Thompkins Atmos Physics Handout

1
Atmospheric Physics
Adrian Tompkins, ICTP, Trieste, Italy

Email: [email protected]
Version release date: November 9, 2015
These course notes and presentations have benefitted enormously from material taken (sometimes
even with permission!) from Stephan de Roode, Keith Shine, John Chiang, Vince Larson, Bill
Cotton, Ulrike Lohmann, Francesca Di Giuseppe, Denis Hartmann, Stephen Lower, in addition to
figures and graphics from many other sources including the IPCC reports and Wikipedia. Apologies
to authors of material that have still not been attributed. To my students, please note that brown
links are citations, blue are cross-document links and red links are for external web resources.
The index and contents page may also be useful. Please contact me for citation corrections,
acknowledgments, suggestions and above, corrections to the numerous mistakes!
2
Contents
1 Dry Thermodynamics 5
1.1 Kinetic theory of heat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Equation of state: The ideal gas law . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 The first law of thermodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4 Rules for differentiating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5 Enthalpy and specific heat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.6 Hydrostatic balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.7 Adiabatic Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.8 Potential Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.9 Entropy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.10 Thermodynamic charts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.11 Buoyancy force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.11.1 Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.11.2 Buoyancy of an air parcel . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.12 Introduction to convection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.13 Atmospheric Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2 Moist Thermodynamics 23
2.1 Saturation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.2 Other measures of water vapour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.3 Water variables in the liquid and ice state . . . . . . . . . . . . . . . . . . . . . . . 31
2.4 Specific heat of moist air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.5 Ways of reaching saturation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3 Atmospheric Convection 39
3.0.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.1 Convection in the atmospheric boundary Layer . . . . . . . . . . . . . . . . . . . . 40
3.1.1 Heat capacity of the surface . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.1.2 Structure of the PBL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.1.3 The laminar layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.1.4 Diurnal cycle of the PBL . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.2 Single cell deep convection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.2.1 Key convective parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.2.2 Convective triggering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.2.3 Updraught structure and entrainment . . . . . . . . . . . . . . . . . . . . . 55
3.2.4 Downdraughts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.3 Organised deep convection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.4 Summary of convection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4 Cloud Physics 65
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.2 Cloud drop formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.2.1 The energy barrier and Kelvin’s equation . . . . . . . . . . . . . . . . . . . 69
4.3 Diffusional growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4.4 Terminal velocity of particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
4.5 Collision and coalescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
4.6 Ice crystal nucleation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3
4 CONTENTS
4.7 Ice saturation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

4.8 Ice nucleation mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
4.9 Homogenous nucleation from the liquid phase . . . . . . . . . . . . . . . . . . . . . 84
4.10 Ice crystal growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.11 Competition between ice nucleation mechanisms . . . . . . . . . . . . . . . . . . . 91
4.12 Aggregation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.13 Riming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.14 Ice particle fall-speeds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.15 Ice multiplication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5 Radiation 99
5.1 Definitions of the radiative field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
5.2 Energy balance models of the atmosphere . . . . . . . . . . . . . . . . . . . . . . . 106
5.3 Sun and Earth Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
5.4 Radiation interactions with a slab . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.4.1 Direct Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.4.2 Emission from Slab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
5.4.3 Scattering from other directions . . . . . . . . . . . . . . . . . . . . . . . . . 112
5.5 Absorption by atmospheric gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
5.6 Scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
5.7 Radiation budget of clouds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
5.8 Summary of Earth’s radiation budget . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.9 Climate change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
6 Exercises 129
6.1 Thermodynamics Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
6.2 Exercises: Deep Convection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
6.3 Radiation Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
6.4 Cloud Physics Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
Chapter 1
Dry Thermodynamics
1.1 Kinetic theory of heat

Kinetic theory of heat
Let us consider a system at temperature T , consisting of N molecules. These move randomly in
all directions in Brownian motion. This randomness implies the internal energies of the molecules
are not equal, and they change in time. However, the mean internal energy of the molecules (i.e.
of the system) is constant in time, and is proportional to the temperature of the system T .
A formal definition of a mole is the amount of a substance that has the same number of particles
as there are in 12 grams of 12 C. This number N = 6.022 × 1023 is known as Avoggadro’s number.
Avoggadro’s hypothesis
In 1811, Avoggadro proposed that a volume of gas is directly proportional to the number of
molecules of gas. Thus Avoggadro’s hypothesis states that a mole of any gas at constant temper-
ature and pressure occupies the same volume
1.2 Equation of state: The ideal gas law

Ideal Gas
The ideal gas model is a model of matter in which the molecules are treated as non-interacting
point particles which are engaged in a random motion that obeys conservation of energy.
The model tends to fail at lower temperatures or higher pressures, when the molecules come
close enough that they start interacting with each other, and not just with their surroundings.
This is usually associated with a phase transition.
The equation of state for an ideal gas relates its pressure p, volume V and temperature T by
pV = N kT = νR∗ T, (1.1)
where
• N is the number of identical molecules
• ν is the number of moles of gas
• k = 1.3806 × 10−23 J K −1
is Boltzmann’s constant,
• R∗ ≡ Na k = 8.341 J mol−1 K−1 is the universal gas constant,
−1
• with Na = 6.022 × 1023 mol Avogadro’s number.
As pointed out in the introduction, the Earth’s atmosphere is a mixture of gases, mostly
nitrogen, oxygen and argon, trace gases like carbon dioxide, ozone and methane, and variable
amounts of water in its three physical phases (see Table 1).
Neglecting water vapour for the moment, the dry atmosphere can be taken to a very good
approximation as an ideal gas.
5
6 CHAPTER 1. DRY THERMODYNAMICS
In general, it is useful to formulate the thermodynamic relations in terms of intensive variables.

An intensive variable is one whose value does not depend on the amount of matter in the system,
like the temperature or pressure. In contrast, an extensive variable, depends on the size of the
system. For instance, the internal energy of a gas is an extensive variable since if we double its
size, all else being kept equal, its internal energy will double.
Given a system whose volume V contains an amount of mass M ,
V
v= (1.2)
M
denotes the specific volume of the system. In principle, every extensive variable can be converted
to its corresponding intensive form by normalizing it by the amount of matter it describes.
We will (generally, but not always!) use lower case letters to denote specific intensive quantities,
as opposed to their non-intensive counterpart for which we will use capital letters, e.g. the specific
volume v and the volume V .
We would like to write the gas law in terms of kilograms rather than moles and use the molecular
weight of a species i, mi (g/mol), which is the mass of 1 mole of identical molecules in grams.
If the total mass of the gas is Mi , we can express the gas law as,
Mi ∗
pV = R T. (1.3)
mi
The molecular weight can be substituted out by using the specific gas constant Ri for a species
i, which is defined as
Ri ≡ R∗ /mi . (1.4)
Partial Pressure
For a mixture of ideal gases the partial pressure pi of the ith gas is defined as the pressure pi that
it would have if the same mass (Mi ) existed alone at the same temperature T and occupying the
same volume V .
By (3) and (4) the partial pressure for an ideal gas can thus be expressed as
T
pi = Mi Ri . (1.5)
V
Dalton’s Law
According to Dalton’s lawP of partial pressures, the total pressure p of a mixture of (ideal) gases
is the sum of the pressures pi of each species i as if it alone occupied a volume V ,
T X
p= (Mi Ri ) = ρRm T. (1.6)
V i
P P
If Mtot = i Mi is the total mass, then Rm = 1/Mtot i Mi Ri the specific gas constant for
the mixture, and ρ = Mtot /V is the density. Equation. (6) is the form of the gas law generally
used in meteorology, although sometimes the specific volume is used instead of density (v = 1/ρ),
thus
pv = Rm T. (1.7)
Table 1 shows the value of the gas constant for dry air, Rd = 287.05 J kg−1 K−1 .
Reminders
1. What is the definition of a Ideal Gas?
2. Can dry air be considered an ideal gas?
3. What is Dalton’s law?

1.3. THE FIRST LAW OF THERMODYNAMICS 7
Gas Molecular Molar Mass Specific (Mi /Mtot )Ri

weight fraction fraction gas constant
mi [g mol−1 ] Mi /Mtot Ri [J kg−1 K−1 ] [J kg−1 K−1 ]
NO2 28.013 0.7809 0.7552 296.80 224.15
O2 31.999 0.2095 0.2315 259.83 60.15
Ar 39.948 0.0093 0.0128 208.13 2.66
CO2 44.010 0.0003 0.0005 188.92 0.09
total 1.0000 1.0000 287.05
Table 1.1: Main components of dry atmospheric air (source Smithsonian Meteorological Tables).
1.3 The first law of thermodynamics

Heat and Temperature
Heat and temperature are not the same thing! The temperature of a body is a measure of the
average thermal energy of the molecules in that body and does not depend on its mass. Heat is a
form of energy that may be transferred from one body to another. The amount of heat required
to raise 2kg of water by 1K is twice the amount required to raise 1kg.
first law of thermodynamics

The first law of thermodynamics is a statement of two empirical facts:
1. Heat is a form of energy that can be transferred
2. Energy is conserved
Thus, the internal energy of a closed system (U ) can change only if heat (Q) is added or if work
(W ) is done on1 the system by its surroundings:
dU = dQ + dW (1.8)
and for a unit mass of gas:

du = dq + dw (1.9)
The rate of work (visualize a piston, as in Fig. 10) is given by
dw = −pdv, (1.10)
such that (8) can be expressed as

du = dq − pdv. (1.11)
Figure 1.1: When heat is added to a gas, there will be an expansion of the gas (work is done) and
an increase in its temperature (internal energy changes) - Atmospheric Sciences: An Intro Survey.
1 Attention! Sometimes defined as work done by the gas thus changing the sign convention for dW
1.4 Rules for differentiating

If the equation of state is governed by three state variables (p, v, T ) then we can write
p = f1 (v, T ) , v = f2 (p, T ) or T = f3 (p, v). (1.12)
In other words, there are only two independent state variables,
∂f1 ∂f1 ∂f2 ∂f2 ∂f3 ∂f3

dp = dv + dT , dv = dp + dT , dT = dp + dv. (1.13)
∂v ∂T ∂p ∂T ∂p ∂v
If we differentiate state variables other than p, v, or T , for instance, the specific internal energy u,
we must specify which set of thermodynamics parameters we use. Otherwise (∂u/∂p) is ambiguous
and depends on the choice of thermodynamic coordinates, for example whether u is defined as a
function of p and v or as a function of p and T .
Notation for differentiating
To express the rate of change of u as a result of an isothermal process, i.e. a change in the pressure
p, we write
∂u(p, T ) ∂u
≡ (1.14)
∂p ∂p T
meaning that p and T are chosen as thermodynamical coordinates and that the the temperature
is held constant for this process.
1.5 Enthalpy and specific heat

A measure of the quantity of heat needed to raise the temperature of a substance by 1 ◦ C is called
the heat capacity C, and is defined as:
C = dQ/dT. (1.15)
But this definition is incomplete. There are many ways to add heat to a system. One could
add heat to a system at constant volume or at constant pressure, or one could add heat as both
the volume and pressure change.
We now define the specific heat capacity at constant pressure, cp and the specific heat capacity
at constant volume, cv . These are very different.

∂q
cv = . (1.16)
∂T v

∂q
cp = . (1.17)
∂T p
For dry air, cp = 1005 J kg−1 K−1 and cv = 718 J kg−1 K−1 . Q: Why is cp greater than cv ?
If one selects the specific volume v and the temperature T as thermodynamic coordinates, then
the specific internal energy u can be expressed as,

∂u ∂u ∂u
du = dv + dT = dT. (1.18)
∂v T ∂T v ∂T v
where the last equality follows from the fact that the internal energy of an ideal gas does not
depend on its volume,
∂u
= 0. (1.19)
∂v T
(19) implicitly assumes that the intermolecular forces are negligibly small, and therefore is appli-
cable only to an ideal gas. Equating (11) and (18) gives

∂u
dq = dT + pdv, (1.20)
∂T v
1.6. HYDROSTATIC BALANCE 9
where dq indicates
the differential amount of heat added. For an isometric process dv = 0, which
∂u ∂q

implies ∂T v
= ∂T from 11, which equates to the definition of cv giving:
v
dq = cv dT + pdv, (1.21)
Eqn. 21 is still not in a form useful in meteorology due to the difficulty of measuring volume of
an air mass. We therefore use the chain-law on the ideal gas law to write check!
dq = cv dT + RdT − vdp, (1.22)
dq = (cv + R)dT − vdp, (1.23)

For an isobaric process (i.e. no pressure change, dp = 0), which placed in Eqn. 23 gives

∂q
cp = = cv + R. (1.24)
∂T p
which leads to the form of the first law commonly used in meteorology since it is formed in
measurable quantities:
dq = cp dT − vdp, (1.25)
We now introduce another state variable that is often used in atmospheric thermodynamics is the
enthalpy h
h = u + pv. (1.26)
Given this definition, the first law can be expressed as
dq = du + pdv = dh − d(pv) + pdv = dh − vdp. (1.27)
The enthalpy gives a measure of total potential energy of the atmosphere.
1.6 Hydrostatic balance

The atmospheric pressure is the force per unit area exerted on a surface by the weight of the air
above. Thus it is clear that pressure will fall with height as shown in Fig. 11. We can derive how
the pressure falls with height if we assume a state of equilibrium known as the hydrostatic balance.
Figure 1.2: Pressure as a function of height force balances the upward acceleration due to the
pressure gradient.
Hydrostatic balance
Hydrostatic equilibrium exists if the force due to gravity is balanced by the vertical pressure
gradient force. From Fig. 12 we see that
(p(z) − p(z + dz))A = ρAdzg (1.28)
giving
dp
= −ρg (1.29)
dz
Figure 1.3: Schematic of a cylindric parcel of air of horizontal area A and height dz that is
hydrostatic balance, i.e. the downward gravity force balances the upward acceleration due to the
pressure gradient.
Hydrostatic equilibrium is widely satisfied over horizontal scales exceeding 10km, and thus the
assumption is used to simplify the governing equations of numerical models of the atmosphere that
use a grid-cell size exceeding this threshold.
Figure 1.4: Graph showing how pressure decreases with height. Dash lines indicate percentage of
atmosphere below the each level (source: Met Today)
Using the ideal gas law, we can write the hydrostatic relation as
dp p
=− g, (1.30)
dz RT
which can be integrated to give

−gz
p = ps exp( ). (1.31)
RT
How do R and T affect the scale height?
Thus we see that the atmospheric pressure decreases exponentially with height with a scale
height factor of RT /g (Fig. 13. Thus the rate at which pressure decreases with height depends
on the gas constant, with heavier molecules leading to a smaller gas constant and a smaller scale
height. On the other hand, higher temperatures result in the more energetic molecules and thus
the scale height increases (Fig. 14.
1.7. ADIABATIC PROCESSES 11
Figure 1.5: Graph showing how temperature affect the pressure scale height
1.7 Adiabatic Processes

Air parcel theory
We introduce the notion of a conceptual ’air parcel’ at this point, which is a finite volume of
air with uniform properties that may move through the atmosphere as a continuous entity. We
may (sometimes) assume that some properties such as the parcel volume and pressure adjust
quickly to stay in equilibrium with the parcel surroundings, but in general this is not the case (e.g.
temperature).
For example, Fig. 15 shows a schematic of a balloon being released. We can imagine our air
parcel to be represented by the balloon. If we assume the pressure of the balloon is always in
equilibrium, then the balloon expands as it ascends.
What will happen to the temperature of the gas inside the balloon?
Figure 1.6: Graph showing how a balloon ’air parcel’ expands during an ascent. A balloon that is
1m in diameter released at sea level will have a 6.7m diameter at 40km.
Before we answer this question, let us briefly consider which processes may alter the temperature
of a gas in a particular location. Some of these are so-called diabatic processes, that involve the
direct transfer of heat energy to/from an air parcel.
Conduction
The process of heat transfer from molecule to molecule, requiring contact. Heat can be transferred
between (from/to) the ground and lowest layer of the atmosphere by conduction. Air is a poor
conductor and conduction is not an efficient mechanism to transfer heat on large scales.
Convection (and advection)

The process of heat transfer by mass motions of a fluid (such as air or water). Heat can be
transferred away from a heat source by the motion. Advection is a similar process but more
commonly refers to heat transfer by horizontal motions. Convection/advection is an efficient
mechanism to transfer heat in the atmosphere
Radiation
Heat transfer by the emission of electromagnetic waves which carry energy away from the emitting
object
Latent heating
Latent heat is the heat released or absorbed when water changes phase, this will be discussed in
more detail later.
A parcel of air can be subject to input of heat and changes in pressure. First we shall consider
the case where no heat is input.
Adiabatic Process
An adiabatic process is one where there is no heat exchange, dq = 0, thus
cp dT = vdp, (1.32)
Thus if the pressure changes, a corresponding temperature change will result.

