of        (1)
horizontal divergence
 p  x y  p
and vertical motion
  V
 is constrained to be zero at
the ground and at the tropopause.
If  is nonzero, its sign is often the
same at all levels in a column in the
 0 troposphere.

Then the sign of must reverse
p
at some level. At this level,

 0  From (1)   0
p
This level is level of non-divergence.
It is found on average near 550-600 mb.
!!! A quantity that is not routinely measured !!!!
Always inferred
 Pressure tendency and Dines decomposition
 Tendency: How a variable changes with time at a particular point or
locally, e.g., pressure tendency.

 Pressure in a column is due to the weight of the air above

 Lower the mass of the air above, lower the pressure

The hydrostatic equation is dp    gdz

z2

p  g   dz  z2  0 top of the atmosphere

z1

Local pressure tendency for a given air column with

thickness dz (assume the thickness is constant)

 p  
   gdz 
 t  z z t
 
 p  
Pr essure tendency:    g  dz
 t  z z t

Let us consider the mass continuity equation

     
t
 
    V      u     v     w  
 x y z 

 p      p p z 
As    g    dz     g     g
 t  z z  t  z  z t t 
 p    
    
    g    u  dz     v  dz   g    w  dz 
t
 z z x z y  z z 
 p    
  

    g     u  dz     v  dz   g  d (  w)
 t  z  z x z y  z

 p    
  z   w  0 near the ground 
    g    u dz       z  0 
 v dz  g  w 
t
 z z x z y   w  0 at z =  
 p    
 
    g     u  dz     v  dz 
 t  z  z x z
y 
Local tendency Horizontal mass convergence
 
 p  
Pr essure tendency:    g  dz
 t  z z t

 p    
 
    g     u  dz     v  dz 
 t  z  z x z y 
Local tendency Horizontal mass convergence
Remembering that 90% of the mass of the atmosphere lies beneath
the tropopause and that we are treating the density as a mean density
for the air column  a constant  , then it can be seen that, at a
synoptic-scale, surface pressure tendencies are directly related
to horizontal divergence patterns in the troposphere.
p
2 
 For synoptic scale, one can write    g    VH dz
t 1
Dines compensation principle
Large scale motions in the atmosphere are in close hydrostatic balance. Hence the
pressure at the base of a fixed column of air is proportional to the mass of air in
that column; if the total mass decreases, so will the surface pressure, and vice versa
Dines showed that low-level convergence is very nearly equal to the divergence
at upper levels and pointed out that upper divergence must exceed the low-level
convergence when a low deepens
ps  
Surface pressure tendency equation:
t 0
 
=  g     V dz

 1
Geostrophic:  V  kˆ  p
f0
 ps
 
  V  0 
t
=0

 It follows that any local change in surface pressure is

associated entirely with ageostrophic motion.
Computation of ω-adiabatic method Alternative: Quasi-geostrophic ω
A second method for inferring vertical velocities, which is not so sensitive to errors in
the measured horizontal velocities, is based on the thermodynamic energy equation:
dT Q     ambient 
 S p  , where S p  static stability =  dry  ,   lapse rate
dt cp   g 
If the diabatic heating Q is small compared to other terms in the heat balance, we can write,
1  T T T  1  T   
   u  v    VH  T 
S p  t x y  S p  t 
 Because temperature advection can usually be estimated quite accurately in midlatitudes by
using geostrophic winds, the adiabatic method can be applied when only geopotential and
temperature data are available.
A disadvantage of the adiabatic method is that the local rate of change of temperature is required.
Is this method suitable for the tropical monsoon region?  weak temperature gradients and
strong diabatic heating are characteristics of monsoon region.
T
Unless observations are taken at close intervals in time, it may be difficult to accurately estimate
t
over a wide area.
This method is also rather inaccurate in situations where strong diabatic heating is present,
such as storms in which heavy rainfall occurs over a large area.
Vorticity equation (Cartesian form) – z coordinates
 ˆ     v u 
 v u 
Absolute vorticity =   f      f
Recall:
t
 
