Dynamics 1
Dynamics 1
Dynamics 1
Lecture Note 3
Eulerian Lagrangian
• rigid control volume, fixed in • control volume moves with
space (though still may be air parcel
rotating with Earth) • control volume stretches and
• must consider flow through shrinks with motion
boundaries of control volume, • “easier” for derivations
thus material inside may be • can be trouble for solving
replaced problems,e.g. the fluid
• easier for solving problems element can become highly
distorted
Conservation of Momentum
We consider an arbitrary macroscopic volume element V, momentum may be
transported into this control volume through convection: the bulk flow of fluid across
the surface. It may also be transported into the control volume through forces which
act on its surfaces. In addition, body forces acting on the material in the volume
change its momentum. We can express these various possibilities verbally as:
d a Va 1
= − ∇p + g a + ν∇ 2Va
dt ρ
Total differentiation
If a variable is a function of space and time, say T(x, y, z, t), then the total (or exact)
differential of T is
Consider the finite-difference form of the above equation (replace d’s with δ’s), divide
both sides by δt and take the limit as δt goes to zero. Because the derivative with
respect to t is
we can write
where we have replaced the small d’s with big D’s to remind ourselves that this is
the time rate of change following the motion. The indices on the partial derivatives
are usually suppressed. Now with u = Dx/Dt, and so forth
U is Holton’s notation for the three dimensional velocity vector (not V as we were
using in earlier discussion). The partial derivative indicates the local time rate of
change at a fixed location (ie Eulerian). It equals the time rate of change following
the air parcel (e.g, which might be warming via condensation) plus advection. The
advection accounts for flow across the control volume boundaries.
Let f = 2 Ω sin φ, which is Coriolis parameter, and then the equations of motion may
be written as:
du ∂p
= −α + vf − 2Ωw cos φ + Fx (1)
dt ∂x
dv ∂p
= −α − uf + Fy ( 2)
dt ∂y
dw ∂p
= −α − 2Ωu cos φ − g + Fz (3)
dt ∂z
where Fx, Fy and Fz are components of frictional force.
The importance of various terms in the equations of motion varies greatly with the
type of motion, and in numerous problems considerable simplification can be
introduced without distortion of essential features.
From equations of motion we see that, in the first place, the pressure terms are
always important, for, otherwise, each parcel of air would move more or less
independently of the adjacent parcels. It follows then that at least one of the other
terms in each of the equations must be comparable in magnitude with the
appropriate pressure term.
Hence, we see that if O(u) ≅ 100 m/s, then the Coriolis force would be about ≅ 0.01
m s-2, which is negligible in comparison to g.
Observations show that the vertical component of the wind velocity of the large scale
currents rarely exceeds 0.1 m/s, i.e., O(w) ≅ 0.1 m/s and that it retains its sign for a
few hours.
If the vertical acceleration (dw/dt) were as small as 0.0001 m s-2, a vertical velocity of
0.1 m/s would be reached in less than half an hour.
Thus, as far as the large scale air currents are concerned, it is permissible to omit
the vertical components of the Coriolis term & the acceleration.
Also,
∂p
O(α ) ~ 9 m s-2
∂z
∂p
0 = −α −g
∂z
∂p
− = ρg (6)
∂z
This is well known hydrostatic equation, stating that the difference in the pressure
between two points, the one above the other, is equal to the weight of the air column
between the two points.
If we are studying the problems of vertical motion, then (dw/dt) can not be neglected.
Observations indicate that vertical accelerations of about 0.1 m s-2 are likely to occur
in strong convection, and the structure of hail stones indicate that the vertical
acceleration as large as 0.5 m s-2 are plausible in extreme cases.
dw ∂p
= −α −g (8)
dt ∂z
Similarly from equation (1), we see that the term containing w may be neglected for
large scale currents.
Thus, near the Earth’s surface the equations of motion are given by (taking into
account the force of friction)
du ∂p
= −α + vf + Fx
dt ∂x
dv ∂p
= −α − uf + Fy
dt ∂y
∂p
0 = −α − g + Fz (9)
∂z
It is seen that the influence of friction decreases with elevation, and is, in general,
negligible above approximately 800 m above ground level. In the free atmosphere it
is satisfactory, except in special problems, to omit the frictional terms. The equations
of motion may then be written as:
du ∂p
= −α + fv
dt ∂x
dv ∂p
= −α − fu
dt ∂y
∂p
0 = −α −g (10)
∂z
We know that the influence of Coriolis force is to cause the motion to deviate from
the direction of the sum of the real forces. With the omission of the frictional force, as
in equation (10), the only remaining real forces are the gravity and pressure force.
The large scale currents are almost horizontal, i.e., almost normal to gravity and the
vertical component of pressure force. Observations from the free atmosphere (where
the frictional force is small or negligible) show that horizontal motion is very nearly at
right angles to horizontal pressure force.
Since, the Coriolis force is at right angles to the motion, it follows that the Coriolis
force will oppose the horizontal pressure force, and observations show that the two
forces are very nearly balanced.
∂p
0 = −α + fv g
∂x
∂p
0 = −α − fu g (11)
∂y
subscript g represents geostrophic wind velocity. We can write
α ∂p
ug = − (12)
f ∂y
α ∂p
vg = (13)
f ∂x
Equations (10) where time derivative is present are known as prognostic
equations.
Equations (12) and (13) where there is no time derivative are known as diagnostic
equations.
(14)
Thus, knowledge of the pressure distribution at any time determines the geostrophic
wind. It should be kept clearly in mind that (14) always defines the geostrophic wind,
but only for large-scale motions away from the equator should the geostrophic wind
be used as an approximation to the actual horizontal wind field.
Now we will prove that (12) and (13) holds Buys Ballot’s Law, i. e., blowing winds
keep low pressure to theleft and high pressure to the right (A law describing the
relationship of the horizontal wind direction in the atmosphere to the pressure
distribution: if one stands with one's back to the wind, the pressure to the left is lower
than to the right in the Northern Hemisphere; in the Southern Hemisphere the
relation is reversed. Also known as baric wind law).
∂p
ug α −
∂y
i.e., ug increases as p decreases with respect to y (or as y increases p decreases)
We can draw the diagram as shown. It is seen from diagram that ug hold BB Law,
i.e., it keeps L at left and H at right.
y
Low
ug
High
p
x
0 ug
From equation (13) we can draw the diagram as follows, which clearly shows that vg
satisfies BB Law.
y
vg
vg
Low High
x
0 p