Adiabatic processes have special significance as many atmospheric motions can be approximated
as adiabatic. Assuming that the atmosphere is in hydrostatic balance (eqn. 27), then (30) can be
written
cp dT = −vρgdz = −gdz, (1.33)
This gives the adiabatic lapse rate as
dT g
=− (1.34)
dz cp
Thus if we lift a dry air parcel without heat exchange its temperature will fall by by 9.8 K km−1
as it expands to keep its pressure in equilibrium with its surroundings. (Fig. 16).
Figure 1.7: Graph showing how an air parcel expands and also cools during its ascent.
Q: Considering the atmospheric diabatic processes, what will determine how well a moving
parcel’s temperature changes is described by the adiabatic lapse rate?
Figure 1.8: Vertical profile of the temperature T for a clear convective boundary layer as observed
by a tethered balloon at Cabauw, Netherlands, around 10:00 h (local time), 23 Aug 2001.
Figure 17 shows the measured lapse rate on one particular day measured in Holland. The lapse
rate is seen to be close to dry adiabatic .
1.8. POTENTIAL TEMPERATURE 13
Conserved quantities
When we describe the atmosphere it is useful to be able to lable air parcels in terms of properties
that are invarient under certain types of motion. In other words we wish to define Lagrangian
conserved quantities.
Q: is temperature a conserved quantity? Now we will try to derive a quantity that is conserved
in adiabatic motion. Using the equation of state we get
RT
cp dT = vdp = dp (1.35)
p
dT R
= dp (1.36)
T cp p
Poisson’s Equation
Eqn. 34 can be integrated to give Poisson’s Equation:
Rc d
T p p
= (1.37)
T0 p0
where T0 and p0 are constants of integration.
Exercise: Derive this equation.
Note that we have reintroduced the subscript d on the gas constant to emphasize that this
considers dry air.
1.8 Potential Temperature

Potential Temperature
If we set the reference pressure p0 to 1000 hPa then T0 is defined as the potential temperature (for
dry air) which is commonly denoted θ:
Rc d
p0 p
θ=T . (1.38)
p
The potential temperature θ can be interpreted as the temperature a parcel would have if it were
displaced adiabatically to a reference height where the pressure is p0 , which is usually taken p0 =
1000 hPa. In other words, a parcel with temperature T at pressure level p will have a potential
temperature θ, which value is equal to the temperature T0 at the pressure level p0 .
The lapse rate of the potential temperature can be obtained by differentiating (36) with respect
to height (exercise)
dθ θ dT Rd T dp
= − (1.39)
dz T dz pcp dz
Assuming that the atmosphere is in a hydrostatic balance,
dp
= −ρg (1.40)
dz
then with aid of the gas law (37) can be written as

dθ θ dT g
= + . (1.41)
dz T dz cp
Since − cgp = Γd is the dry adiabatic lapse rate, we conclude that if the temperature profile follows
the dry adiabatic lapse rate, the potential temperature is constant with height.
Dry static energy
We can define an analogue to potential temperature in height coordinates. Integrating eqn. 31
(which we recall assumes hydrostatic balance) we get
cp (T − T0 ) = −g(z − z0 ). (1.42)
Analogously to the potential temperature, we see that the quantity
s = cp T + gz (1.43)
is conserved in dry adiabatic motion , and is referred to as the dry static energy.
1.9 Entropy
Another variable of relevance is the specific entropy φ which is measure of a system’s energy per
unit temperature available for doing useful work. It is defined as
dq
dφ = . (1.44)
T
Unlike heat and work, the entropy is a state variable, thus we can speak of the entropy in state
A or B. Note that, with the temperature in the denominator of (43), heat and entropy are not
linearly related. We will consider diabatic and adiabatic processes occuring to a parcel in a
thermodynamic cycle. A thermodynamic cyclic process indicates a series processes transferring
heat and work, while varying pressure, temperature, and other state variables, but that eventually
return a system to its initial state. In theH process of going through this cycle, the system may
perform work on its surroundings. We use to indicate an integral over such a cycle. A reversible
process is defined as I
dφ = 0, (1.45)
This integral is independent of the path. An isentropic process (iso = ”equal” (Greek); entropy =
”disorder”) is one during which the entropy of the system remains constant; a reversible, adiabatic
process is isentropic .
The second law of thermodynamics states that a process will tend to increase the entropy of a
system, thus T dφ = du + pdv ≥ 0. The relevance of entropy to meteorologists is clear deriving the
relationship between the potential temperature and entropy.
If we take logs of eqn. 36 we have
 
Rc d
p0 p
 = lnT + Rd (lnp0 − lnp).
lnθ = ln T (1.46)
p cp
Differentiating we obtain,
Rd dT Rd dp
dlnθ = dlnT − dlnp = − , (1.47)
cp T cp p
and using pv = Rd T we get
cp dT − vdp dq
cp dlnθ = = = dφ (1.48)
T T
1.10 Thermodynamic charts

The heat added in a cyclic process is2
I I I
dq = T dφ = cp T d(lnθ), (1.49)
where the last equality follows from using (46). Thus it is clear that a chart that has perpendicular
coordinates of temperature T and entropy φ, or equivalently T versus lnθ will have the properties
of equal areas=equal energy ; useful properties of a thermodynamic diagram.
Tephigrams
One thermodynamic chart is a tephigram . Using (36) one can plot p as a function of the two
ordinates. Note that the lines of constant pressure, (isobars), are not exactly straight lines.
We are used to using orthogonal charts (left, Fig. 19), but no reason for orthogonality (right,
Fig. 20).
Meteorologists are used to using pressure as a vertical height coordinate and thus rotate the
chart to make the curved pressure lines almost horizontal, as shown in Figure 21.
Look at Fig. 23:
1.10. THERMODYNAMIC CHARTS 15
Figure 1.9: The tephigram basic structure
Figure 1.10: An orthogonal chart
Figure 1.11: A non-orthogonal chart
• Is the pressure of the green square greater or lower than the yellow one?
• Is it warmer or cooler?
Points that lie to the right (left) on a isobar are warmer (cooler).
2 This can also be derived directly from the first law and the hydrostatic relation.
Figure 1.12: A tephigram as they are usually used, with the chart rotated to make the pressure
lines horizontal. Note that the lines of θ are not equally spaced - Q: Why?.
Figure 1.13: Reminder tephigram Questions
1.11 Buoyancy force

1.11.1 Concept
A fixed body of fluid has a volume V = δxδyδz and density ρ (Fig. 24). It displaces an equal
volume of ambient fluid having density ρenv . In the surrounging fluid we assume hydrostatic
balance (Eqn. 38) and as density is everywhere constant p = −ρgz.
The total force F on the box is thus the difference in pressure between the upper and lower
faces scaled by the area of the face, and the gravitational force ρgV :
F = ρenv g∆z∆x∆y − ρgV = gV (ρenv − ρ) (1.50)

In other words the buoyancy force is simply the difference between the weight of the body and
1.11. BUOYANCY FORCE 17
Figure 1.14: Reminder tephigram Questions
Figure 1.15: Schematic: parcel of fluid/air “tied” to the bottom in a fluid/gas of density ρenv .
the fluid it displaces.

If the body were released the buoyancy acceleration B would be

F ρenv − ρ
B= =g (1.51)
M ρ
1.11.2 Buoyancy of an air parcel

In the atmosphere3 we are concerned with motions when gravity acts on variations of density in
a fluid and in general the resulting motions will themselves alter the density anomalies through
mixing and advection, which complicates matters and will be discussed later. For the moment, we
consider the concept of buoyancy on a parcel of air and make the assumption that local density
and pressure variations in the parcel are small compared to their mean values in the ambient
environment
The buoyancy is derived from the vertical momentum equation for an ideal inviscid fluid
dw 1 ∂p
=− −g (1.52)
dt ρ ∂z
We now divide fields into the mean ambient values and the local perturbation values,
p = p + p0 (1.53)
3 this discussion is from Emanuel (1994)
ρ = ρ + ρ0 (1.54)
and again assuming the ambient fluid is in hydrostatic balance:
1 ∂p
− −g =0 (1.55)
ρ ∂z
Equation 50 can be rewritten

dw 1 ∂
=− (p + p0 ) − g (1.56)
dt (ρ + ρ0 ) ∂z
The inverse density term can be expanded in an geometric series:
! " 0 2 #
1 1 1 1 ρ0 ρ
=− 0 =− 1− + + ... (1.57)
ρ + ρ0 ρ 1 + ρρ ρ ρ ρ
Now since we assumed perturbation quantities are small, we can drop all terms of second order
or higher, which gives (check)
1 ∂p0 1 ∂p ρ0

dw 1 ∂p
=− −g− + (1.58)
dt ρ ∂z ρ ∂z ρ ∂z ρ
The first two terms on the RHS cancel due to the assumption of hydrostatic balance, which
also allows us to substitute g in the final term on right, giving:
1 ∂p0
0
dw ρ
=− −g . (1.59)
dt ρ ∂z ρ
The first term in eqn. 57 is referred to as the nonhydrostatic pressure gradient acceleration which
arises from dynamical effects of forced momentum changes 4 The second term on the RHS of
eqn. 56 represents the action of gravity on density anomalies and is referred to as the buoyancy
acceleration 0
ρ
B = −g (1.60)
ρ
Buoyancy is thus related to density anomalies, which is turn can be a result of temperature
anomalies, pressure anomalies, or by the presence of dissolved solids in a fluid or suspensions of
small particles.
However the contribution of pressure perturbations may usually be neglected for flows which
are substantially subsonic, thus giving the approximate definition
0
T
B≈g (1.61)
T
1.12 Introduction to convection

Convection
Convection is the name given to any motions that result from the action of the gravitional field
on variations in density
Convection in the atmosphere can take a variety of forms, and often involves the complexity of
phase changes of water vapour.
Neglecting these for the moment, we see from Eqn. 59 that convection can arise when temper-
ature perturbations result in density anomalies and conseqently a buoyancy acceleration.
Many earlier studies of convection concerned motions arises from a point heat source. An
example is the smoke rises from a lit cigarette (Fig. 25). In general, convection in geophysical
flows almost always arises from bouyancy sources distributed over areas that are large relative to
the depth of the convecting layer.
4 When we discuss convection later, recall that in most atmospheric models this term is neglected.
1.13. ATMOSPHERIC STABILITY 19
Figure 1.16: Cigarette smoke acts as a conventient marker for the convective motions arises from
the point heat source
Figure 1.17: Examples of bénard cells (http://www.catea.gatech.edu)
A classical example of this kind of convection was studied at the turn of the last century and
is referred to as Rayleigh Bénard convection.
Rayleigh Bénard convection
If a fluid is placed between two plates of equal temperature, the fluid will remain at rest. If
the temperature of the lower plate is increased with respect to the upper plate, heat transfer by
molecular diffusion will occur. Increasing the temperature of the lower plate further, eventually
a critical temperature difference is reached at which the unstable distribution of mass (density
perturbations) is such that convective overturning commences. The convection takes the form of
cells (Fig. 26
Convection cells will thus occur if a fluid is heated from below. An obvious example of this
in the atmosphere is when the ground is heated by solar radiation, which subsequently is warmer
than the overlying atmosphere.
The surface will heat the lowest layer of the atmosphere in contact with the ground, and this
will induce a similar unstable vertical profile of mass, with convection ensuing (Fig. 27) .
These convective cells are usually restricted to the lowest few hundred metres of the atmosphere
for reason we will see later, and thus this layer, subjected to the strong vertical mixing of convection
is referred to commonly as the boundary layer . Above the boundary layer the effect of earth surface
on winds is minimal.
1.13 Atmospheric Stability

We now introduce the concept of stability in the atmosphere. What do we mean by stability? We
refer to an unstable situation as one where a small initial perturbation will grow rapidly in time
28.
If a parcel of air is warmer(cooler) than its environment it will be subject to a net upwards
(downwards) buoyancy force and will accelerate upwards (downwards), and is usually referred to
Figure 1.18: The development of a thermal. A thermal is a rising bubble of air that carries heat
energy upward by convection (source Meteorology today)
Figure 1.19: Cartoon illustrating stable and unstable situations (from Meteorology today)
as positively (negatively) buoyant. Equal temperatures are referred to as neutral buoyancy.

A parcel of air that is neutrally buoyant and subjected to a small vertical displacement may
become positively or negatively buoyant, depending on whether the parcel’s temperature changes
changes more or less rapidly than the environmental lapse rate.
The environmental θ lapse rate is dθdz
env
and we will assume the lifted parcel cool adiabatically
(reminder: what does this mean?) and thus θ of the parcel is invariant.
We can thus define three situations:
dθenv
dz < 0 : Unstable
dθenv
dz = 0 : Neutral
dθenv
dz > 0 : Stable
With the example stable and unstable profiles shown in Fig. 29 This is illustrated schematically
in Fig. 30
Without instrumentation to measure the profile of temperature one can not say whether a boundary
layer is well mixed, however one clue is the presence of a gusty intermittent wind associated with
the turbulent eddies, while if the layer is deep enough such that the top of the boundary layer
1.13. ATMOSPHERIC STABILITY 21
Figure 1.20: Two idealized profiles showing (dotted/dashed) stable/unstable environmental tem-
perature profiles
becomes saturated and forms “fair weather cumulus” cloud (see next section) then this is a sure
sign that the boundary layer is well mixed and close to neutral stability.
Thus unstable layers are quite uncommon except in the lowest few metres of the atmosphere,
neutral layers are quite common and result from the convective/turbulent reaction to processes
that produce instability, and stable layers are also common. In this case displaced parcels of air
will undergo simple harmonic motion.
Lenticular or lee-wave cloud indicating atmospheric stable

layer. Courtesy UCAR
Reminders:
• If we express the gas law in terms of mass instead of moles, we switch from using the universal
gas constants to a specific gas constant whose value depends on the molecular weight.
• The dry atmosphere and water vapor can be taken to a good approximation as ideal gases.
• The state of the atmosphere can be expressed by three state variables, p, v and T , of which
two are independent.
• The first law of thermodynamics dictates that energy is conserved.
Figure 1.21: Cartoon illustrating dry stable situations (from Meteorology today)
Chapter 2
Moist Thermodynamics
Moist thermodynamics
In the previous section we restricted our analysis to dry air, we now extend this to include water,
which unlike other atmospheric constituents, can appear in all its three phases: solid, liquid and
vapour (Fig. 31).
Figure 2.1: The structure of the three states of water (from Meteorology Today)
In water1 , each hydrogen nucleus is bound to the central oxygen atom by a pair of electrons
that are shared between them (Fig. 32). In H2O, only two of the six outer-shell electrons of oxygen
are used for this purpose, leaving four electrons which are organized into two non-bonding pairs.
The repulsions of the electrons lead to the assymetry of the molecule giving it a charge dipole.
The four electron pairs surrounding the oxygen tend to arrange themselves as far from each
Figure 2.2: The structure of the water vapour molecule(www.chem1.com)
other as possible in order to minimize repulsions between these clouds of negative charge. This
1 the following discussion is adapted with permission from the from the excellent Chem1 Virtual Textbook by
Stephen Lower. The full link is http://www.chem1.com/acad/webtext/virtualtextbook.html with an abbreviated

link given in figure sources
23
24 CHAPTER 2. MOIST THERMODYNAMICS
would ordinarly result in a tetrahedral geometry in which the angle between electron pairs (and
therefore the H-O-H bond angle) is 109.5o . However, because the two non-bonding pairs remain
closer to the oxygen atom, these exert a stronger repulsion against the two covalent (i.e. bonds
that share electrons) bonding pairs, effectively pushing the two hydrogen atoms closer together.
The result is a distorted tetrahedral arrangement in which the H-O-H angle is 104.5o . The H2O
molecule is electrically neutral, but the positive and negative charges are not distributed uniformly
(Fig. 33). The electronic (negative) charge is concentrated at the oxygen end of the molecule,
Figure 2.3: The charge dipole (www.chem1.com)
owing partly to the nonbonding electrons (solid blue circles), and to oxygen’s high nuclear charge
which exerts stronger attractions on the electrons. This charge displacement constitutes an electric
dipole, represented by the arrow at the bottom. Opposite charges attract, so it is not surprising
that the negative end of one water molecule will tend to orient itself so as to be close to the positive
end of another molecule that happens to be nearby (Fig. 34). The strength of this dipole-dipole
Figure 2.4: The Hydrogen bond between two water vapour molecules. (www.chem1.com)
attraction is less than that of a normal chemical bond, and so it is completely overwhelmed by
ordinary thermal motions in the gas phase.
When the H2O molecules are crowded together in the liquid, these attractive forces exert a
very noticeable effect, which is referred to as hydrogen bonding.
• The hydrogen bond is somewhat longer than the covalent (a chemical bond involving the
sharing of electrons) O-H bond, and is also weaker; about 23 kJ mol−1 compared to the O-H
covalent bond strength of 492 kJ mol−1
• Owing to disruptions of these weak attractions by thermal motions, the lifetime of any single
hydrogen bond is very short
Equation of state for water vapour

Water vapour can be treated as an ideal gas to a good approximation and thus from the gas law
it follows that
e = ρv Rv T, (2.1)
where ρv is the density of water vapor, and Rv is the specific gas constant for water vapour which
is equal to 461.5 J kg−1 K−1 .
2.1. SATURATION 25
In the literature the ratio between the gas constants for dry air and water vapour is used:
Rd mv
= = (2.2)
Rv md
where = 0.622 from the values of the gas constants.
Atmospheric air is a mixture of the ideal gases of dry air and water vapour (Fig. 35). From
Dalton’s law, the total pressure of moist air is the sum of the partial pressures of the dry air and
the vapour
p = pd + e. (2.3)
Figure 2.5: Schematic of molecules in moist air (Meteorology today)
2.1 Saturation
Evaporation
Imagine a closed system consisting of a water body and a vacuum above. The water molecules in
the water body are in a state of thermal agitation and the most energetic ones will overcome the
inter-molecular hygrogen bond attraction and break free of the water body surface. The rate at
which vapor molecules leave the surface depends upon the characteristics of the surface.
The process is known as evaporation.
The vapour vapour pressure increases as a result, which will be denoted e.