k  V      
t t  x y 
 x y 
  u u u u p  1
  u  v  w  fv    Fx 0 (1) specific volume  
y  t x y z x  
f 1  2 cos 
  v v v v p    2 sin    
  u  v  w  fu    Fy  0 (2) y a  a
x  t x y z y  a  radius of the Earth
Subtracting (1) from (2),
     u v   w v w u    p  p  ˆ 
t
 V    w
z
  v  (  f ) 
x
 
y x z

y z

x y

y x

  k    F fric 
     

d   w v w u  ˆ 
dt
  f      f    VH    ˆ
  k   p     k    F fric  
 x z y z 
Stretching Tilting Baroclinicity Friction
 Rate of change of absolute vorticity following the motion
Non-divergent vorticity equation (RHS  0)  
  VH     f   0
(valid for barotropic atmosphere) t
 Non-divergent barotropic vorticity equation
In the case of a barotropic fluid (purely horizontal flow, w  0)
with constant fluid depth H d   w v w u  ˆ
  f      f    VH    ˆ 

  k   p     k    F fric 
dt  x z y z 
d
  f   0  absolute vorticity conservation following the horizontal motion.
dt
For horizontal motion, that is non-divergent, the flow field can be represented by
In the absence of sources such as
  stretching, tilting, baroclinicity,
stream function ( ) such that u   , v 
y x friction (non-divergent, barotropic,
inviscid fluid)
d H (  f )    f
  u  v  v 0 Absolute vorticity is conserved
dt t x x y
following the motion
 2           f 
t
       0 f 1 
  2 sin   
2 cos 

 y x   x y  x  y  y a  a

 2         2 
t
        0       J ( ,  )   0
 x y y x  x t x
The flow in the mid-troposphere is often nearly nondivergent on the synoptic scale, the
barotropic vorticity equation provides a surprisingly good model for short-term
forecasts of the synoptic-scale 500-hPa flow field.
 Westward dispersion
Linear model No vortex movement
   2 westward dispersion

 2 
t
     
x
0
  effect
Linear

J. Atmos Sci, Volume 44

r   1   r b  
V (r )  Vm   exp  1     
 rm   b   rm   
b b
2Vm  1 r   
1   r   
Initial conditions  (r )  1     exp  1     
r m  2  rm  
   b   rm   
Axi-symmetric vortex
(no radial wind)
Non-linear model
 2 
t
     J ( ,  )  
x
0   effect

Nonlinear Linear
Northwest movement of the vortex
 Divergence (stretching/shrinking) term
The divergence term represents the fluid analog to conservation Coriolis force (small vectors), convergent/
of angular momentum in rigid-body mechanics. divergent wind field (large vectors)

 dw   dw 
This represents the stretching   0  or shrinking   0
 dz   dz 
of an air column and its effects on Earth's vorticity and relative
vorticity.

Its effects are analogous to the increase and decrease in

rotational speed experienced by an ice skater whose arms
are brought in or are extended outward;

A purely convergent flow field acquires cyclonic relative

vorticity from the Earth's vorticity as the air is accelerated
to the right in the Northern Hemisphere by the Coriolis force
(and to the left in the Southern Hemisphere). Similarly a purely  u v  dw
   
divergent flow acquires anticyclonic vorticity.  x y  dz

d  w v w u  ˆ 
dt
    f      ˆ
  k   p     k    F fric  
 x z y z 
Tilting/twisting v
 0
(a) Suppose that initially the vorticity vector (  V ) points Twisting z
in the x - direction owing to a decrease in v with height.
Suppose also that there is a rising motion and sinking
motion at large and small values of x, respectively.
Then the vorticity vector will become tilted about the y -axis,

(b) so that the vorticity vector becomes more aligned

with the z - axis. Thus, the vorticity about the x-axis has
Tilting
been converted into vertical vorticity.
Mathematically, the tilting term in this case,

 V   v w u w  w
kˆ    w      0
 x
 z   z x z y 

d   w v w u  ˆ 
dt
    f    VH  
x z

y z
ˆ 
  k   p     k    F fric  Critical for
 
tornadoes
Vertical shear in the horizontal motion is going to twist
Horizontal shear in the vertical motion is going to tilt
 Vorticity equation – Tilting/twisting term
Vertical shear in the horizontal motion is going to twist
Horizontal shear in the vertical motion is going to tilt

 w v w u 
   w
 x  z y z  0
y
w u u
 0, 0 0
y z z

d
  f   0
dt
twisting arises due to vertical shear
in the horizontal motion, and horizontal
shear in the vertical motion

d   w v w u 
  f      f    V    
dt  x z y z 

 
 kˆ   p     kˆ    F fric
 Solenoidal or baroclinicity dC 
      p   dA
dC dt
0 A
dt dC
Solenoidal term:    p sin 
dt

