DV DT FK V P: Isobaric Coordinates

Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

ISOBARIC COORDINATES

How do we write the horizontal equations of motion in pressure coordinates? Starting with the
vector equation in spherical coordinates,

dV
------- + fkˆ × V = – --1- ∇p
dt ρ

What have we neglected above?


Above,

V = uiˆ + vjˆ horizontal velocity vector


and,
∂ ∂
∇ = ˆi ----- + ˆj ----- horizontal gradient operator
∂x ∂y

The pressure gradient force (PGF) in isobaric coordinates is given as:

1
– --- ∇p = – ∇ p Φ
ρ

The total derivative in Cartesian coordinates is

z-fixed dz/dt
d ∂ ∂ ∂ ∂
----- = ---- + u ----- + v ----- + w -----
dt ∂t ∂x ∂y ∂z

and the total derivative in isobaric coordinates is

p-fixed ω=dp/dt
d- ∂ ∂ ∂ ∂ ∂ ∂
---- = ---- + u ----- + v ----- + ω ------ = ---- + V h ⋅ ∇ p + ω ------
dt ∂t ∂x ∂y ∂p ∂t ∂p

Looking at ω, we have

dp ∂p ∂p ∂p ∂p ∂p
ω ≡ ------ = ------ + u ------ + v ------ + w ------ ≈ w ------
dt ∂t ∂x ∂y ∂z ∂z

from scaling arguments that ∂ p/ ∂ t, u ∂ p/ ∂ x and v ∂ p/ ∂ y are small, so we retain the w ∂ p/ ∂ z


term only. We can substitute in the hydrostatic approximation (good for what scale flow?),
∂p
ω ≈ w ------ = – wρg
∂z

Note that the ‘-’ sign is consistent with the idea that a rising parcel experiences decreasing pres-
sure, i.e. w > 0, ω < 0.

Now is a good time to return to mass conservation and take a look at that equation in pressure
coordinates (we skipped this section - Holton 3.1.2, pg 59)

Consider an infinitesimal fluid parcel confined between two pressure surfaces p0 and p0 + δp,

p0 + δp

δp<0 δz

p0 δy
δx

We have

dM ( ρδxδyδz )-
d-----------------------------
-------- = 0, M=ρV = ρδxδyδz → = 0
dt dt

Assume that atmosphere is hydrostatic, δp = -ρgδz OR δz = -δp/ρg, substitute in above

1 d ( δxδyδp ) ( δxδyδp )-
d-------------------------- dδp dδx dδy
– --- --------------------------- = 0 ∴ = δxδy --------- + δyδp --------- + δxδp ---------
g dt dt dt dt dt

and
1 - dδp dδx dδy  1 dδp 1 dδx 1 dδy
------------------  δxδy --------- + δyδp --------- + δxδp ---------  = ------ --------- + ----- --------- + ----- --------- = 0
δxδyδp  dt dt dt  δp dt δx dt δy dt

Recall from our lecture on (???)

--------- = ∂u
1 dδx
----- ------, etc....
δx dt ∂x

we arrive at the mass conservation equation in pressure coordinates,


 ∂u
------ ∂v
----- ∂ω
-------
 ∂x + ∂y p + ∂p = 0

Note that:

*ρ does not appear in this equation directly but it is in there implicitly!


*ONLY one assumption made (hydrostatic)

The mass conservation equation is simplified when expressed in pressure coordinates, recall the
Cartesian form,

∂u- ∂v ∂w 1 dρ ∂u ∂v ∂w 1 ∂ρ ∂ρ ∂ρ ∂ρ
----- + ----- + ------- + --- ------ = ------ + ----- + ------- + ---  ------ + u ------ + v ------ + w ------ = 0
∂x ∂y ∂z ρ dt ∂x ∂y ∂z ρ ∂t ∂x ∂y ∂z

and the anelastic form

∂u
------ + ∂v
----- + ∂w w- dρ
------- + ---- --------0- = 0
∂x ∂y ∂z ρ 0 dz

both of which involve density and are more mathematically complicated!

Isobaric form is simpler! This is another major reason why isobaric coords are used. [PGF, geo-
strophic wind, thermal wind and several vorticity theorems (oh goody!) are also simpler in iso-
baric coords].

We can back out ω from the pressure coordinate mass conservation equation, i.e.

ω(p) p p
∂u ∂v ∂u ∂v
∫ ∂ ω′ = – ∫  ------ + ----- ∂p′ → ω ( p ) – ω ( p s ) = – ∫  ------ + ----- ∂p′
 ∂x ∂y  ∂x ∂y
ω ( ps ) ps p ps p

ω(ps) is frequently set to zero (it’s the ground), but this is not really precise, recall

∂p ∂p ∂p ∂p
ω = ------ + u ------ + v ------ + w ------
∂t ∂x ∂y ∂z
Note that w = 0 at the ground, thus

∂p ∂p ∂p
ω = ------ + u ------ + v ------
∂t ∂x ∂y

Thus, the w ∂ p/ ∂ z term is NOT dominant at the surface when w = 0!


HOLTON CH 4 VORTICITY

Vorticity is a measure of rotation in a fluid and has both a scalar magnitude and direction (i.e. it is
a VECTOR!), it is defined as

∇×U≡ω

The vorticity vector is 3-D, i.e. it has hz and vertical components,

ˆi ˆj kˆ ∂w ∂v ∂u ∂w ∂v ∂u
∇ × U = ω = ∂ ⁄ ∂x ∂ ⁄ ∂y ∂ ⁄ ∂z =  ------- – ----- ˆi +  ------ – ------- ˆj +  ----- – ------ kˆ
 ∂y ∂z  ∂z ∂x   ∂x ∂y
u v w

To get any scalar component of the vorticity take the dot product with the respective unit vector,
so for example to get the vertical component

∂v ∂u
kˆ ⋅ ω = ----- – ------ ≡ ζ
∂x ∂y

Recall in our Coriolis discussions in chapter’s 1 and 2, that u is the relative velocity component,
thus ∇ × U = ω is the ‘relative’ vorticity. Similarly, recall that the absolute velocity (i.e. the

velocity viewed from a fixed reference frame) is U a = U + Ω × r , hence the ‘absolute’ vorticity
is given by

ω a = ∇ × U a = ∇ ×  U + Ω × r = ∇ × U + ∇ ×  Ω × r = ω + 2Ω
   
Thus,

ω a = ω + 2Ω

In words then, the absolute vorticity is just the relative vorticity + the earth’s vorticity (which is
just twice the angular velocity of the earth). We shall see that the ‘relative’ vorticity is twice the
local angular momentum. An air parcel has vorticity if it is rotating about an axis through itself.
The flow doesn’t have to have an apparent vortex to have vorticity, the best way to see this is to
invoke the concept of a pinwheel. Assume unidirectional wind shear profile, u = αZ, v = 0, w = 0,
where α is assumed constant
z
y
x

pinwheel rotating about j axis

You might also like