Some of the water molecules will subsequently collide with the water surface and stick to the
surface. This process is known as condensation, and decreases the vapour pressure. The rate at
which vapor molecules arrive at a surface of liquid (cloud drop) or solid (ice crystal) depends upon
the vapor pressure. Since the condensation depends on the vapour pressure, it is apparent that
in the absence of external perturbations a state of dynamic equilibrium will eventually be reached
in which the rates of condensation and evaporation equal each other (considering only liquid for
now). The vapour is said to be saturated in this case, and the vapour pressure e is equal to the
so-called saturation vapour pressure es with respect to water.
As the evaporation rate is only dependent on temperature, and the condensation rate on the
vapour pressure, there are two important consequences to note concerning es :
1. es depends on T and T only.
2. Saturation is independent of the pressure of other gases
Supersaturation
If the saturation vapour pressure exceed the saturation value e > es then the air is said to be
supersaturated with respect to liquid water. We shall see that greatly supersaturated states with
respect to liquid water in the atmosphere are not observed, as the excess water vapour quickly
condenses to form liquid cloud droplets. The phrase liquid water is emphasized as the situation is
not the same when it comes to cloud ice crystals as we shall see.
We now derive the dependence of es on T . Heat must be supplied to change a unit mass of
liquid to vapour at a constant temperature; this is known as the Latent Heat of Vaporization,
denoted Lv . Throughout the process of evaporation the es remains constant as it is a function of
T only. For the transition from liquid phase q1 to vapour phase q2 ,
Z q2 Z u2 Z v2
Lv = dq = du + pdv = u2 − u1 + es (v2 − v1 ). (2.4)
q1 u1 v1
However, since temperature is also constant we can write

Z q2
dq
Lv = T = T (φ2 − φ1 ) (2.5)
q1 T
Equating (63) and (64) shows
u1 + es v1 − T φ1 = u2 + es v2 − T φ2 . (2.6)
Gibbs Energy or Gibbs Function

This implies that the so-called Gibbs Function/Energy
G = u + es v − T φ (2.7)
of a system is constant during isothermal, isobaric changes of phase.

The Gibbs function is a measure of the maximum obtainable work from a system at constant
temperature and pressure.The Gibbs function does vary however as a function of temperature and
pressure, and differentiating (66) gives:
dG = du + vdes + es dv − T dφ − φdT. (2.8)
But T dφ = dq = du + es dv giving
dG = vdes − φdT. (2.9)
The Gibbs function is independent of phase, such that dG1 = dG2 :
v1 des − φ1 dT = v2 des − φ2 dT. (2.10)
Clausius Clapeyron Equation

We can now rearrange (69) to derive the Clausius Clapeyron Equation:
des φ2 − φ1 Lv
= = , (2.11)
dT v2 − v1 T (v2 − v1 )
2.1. SATURATION 27
where the last relationship uses (64). Under ordinary conditions the specific volume for water
vapour is much greater than that of liquid, v2 v1 allowing the Clausius Clapeyron equation to
be approximated as
des Lv Lv es
≈ = , (2.12)
dT T v2 Rv T 2
where the final relationship uses the Gas Law once again.
At first sight it would appear to be straightforward to integrate (71) to give es as a function of
T , but this is complicated by the fact that the Latent heat of vaporization Lv is itself a function of
temperature. Fortunately this dependence is weak, with Lv varying by 6% from -30◦ C to +30◦ C.
Neglecting this dependence as a first approximation, integrating (71) gives

Lv 1 1
es = es0 exp − . (2.13)
Rv T0 T
es0 is the vapour pressure at temperature T0 , and at T = 0◦ C we have es0 =6.11 hPa, and a value
of 2.50 × 106 J kg−1 can be assumed for Lv . Substituting these values gives an approximate for
for es as
−B
es (T ) = Ae T (2.14)
where the constants are A = 2.53 × 108 kPa and B = 5.42 × 103 K. A more accurate empirical
form is given by Bolton’s formula for the saturation vapour pressure as a function of T in o C:

17.67T
es (T ) = 611.2exp (2.15)
T + 243.5
es doubles for every 10◦ C increase in temperature. Note that the nonlinearity of the saturation
Figure 2.6: The saturation vapour pressure as a function of temperature
vapour pressure has important consequences for mixing of air parcels as is thus relevant for processes
such as atmospheric convection. Below the freezing point at T = 0◦ C the Clausius Clapeyron
equation describes the saturation pressure of supercooled liquid water (liquid water at temperatures
below 0◦ C), and this is still relevant as we will see later that ice crystals do not form at T = 0◦ C
and supercooled liquid cloud droplets are common. Of course at temperatures below freezing ice
crystals may be present, and the saturation vapour pressure of ice, denoted ei is also described by
(71) but with Lv replaced by Ls , which is the Latent heat of Sublimation, for which a value of
2.83 × 106 J kg−1 can be used. The ratio between the two at subfreezing temperatures is

es (T ) Lf T0
= exp −1 , (2.16)
ei (T ) Rv T0 T
where Lf is the Latent Heat of Fusion and is equal to Ls − Lv . Thus the saturation with respect
to ice is lower than that with respect to (supercooled) liquid water (Fig. 37). Note that while
Figure 2.7: The saturation vapour pressure as a function of temperature with respect to ice and
liquid (source Atmospheric Thermodynamics by Craig F. Bohren and Bruce A. Albrecht)
Figure 2.8: The difference in saturation vapour pressure as a function of temperature with respect
to ice and liquid (source Atmospheric Thermodynamics by Craig F. Bohren and Bruce A. Albrecht)
the ratio of ei and es increases as temperature decreases, the difference maximizes at around -12C
(Fig. 38).
The dependency of the latent heat of vaporization on the temperature
To calculate the dependency of Lv on T we use (63) and note again that v2 v1 and that
es v2 = Rv T . Then differentiating we get
dLv du2 du1

= − + Rv (2.17)
dT dT dT
Now dudT = cvv which is the specific heat capacity of water vapour at constant volume and
2 du1
dT = cvl
is the specific heat capacity of liquid water at constant volume.
We again note the relationship that the specific heat of water vapour at constant pressure
cpv = cvv + Rv , and then integrate to get
Lv (T ) = L0 − (cvl − cpv )(T − T0 ), (2.18)
where L0 = L(T0 ) is the constant of integration. Both cvl and cpv vary as a function of temper-
ature and pressure, but this variation is weak amounting to less than 3% over the full range of
tropospheric conditions, and the constant values of cpv =1870 J kg−1 K−1 and cvl =4187J kg−1
K−1 will be assumed here, while cvv '1410 J kg−1 K−1 . For a complete list of gas constants and
specific heat values refer to Appendix 2 of Emanuel (1994).
2.2. OTHER MEASURES OF WATER VAPOUR 29
2.2 Other measures of water vapour

In addition to the vapour pressure e and the vapour density ρv , there are alternative ways to
describe the water vapour content of air that are more commonly used in meteorology.
Mixing Ratio rv
This is defined as the mass of water vapour per unit mass of dry air
Mv ρv
rv = = . (2.19)
Md ρd
p−e
From the equation of state, ρd = Rd T , so that
Rd
ρv Rd T ρv Rv T Rv
e
rv = = = (2.20)
p−e p−e p−e
The saturation mixing ratio rs , with respect to liquid water, is defined by replacing e with es ,
and is a function of both pressure and temperature.
The more accurate Teten’s empirical formula for the saturation mixing ratio rs as a function
of pressure p and temperature T is:

380 (T − 273.16)
rs (T ) = exp 17.5 (2.21)
p (T − 32.19)
which can be differentiated to give:
drs (T ) 4217
= rs (2.22)
dT (T − 32.19)2
Specific Humidity qv
The is defined as the mass of water vapour per unit mass of moist air
ρv ρv
qv = = , (2.23)
ρ ρd + ρv
and using the same substitution as above (exercise: show)

e
qv = (2.24)
p − (1 − )e
The is defined by replacing e with es . At all normal atmospheric conditions e p, implying

that in practice
e
qv ≈ rv ≈ , (2.25)
p
The difference between rv and qv is greatest at the surface in the tropics and roughly 2% there2 .
Relative Humidity RH
The ratio of the vapour pressure to its saturated value:
e
RH = . (2.26)
es
Note that the relative humidity can be approximated as

r
RH ≈ , (2.27)
rs
however often this is used as an exact relationship in the literature. Some texts such as Rogers
and Yau (1989) break this convention altogether and define RH in terms of mixing ratio and state
2 This similarity leads to the two terms being used almost interchangeably in meteorology, not good practice, and
one that is not helped by the fact that the notation for the specific humidity and mixing ratio are not standard,
with qv often used for mixing ratio in the literature. Note also that the subscript v is not always used.
(85) as an approximate relationship Adding water vapour to dry air changes its density. Q: in
which direction? The molecular weight of water is 18.02 g mol−1 , less than that of dry air at 28.97
g mol−1 . Therefore the density of water vapour at standard temperature and pressure (1000 hPa
and 0o C) is lower than that of dry air Exercise: calculate these densities. So a sample of moist air
will be less dense than the equivalent dry sample.
Virtual Temperature Tv
A hypothetical temperature, the virtual temperature is the temperature a sample of dry air would
have to have in order to have the same density an identical volume of the moist air in question, at
the same pressure.
To derive Tv we recall:
ρv Rv
p = pd + e = ρd Rd T + ρv Rv T = ρd Rd T + ρd Rd T (2.28)
ρd Rd
Recalling the definitions of = Rd /Rv and rv = ρv /ρd , (87) can be simplified to
rv
p = ρd Rd T 1 + (2.29)

ρ
and then using the relation ρd = 1+rv ( Exercise: show )
1 + rv

p = ρRd T (2.30)
1 + rv
Thus the equation of state for moist air can be written as the equation of state for dry air, but
using the adjustment factor in brackets. We thus define the virtual temperature as
1 + rv

Tv ≡ T (2.31)
1 + rv
Allowing us to write the equation of state for moist air as
p = ρRd Tv . (2.32)
Since rv 1, by ignoring second order terms the definition of Tv can be simplified to Exercise:
show
1−
Tv = T (1 + rv ) (2.33)

We can see that for moist air Tv > T always since a unit volume of moist air is less dense than a
unit volume of dry air. Thus for moist air it is Tv that should be used in the definition of buoyancy
to account for this effect. We define the virtual potential temperature in terms of Tv
Rc d
p0 p
θv ≡ Tv . (2.34)
p
and this quantity is conserved in adiabatic motion of moist air. We can apply the equation of state
to moist air by replacing T with Tv , or we can alternatively use the gas constant for moist air Rm :
pv = Rd Tv = Rm T (2.35)
Thus we can see that Rm is related to Rd by

1 + rv

Rm = Rd . (2.36)
1 + rv
Take care not to confuse Rm with Rv .

Water Vapour in Tephigrams
We now return to the tephigram introduced in the previous section and add lines of constant
(isopleths of) saturation mixing ratio to the tephigram, recalling from (79) that rs is a function
of pressure and temperature (Fig.39). Q: This tephigram has an error, such that it is not a true
thermodynamic chart. What is it?
2.3. WATER VARIABLES IN THE LIQUID AND ICE STATE 31
Figure 2.9: A blank tephigram with isopleths of rs highlighted.
Figure 2.10: A correct tephigram. Courtesy EWG at McGill.
As an aside we reproduce a scanned correct tephigram in Fig. 40, as used and produced by
the UK Meteorological Office, unfortunately only for the lower half of the troposphere to 500
hPa. Thus if at a pressure of 950 hPa we measure a temperature of 20◦ C, we can plot this
on the tephigram, marked as a cyan star in Fig. (41. Likewise if we measure a mixing ratio of
r = 10gkg −1 we can also plot this using the mixing ratio isopleths, marked as an orange star.
Making measurements through out the atmosphere (e.g. by a balloon) we can plot a sounding of
temperature and humidity, as show by the yellow and blue stars.)
attempt tephigram exercise I
We will now return to the Cabauw profiles (Fig. 17) and examine the humidity measurements
taken during the balloon ascent, which are shown in Fig. 42. These show the sharp drop in
humidity at the boundary layer top at 400m and that the humidity is roughly constant through
the boundary layer. From the RH profile, we can see relative humidity increasing through the
boundary layer and that the top of the boundary layer is almost saturated.
2.3 Water variables in the liquid and ice state

Analogous to the water vapour variables described above we can define similar variables that
describe the quantity of liquid water or ice in air
Figure 2.11: tephigram with T and r soundings plotted
Figure 2.12: Vertical profile of RH and θ for the Cabauw profile in Fig. 17.
• Absolute liquid (ice) water ice density3 : ρl or L kg m−3 (or ρi )

ρl
• Liquid (ice) water mixing ratio: rl = ρd kg kg−1 (or ri )
ρl
• Specific liquid (ice) water content: ql = ρ kg kg−1 (or qi )
The total water mixing ratio is the sum of the three phases rt = rv +rl +ri . Additionally, analogous
to the Tv we can define the density temperature4 Tρ which is the temperature dry air would have
to have equal density to moist cloud air (exercise: show):
1 + rv

Tρ ≡ T (2.37)
1 + rt
Similarly to Tv , if we assume rl 1 and ri 1 and ignore second order terms we get
1−
Tρ ' T (1 + rv − rl − ri ) (2.38)

The last two terms show how liquid and ice make a parcel more dense, and this effect is referred
to as the water loading effect.
3 Attention! The absolute liquid water density here is the total mass of liquid water droplets divided by a unit
volume of cloudy air - this is not to be confused with the density of liquid water which will denoted ρL . To avoid
confusion, the rest of this course always uses liquid water mixing ratio or specific water content, or we will use the
notation L instead of ρl
4 Again, there is not agreement in the literature concerning the virtual temperature T which is sometimes defined
v
to include liquid and ice. Here we follow the notation of Emanuel (1994)
2.4. SPECIFIC HEAT OF MOIST AIR 33
2.4 Specific heat of moist air
Specific heat of moist air cvm and cpm

The presence of moisture alters the specific heat capacity of air. We add heat to a sample of air
consisting of a unit mass (1kg) of dry air and r kilograms of water vapour
(1 + r)dq = cv dT + rcvv dT (2.39)
this gives
cvv
1+ cv r

dq
cvm = = cv (2.40)
dT 1+r
As cvv /cv = 1.96 ≈ 2 then ignoring second order terms,
cvm ≈ cv (1 + r) (2.41)
Likewise the specific heat of moist air at constant pressure can be approximated
cpm ≈ cp (1 + 0.9r) (2.42)
As r < 10−2 these correction factors can generally be neglected.
Reminders:
• What does it mean for air to be saturated?
• What is es a function of?
• What was one of the problems to get an expression es = F (T )?
• Q: Can you think of some processes by which air can become saturated?
2.5 Ways of reaching saturation
Figure 2.13: We will now start to consider processes relevant to deep moist convection so here is
a superfluous photo to motivate you (source unknown)
Ways of reaching saturation

There are several processes by which a parcel of air may become saturated which are all relevant
to cloud physics:
• Diabatic cooling (e.g. radiation)
• Adiabatic cooling (e.g. ascent)
• Evaporation (e.g. of precipitation falling through parcel)

Figure 2.14: Figure from Rogers and Yau (1989) showing the dew-point and condensation temper-
atures. See text for details
Diabatic Cooling: Dew Point Temperature Td

As air is cooled isobarically, r is conserved, and the air will reach saturation when T is such that
rs (T ) = r. (2.43)
This temperature is known as the Dew Point Temperature Td . One can write equivalently e(T ) =
es (Td ).
Adiabatic Cooling: Condensation Temperature Tc

As air is cooled adiabatically, θv is conserved, and the air will reach saturation at the isentropic
condensation temperature and pressure. This pressure level is sometimes referred to as the lifting
condensation level or LCL.
If ascent and expansion continues condensation will occur (we will see why air does not become
supersaturated with respect to liquid later), thus the temperature will decrease at a slower rate.
Now in this conceptional model of a parcel of air undergoing ascent in a cumulus cloud, we need
to decide what happens to the condensed water: Does it remain in the parcel or will it fall out?
Q: If the droplets remain in the parcel what do we need to consider?
Figure 2.15: Schematic of parcel ascending and reaching saturation, with condensed water falling
out instantaneously in the so-called pseudo-adiabatic assumption.
Pseudo-adiabatic process
2.5. WAYS OF REACHING SATURATION 35
If it is assumed that the condensed water remains in the parcel of air then we would need to account
for its water loading effect and its modification of the heat capacity. Moreover when the freezing
point is reached we need to consider if and how some of the liquid droplets will freeze, invoking
ice processes which are complex, as we shall see. These are issues concerning microphysics and
cloud dynamics, thus involving a cloud model.
For now we take the simplest case and assume that all condensate is immediately lost as
precipitation, known as the pseudo-adiabatic assumption (Fig. 45).
For moist saturated ascent and neglecting the correction factor for the specific heat (so we can
use cp ),
cp dT − vdp + Lv drs ' 0 (2.44)
Thus the saturated moist adiabatic lapse rate Γs is
dT v dp Lv drs Lv drs
Γs = = − = Γd − (2.45)
dz cp dz cp dz cp dz
v dp
(Exercise: show Γd = cp dz ).
As temperature falls with height, drdz is negative implying that Γs > Γd . Recalling that Γd =
s
-9.8 K km , Γs is generally between -3 to -4 K km−1 in the lower troposphere. Q: In the upper

−1
troposphere Γs ∼ Γd , why is this the case?