If the atmosphere is barotropic
kˆ   p     0
Pressure and density surfaces lie
on each other

If the atmosphere is baroclinic

kˆ   p     0  cyclonic

 kˆ   p     0  anticyclonic
Pressure and density surfaces do not
lie on each other. Solenoidal term
acts as a circulation mechanism to
dC bring the pressure and density surfaces
0
dt lie on each other
In general the circulation that develops

would be such that the density and d   V  ˆ 
pressure surfaces would become parallel dt
   f  
    f    V H  ˆ
k     w   k    p      ˆ
k   F friction 
 z 
(baroclinic barotropic)
Stretching Tilting Solenoidal Friction
 Effect of friction Example: western boundary
Illustration of frictional generation of vorticity currents in the ocean
alongside a wall

(a) Air is forced to flow (solid-line vectors) along the

edge of a vertical wall (hatched area). Friction
(dashed vectors) acts in the direction opposite to that
of the wind; the magnitude of the friction force is
greatest along the edge of the wall, and decreases
in magnitude with distance from the wall. Thus the
wind is slowed down the most along the wall.

(b) and least away from the wall.


ˆ 
In this figure, k    F fric  0 

d   V  
dt
    f    VH  kˆ   
 w   kˆ   p     kˆ    F friction 
 z 
Stretching Tilting Solenoidal Friction
Vorticity equation (Cartesian form): (x,y,p,t)
 v u  dp
Absolute vorticity p =    f;   geopotential height ( h   g )
 x y  p dt
 height of isobaric surface
from the sea level
  u u u u  
 u v   fv   Fx   0 (1)
y  t x y p x 
  v v v v   h geometric
  u  v    fu   Fy 0 (2) height (z )
x  t x y p y  sea level
Subtracting (1) from (2),
      v   u  ˆ 
t
 VH   p  
p
 v   (  f )  VH    
  k    F fric 
 p x y p 
Absolute vorticity (p -surface)
Stretching Tilting/Twisting Friction
following the motion

Advantage: Solenoidal/baroclinity
terms are implicit
sˆ direction is parallel to flow, positive in direction of flow natural coordinate
nˆ direction is perpendicular to the flow, positive to left of flow. viewpoint
Calculate circulation
Denote the distance along the top leg as  s
Denote the distance along the bottom leg as  s  d  s 
Denote the velocity along the bottom leg as V
Note: only curved sides of this box will contribute to the
circulation, since the wind velocity is zero on the sides
in the nˆ direction.
 V 
Using Taylor's expansion, velocity along the top:   V  n V V
 n    
d  s    n R n
  V 
C   V  dl  V ( s   n)   V   n  s
 n 

V V curvature
C  V  n   n s 
n R vorticity
 C   V V V V shear
  Lim    V    
 n s  0  n s s n R n
  n vorticity
Physical interpretation V V
  
R n
300 hPa winds (straight jet)

V
Shear vorticity
v u n
   300 hPa heights
x y

V
Curvature vorticity
R
Remember: For adiabatic motions
 a    f is conserved for 2D
inviscid barotropic fluid motions
In stratified (baroclinic) 3D fluid,
equivalent is potential vorticity
potential for
 a   
q     generating
  z 
vorticity
 Potential vorticity in barotropic fluids
A model that has proved useful for elucidating some aspects of the horizontal structure of large-scale
atmospheric motions is the barotropic model. In the most general version of this model, the atmosphere
is represented as a homogeneous incompressible fluid of variable depth, h( x, y, t )  z2 - z1 , where
z2 and z1 are the heights of the upper and lower boundaries, respectively.
In a barotropic (incompressible) fluid, the vorticity equation (combined with the continuity equation) for
can be written as:
d  w w( z )  w( z1 )
  f      f    VH    f     f  2
dt z h
d  dz dz  dh
h   f     f   2  1     f 
dt  dt dt  dt
1 d 1 dh d d ln h d   f 
  f    ln   f    ln  0
  f  dt h dt dt dt dt  h 
d   f  d   f 
ln    0 implies that  0 BAROTROPIC VORTICITY EQUATION
dt  h  dt  h 
f
 = potential vorticity  potential vorticity is conserved following the motion in a barotropic
h
atmosphere. This is also called 'Rossby potential vorticity'.