Isobaric Equivalent temperature Tie
Let us take two parcels of air at the same pressure, but that differ in both their temperatures
and humidity contents; how can we compare them energetically? One measure is the isobaric
equivalent temperature, Tie . Tie is the temperature that a parcel would have if all of the humidity
would condense isobarically.
As this is an isobaric process then dp = 0 and integrating eqn. 103:
Lv
Tie = T + rv (2.46)
cp
neglecting the temperature dependence of Lv as always.
Adiabatic Equivalent temperature Te

As this isobaric process of Tie can not occur in the atmosphere, we can instead define the tephi-
gram related adiabatic equivalent temperature Te , by raising the parcel to the upper troposphere,
condensing the water and then descending to the original pressure.
From (103), we divide by cp T , use the first law in the pressure term to get,
Z Te Z p Z 0
dT Rdp Ldrs
− + = 0, (2.47)
T T p C p p rv C p T
and then assume T is a constant in the humidity term (a fairly reasonable approximation as T
varies by at most 30%) to permit the integration and get

Lv rv
Te = T exp . (2.48)
cp T
Sometimes you will see T set to Tc in the humidity term (why?), giving Tc in the exponential
term on the RHS.
Equivalent potential temperature θe

We found it useful to derive θv which was conserved in dry adiabatic motions. An analogous quan-
tity (approximately) conserved in moist adiabatic motions is the equivalent potential temperature
θe .
We will see that each moist adiabat is uniquely labelled by one value of θe . The graphical
determination of this is given in Fig. 47: after condensation the parcel follows a moist adiabat
until all moisture is condensed, and then descends dry adiabatically to the reference pressure. θe
can be approximated by
Lv rv
θe = θexp (2.49)
cp T
Figure 2.16: A blank tephigram with moist adiabats highlighted.
Figure 2.17: A blank tephigram showing the definition of Te and θe .
Evaporation: Wet Bulb Temperature Tw

In a Stevenson screen there is usually a wet bulb thermometer to measure the wet bulb temperature
Tw , and determine the atmospheric water vapour content. The thermometer bulb is wrapped in
a muslin cloth kept damp by a wick to a liquid reservoir, and the bulb is thus cooled by the
evaporation process (Fig. 48). Tw is the temperature reached if water vapour is evaporated into it
until it becomes saturated, with the latent heat provided by the air. Note that Td ≤ Tw ≤ T , and
that the difference between Tw and Td provides a measure of the saturation of the atmosphere
To calculate the wet-bulb temperature we consider an air parcel consisting of unit mass of
dry air with mixing ratio rv of water vapour. The heat associated with the evaporation of drv is
−Lv drv . Thus
(1 + rv )Cpm dT = −Lv drv (2.50)
Using the definition of Cpm we get
−Lv drv
Cp dT ' (2.51)
(1 + rv )(1 + 0.9r)
2.5. WAYS OF REACHING SATURATION 37
Figure 2.18: Method of measuring the wet-bulb temperature
For most purposes, we can neglect the correction factors and assume
Cp dT ≈ −Lv drv (2.52)
Treating Lv as constant we can integrate to get

Lv
Tw = T − (rs (p, Tw ) − rv ) (2.53)
Cp
Q: what is the problem to solve this equation for Tw ? To get Tw we substitute (73) to give a form
that can be solved by iteration:
Lv −B
Tw = T − (Ae Tw − rv ) (2.54)
Cp
However we can estimate Tw directly from the tephigram as we shall see now. Figure 49 shows the
construction to derive the wet bulb temperature Tw from a tephigram by lifting a parcel adiabatic
to reach saturation and then following a moist adiabat back down to the original pressure. If the
moist adiabat is followed to a reference pressure P0 =1000 hPa, then the resulting temperature is
known as the wet bulb potential temperature θw . Note that there is a one-to-one unique mapping
between θw and thetae . With each referring to a unique moist adiabat.
Exercise: attempt tephigram exercise II
Summary
We have introduced variables that described the thermodynamic state of air and allow us to com-
pare the energetics of two air masses equivalently that may have differing properties of temperature
and humidity. We discussed a number of conserved variables, that act as “markers of air” and are
conserved under adiabatic motion. θ was conserved in dry adiabatic ascent of a dry air parcel, and
for a moist air parcel undergoing unsaturated ascent θv and rv were conserved. For an air-mass
undergoing moist saturated ascent θe or equivalently θw are (approximately) conserved. If the
condensed water falls out of the air parcel instantaneous the process is called pseudo-adiabatic. If
the condensed water remains in the parcel, rt is conserved, and the water loading and condensate
heat capacity must be accounted for. The vertical gradients of θ and θe indicate the likelihood of
convection.
Figure 2.19: Figure from Rogers and Yau (1989) showing the calculation of wet-bulb temperature
from a tephigram. This is known as Normand’s Construction.
Chapter 3
Atmospheric Convection
3.0.1 Overview
In this section we will briefly introduce various types of convection that can occur. The detailed
study of the dynamical equations concerning the each regime type is beyond the scope of this
introductory course, however it is important to be able to place the convection regime types into
context, which are given in Fig. 50.
Figure 3.1: A composite infra-red cloud image where bright white indicates cold cloud tops. Some
example convective cloud regimes/types are highlighted
The chart of vertical cloud types (Fig. 51) illustrates common cloud types as a function of
height. It is seen that many of these are of type “cumulus” which are associated with vertical
atmospheric instability and thus are vertically extended.
Figure 3.2: Chart of cloud types as a function of height (left) schematic chart with height scale
and (right) photos (source unknown)
Photos of tropical convective clouds are given in Fig. 52.

Convection
39
40 CHAPTER 3. ATMOSPHERIC CONVECTION
Figure 3.3: Tropical convective cloud types (source UCAR)
Convection is the name given to any motions that result from the action of the gravitional field
on variations in density
Convection in the atmosphere can take a variety of forms, and often involves the complexity of
phase changes of water vapour.
3.1 Convection in the atmospheric boundary Layer

The lower boundary of the atmosphere is the solid or liquid surface of the Earth.
The planetary boundary layer
The part of the atmosphere that is directly influenced by the presence of this surface and responds
to the presence of this surface on timescales shorter than one hour is generally referred to as the
atmospheric or planetary boundary layer, which we will refer to as the PBL.
SKIP to PBL
3.1.1 Heat capacity of the surface

The sea surface in low wind conditions is relatively simple in its interaction with the atmosphere,
while land surfaces, with complicated topography and varying land surface properties is more
complex. One major difference between the two is the effective heat capacity which governs the
daily (diurnal) and seasonal cycles of temperature.
Heat Capacity of water
Water is very transparent to solar radiation, especially in the visible range of wavelengths. sunlight
is attentuated as it travels downwards from the surface, with a small fraction reaching depths of
tens of metres, depending on the turbidity (cloudiness) of the water.
Turbulent mixing can spread the heating deeper, especially over longer (seasonal timescales),
implying daily and seasonal equivalent depths of the absorbing layer of 10 and 50m. The resulting
heat capacity of this layer, and the difference in net heat loss/gain of the diurnal and seasonal
3.1. CONVECTION IN THE ATMOSPHERIC BOUNDARY LAYER 41
cycles of radiation input, imply a diurnal temperature range of about 0.1K and seasonal range of
approximately 2.5 K.
In the absence of vegetation the surface heat balance is relatively straightforward:
Rn − G = H + LE (3.1)
where Rn is the net radiative flux density, G is the heat flux into the soil, and H and LE are the
sensible and latent heat fluxes, respectively (Noilhan and Planton, 1989).
Bowen Ratio
H
The ratio of the sensible to latent surface heat fluxes is known as the Bowen ratio: LE
The land surface is complex, with a highly varying surface characteristics. The albedo (reflec-
tivity) of the surface, for example, can change dramatically from low values over forest or swamp
to very highly reflective snow or ice covered surfaces. We do not cover the details here, but the
depth of the layer heated by the daily and seasonal cycles is determined by the thermal diffusivity
of the surface material. Even a relatively conductive wet soil is effected by the annual cycle to a
depth of no more than a few metres, the daily cycle affects a much thinner layer, and thus the
daily cycle of surface temperature over land is far greater than that over ocean.
This is seen in Fig. 53. A worm living at a depth of 6m only experiences at most 10% of the
annual temperature cycle of the surface, and usually a lot less! the daily temperature cycle only
affects a depth of tens of centimetres.
Figure 3.4: The reduction in temperature cycle as a function of depth, note how the temperature
wave phase changes as a function of depth.
This difference in thermal capacity is summarized in table 2.
material dry peat wet sand water air units

density 300 2000 1000 1.2 kg m−3
specific heat capacity 1910 1480 4180 1000 J kg−1 K−1
thermal conductivity 0.06 2.2 0.57 0.025 W m−1 K−1
damping depth
diurnal 5.2 14 500A 30000A cm
annual 1.0 2.7 100A 7000A m
heat capacity
diurnal 0.04 0.6 31A 0.63A M J K−1 m−2
annual 0.81 11.4 590A 12A M J K−1 m−2
A
Table 3.1: Thermal properties of surface materials taken from McIlveen (1991). marks turbulent
eddy values.
3.1.2 Structure of the PBL

The boundary layer can be divided into the interfacial layer, the surface layer and the turbulent
layer above (Fig. 54)
3.1.3 The laminar layer

Laminar Layer
Figure 3.5: A sketch of the layers that make up the boundary layer.
Air molecules in contact with a liquid or solid surface are at rest, due to the forces of intermolecular
attraction. At a short distance from the surface, air can move, but the flow is orderly and laminar
due to the viscous friction, hence the name laminar layer.
The depth of the laminar layer is extremely thin, on the order of 1mm and the viscosity implies
gradients of meteorological quantities are very high in this layer. For example over a strongly
heated land surface, temperature can drop by 2K over the laminar layer.
Further from the surface, flow speeds increase and the impact of the surface roughness leads
to non-laminar or turbulent flow. In the lowest tens of metres, the turbulence energy transport
can not remove the instability that results from the strong surface heating, and the temperature
profile is super-adiabatic, in this surface layer.
Further from the surface the impact of the heating reduces, and the turbulence results in a well
mixed layer that is close to neutral.Thus the turbulent boundary has a structure along the lines of
Fig. 55, which also shows the entrainment zone at the boundary layer top where turbulent eddies
encounter the temperature inversion.
Figure 3.6: A sketch of the deeper layers that make up the boundary layer, also showing the
entrainment zone at the boundary layer top.
Their upward momentum can carry them through the inversion where they mix with air above
and within the inversion layer and then sink to their level of neutral buoyancy. This mixing process
is important and weakens the inversion strength, while also bringing more mass into the boundary
layer, causing it to deepen.
3.1.4 Diurnal cycle of the PBL

The diurnal cycle of the PBL is thus depicted in fig. 56. Note how the stable noctural layer leaves
evidence of a residual inversion from the day before.
Note that the boundary layer is highly turbulent during the day as alternatively emphasized in
Fig. 57.
We now examine the Cabauw profiles (Fig. 58). Q: Why do you think the profile is near neutral?
Figure 58 shows observations made in a clear convective boundary layer (CBL). Because of solar
radiative heating turbulent eddies are driven from ground surface. The thermals can penetrate into
the thermal inversion, which for this case is located at about 380 m, above which they are damped
3.1. CONVECTION IN THE ATMOSPHERIC BOUNDARY LAYER 43
Figure 3.7: A smoke plume in the PBL showing the nature of turbulence
Figure 3.8: A sketch of the diurnal evolution of the PBL, also showing the entrainment zone at
the boundary layer top.
Figure 3.9: A second sketch of the diurnal evolution of the PBL, also showing the entrainment
zone at the boundary layer top. Source: Antti Piironen, http://lidar.ssec.wisc.edu/
by the stable stratification. The turbulent eddies warm the upper boundary layer thus bringing an
unstable stratification profile back towards neutrality. The fact that the profile of temperature is
“well-mixed”, with the temperature following approximately the dry-adiabatic lapse rate, implies
that the warming timescale associated with the turbulent eddy mixing is fast compared to the
processes that destabilize the profile (such as radiative cooling of the boundary layer). Were this
not the case, super-adiabatic unstable layers could form, such as in the surface layer when strongly
heated by the sun.
Without instrumentation to measure the profile of temperature one can not say whether a
boundary layer is “well mixed”, however one clue is the presence of a gusty intermittent wind
associated with the turbulent eddies, while if the layer is deep enough such that the top of the
boundary layer becomes saturated and forms “fair weather cumulus” cloud (Fig. 59) then this is
a sure sign that the boundary layer is well mixed and close to neutral stability.
Thus unstable layers are quite uncommon except in the lowest few metres of the atmosphere,
Figure 3.10: Vertical profile of the potential temperature θ for the Cabauw profile (re. Fig. ??.
Figure 3.11: Fair weather cumulus: evidence of a well mixed boundary layer
neutral layers are quite common and result from the convective/turbulent reaction to processes
that produce instability, and stable layers are also common. In this case displaced parcels of air
will undergo simple harmonic motion (can result in Lenticular cloud Fig. 60).
Figure 3.12: Lenticular or lee-wave cloud indicating atmospheric stable layer. Courtesy UCAR
If the turbulent boundary layer becomes deep enough to cause saturation then cloud will form.
We saw an example of this fair-weather cumulus earlier in the course. If the well mixed layer is
capped by a strong stable temperature inversion then the buoyant turbulent updraughts will be
stunted quickly when rising into the base of the stable layer. Thus strong temperature inversions
forming under regions of subsidence (such as associated with anti-cyclone or over the Eastern
Pacific) lead to stratocumulus with a high cloud cover
In the literature we talk of stratus or stratocumulus; the accepted division being based on the
3.2. SINGLE CELL DEEP CONVECTION 45
Figure 3.13: Example stratocumulus
optical thickness (i.e. thickness measured in radiative terms) being greater (less) than 23. However
the terms are used interchangeably in reality.
This schematic in Fig. 61 illustrates the main features of Marine stratocumulus (source Bjorn
Stevens). We remind ourselves that θ and qv are near constant in the well mixed sub-cloud layer.
Part of the entrainment is radiatively driven by cloud top cooling (see radiation component of PA).
The strong temperature inversion acts as a cap to parcel ascent. But what happens if the
inversion reduces or a parcel is particularly energetic?
Figure 3.14: Example stratocumulus (source Bjorn Stevens)
3.2 Single cell deep convection

Deep convection involves parcel ascent from the boundary layer to the tropopause. We introduce
the concepts of stability measures relevant for deep convection.
If the strength of the inversion erodes, some energetic parcels of air may penetrate the inversion,
and then undergo moist saturated ascent to the tropopause, forming thunderstorms, or deep moist
convection. These have three distinct development stages (ca. 1 hour lifetime). (Fig. 62)
• Cumulus growth stage: successive moistening of environment
• Mature stage: droplets suspended by updraughts, precipitation formation, drop evaporation
and rainfall drags forms downdraughts.
• Dissipation stage: intense precipitation, downdraughts dominate updraughts, updraughts
cease (low shear), storm dissipation.
Note: The following more detailed description of thunderstorm development is
taken from Meteorology Today by C. D. Ahrens. Ordinary cell (air mass) thunderstorms or,
simply, ordinary thunderstorms, tend to form in a region where there is limited wind shear - that
is, where the wind speed and wind direction do not abruptly change with increasing height above
the surface. Many ordinary thunderstorms appear to form as parcels of air are lifted from the
Figure 3.15: Schematic of the development and dissipation of a single-cell thunderstorm convective
event (Doswell)
.
surface by turbulent overturning in the presence of wind. Moreover, ordinary storms often form
along shallow zones where surface winds converge. Such zones may be due to any number of
things, such as topographic irregularities, sea-breeze fronts, or the cold outflow of air from inside a
thunderstorm that reaches the ground and spreads horizontally. These converging wind boundaries
are normally zones of contrasting air temperature and humidity and, hence, air density.
Extensive studies indicate that ordinary thunderstorms go through a cycle of development from
birth to maturity to decay. The first stage is known as the cumulus stage, or growth stage. As
a parcel of warm, humid air rises, it cools and condenses into a single cumulus cloud or a cluster
of clouds (Fig. 62a). If you have ever watched a thunderstorm develop, you may have noticed
that at first the cumulus cloud grows upward only a short distance, then it dissipates. The top of
the cloud dissipates because the cloud droplets evaporate as the drier air surrounding the cloud
mixes with it. However, after the water drops evaporate, the air is more moist than before. So, the
rising air is now able to condense at successively higher levels, and the cumulus cloud grows taller,
often appearing as a rising dome or tower. As the cloud builds, the transformation of water vapor
into liquid or solid cloud particles releases large quantities of latent heat, a process that keeps the
rising air inside the cloud warmer (less dense) than the air surrounding it. The cloud continues
to grow in the unstable atmosphere as long as it is constantly fed by rising air from below. In
this manner, a cumulus cloud may show extensive vertical development and grow into a towering
cumulus cloud (cumulus congestus) in just a few minutes.
During the cumulus stage, there normally is insufficient time for precipitation to form, and
the updraughts keep water droplets and ice crystals suspended within the cloud. Also, there is
no lightning or thunder during this stage. As the cloud builds well above the freezing level, the
cloud particles grow larger. They also become heavier. Eventually, the rising air is no longer able
to keep them suspended, and they begin to fall. While this phenomenon is taking place, drier air
from around the cloud is being drawn into it in a process called entrainment. The entrainment
of drier air causes some of the raindrops to evaporate, which chills the air. The air, now colder
and heavier than the air around it, begins to descend as a downdraught. The downdraught may
be enhanced as falling precipitation drags some of the air along with it. The appearance of the
downdraught marks the beginning of the mature stage. The downdraught and updraught within
the mature thunderstorm now constitute the cell. In some storms, there are several cells, each of
which may last for less than 30 minutes. During its mature stage, the thunderstorm is most in-
tense. The top of the cloud, having reached a stable region of the atmosphere (which may be the
stratosphere), begins to take on the familiar anvil shape, as upper-level winds spread the cloud’s
ice crystals horizontally (see Fig. 62b).
The cloud itself may extend upward to an altitude of over 12 km and be several kilometers
in diameter near its base. Updraughts and downdraughts reach their greatest strength in the
middle of the cloud, creating severe turbulence. Lightning and thunder are also present in the
mature stage. Heavy rain (and occasionally small hail) falls from the cloud. And, at the surface,
there is often a downrush of cold air with the onset of precipitation. Where the cold downdraught
reaches the surface, the air spreads out horizontally in all directions. The surface boundary that
separates the advancing cooler air from the surrounding warmer air is called a gust front. Along
the gust front, winds rapidly change both direction and speed. Look at Fig. 62b and notice that
the gust front forces warm, humid air up into the storm, which enhances the cloud’s updraught.
In the region of the downdraught, rainfall may or may not reach the surface, depending on the
relative humidity beneath the storm. In the dry air of the desert Southwest, for example, a mature
thunderstorm may look ominous and contain all of the ingredients of any other storm, except that
the raindrops evaporate before reaching the ground. However, intense downdraughts from the
storm may reach the surface, producing strong, gusty winds and a gust front.
After the storm enters the mature stage, it begins to dissipate in about 15 to 30 minutes. The
dissipating stage occurs when the updraughts weaken as the gust front moves away from the storm
and no longer enhances the updraughts. At this stage, as illustrated in Fig. 62c, downdraughts
tend to dominate throughout much of the cloud. The reason the storm does not normally last very
long is that the downdraughts inside the cloud tend to cut off the storm’s fuel supply by destroying
the humid updraughts. Deprived of the rich supply of warm, humid air, cloud droplets no longer
form. Light precipitation now falls from the cloud, accompanied by only weak downdraughts. As
the storm dies, the lower-level cloud particles evaporate rapidly, sometimes leaving only the cirrus
anvil as the reminder of the once mighty presence. A single ordinary cell thunderstorm may go
through its three stages in one hour or less. End of book quote.
Figure 3.16: Schematic of other processes to consider in deep convection
Processes in deep convection