For baroclinic fluids, potential vorticity is a function of all dependent


variables of the fluid (V, and  ).
 Conservation of potential vorticity is the air’s equivalent of the conservation of angular momentum
POTENTIAL VORTICITY IN A BAROTROPIC FLUID
d   f 
0 When a spinning ice
dt  h 
f
skater has her arms
 is conserved following the fluid parcel in a barotropic fluid
h spread out laterally, she
POTENTIAL VORTICITY IN A BAROCLINIC FLUID spins slowly. When she
On isentropic surfaces,
contracts her arms, her

P   g (  f )
p
( Ertel ' s potential vorticity ) rate of spin accelerates.
P is conserved following an air parcel in adiabatic flow, and is therefore a good tracer of air parcels under
conditions where diabatic heating (latent heat of condensation, radiation, etc.) can be neglected.

When air converges, the column stretches. To maintain potential vorticity,

the air spins faster (vorticity increases), resulting in the stretched vortex on
the right. Divergence, on the other hand, causes vortex spreading and slows
down the rate of spin.
 Generalized potential vorticity equation
The concept of potential vorticity (usually denoted by a variable q ) was generalized by Ertel, during
World War II, to include the complete 3D vorticity vector. Ertel's now-famous analysis, however was
apparently done independently of Rossby's work.
Ertel's work included the effects of friction and diabatic heating, the potential Compressible
vorticity equation was formulated with height as a vertical coordinate. fluids

V    1 
Equation of motion in 3D
t
 

  V   V  fkˆ  V  p  gkˆ  F friction

  1     
     
Using the vector identity, A   A   A  A  A    A , A  arbitrary vector, and let   gz
2

V    1  1 
We can write,
t
 

 V    V  fkˆ      V  V   p  F friction (1)
2  
dp   1 
The differential form of thermodynamic energy equation is Tds  c p dT   T  s  d r  c p T  d r  p  d r
 

s  specific entropy per unit mass = c p ln  ; dr  dxˆi  dyˆj  dzkˆ
1  
It follows that T s  c pT   p, substituting this in (1), we get absolute vorticity   V  fkˆ  ωa


V    1  
t
 

 V    V  fkˆ      c pT  V  V   T s  F friction
2 
     1   
  V      V    V  fkˆ         c pT  V  V      T s    F friction
 
t       2  READ Sec. 3.8 of
Atmospheric dynamics – Mankin Mak (2011)
 Generalized potential vorticity equation
    
t      
  V      V    V  fkˆ     T s    Ffriction

Following the vector identities

         
    
  A  B  A (  B)  B   A  A   B  B   A   
         
     
 A  B = B   A  A   B  B    A +A    B    
   
 
  cA  c  A  c  A,     A  0,   a  0


 V      V  fkˆ   V  V     V  fkˆ +    V  fkˆ    V  T  s    F
t
        friction

d    1 d  
dt
  
  V  fkˆ     V  fkˆ    V 
  
 dt
 
  V  fkˆ  T  s    F friction

Dividing both sides by 

1 d  1 d  1   1 1 
 dt

  V  fkˆ  2 
 dt
   
  V  fkˆ     V  fkˆ    V  T  s    F friction
   

  V  fkˆ
Define the Ertel's potential vorticity by q, q   s

 
dq d    V  fkˆ    V  fkˆ d
 s     s
dt dt     dt
d s 
dt
s    V   s
t
 
 Generalized potential vorticity equation
1 d  1 d  1   1 1 
 dt

  V  fkˆ  2
 dt
   
  V  fkˆ     V  fkˆ    V  T  s    F friction
   
(i )


  V  fkˆ
Define the Ertel's potential vorticity by q, q   s

 
dq d    V  fkˆ    V  fkˆ d
 s     s (ii )
dt dt     dt
d s 
dt
s    V   s
t
 
   
   