There are several factors to consider in a model of deep convection (Fig. 63):
• Mixing between convection and its environment
• Ultimate depth of the cloud
• Precipitation and convective moisture transport
• Convective heating rates
• Downdraught formation
• Convective ’triggering’
3.2.1 Key convective parameters

The presence of instability is of great importance in weather forecasting and to understand the
various climatic regimes. Earlier we classified three stability states in a dry environment. For a
moist environment there are five stability categories.
Figure 3.17: Tephigram showing 5 possible stability categories for environmental lapse rate. The
red lines highlight the parcel dry and moist adiabats.
.
• AB: Absolutely unstable (even if unsaturated, parcel is warmer),
• AC: Dry Neutral (parcel is neutral to dry ascent),
• AD: Conditionally unstable (for unsaturated ascent profile is stable, while it is unstable for
saturated ascent),
• AE: Saturated Neutral (for saturated ascent parcel neutral),
• AF: Absolutely stable
Key convective parameters

We now examine the case of conditional instability further to define a range of parameters that
are important for describing the potential for, and characteristics of, convection
. Let us examine the case of the surface parcel in the blown up tephigram in Fig. 65. The parcel
is unsaturated and thus if lifted follows a dry adiabat. The profile is stable to dry adiabatic ascent,
but let us assume that the mechanical lifting and/or the parcel initial momentum is sufficient such
that ascent can continue. The parcel continues until it reaches the condensation temperature and
pressure that we defined earlier. This pressure level is called the lifting condensation level, or
LCL.
If ascent continues, after the LCL the parcel follows a moist adiabat. Since the profile is con-
ditionally unstable, the moist adiabat must eventual cross the environmental temperature profile,
implying that the parcel becomes warmer than the environment. After this point the saturated
parcel accelerates upwards, undergoing free convection in meteorology terminology. Thus the level
at which the parcel becomes positively buoyant is called the level of free convection or LFC.
Convective inhibition, CIN
The area enclosed on the tephigram between the parcel trajectory and the environmental profile
when the parcel is negatively buoyant (i.e. colder than the environmental) is proportional to
the energy that the parcel must be supplied to overcome this inhibition barrier to undergo free
convection. This area is thus called the convective inhibition, more commonly abbreviated to
CIN.
After the LFC, the parcel accelerates upwards following a moist adiabat. Reminders: What
assumptions does this make concerning the parcel? The ascent may continue to the upper tro-
posphere where it recrosses the environmental temperature line. This level is called the level of
Figure 3.18: Tephigram showing the boundary layer profiles of temperature and humidity and the
pseudo-adiabatic ascent of the surface parcel. See text for details of definitions.
.
Figure 3.19: Full tephigram from Fig. 65 additionally showing the full parcel ascent and the
positive energy area of CAPE.
.
neutral buoyancy or LNB . The parcel will overshoot and undergo oscillatory motion, but the LNB
marks the cloud top approximately.
CAPE
The positive area demarked is proportional to the energy that can be potentially gained by the
parcel. We will call the total1 energy available to a parcel starting from level i the convective
available potential energy or CAPE.
If we concern ourselves only with the motions in the vertical then CAPE calculated for a parcel
starting from pressure level p is thus the integral of the buoyancy acceleration
Z LN B
CAP Ep = Bdz, (3.2)
i
and substituting the approximation of buoyancy using the virtual temperature (B = g(Tv,u −
Tv,env )/Tv,env ), we get
Z LN B
Tv,u − Tv,env
CAP Ep = g dz, (3.3)
i Tv,env
where Tv,u and Tv,env are the updraught/environment parcel virtual temperatures. We recall that
equall areas are proportional to equal energies on a thermodynamic chart. If we follow a parcel
1 i.e. including CIN. Often the literature refers to the CAPE of a sounding as only the positive area, as marked
in Fig. 66, but here we follow Emanuel (1994).

through its moist adiabatic ascent and then then trace the environmental temperature profile back
down to the original pressure, the area enclosed by this trajectory, the CAP E, is equivalent to the
energy available to do work on the parcel. If we assume that all of the potential energy represented
by the CAP E is converted to the parcel kinetic energy, we can estimate the peak updraught
velocity as p
wmax = 2CAP Eps , (3.4)
where we are assuming a parcel starting from the surface (ps is the surface pressure). In reality
frictional forces and updraught mixing greatly reduce the actual velocities achieved in convective
clouds.
Q: updraught velocities tend to be higher in convection over land than oceans, why do you think
this is?
To summarize this, we have defined
• LCL: Lifting Condensation Level
• LFC: Level of Free Convection
• CIN: Convective Inhibition
• LNB: Level of neutral buoyancy
• CAPE: Convective available potential energy
Figure 67 shows the anvil cloud spreading out from a thunderstorm over Mali in West Africa.
Notice how flat the anvil top is, marking the LNB. Over the convective updraught region, the
cloud top is undulating and higher, revealing the over-shooting updraughts.
Figure 3.20: View from space of a convective even over Mali (image courtesy of NASA)
3.2.2 Convective triggering

We can see from the tephigram in Fig. 66 that surface parcels are stable to small displacements,
but unstable for large displacements. This is why the profile is termed conditionally unstable
when Γd < Γenv < Γ. The triggering of deep convection is a rather discrete occurrence. If
the atmospheric profile has positive CAPE, then if a boundary layer parcel can overcome CIN,
it will then often proceed to form deep convection, which will reach teh LNB near the radiative
troposphere.
This is apparent when examining the PDF of cloud top heights which lead to this picture of
trimodal convection in Johnson et al. (1999) (Fig. 68). There are a number of mechanisms by
which the CIN can be overcome:
• Mechanical lifting
– Flow over orography
Figure 3.21: Schematic of convection structure in the Tropics (from Johnson et al., 1999)
– Air rising over colder, denser air masses at fronts

– Air rising over denser cold-pools from other convective events.
• Surface convergence due to low pressure perturbation
– land-sea breeze
– elevated heat low
– meso-scale pressure perturbations
• Increasing boundary layer h (moist static energy)
– Surface heating
– or moistening
Triggering: Flow over orography

A schematic of flow-induced orographic triggering is given in 69 from Banta and Barker Schaaf
(1987). They use radar and satellite imagery to locate common triggering locations of convection
over the Rocky Mountains.
Triggering: Elevated heat low

However, triggering over orography can also be due to their creation of elevated heat lows, as the
mountain top warms more than the surrounding air.
An example of this during summer 2015 observed with a precipitation radar is shown in Fig.
70 from the Carnic Alps.
Triggering: Coldpool dynamics
Downdraughts and their associated coldpools are an important component of convective systems
since the spreading coldpools can lift environmental air and trigger new convective events.
This is shown in the schematic of Fig. 71 The strength of the wind shear in the boundary layer
is key to this role, as it determines whether the coldpool-induced vorticity is enhanced or offset.
The impact of coldpools on convective triggering can be clearly seen in the visable satellite
imagery (Fig.72). Note the small cumulus at the leading edges and that some of these are able
to overcome CIN and trigger new events. .
Triggering at fronts
The lifting of air at cold fronts can triggering convection, as shown by the embedded cumulus
towers in the schematic (Fig. 73).
Convective Trigger Temperature

Figure 3.22: Schematic of orographically-forced mechanical triggering mechanisms (from Banta

and Barker Schaaf, 1987)
Figure 3.23: Series of radar images taken between 1310 and 1410 CEST over the Carnic alps
(courtesy of Osmer)
Concerning thermodynamic triggering mechanisms, the convective trigger temperature is the tem-
perature the surface layer would have to be heated to in order to remove all CIN, assuming no
change in boundary layer mixing ratio during this heating process.
Q: is this a good assumption and when is the trigger T relevant? . The convective trig-
ger temperature is found by tracing a humidity isopleth of the surface layer mixing ratio to the
environmental temperature curve and then following a dry adiabat back to the surface pressure
(Fig. 74). For this particular example the trigger temperature is 35◦ C, requiring a warming of
roughly 5◦ C. The convective trigger temperature is more relevant for cases where convection is
not mechanically forced, for example whether convection will occur in undisturbed conditions as
a result of the diurnal heating, particularly over land in the tropics over mid-latitude summers.
If the soil conditions are wet then the surface heating can also lead to a significant latent heat
(humidity) flux, which will lower the trigger temperature. ENTRAINMENT MCS It is thus clear
to see why dry situations over land can become “locked-in” through a positive feedback. If a
blocking high causes persistent dry conditions, the soil will dry out and the trigger temperature
will be higher, making future convective storms less likely. Such a feedback has been shown to be
relevant in Africa (Taylor et al., 1997) and has been also highlighted as playing a role in increasing
Figure 3.24: Schematic from Rotunno et al. (1988) showing how air lifted by spreading coldpools
can lead to new convective cells being formed
.
Figure 3.25: Visible satellite image clearly showing the inhibition of all convection within the wake
itself but highlighting new convection at the wake boundaries. (source isites.harvard.edu)
Figure 3.26: Schematic cross section of convection embedded in a front (source).
the severity of the 2003 summer heatwave over Europe (Fischer et al., 2007).
Q: attempt tephigram exercise III and IV
Triggering: Coldpool thermodynamics

As well as a mechanical role, in low wind-shear conditions coldpools can also play a thermodynam-
Figure 3.27: Tephigram showing the derivation of the convective trigger temperature
ical role in convective triggering (see Tompkins, 2001b).

If Fig. 86 is examined closely it is seen that new events are triggered at the edges of old, virtually
recovered coldpools. This is due to a thermodynamical process. Tompkins (2001b) questioned
why the new triggering event seemed to often occur at the edges of coldpools that had seemed
to have almost recovered, rather than earlier in their lifetimes, when negative temperature and
bouyancy perturbations are higher, the coldpools are deeper and faster moving, and consequently
mechanical lifting would be greater. Examining timeslices through an individual coldpools shows
the thermodynamic and dynamic perturbations associated with the coldpool in low shear and low
mean wind conditions.
Figure 3.28: Cross sections at various times through cold pool B in Fig. 86 for the water vapor
mass mixing ratio (g kg−1 ), potential temperature (K), horizontal velocity (m s−1 ), moist static
energy (KJ kg−1 ), buoyancy (m s−2 ), vertical velocity (m s−1 ), latent heat flux (W m−2 ), sensible
heat flux (W m−2 ), and rain rate (mm h−1 ). The solid, dotted, dashed, and dot-dash lines show
the cross section at 0, 30, 60, and 90 min into the simulation. The solid line thus shows the
“prestorm” state, and the dotted line corresponds to the first panel in Fig. 86
In Fig. 75 the slices are taken in an left-right direction and show the prewake conditions and
the development after 30, 60, and 90 min respectively. The graphs for 30 min (dotted lines) show
clearly the signal of the new downdraft. The vertical velocity, for example, peaks at about -1.4 m
s−1 and surface velocities are enhanced over a distance exceeding 3 km with a maximum of 4 m
s−1 . Rainfall is strong at this time, exceeding 100 mm h−1 . It is interesting to note that although
the potential temperature has been clearly depressed by more than 1 K, the water vapor shows no
signs of any decrease and has, in fact, actually increased in value, as seen in the previous section.
The increased surface wind speed causes an enhancement of both the sensible and latent surface
heat fluxes, with a much larger percentage increase in sensible heat flux, as previously noted in the
observational studies, due to the addition increase in the air-sea surface temperature difference.
Half an hour later (dashed lines) the Î¸ graph shows that the wake has spread to a diameter
of almost 10 km, with the negative perturbation reaching -1.4 K. The horizontal velocity actually
shows the highest velocities near the edges of the cold pool, with values exceeding 5 m s−1 , and
in association with this the vertical velocity shows considerable uplift at the cold pool edges, with
an upward velocity of more than 0.5 m s−1 on the western edge. By this time, the water vapor
now also shows a significant depression, dropping by around 2 g kg−1 . The depression is seen to
be more restricted in horizontal scope than that of Î¸ and toward the edges of the wake the vapor
appears to in fact be slightly greater than the surrounding environment. The dot-dashed lines on
these figures, showing the wake state after 90 min, indicate that the convective downdraft that
initiated the wake has expired, as one would expect. The potential temperature in the center of the
wake has recovered rapidly to the prewake conditions, whereas the water vapor here has fallen even
lower to a very dry 11 g kg−1 . This indicates that the central temperature recovery is not due to
surface fluxes, but is instead due to environmental air from above the boundary layer being drawn
down in the wake of the dying downdraft. The moisture perturbation is sufficient to maintain a
significant negative buoyancy perturbation throughout the central region of the wake. The rainfall
rate associated with this system has fallen to around 5 mm h−1 associated with the anvil cloud,
which, due to the lack of mean background vertical wind shear, is restricted in horizontal extent
to around 6 km in diameter. The surface velocity away from the center of the cold pool is largely
above 3 m s−1 and thus both the sensible and latent surface heat fluxes remain enhanced.
Figure 3.29: Composites of the mean cold pool properties, calculated for the points corresponding
to the last 33% of the wake radius lying inside the gust front. The composite is here calculated
between the time of cold pool detection and the time at which the maximum radius is achieved
(i.e., decaying phase ignored). The 44 wakes that require at least 1 h to achieve their respective
maximum radii are composited.
By compositing many simulated coldpools (Fig. 76), it is seen that new events trigger when
the buoyancy perturbation has recovered at the edges of old coldpools, which is where CAPE is
maximum and CIN is minimum as a result (see Tompkins (2001b) for CAPE and CIN plots) Note:
mechanical lifting can play a role when 2 wakes collide.
3.2.3 Updraught structure and entrainment
Entrainment
In deep convection, updraught air may ascent to close to the tropopause without mixing with the
environment. However, much of the air at the cloud boundaries undergoes turbulent mixing events
with environmental air, a process known as entrainment.
The process of mixing was proven in early laboratory studies using buoyant plumes of dyed
liquids mixing, however these studies were unable to adequate represent the effect of latent heating.
The understanding and representation of mixing in convection is a long-standing problem. The
Figure 3.30: Example of a laboratory experiment of buoyant plumes in a tank experiment of Turner
(1963)
highly turbulent nature of clouds was evidenced in the first aircraft measurement that were taken
after the second world war, such as by (Byers and Braham Jr., 1948; WARNER and Newnham,
1952; Warner, 1955), with example aircraft and instruments of this period shown in (Fig. 78).
Figure 3.31: (Left) Early post-war aircraft fitted out with cloud observation instrumentation
(source www.egoaltar.com), and (right) example of the instrumentation (Telford and Warner,
1962)
Examples of in-cloud measurements from Telford and Warner (1962) (Fig. 79), show the
highly turbulent nature of the cloud and reveal the differences between updraught and downdraught
regions. SKIP MATHS If the figure is closely examined, it is seen that the main updraught is
towards the centre of the cloud region while the main (saturated) downdraught areas are at the
edges of the cloudy region. The value of the liquid cloud water content is fairly similar in much of
the central area of the cloud, indicating that this air ascended to this level from the boundary layer
with little mixing occuring with the environment. The liquid cloud water is said to be close to its
adiabatic value. Close to the cloud edges however, the values are much smaller, indicating turbulent
mixing which results in evaporative cooling. This can lead to negatively buoyant downdraughts
(especially in the presence of precipitation) which are discussed in the following section.
Mixing in Cumulus clouds