V   s   V  s   s    V  s    V  
d ds  
dt dt

s     s    V  s    V  (iii )

Substituting (iii), (i) in (ii)


dq  1    1 1     V  fkˆ  ds   
 
 s     V  fkˆ    V  T  s    F friction    
     s    V  s    V 
dt         dt 
1 1 ds d ln 
Given that, s  T  s  0 and using Q   cp
 T dt dt
 
dq 1      V  f ˆ
k  Q    V  fkˆ 

dt 
   
s     V  fkˆ    V  s    F friction +  
  
T  
  s    V 

1 ˆ 


fk   s    V 
 Generalized potential vorticity equation
 
dq 1      V  f ˆ
k      V  fkˆ
Q 

dt 
 
  
s     V  fkˆ    V  s    F friction + 
  
T  
  s    V 

1 ˆ 

f k  s    V


   
   
It is a vector property that s    V   V    V   s    V
  
   
It is left as an exercise for the reader to verify that s  fkˆ   V  fkˆ   s    V  fkˆ  s    V  0

Upon substitution,
 1 
dq 1  1   Q  ω  Q 
 s    F friction 
dt  
 ˆ 
  V  fk      s    F friction 
T   
a
 
T 
 dq
For frictionless, F friction  0, and adiabatic Q  0   0 (Rossby's potential vorticity equation)
dt
The derivation is valid even if the atmosphere is non-hydrostatic as the derivation did not make use of
hydrostatic assumption. Thus, the equation for Ertel potential vorticity (q ) has broad applications
 
for flow in the atmosphere. absolute vorticity   V  fkˆ  ωa
q of a fluid parcel would increase if

(i) the gradient of heating has component in the direction of absolute vorticity vector ω a
(ii) the curl of the frictional force has a component in the direction of the gradient of potential temperature.
Rossby, C. G., (1940): Planetary flow patterns in the atmosphere. Quart. J. Roy. Met. Soc., 66, 68-87.
Ertel, H., (1942): Ein Neuer hydrodynamischer Wirbelsatz. Met. Z., 271-281.
 PV of a compressible fluid
 
dq ωa  Q    F friction
      s  
dt  T   q
ω a  


z    f 
 z

 z

s  c p ln   specific entropy per unit mass  q   g (  f )
p
= constant.

q is often referred to as the IPV

q  Ertel potential vorticity of a compressible fluid
  
dq d  ω a    ω a  Q     F friction
 
 
 
dt dt     

The PV of a fluid parcel would increase if

(a) the gradient of heating has a component in the direction of the absolute vorticity vector
and/or
(b) the curl of thefrictional force has a component in the direction of the gradient of  .

q is conserved under adiabatic and inviscid conditions

 Why vertical component is important
for synoptic scale motions?
 
dq ω a  Q    F friction
      s 
dt  T  
s  c p ln   specific entropy per unit mass.

ω a  s
q  Ertel potential vorticity of a compressible fluid,

q is conserved under adiabatic and inviscid conditions.
For large-scale atmospheric motions,
Vertical Horizontal
  v u   u v 
ωa 104  105 s 1   f    10 3 s 1   , 
 x y   z z 

= s 105  102 s 1 108 s 1


This suggests that the vertical component of ωa is rather important for large - scale (synoptic scale)
motions.
Therefore, a simplified Ertel's formula can be written as follows:

ω a    z    f  
q    q   g (  f ) = constant.
  z  z p
q is often referred to as the "Isentropic Potential Vorticity (IPV)"

The potential vorticity is a quantity that is related to the absolute vorticity (ωa ) and the stratification
(θ) that is materially conserved in the absence of friction or diabatic heating.
 Use of potential vorticity
The real atmosphere is usually baroclinic, so it is more appropriate to use conservation
of Ertel's potential vorticity.
 
q   g (  f ) =  g a   a  component of absolute vorticity
p p
normal to an isentropic surface. P is also referred to as the ísentropic potential vorticity (IPV)