Consider two masses of moist air at pressure p with diverse properties (we use subscripts 1 and 2)
of temperature and humidity and are subsequently thoroughly mixed isobarically.
The specific humidity of the mixture is
M1 M2
qv = qv1 + qv2 (3.5)
M1 + M2 M1 + M2
Figure 3.32: Some of the instrumentation and an example transect through a fair weather cumulus
cloud from Telford and Warner (1962).
We can write this relationship approximately using mixing ratios and vapour pressure
M1 M2
rv ≈ rv1 + rv2 (3.6)
M1 + M2 M1 + M2
M1 M2
e≈ e1 + e2 (3.7)
M1 + M2 M1 + M2
We will assume that there is no net loss of gain of heat during mixing so that the change in heat
of the two parcels is equal
M1 (cp + cpv rv1 )(T − T1 ) + M2 (cp + cpv rv2 )(T − T2 ) = 0. (3.8)
If we neglect the contribution of water vapour in the heat capacities then we can see that temper-
ature also approximately is linearly mixed:
M1 M2
T ≈ T1 + T2 (3.9)
M1 + M2 M1 + M2
As the saturation vapour pressure is a nonlinear function of temperature, then the approximate
linear mixing property of vapour pressure and temperature imply that the mixing of two unsatu-
rated parcels of air can result in a mixed parcel that is saturated (draw sketch). This phenomenon
is commonly seen when your breath is visible on a winter’s day. In fact it was from this observation
that James Hutton deduced in 1784 the concave shape of the Clausius Clapeyron curve a century
before it was derived.
The degree of mixing impacts the eventual cloud top height (Fig. 80). Note the strongly
nonlinear vapour mixing ratio. Paluch (1979) showed strong evidence that most air in clouds
had undergone one or two mixing “events”, suggesting an episodic mixing model as a result (Fig.
81).
• es doubles for every 10◦ C increase in temperature.
• All vapour in excess of saturation condenses (ice more complex)
• Formation of precipitation efficient in fast updraughts (ice)
• precipitation efficiencies high (60-80%)
What does this mean?
• Essentially 60 to 80% of the water vapour in the parcels of the boundary layer that undergo
ascent in deep convective clouds, falls back to the surface as precipitation.
• The number is much higher for the parcels that rise undilute to the LNB.
Figure 3.33: Mixing in deep convection
Figure 3.34: Comparison of the total mixing ratio and the wet equivalent potential temperature
computing from data collected inside a growing cumulus, taken from Paluch (1979)
3.2.4 Downdraughts
From the transect through a fair weather cumulus cloud it was seen that clouds consist of both
updraught and also downdraught components.
Coldpools
In deep convective systems these saturated downdraughts form at mid-tropospheric levels and can
penetrate all the way to the surface, spreading out to form a convective Coldpool , sometimes
referred to as a convective wake.
Coldpools are most visible in dusty environments as seen in this example from NOAA of a dust
storm in Iraq (Fig. 85). See also (Seigel and van den Heever, 2012) for high resolution simulations
Figure 3.35: The saturation vapour pressure as a function of temperature
Figure 3.36: Schematic of precipitation formation in deep convection (source COMET goes-r.gov)
Figure 3.37: Schematic of a coldpool (source: meted.ucar.edu)
of dust lofting in coldpool structures, and also further details available by clicking here.
• Downdraughts are formed both by the frictional drag of falling precipitation and are also
buoyancy driven.
• The buoyancy effect derives from the nonlinearity of the saturation vapour pressure curve
and the fact that temperature and mixing ratio are approximately linearly mixing.
• This means that a mixture of updraught cloudy air and its environment air, subsequently
brought to saturation by evaporation of precipitation, can be negatively bouyant with respect
to the surrounding environment (see tephigram exercise sheet).
Figure 3.38: Photo of dust lifted in a coldpool outflow in Iraq (source: www.see.leeds.ac.uk)
Figure 3.39: Four time slices of a coldpool evolution from (taken from Tompkins, 2001b). Click
for movie.
Fig. 86 shows coldpool development in a domain of 90 by 90km, using a horizontal resolution

of 350m. Note the structure of the coldpools with temperature depressions ranging up to 1-2K.
The tephigram exercises show how important the mixing and precipitation evaporation (wet
bulb) processes are to forming convective downdraughts. Downdraughts and coldpools can be
dangerous for aircraft landings (Fig. 87) and thus their short-term prediction is a crucial task of
short-term aviation nowcasting in potentially convective situations. Severe events are sometimes
referred to as microbursts.
3.3 Organised deep convection

Mesoscale convective systems
If convection becomes self-propagating and one can speak of organised mesoscale convective sys-
tems (or MCS) consisting of numerous convective cells. Such systems can have a life time that
greatly exceeds the life-span of an individual convective events.
Examples of organised convective in mesoscale systems are
• Multi-celluar systems
• Squall-lines
3.3. ORGANISED DEEP CONVECTION 61
Figure 3.40: Schematic showing how a downdraught or severe microburst can cause an aircraft
stall on approach (source: electronicdesign.com)
.
• Madden Julian Oscillation
• Convectively-coupled waves
In the presence of wind shear the convective can become linearly organised and is referred
to as a squall line, with examples shown in Fig. 88. In addition to the saturated convective
downdraughts, larger mesoscale downdraught also play a thermodynamical and dynamical role
in squall line evolution. Note that impact convection organisation is rarely represented in the
Figure 3.41: Left: Radar image of a squall line over the United States (source: NOAA). Right:
Photo of a squall line over Mexico taken from the Space Shuttle (source: NASA)
.
representation of convection in numerical weather prediction or climate models. Conclusions An
Figure 3.42: Series of radar images taken between 1310 and 1410 CEST over the Carnic alps
(courtesy of Osmer)
Figure 3.43: Simulations of radiative-convective equilibrium showing evolution from relatively

unorganised state after ten days of simulation (left) to strongly organised state after 50 days
(right).
Figure 3.44: Simulations of radiative-convective equilibrium over a SST gradient. The gradient is
reversed after 5 days, but the dry free troposphere prevents deeep convection from immediately
reestablishing over the warmest SSTs in day 6-10.Adapted from Tompkins (2001c,a)
example of a squall feature during summer 2015 observed with a precipitation radar is shown in
Fig. 89 from the Carnic Alps.
High resolution simulations with convective resolving models have been used to determine the
mechanisms by which convection can spontaneously organise, including:
• Water vapour feedback
• Cold pools
• Cloud-radiation feedback
• interaction with the surface
see for example Held et al. (1993); Tompkins and Craig (1998); Grabowski et al. (2000); Tomp-
kins (2001c,b); Grabowski and Moncrieff (2002, 2004) Often in idealized high resolution models
of convection, the domain organises into a moist precipitating area surrounded by a very dry
atmosphere, such as (Bretherton et al., 2005) (Fig. 90).
For example water vapour perturbations can strongly affect organisation (Fig. 91).
Convection → locally moistens→reduced evaporative cooling on mixing.
Figure. 92 show the normalized water vapour in the “switched SST” experiment of Tompkins
(2001c) and reveals how the dry zone created by the previous subsidence regime persists and
supporesses deep convection. Convection gradually encroaches on the areas over the warmest
SSTs by moistening from the edges. Conclusions
Differences between radiative heating in dry, cloud-free and moist, cloudy zones can also drive
secondly circulations and convergence in convective areas (Gray and Jacobson Jr, 1977)
3.4. SUMMARY OF CONVECTION 63
Figure 3.45: Normalized water vapour, showing how the dry air previously subject to subsidence
subsequently suppresses convection. Tompkins (2001b)
Conclusions
The lifting of convectively unstable layers can lead to the commonly observed cloud types of
altocumulus (mid tropospheric, up to around 6000m) and cirrocumulus (upper tropospheric). The
vertical division between the two cloud types is not purely taxonomic as cirrocumulus are cloud
consisting purely of ice crystals while altocumulus are generally mixed phase clouds (consisting of
both liquid and ice). Altocumulus normally occurs when a large air mass is lifted to middle levels
by a landmass or an incoming frontal system.
Figure 3.46: Examples of clouds formed by convectively unstable layers. (a) Altocumulus (b)
Altocumulus undulus, resulting from vertical wind shear (c) cirrocumulus (source: theairlinepi-
lots.com)
Layer Convective Instability

If a layer of the atmosphere is lifted that is absolutely stable, but which becomes conditionally (or
absolutely) unstable through lifting then the layer is termed convectively unstable
Examining the tephigram it is clear that a criterion for convective instability can be understood
in terms of θe (or equivalently θw ).
Three criteria for convective Instability:
dθe
dz < 0 : Convectively Unstable
dθe
dz = 0 : Convectively Neutral
dθe
dz > 0 : Convectively Stable
This kind of layer instability can lead to altocumulus or cirrocumulus clouds, which can be
signs of impending arrival of a front, and are illustrated in Fig. 93. This layer instability can be
best understood using the tephigram, which show a case where dθ dz < 0. Lifting the layer through
e
a sufficient distance to result in saturation leads to absolute instability within the layer. (fig. 94
3.4 Summary of convection

• Convection is the name given to any motions that result from variations in density.
• Convection in the atmosphere can be considered as three modes:
– turbulence in the boundary layer (dry and moist) capped by an inversion

– deep moist convection penetrating to the tropopause
Figure 3.47: Illustration of layer instability
– mid level convection arising from frontal lifting

• Convection components:
– turbulent updraughts
– entrainment and detrainment
– condensation and formation of precipitation
– downdraughts and coldpools
• While the “parcel” view of convection is very informative, it neglects the complexities of
convective organisation.
Chapter 4
Cloud Physics
4.1 Introduction
Introduction to cloud physics
In all our discussions so far of convection, we have readily assumed that in updraught motions,
once saturation is reached cloud drops readily form. However it is not obvious that this is the case.
Indeed, we shall see that this is not the case for the formation of ice crystals.
In this part of the course we move to the small scale physics that occurs on the microscale of
droplets within clouds, known as cloud microphysics. What are the processes that we need to
consider in clouds therefore?
Cloud processes
We will consider the
• change of phase from water vapour to liquid droplets or ice crystals
• transformation of small cloud droplets to larger rain drops
• freezing of cloud droplets
• formation of ice crystals from water vapour
• advection/falling of the larger sized droplets (precipitation)
• evaporation/sublimation of precipitation and cloud
Cloud particle modes

Condensed droplets of water can obviously have a range of sizes, or droplet radius. However the
probability density function describing the droplet sizes is not a smooth function from the very
small to the very large (large being a relative term!), but instead we shall see that diverse and
discrete cloud processes contrive to produce distinct peaks in the droplets size spectra, termed
modes. We shall follow the convention of considering each mode as a bulk quantity.
For example, Fig. 4.1 shows typical drop size spectra for a range of cloud types (Quante, 2004).
We will consider modes of
• cloud drops
• rain drops
• ice crystals
• snow flakes
For ice processes the division is less obvious since ice crystals can form many different shapes
or habits, with differing radiative and microphysical (e.g. fall-speed) properties despite similar
particle sizes which may be treated as a separate category. It is always possible to divide each
mode into finer size categories of course, for instance considering small and large cloud ice particles
65
66 CHAPTER 4. CLOUD PHYSICS
Figure 4.1: Measured drop sizes from various types of clouds, from Quante (2004).
Figure 4.2: Typical drop sizes, from Mcdonald 1958 ?, reproduced from Rogers and Yau (1989).
Figure 4.3: Schematic of cloud process pathways between various cloud particle modes (source
unknown)
separately. Figure 4.2 shows a schematic of the typical sizes of cloud particles in warm
phase (i.e. no ice processes involved) clouds. The cloud droplet has a typical radius of 10µm, (a
micron=1 µm), 100 times smaller than a typical raindrop. In general we will consider cloud
4.2. CLOUD DROP FORMATION 67
microphysical processes as pathways that can either convert particles from one or more discrete
modes to a different particle mode, or can change the mass or size distribution within one particular
mode. Figure 4.3 reveals a bewildering array of such processes. A full description of clouds
at this level of complexity is beyond the scope of this course, but the diagram reemphasizes the
complexity of the task to represent such small-scale complex processes in global climate models
with 100km size grid-boxes.
4.2 Cloud drop formation

In our earlier lecture we assumed that supersaturated states could not exist and water vapour in
excess of the saturation mixing ratio was immediately condensed into cloud droplets. It is not
obvious that this should be the case. We will now consider the effects imports for cloud particle
activation.
Water molecules exhibit two types of interactions in the liquid and solid phases: strong covalent
bonds within the molecule (O-H bonds) and relatively weak hydrogen bonds between them (Fig.
4.4).
Figure 4.4: Schematic of liquid water structure
Surface tension
A molecule within the bulk of a liquid experiences attractions to neighboring molecules in all directions,
but since these average out to zero, there is no net force on the molecule (Fig.4.5). For a molecule that
finds itself at the surface, the situation is quite different; it experiences forces only sideways and downward.
As a consequence, a molecule at the surface will tend to be drawn into the bulk of the liquid. But since
there must always be some surface, the overall effect is to minimize the surface area of a liquid. This is
what creates the stretched-membrane effect known as surface tension, σl,v .
Figure 4.5: Schematic of the surface tension effect (source: www.chem1.com)
The surface tension is the free energy per unit surface area of the liquid and can be viewed
as the work per unit area required to extend the surface of liquid at constant temperature. The
Figure 4.6: Schematic of Gibbs free energy as a function of the substance volume at three different
pressures that are respectively less than, equal, or exceed the saturation pressure ps . Local equi-
libria (energy minima) are shown with circle, with the larger circles indicating absolute minima.
From Kashchiev (2000)
formation of a liquid drop needs an energy:
∆E = 4πr2 σl,v (4.1)
r is the drop radius. σl,v ≈ 7.5 × 10−2 N m−1 for usual conditions. The distinction between
molecules located at the surface and those deep inside is especially prominent in water, owing to
the strong hydrogen-bonding forces. Thus, compared to most other liquids, water also has a high
surface tension. The geometric shape that has the smallest ratio of surface area to volume is the
sphere, so very small quantities of liquids tend to form spherical drops. The fact that energy
is required to form the drop interface implies an energy barrier. The implication is that phase
transitions are not spontaneous, even if the Gibbs free energy would be lower as a result.This
is illustrated in Fig. 4.6. Even if e > es and thus liquid is the “preferred” phase, exhibiting
a Gibbs energy minimum, phase transition requires an energy barrier to be overcome. There
is no thermodynamical reason why a system in a local stable state may increase its Gibbs free
energy and cross the barrier to change phase. Thus processes of phase transition require concepts
from statistical mechanics that describe molecular-scale fluctuations of the system, in addition to
thermodynamics. Random fluctuations may lead to some molecules overcoming the Gibbs energy
barrier. The size of the barrier depends on the path. For example, in Fig. 4.6 we can view the
path along the line p = ps as the result of all molecules changing phase from gas to liquid, which
would involve a high energy barrier. Instead, the barrier energy is much lower if the new phase
occurs as the result of the formation of a small stable nucleus, involving a subset of n molecules
(Fig. 4.7). Hence phase transition in this way is referred to as a nucleation event . The nucleation
event in this example is referred to as homogeneous, since no foreign surface is present to lower the
energy barrier.
Figure 4.7: Schematic of the a nucleation event (source: Lohmann U, ETH)

4.2.1 The energy barrier and Kelvin’s equation

We now examine the energy barrier associated with the formation of a nucleus of n molecules in
the new liquid phase formed from a parent phase of N molecules as illustrated in Fig. 4.7. The
Gibbs State 1 has a Gibbs free energy of G1 = N gv , where gv is the Gibbs free energy per molecule
in the vapour phase. G2 = (N − n)gv + G(n), where G(n) is the Gibbs free energy of the liquid
phase nucleus, and has two contributions: G(n) = ngl + Gsurf (n). The first term is the Gibbs free
energy associated with the molecules in the liquid phase, while Gsurf (n) represents the work that
has to be done to form a surface around a volume containing n molecules. The work needed to
form the cluster is therefore
∆W (n) = n(gl − gv ) + Gsurf (n) (4.2)
To calculate the first term, we integrate dg from es for liquid to e the vapour pressure of the
surroundings at constant temperature:
Z e Z e Z e
kT
gv − gl = dg = vdp = dp (4.3)
es es es p
where we use pv = kT for a single molecule to derive the final term. Integration of this gives
gv − gl = kT ln(e/es ) (refer to derivation of Clausius Clapeyron in the thermodynamics notes).
This is per molecule, thus for n molecules we can write
Rv T 4 3
n(gl − gv ) = nkT ln(S) = πr ln(S), (4.4)
vl 3
where vl is the specific volume of liquid water and r is the nucleus radius, and S = ees . The
second term of eqn. 4.2 should account for the pressure difference within the droplet, but liquid
droplets are approximately incompressible, and thus the energy is given by Eqn 4.1
Gsurf = 4πr2 σl,v (4.5)
Combining eqns 4.4 and 4.5, we obtain the equation for Gibbs free energy for a cluster formation
in a parent phase.
Rv T 4 3
∆G = 4πr2 σl,v − πr ln(S) (4.6)
| {z } vl 3
surface term
| {z }
volume term
The first term on the right is referred to as the surface term and the second is the volume term.
Note the different r dependency. We assumed a constant temperature in this derivation and thus
latent heating is neglected. Equation 4.6 is illustrated in Fig. 4.8. If S < 1 then the volume and
surface terms are both positive, and ∆G increases monotonically with r. If S = 1 the volume term
is zero, but ∆G increases due to the surface term. For S > 1, there is a peak value of ∆G occuring
at a critical radius, which is an unstable equilbrium and marks the magnitude of the energy barrier.
If the drop radius is smaller than the critical radius the drop will tend to dissipate, while for a
radius exceeding this critical threshold the drop should grow (in theory infinitely). To calculate
the critical radius, we can differentiate eqn. 4.4 to get dG dG
dr (exercise!) and then set dr = 0. The
critical radius rc is then
2σl,v vl
rc = (4.7)
Rv T ln(e/es )
Thus we have seen that there is an energy barrier resulting from the surface tension of a droplet,
which was defined as the work per unit area required to extend the surface of a drop. A more phys-
ical interpretation for this is as follows. We saw that surface tension resulted from the asymmetry
of the forces of attraction at the drop surface. Work is thus required to bring a molecule from the
interior of a drop to its surface, as it requires bonds to be broken. A molecule on the surface will
then require less energy to overcome the remainding binding H-bonds and be released in the gas
phase. The smaller the droplet, the more curved the surface and the stronger this effect becomes.
Saturation over a curved surface
The relationship given earlier in the thermodynamics course for the saturation vapour pressure
was for a planar water surface. We have seen that on the scale of a cloud droplet the curvature of
the surface is sufficient to reduce the number of (attracting) neighbouring molecules.
Figure 4.8: Gibbs free energy for homogeneous water droplet formation of radius r at T=273.15K
for saturation ratios of S= 0.99 (solid), 1 (dotted), 1.01 (dashed) and 1.02 (dot-dash). The critical
radii of 0.06 and 0.12 µm are shown for the latter two saturation values (source U. Lohmann).
Figure 4.9: Schematic of evaporation process showing how water molecules have fewer neighbours
in a curved drop relative to a planar surface.
The saturation vapour pressure is thus higher over a curved surface. The saturation vapour
pressure of a liquid droplet of radius r given by inverting Eqn. 4.7, thus giving the minimum
supersaturation S = e/es that is required for a droplet of radius r to exist. :

es (r) 2σl,v
= exp , (4.8)
es (∞) rρl Rv T
1
. This is known as the Thomson or Kelvin effect. Although T is in the denominator on the
RHS, es is also a function of T and the temperature dependence is in fact weak. Note that the
surface tension in the above equation refers to the surface tension of a pure liquid water droplet
in water vapour. Presence of aerosols will change this value. Moreover, the above formulations
1 We will use notation es (r) = ers interchangeably.
are also valid for other phase changes, e.g. ice germ formation in a liquid, but the surface tension
value are less well defined from experimentation. We can simplify eqn. 4.8 to
es (r) a
= exp , (4.9)
es (∞) rT
where a = σl,v /Rv ρL which is (almost) constant at 3.3 × 10−7 m K.