If the flow is adiabatic, any change in IPV must be due to the advection of IPV.
If the flow is not adiabatic, IPV can be used to diagnose where and when the diabatic
processes are acting to influence the flow.
Since diabatic processes are associated with creation and destruction of PV, the
Lagrangian rate of change of PV is
dq   d 
  g   f     PV is increased when the vertical gradient of diabatic heating is
dt p  dt 
positive.
 What is barotropic or baroclinic?
– The atmosphere is both barotropic and part baroclinic. Barotropic is very consistent, no air
masses, no fronts and is characteristic of the "tropics“. Baroclinic is much more variable.
Different air masses, cold fronts, development of cyclones. Baroclinic is characteristic of extra-
tropical regions.
• BAROTROPIC
– Region of uniform temperature distribution; A lack of fronts. Everyday being
about hot and humid (with no cold fronts to cool things off) surroundings
would be a barotropic type atmosphere. Part of the word barotropic is tropic.
The tropical latitudes are barotropic as there are no fronts in the tropics.
• BAROCLINIC
– Distinct air mass regions exist. Fronts separate warmer from colder air. In a
synoptic scale baroclinic environment you will find the polar jet in the
vicinity, troughs of low pressure (mid-latitude cyclones) and frontal
boundaries.
– There are clear density gradients in a baroclinic environment caused by the
fronts. Any time you are near a mid-latitude cyclone you are in a baroclinic
environment.
– Part of the word baroclinic is clinic. If the atmosphere is out of balance, it is
baroclinic, just as if a person felt out of balance they would need to go to a
clinic.
 Barotropic and baroclinic
With solenoidal
No solenoidal

A barotropic atmosphere is one in which An atmosphere in which density depends

the density depends only in the pressure, so on both temperature and pressure is called
that isobaric surfaces are also surfaces of an baroclinic atmosphere.
constant density.
 Idealized barotropic and baroclinic structures
No vertical shear Vertical shear

Temperatures in a barotropic environment are Baroclinic motions :

homogenous and uniform. Warm and humid
throughout the year with little fluctuation in Temperatures in a baroclinic environment
temperature. are heterogenous – which means large
In a barotropic environment in the tropics there temperature differentials.
is NO advections such as cold air advection or
warm air advection, Lower and upper troposphere communicates
 No fronts such as cold or warm fronts, nor with a finite vertical velocity.
In a Baroclinic environment the upper levels (500
any occluded fronts are present here. mb and above) are characterized by large WAVES –
Ridges and Troughs. Primarily the Polar Jetstream
governs the troughs and Ridges
 Taylor-Proudman Theorem
If the fluid is homogeneous ( uniform) then, the geostrophic flow is two-dimensional

and does not vary in the direction of the rotation vector, Ω.
If  and f are constant, then taking
the vertical derivative of the geostrophic
 ug , vg   f1   py , px   flow components and using hydrostatic
 
 u g vg 
balance, we see that  , 0
 z z 
If the flow is sufficiently slow and steady (Ro << 1) and friction
 1
is negligible, then 2Ω  V  p  gkˆ  0

   1  The Taylor-Proudman theorem,
Taking the curl   2Ω  V  p  gkˆ  , we find that
    states that slow, steady, frictionless
if the fluid is barotropic [i.e., one in which    ( p ), then flow of a barotropic, incompressible
 fluid is two-dimensional and does
 Ω    V  0.
not vary in the direction of the
Since Ω   is the gradient operation in the direction of Ω, i.e., kˆ
 rotation vector Ω
V
  0  barotropic = no vertical shear
z
 Baroclinic motions : lower and upper
troposphere communicates with a finite
vertical velocity.
p    0 p    0

Vχ
Baroclinic Baroclinic
Barotropic


Vψ

Cross-section of the zonally averaged isopycnals plotted against pressure for the NH
winter (1998.2010: December.February).
 In the middle latitudes, in the vicinity of the westerly maximum, the atmosphere is
highly baroclinic, as indicated by the large angle between the  and p gradients
(vectors p and  ).
 In the tropics, however, the gradients are almost parallel, indicating a lack of
.
baroclinicity. The root word of Barotropic is tropic.
p     0 p     0
 Equivalent Barotropic System
• Barotropic systems are characterized by a lack of wind shear
(temperature is uniform, no temperature gradient). Usually,
in operational meteorology, references to barotropic systems
refer to equivalent barotropic systems - systems in which
temperature gradients exist, but are parallel to height
gradients on a constant pressure surface. In such systems,
height contours and isotherms are parallel everywhere, and
winds do not change direction with height.