The right hand side of the expression can be usefully approximated to
es (r) a
≈1+ , (4.10)
es (∞) rT
Homogeneous nucleation of liquid droplets

A chance collision of molecules to form a droplet is called homogeneous nucleation.
Let us consider such an event involving a relatively rare event: the chance collision contempo-
raneously of N =183 molecules of water vapour.
To calculate the radius of this newly formed droplet we note that the droplet volume is given
by
4 mv N
V = πr3 = . (4.11)
3 Na ρ L
Recall that the molecular weight mv is the mass per mole, while Na is the number of molecules
per mole, thus m Na is the mass per molecule.
v
Solving eqn. 4.11 for r, tells us that the droplet of N =183 molecules has a radius of approxi-
mately 10−3 µm. At a temperature of 273K, the ratio of the equilibrium vapour pressure es (r) is
three times greater than the value over a planer surface es (∞). The rate of growth of the droplet
is proportional to the difference e − es (r). If e < es (r) then the droplet will evaporate, while it will
grow if e > es (r). If we define the saturation ratio S as
e
S= (4.12)
es (∞)
so that S =3 equates to a relative humidity of 300%, then for the nascent droplet to grow would
require a saturation ratio of S >3. Q: Do we observe such values of relative humidity?. In fact,
such high values are never measured in the atmosphere, thus homogeneous nucleation is not a
relevant mechanism for cloud formation. So what is it?
Aerosols
Aerosols in the atmosphere can range in size from 10−4 to 10 µm radius with particle concentrations
also widely varying from 103 cm−3 in a remote location to > 105 cm−3 in an urban environment
such as London.
There are a number of natural and anthropogenic sources for aerosols:
• mineral soil dust
• sea salt (e.g. Fig. ??)
• gas to particle conversion SO2 →sulphate aerosols
• combustion of fossil fuels / biomass burning
Small aerosols with radii < 0.2 µm are referred to as Aitken particles, 0.2 < r < 2 µm are large
aerosols, and r > 2 µm are giant aerosols. Both dry deposition (sedimentation) and wet deposition
(removal by precipitation) are the sinks of aerosols, with a typical aerosol lifetime being around 1
week.
Heterogeneous nucleation of liquid droplets
Many aerosols in the atmosphere are hydrophilic (A hydrophilic molecule is one that has a tendency
to interact with or be dissolved by water) and thus water molecules can collect on their surface.
Clouds can thus form by a process known as heterogeneous nucleation, where water molecules
collect on a foreign substance. These hydrophilic aerosols are called Cloud Condensation Nuclei
Figure 4.10: Formation of sea salt aerosol, from Pruppacher and Klett (1997).
Figure 4.11: Schematic showing dissolution of sodium ion in water (wikipedia)
or CCN. These CCN are always present in sufficient numbers in the lower and middle troposphere
to initiate cloud growth.
CCN aerosols can be insoluble but wettable, which means that the surface tension between
their nucleating surface and water is sufficiently low and water can form a spherical cap completely
surrounding the aerosol. Thus the physics of drop nucleation is the same as for pure water, and
only large or giant aerosols (r > 0.2 µm) generally have a low enough curvature to form cloud
droplets at observed supersaturations. However, aerosols can instead be soluble, in which case
aerosols with much smaller radii can act as CCN. Solvation, also sometimes called dissolution,
is the process of attraction and association of molecules of a solvent with molecules or ions of a
solute. As ions dissolve in a solvent they spread out and become surrounded by solvent molecules
(Fig. 4.11).
The solution term
The presence of dissolved substances implies that the some water molecules are replaced from the
droplet surface (Fig. 4.12). Thus the saturation vapour pressure is reduced for a solute.
Figure 4.12: Schematic showing how saturation vapour is instead reduced when there are molecules
of dissolved substance (red) present in the droplet.
If es (sol) is the saturation vapour pressure over a solute and nw and ns are the number of
4.3. DIFFUSIONAL GROWTH 73
water and solute molecules, respectively, then the fraction of surface which is occupied by water
molecules is simply:
−1
es (sol) nw ns ns
∞
= = 1+ '1− , (4.13)
es nw + ns nw nw
where the final approximation assumes ns nw . In a droplet the number of water molecules is
proportional to r3 , thus for a fixed mass of aerosol this effect adjust the saturation vapour pressure
by a factor
es (sol) b
=1− 3 (4.14)
e∞
s r
where b is a constant that depends on the aerosol mass and type. The curvature (4.10) and solute
(4.14) effects can be combined to give the resultant equilibrium curve referred to as the Köhler
curve given by:
 
ers (sol)

b a  a b 
= 1 − exp ≈
1 + −  (4.15)
e∞s r 3 rT rT
|{z} r 3
|{z}

curvature term solute term
Figure 4.13 shows a normalized curve, and the actual shape depends on the mass and type of
solute. For a solute formed with 10−16 g of ammonium sulphate the solution term is ineffective
for radii above 0.3 µm. Due to the r3 factor the solute term dominates at small droplet radii.
If RH increases starting from a low value, water vapour start to condense on aerosol parcels as
Figure 4.13: Köhler curve for the equilibrium saturation vapour pressure for a liquid solute droplet.
RH reaches about 80% (S =0.8), referred to as haze particles. At these small radii the droplet is
stable as an increase in RH will cause the haze particle to grow until a new equilibrium radius is
reached. However, if the RH continues to increase a critical radius R∗ is reached at the critical
supersaturation S ∗ value at which the droplet becomes unstable, and grows rapidly by diffusion.
The droplet is said to be activated. It is the haze particles that are responsible for reducing
visibility on a sunny (humid) day.
In summary, about 10 to 20% of aerosols over oceans can act as CCN, while over land only
about 1% can act as CCN. Nevertheless, the total concentration of CCN is still higher over land
with a typical value of 500 cm−3 compared to 100 cm−3 over oceans, but these numbers are highly
temporarily and spatially variable. Therefore a cloud air parcel brought to saturation over ocean
shares the available water between fewer CCN, so we would expect fewer but larger cloud droplets.
The consequence of this is that maritime clouds are more likely to rain. We will see that with
larger droplets it is easier to grow raindrops.
4.3 Diffusional growth
Diffusional Growth of droplet

Once a cloud particle is activated it grows rapidly by diffusion of water vapour. As is usual, the
local diffusive flux is assumed proportional to the vapour gradient. Integrated over a sphere of
radius n (Fig 4.14) the total diffusive flux F (kg s−1 ) is:
dρv
F = 4πn2 D , (4.16)
dn
where D is the diffusion coefficient (≈ 2.2 × 10−5 m2 s−1 at 273K) and ρv is the vapour density.
Figure 4.14: Sketch of diffusion growth of a cloud drop.
In a steady state the diffusion rate is balanced by rate of increase of mass of droplet M :
dM dρv
= 4πn2 D . (4.17)
dt dn
We will assume that the growth is constant. Thus
dM ∞ dn
Z Z ρv (∞)
= 4πD dρv . (4.18)
dt r n2 ρv (r)
giving
dM
= 4πDr(ρv (∞) − ρv (r)) (4.19)
dt
We want the rate of change of radius, so we need the expression
4 3
M= πr ρL , (4.20)
3
which we differentiate to give
dM dr
= 4πr2 ρL . (4.21)
dt dt
Substituting (4.21) into (4.19) gives
dr D
= (ρv (∞) − ρv (r)). (4.22)
dt ρL r
We now use the ideal gas law (e = ρv (∞)Rv T ) to change the density to vapour pressure, noting
that at the droplet surface the air is exactly saturated:
dr D
= (Se∞ r
s − es (sol)) (4.23)
dt ρL rRv T
Now we use our earlier approximate expression for the Köhler curve given in (4.15):
dr De∞ s a b
= (S − 1 − + 3) (4.24)
dt ρL rRv T rT r
This equation is not tractable, but we can solve for r > 1µm, with the subsequent simplification:
dr De∞ s
' (S − 1) (4.25)
dt ρL rRv T
4.3. DIFFUSIONAL GROWTH 75
This expression is approximate and ignores an important effect. We have neglected the fact that
latent heat is warming the droplet, and to be strict the expression should account for the diffusion
of heat away from the droplet. This “complication” reduces the droplet growth rate very roughly
by a factor of 2.
Availability of water vapour
For a hypothetical cloud parcel there are two main terms that affect the supersaturation (or
equivalently relative humidity):
• increase due to the parcel cooling: rate proportional to the parcel velocity
• decrease due to the diffusion process
The evolution of an air parcel containing a spectra of diverse CCN can be calculated numerically
using a Lagrangian parcel model (see Ren and Mackenzie, 2005, for an example in ice clouds).
We will consider the hypothetical evolution of the parcel:
1. All CCN start to form haze particles at RH ≈ 80%,
2. the larger droplets formed on giant CCN (and thus have lower activation S) become activated
first.
3. These giant nuclei are few, thus have limited effect on the water availability, RH continues
to increase,
4. and we start to activate drops with higher activation S ∗ . There are many of these and thus
water vapour is consumed, eventually balance is reached where S is at a maximum.
5. After this point particles with lower S ∗ will continue to grow, while those with higher S ∗ will
decay as S decreases.
The final points to note are that
• The maximum supersaturation between (0.1 to 0.5%) occurs within 10 to 100 metres of cloud
base,
• Smax defines the number concentration of cloud droplets, and is thus also determined close
to cloud base,
• higher updraught speeds give higher Smax , thus higher number of activated droplets leading
to a higher number of final cloud droplets,
• Growth of cloud droplets from an initial aerosol of 0.5 µm takes only a few seconds explaining
why the cloud base is well defined,
• Calculated drop size spectra from modelling this process are narrower than observed (i.e.
less size variability). This is because we have neglected the processes of cloud mixing and
coalescence of droplets.
Time taken for diffusional growth

We have an expression for the rate of change of r as a function of time and supersaturation. If
we assume S is time-independent we can find out how long it takes to grow cloud droplets of a
certain size by the diffusion process by integrating (4.25).
This gives (exercise: check!)
ρL Rv T
t= (r2 (t) − r2 (0)). (4.26)
2De∞
s (S − 1)
If we take an example of T =284K and r(0)=0.5 µm and S − 1 =0.002 (a supersaturation of 0.2%)

then table 4.1 gives the time taken to grow a cloud droplet of the given sizes Which size is a cloud
droplet? Can drops grow by diffusion? The nonlinearity of the diffusional growth is emphasized
in Fig. 4.15. We recall that 10 µm was a typical cloud radius and thus growth by diffusion is fast
enough to produce such droplets. However 100 µm is a small raindrop size, and it is seen that
over 4 hours are required to achieve this initial raindrop, far longer than the observed life-cycles
of precipitating clouds.
r (µm) 1 5 10 100
t (seconds) 1 36 150 15000
Table 4.1: Time taken to grow a cloud drop of given radius by diffusion.
Figure 4.15: Diffusion growth curves, from Lohmann U..
4.4 Terminal velocity of particles

Terminal velocity
We now introduce the concept of the terminal fallspeed of particles in a simplified way. The
terminal velocity is achieved when the force due to gravity F = 43 πr3 g(ρL − ρ) is balanced by the
drag on the droplet.
The drag strongly depends on the drop size, as seen in Fig. 4.16. The drag on a droplet is
Figure 4.16: Changes in the air flow patterns round falling water drops as the Reynolds number
increases (source: Mason, 1978)
related to the particle velocity V and the radius r and can be divided into three regimes, leading
to three distinct terminal fall speed Vt relationships:
4.5. COLLISION AND COALESCENCE 77
• r < 30µm: Drag ∝ V r giving Vt = X1 r2 where X1 ∼ 1.2 × 108 s−1 m−1 .
• 30 < r < 1000µm: Drag ∝ V r2 giving Vt = X2 r where X2 ∼ 8 × 103 s−1 .

√
• r > 1000µm: Drag ∝ V 2 r2 giving Vt = X3 r where X3 ∼ 250s−1 m0.5 .
Table 4.2 gives typical fallspeeds as a function of radius. From these fallspeeds we notice two
r (µm) 1 10 20 100 1000

Vt (m s−1 ) 1.2 × 10−4 0.012 0.048 0.8 8
Table 4.2: Time taken to grow a cloud drop of given radius by diffusion.
facts:
• The fallspeed of typical cloud droplets (2 to 20 µm) is negligibly small compared to typical
updraught and downdraught velocities, with a 10 micron droplet requiring a day to fall 1km.
This means that to a good approximation we can assume that cloud droplets are in suspension
in the air.
• The differential terminal fallspeeds also implies that larger droplets falling faster than smaller
ones may collide and collect smaller droplets during their descent (see schematic in Figs. 4.17)
Figure 4.17: Schematic of larger raindrops colliding with smaller drops during descent (source
hrefwww.islandnet.comislandnet.com).
The sequence of laboratory photos shown in Fig. 4.18 show a droplet collision and cohesion
event.
4.5 Collision and coalescence

We consider a large drop of radius R is falling through a cloud of smaller droplets radius r. If we
take a simple view and assume that any small droplet in the path of the large droplet comes into
contact and is collected (as in the sequence of Fig. 4.19, the volume of droplets collected per unit
time is π(R + r)2 (V − v). The liquid water content of the small droplets is L = ql ρ in kg m−3 ,
giving a mass accumulation rate of
dM
= Lπ(R + r)2 (V − v) (4.27)
dt
We will simplify the equation by assuming R >> r and V >> v, and then use (4.21) which we
recall states dM 2 dr
dt = 4πR ρL dt to give
dR LV
= (4.28)
dt 4ρL
Figure 4.18: Sequence of shots showing droplet collision and subsequent cohesion (source unknown)
Figure 4.19: Schematic of collision and coalescence. Larger drops of radius R fall at terminal
velocity V , collecting all smaller droplets of radius r within a cylinder of radius R + r.
If we take the case of the initial growth of small droplets (r < 30 µm) then the terminal velocity
was given by V = X1 R2 giving
dR LX1 R2
= (4.29)
dt 4ρL
Thus the growth rate due to collisions is proportional to the square of the droplet radius in this
regime, while we recall that the radius rate change due to the diffusive process was proportional
to the inverse of radius. This implies that there is a changing balance between the competition of
these two processes, with the dominance of diffusion overcome by the importance of the collection
process as the radius increases (see 4.20)
We can simply integrate (4.29) to show that for L = 10−3 kg m−3 a drop can grow from 20 to
30 µm in approximately 10 minutes. We then need to introduce the relationship Vt = X2 R into
(4.28) and integrate to calculate that the raindrop can attain a radius of around 300 µm in 20
minutes.
These times appear to be reasonable compared to cloud lifetimes, but we have ignored two
effects in this simple view, Q: Can you think what they might be? We assumed simplistically that
all droplets with in the large droplet trajectory collided with the large drop and that all collisions
led to coalescence. Neither of these two assumptions are very accurate.
Collection efficiency E(R, r)
4.5. COLLISION AND COALESCENCE 79
Figure 4.20: Schematic of droplet radius as a function of time resulting from growth by diffusion and
collisions processes. NOTE: Y-AXIS LABLE IS INCORRECT AND SHOULD READ “DROPLET
RADIUS”, FIGURE WILL BE CORRECTED SHORTLY.
We saw earlier in Fig. 4.16 how the streamlines around various raindrop sizes looked. The lack of
inertia of very small droplets implies that they will tend to get swept around the larger droplets
if r/R is small (see Fig. 4.21). To describe this we introduce the collision efficiency parameter
E(R, r). For small r/R, E(R, r) can be as low as 0.1, while when R ∼ r the flow fields can interact
in a complex way and result in E(R, r) > 1.
Coalescence efficiency,
The Coalescence efficiency is often assumed to be ∼1, but observations show that it can be lower
than 0.5.
Figure 4.21: Schematic showing more realistic path of small droplet (source Mason, 1978).
Figure 4.22 shows two methodologies for the calculation of E(R, r) from Klett and Davis (1973)
which differ greatly for R < 30 µm in this case. Taking the collision efficiency and Coalescence
efficiency into account modifies our radius growth rate equation to
dR LV E(R, r)
= (4.30)
dt 4ρL
and growing a droplet from 20 to 30 µm can now take as much as 100 minutes instead of 10
minutes, and we no longer grow rain drops in the observed time of cloud development. There are
a number of mechanisms that lead to an accelerated growth rate to get us out of this impassé:
• Statistical models of raindrop growth: we assumed that droplet growth was a discrete function
of time, whereas it consists of a serious of discrete events. If in a time ∆t 10% of raindrops
collide and coalesce then after time 2∆t there will be one large drop from 100 initial droplets
(see Fig. 4.23). But this larger droplet will then be favoured for further growth as the growth
rate is ∝ r2 . A spectrum of drop sizes, often bimodal, is therefore generated (see Fig. 4.24).
• It was also assumed that droplets were evenly distributed within the cloud, whereas in reality
cloud are inhomogeneous, and regions of higher liquid water content will favour droplet
growth.
Figure 4.22: Graph of theoretical collection efficiencies E(R, r) for two calculation methodologies
with the x-axis showing the radii ratio r/R. Each line is labled with the radius of the larger droplet
(source Klett and Davis, 1973).
• The process of entrainment cause the partial evaporation of droplets broadening the droplet
size distribution.
Figure 4.23: stochastic model of sweep out (source Berry, 1967).
Raindrop size distributions