While some systems (such as closed lows or cutoff lows) may

reach a state that is close to equivalent barotropic, the term
barotropic system usually is used in a relative sense to
describe systems that are really only close to being equivalent
barotropic, i.e., isotherms and height contours are nearly
parallel everywhere and directional wind shear is weak.
What is equivalent Baroclinic
barotropic?
In the atmosphere the isotherms
are sometimes parallel to the
height contours. Often times
meteorologists say “barotropic”
when they really mean “equivalent
barotropic”

 If the isotherms are very widely spaced then

the region or level is close to barotropic.
 If the isotherms are parallel to the height
contours then the region or level is
equivalent barotropic.
 If the isotherms cross the height contours
the region or level is baroclinic
Variation of wind with height is vertically
averaged assuming that the thermal wind is The low over Kansas at 850 mb is nearly equivalent barotropic since
in the same direction as the geostrophic wind the height and thickness lines are nearly parallel
at all heights

Barotropic

Baroclinic
Equivalent barotropic:
 Thermal/contour trough axes in phase.
 Thermal/contour ridge axes in phase.
 Longwave troughs - cold core
 Longwave ridges - warm core.

Baroclinic:
The state of the atmosphere where
isotherms exist on isobaric charts
and these isotherms intersect the
height contours (i.e., isotherms and
height contours are “out-of-phase”
with one another).

Vertical shear is allowed. Wind

direction changes with height, and is
usually accompanied by speed
changes.
 Highly baroclinic

Equivalent barotropic

Equivalent barotropic:
Weak wind shear and directional change

Tropics: height contours and thickness contours are spread very far apart
Note that equivalent barotropic low near the Philippines.
 Barotropic Baroclinic Equivalent barotropic
Dines compensation
 0
D
D D

No tilt Westward tilt Eastward tilt

 drift
No thermal wind No significant
wind direction change

C C
C
 0
Summary: Vorticity and Circulation

Vorticity is defined to be the curl of velocity, and normally denoted by the symbol ω
The three-dimensional vorticity vector is given by:
ˆi ˆj kˆ
      w v  ˆ  u w  ˆ  v u 
ω  V   ˆi     j   k  
x y z  y z   z x   x y 
u v w

In atmospheric and oceanic sciences, we are primarily concerned with circulations in

the horizontal plane, vorticity implies the vertical component (unless otherwise stated.)
 v u
  kˆ  ω   (vorticity has the units analogous to "rotations" per second).
x y

Circulation is defined to be the integral of velocity around a closed loop C   V  dl
The circulation round any reducible closed curve is equal to the integral of vorticity over
an open surface bounded by the curve and, equivalently, is equal to the strength of the
vortex-tube formed by all the vortex lines passing through the curve.
  
 
C   V  dl     V  nˆ dA
line Area
Circulation Theorems
 Bjerknes: Absolute circulation is changed by solenoidal term
 Kelvin: Absolute circulation is conserved in barotropic fluids
dC d  In a barotropic fluid,    ( p) therefore an exact differential  Kelvin's theorem

dt dt  V  dl In a baroclinic fluid,    ( p, T ) Not exact differential  Bjerknes theorem
Kelvins circulation theorem
dC states that the circulation around a closed curve moving with a frictionless,
  dp   f ( p)dp  0 
dt
0
barotropic fluid (no solenoids) is constant    f  conserved, gives rise
dp 1 1
  ( p)   p2    p1   0
  to Rossby waves in barotropic fluids
 If baroclinicity arises, solenoid brings the conditions back to barotropic
circulation conservation
through a closed path of circulation
Cabs  Crel +(2sin ) A  constant
Bjerknes circulation theorem
dp
In a baroclinic fluid     RT ln p

Ca  Cr  (2 sin  ) A
dC
 (p   )  dA    p sin  dA
dt 
A A
Crel
dC
  R  (T   ln p)  dA
dt A

Change of circulation is by baroclinicity

fA
(p   )  solenoidal term
Bulletin of American Meteorological Society (2003),
Pages 471-480