In many parts of the globe frozen precipitate melt before they reach the surface and thus precip-
itation reaches the surface in the form of rain. There is a natural limit to the raindrop size due
to droplet breakup, and it is usually very rare to measure raindrops with radii larger than 3mm -
with 2mm a more common limit.
Q: If you stand underneath a tree during a shower, you may get much larger drops fall on you.
Why?
The reduction in occurrence with droplet size implies that raindrop radii (or diameters) tend
to follow an inverse exponential distribution, known as the Marshall-Palmer distribution (Marshall
4.6. ICE CRYSTAL NUCLEATION 81
Figure 4.24: Example of the development of a droplet spectrum by stochastic coalescence (source
Berry and Reinhardt, 1974).
and Palmer, 1948), after the first authors to suggest the relationship from observations
N (D) = N0 e−ΛD , (4.31)
where N (D)dD is the number of drops per unit volume with diameters between D and D + ∆D.
The slope factor Lambda depends on rainfall rate R and is given by
Q: what properties do you notice from the rainfall size distribution?
Figure 4.25: Marshall-Palmer distributions (from Marshall and Palmer, 1948)
− Λ(R) = 41R−0.21 , (4.32)

where the units of R and Lambda are mm hour−1 and cm−1 , respectively. It was also found
that N0 was independent of rainfall rate, taking a value of N0 = 0.08 cm−4 . Fig 4.25 shows the
Marshall-Palmer distribution.
Of course, these relationships are empirical and valid over a large-number of observations.
Exact relationships will vary with location and time and departures from the MP distribution will
always occur (Fig. 4.26).
4.6 Ice crystal nucleation

Introduction to ice processes
Figure 4.26: An example of raindrop size distributions from Willis (1984)
The first thing that should be emphasized when considering the ice phase is that ice phase processes
in clouds are far more complicated (Fig. 4.27) and consequentially poorly understood relative to
the warm phase cloud processes. This course can only skim the surface of the relevant mechanisms.
Figure 4.27: Example of an ice crystal from http://www.its.caltech.edu/)
Liquid water was described as a state where water vapour molecules were jumbled together and
weak hydrogen bonds were formed between molecules, which are constantly overcome by thermal
agitation. If temperatures become low enough to prevent the disruptive effects of thermal motions,
water freezes into ice in which the hydrogen bonds form a rigid and stable network (Fig. 4.28).
In this well-defined structure of ice each water molecule is surrounded by four neighboring H2Os.
Two of these are hydrogen-bonded to the oxygen atom on the central H2O molecule, and each of
the two hydrogen atoms is similarly bonded to another neighboring water molecule. This lattice
arrangement requires that the molecules be somewhat farther apart than would otherwise be the
case in liquid and as a consequence, ice, in which hydrogen bonding is at its maximum, has a more
open structure, and thus a lower density than water (Fig. 4.29).
Figure 4.28: The ice latice structure (www.chem1.com)

4.7. ICE SATURATION 83
(a)
(b)
Figure 4.29: Schematic of (a) liquid and (b) ice molecular structures (source edinformatics.com)
4.7 Ice saturation

Saturation over a plane surface of pure ice
Before we continue with the discussion of ice we must first consider the concept of saturation over
a planar ice surface. Analogous to the liquid water saturation, the air is said to be saturated if the
deposition (vapour→ice) rate equals the sublimation (ice→vapour) rate.
As the intermolecular bonding energy of molecules in ice is greater than that in liquid water,
at a given temperature, the evaporation rate is larger than the sublimation rate. Thus it is clear
that esi (T ) < esw (T ) (see Fig. 4.30) where the saturation vapour pressures are
• esi over ice
• esw over liquid water.
Figure 4.30: Water and ice saturation vapour pressures (from Berner and Berner 1987)
Note that while the ratio eeswsi

increases with decreasing temperature, the highly nonlinear
saturation curves implies that this is not true of the absolute difference between the two esw − esi ,
which reaches a maximum at a temperature of around -15o C.
4.8 Ice nucleation mechanisms

There are three potential mechanisms by which ice clouds can form:
• Homogeneous freezing from the liquid phase
• Homogeneous nucleation from the vapour phase
• Heterogeneous nucleation from the vapour phase
Homogeneous nucleation from the vapour phase requires very high relative humidities for similar
reasons to those given for the warm phase, and thus this is not a viable mechanism for ice crystal
formation in the atmosphere. Thus the two key methods by which ice can form is by homoge-
neous freezing of a liquid droplet, or by direct vapour deposition on an ice nucleus (heterogeneous
deposition nucleation). Homogeneous freezing is the dominant process in cirrus clouds, which het-
erogenous nucleation from the vapour phase dominates crystal formation in the mixed phase and
in clouds with fast vertical updraught speeds.
4.9 Homogenous nucleation from the liquid phase

When the temperature falls below 0◦ C there is no guarantee that freezing of liquid cloud droplets
will immediately occur. Freezing begins when an initial crystal, termed ice germ, is formed by
statistical fluctuations of the liquid molecular arrangement to form a stable ice-like lattice structure.
As in all nucleation processes, energy is required for the formation of the ice germ surface:
4πr3 Rv T es,w
∆Gi,w = 4πr2 σi,w − ln (4.33)
3vi es,i
e
Thus, the lower the temperature, the larger es,w
s,i
becomes, which lowers the energy barrier to form
a critical ice germ. From Fig. 4.31 it is seen that at colder temperature the energy barrier is far
lower than that for homogenous liquid or ice crystal nucleation from the vapour phase. If the
Figure 4.31: Gibbs free energy barrier for homogeous nucleation phase transitions (from Lohmann
U.)
germ is over a critical size, then other water molecules will bind to the ice germ rapidly and the
water body will freeze rapidly. Again, we refer to concepts of statistical mechanics. The larger
a body, the more likely it is that random energy fluctuations will occur to create an ice germ of
a critical size for spontaneous freezing to occur. Large drop are more likely to freeze than small
drops. It is clear that a pond therefore freezes more readily than a cloud of liquid droplets as the
formation of an ice germ of critical size only has to occur once in the pond, while the event needs
to occur in each cloud droplet which is far less likely. Thus it is possible and indeed very common
to find liquid cloud drops existing at temperatures much below the freezing point; referred to as
supercooled droplets (see Fig. 4.32). Theoretical results imply that a liquid drop of 5 µm will
spontaneously freeze at temperature of around -40◦ C, however, these calculations are uncertain as
σi,w is poorly known. Empirical measurements inside real clouds show that they contain no liquid
4.9. HOMOGENOUS NUCLEATION FROM THE LIQUID PHASE 85
Figure 4.32: Liquid water to ice fraction in the ECMWF cloud scheme (Forbes et al. 2011).
drops once temperatures of around -35◦ C to -40◦ C are attained (depending on the cloud drop size
spectra). Such clouds are said to be completely glaciated.
Homogeneous nucleation of ice from vapour
It is also theoretically possible to nucleate ice crystals directly from the vapour phase, however
theory shows that very high supersaturations (> 1000%) with respect to ice would be required. At
these high supersaturations, air would also be supersaturated with respect to liquid water, forming
droplets that would freeze. Thus homogeneous nucleation of ice directly from the vapour phase is
not a relevant mechanism for creating ice cloud.
Ice supersaturation in the upper troposphere

One consequence of the above is that cloud-free air can be supersaturated with respect to ice and
not form ice cloud. However there is an upper limit to the amount of ice supersaturation set by
esw since at this point liquid droplets form which will homogeneously freeze if the temperature
is below -40◦ C. A common sign of the upper troposphere being supersaturated is the presence of
permanent contrail cloud (Fig. 4.33).
Figure 4.33: Photo and satellite picture of permanent contrails over the UK. (source unknown).
As temperature get colder, esw and esi diverge, and larger ice supersaturations are possible. In
fact it is more complicated and the limit is below esw since we recall that aqueous solution droplets
(haze) forms at RH with respect to water substantially below 100%.
Koop et al. (2000) investigates this limit in detail and shows an upper limit of 45% ice super-
saturation at T = 235 K increasing to 67% at T = 190K.
In Fig. 4.34 Spichtinger et al. (2003) shows using retrievals from microwave limb sounder data
that ice supersaturated states are quite common in the upper troposphere. We have seen that
homogeneous ice nucleation occurs at temperature below -40◦ C and that supercooled liquid water
is very common below 0◦ C. However it is also true that ice is observed in clouds between 0◦ C and
-40◦ C. Clouds Q: How could it get there?.
• Air was previously at much colder temperatures: parcel history
• Ice has fallen from above: ice sedimentation
• Ice has nucleated in situ: how?

Figure 4.34: Frequency of ice supersaturation at 215hPa (source Spichtinger et al., 2003)
Ice can form by collision of liquid water droplets with other ice crystals, freezing on contact.
Alternatively, ice crystals can form with the aid of aerosols in a heterogeneous nucleation process.
Heterogeneous nucleation of ice
Aerosols in the atmosphere can also act as ice nuclei (IN) if their molecular structure is close
enough to the lattice structure of ice. The first point to emphasize is that it is much less common
for aerosols to have this property and therefore ice nuclei are much less common than CCN.
At T = −20◦ C a typical number concentration (NC) of ice nuclei might be 10−3 cm−3 compared
to CCN of roughly 102 cm−3 .
Whether an aerosol can act as a cloud nuclei depends on the ice supersaturation and the
temperature.
As ice supersaturation increases and temperature reduces more aerosols take on the property
of being ice nuclei. Fletcher (1962) found that for each 4◦ C of cooling the number of ice nuclei
increases by a factor of ten (Fig. 4.35).
Figure 4.35: Ice nuclei concentration as a function of temperature (source Fletcher, 1962)
Common ice nuclei

Some ice nuclei are given in table 4.3 along with their maximum nucleation temperature.
Demott et al. (2003) tested air samples taken from mountain top in mid-Western USA and
documented the distribution of IN depicted in Fig. 4.36. The wide range of supersaturation
and temperature thresholds for various classes of IN is shown in Fig. 4.37 which is taken from
4.9. HOMOGENOUS NUCLEATION FROM THE LIQUID PHASE 87
Substance Nucleation T (◦ C) Notes

ice 0
silver iodide -4 used in cloud seeding
clay -9 often seen in snow crystals
cholesterol -2
bacteria(!) close to 0 used in snow machines
Table 4.3: ice nuclei and their nucleation temperature maximum
Figure 4.36: From Demott et al. (2003), distribution of IN type from air samples taken over the
mid-West USA.
Figure 4.37: The onset temperatures and relative humidities for IN from a compilation of exper-
imental data of sub- and supermi- cron aerosol particles in the literature from Hoose and Möhler
(2012). The wide variability in the thresholds is emphasized.
Hoose and Möhler (2012). It should be emphasized that knowledge of which aerosols can act
as ice nuclei is uncertain and the subject of current research. Ice nuclei concentrations are highly
variable, and although the Fletcher curve may be accepted as typical, concentrations can vary by
orders of magnitude. Aerosols may also be carried a long distance in the atmosphere before being
involved in ice cloud nucleation events. Demott et al. (2003) and others have shown that mineral
aerosols from Western Africa are a significant source of ice cloud nuclei over the United States
for example! In fact many snow or ice particles contain clay mineral aerosols indicating that the
desert regions of the world are in general an important source of IN (see Fig. 4.38).
General properties of ice nuclei

Observations and laboratory experiments indicate that aerosol particles usually satisfy a list of
What kind of cloud is this figure showing?
Figure 4.38: Modis image over Europe (source MODIS website)
criteria if they are to serve as IN (Pruppacher and Klett, 1997).
• Insolubility criterion: In general, IN must be very insoluble. The disadvantage of a soluble

substrate is that it disintegrates under the action of water. Hence the molecular structural
requirement for ice nucleation can not be maintained.
• Size criterion: IN must have a size comparable to, or larger than, that of a critical ice
embryo. Generally speaking the radius of the IN must be greater than 0.1 µm; i.e. Aitken
particles do not tend to be ice nuclei. At sizes less than this critical radius (0.1 µm), the nu-
cleating ability of the particle decreases rapidly (and becomes increasingly more temperature
dependent).
• Chemical Bond criterion: The chemical nature of an IN, that is the type and strength of
the chemical bonding sites at the IN surface, affect nucleation. In view of this bond criterion,
certain complex organic molecules (i.e. aerobic bacteria) exhibit good ice nucleation abilities.
The hydrogen bonding groups in ice are similar to the hydrogen bonding groups in many
organic molecules.
• Crystallographic criterion: Since ice nucleation on a substrate is actually an overgrowth

of ice on the substrate, it is reasonable to expect the nucleating ability of the substrate to
increase when the lattice structure of the substrate is similar to the hexagonal lattice structure
of ice. In this way, molecular matching between the molecules of ice and the substrate may be
achieved (Fig. 4.39). The crystallographic matching reduces the “misfit” and elastic strain
in the ice.
Figure 4.39: An ice-embryo (dark points) growing upon a crystalline substrate with a misfit of
10%. The interface is dislocated, dislocations being indicated by arrows. (from Fletcher, 1969) ?.
Both of the properties of misfit and elastic strain reduce the nucleating ability of the substrate.
Kaolinite forms good ice nuclei because it has a pseudo-hexagonal structure. AgI (silver iodide)
also has a hexagonal form. Other molecules have a non-hexagonal shape (e.g. cubic), but still have
crystal structures close enough to ice to aid nucleation. The crystallographic requirement has to
do with the arrangement of molecules and is distinct from the chemical bond criterion.
• Active Site criterion: Ice nucleation generally occurs at distinct sites on the substrate.
Laboratory studies have indicated that ice nucleation is preferred on those substrates which
4.10. ICE CRYSTAL GROWTH 89
have pits and steps on their surfaces. Those nuclei that have smooth surfaces are less efficient
nuclei, that is, nucleation occurs at lower temperatures. An example is shown in Fig. 4.40
where ice crystals preferentially formed on steps etched onto the surface of a cadmium iodide
crystal.
Figure 4.40: An epitaxial deposit of ice crystals growing at the steps of a hexagonal growth spiral
on cadmium iodide (source ?).
Mechanisms for Heterogeneous Nucleation

There are several mechanisms for ice crystal nucleation, schematically illustrated in Fig. 4.41.
Figure 4.41: Schematic of heterogeneous nucleation pathways (source: Rogers and Yau, 1989).
These are
• heterogeneous deposition nucleation: the vapour undergoes deposition directly onto the IN
• condensation nucleation: the IN acts first as a CCN to form a liquid droplet, and then acts
as a IN to initiate freezing
• contact nucleation: a supercooled liquid droplet undergoes freezing immediately on contact
with the IN
• immersion nucleation: IN absorbed into a liquid droplet, and later initiates freezing event,
possibly after droplet cooling.
4.10 Ice crystal growth

We divide the consideration of ice crystal growth by diffusion of water vapour into two regimes:
mixed phase and glaciated clouds. For mixed phase clouds we recall that at subfreezing temper-
atures the ice and water vapour saturation pressures diverge, and thus an air volume at water
saturation is supersaturated with respect to a planar ice surface, reiterated in Fig. 4.42.
Figure 4.42: Diagram showing relative degrees of ice and liquid water super or sub-saturation as a
function of temperature (source: Pruppacher and Klett, 1997))
Bergeron-Findeison Effect
The Bergeron-Findeison effect is an important growth enhancing mechanism in mixed phase clouds.
In a supercooled liquid water cloud imagine one liquid droplet freezes (Fig. 4.43):
• The cloud is initially saturated with respect to liquid water.
• It is thus ice supersaturated.
• The ice crystal will grow by diffusion of water vapour towards the crystal,
• This reduces the vapour pressure below the liquid water saturation value.
• The liquid droplets evaporate on a fast timescale maintaining the vapour pressure close to
the liquid water saturation.
Figure 4.43: Schematic of Bergeron-Findeison process, with the cyan hexagon representing an ice
crystal in a liquid water cloud
The outcome of the Bergeron-Findeison effect is that the growth rate of the ice crystal is faster
than it would be in the absence of liquid cloud droplets. For a water saturated environment the
growth rate is fastest when the difference between the saturation limits is largest in absolute (not
relative!) terms. This occurs around T = −15◦ C.
A laboratory photo shows the result of the BF-effect, with the lack of liquid droplets surrounding
the ice crystal on the plate (Fig. 4.44)
Ice Habits
Ice habits can be very complex and depend on the temperature and ambient ice supersaturation
at which ice nucleation takes place (Fig 4.45). The crystal shape has a strong impact on radiative
properties (Macke et al., 1996) and the fall speeds of the ice particle.
Ice forms crystals having a hexagonal lattice structure, which in their full development would
tend to form hexagonal prisms very similar to those sometimes seen in quartz. This does occa-
sionally happen and depends on the ambient temperature and supersaturation. More commonly,
4.11. COMPETITION BETWEEN ICE NUCLEATION MECHANISMS 91
Figure 4.44: Photo of BF-Effect (source R. Potter from ?)
Figure 4.45: Ice crystal types as a function of Si and T (source caltech.edu)
crystals form in a flattened fractal-like hexagonal structure. Ice that grows under warmer condi-
tions grow more slowly, resulting in smoother, less intricate shapes. A classification of ice shapes
in shown in Figure 4.46.
4.11 Competition between ice nucleation mechanisms

Competition between ice nucleation mechanisms
At temperatures warmer than -40◦ C the only nucleation mechanism is heterogeneous. However at
temperatures colder than this both heterogeneous and homogeneous nucleation can occur contem-
poraneously.
As many IN become active at ice supersaturations of 10 to 30%, well below the threshold range
for homogeneous nucleation, then homogeneous nucleation has the potential to be the dominant
nucleation mechanism, if and only if IN are present in sufficient numbers to prevent the RH
reaching the homogeneous nucleation threshold in a rising (cooling) parcel of air.
Q: Both mechanism create ice crystals, so why do we care which one dominates? Demott
et al. (2003) (Fig. 4.47) documents the concentration of nucleated ice crystals as a function of
water saturation for air samples collected over the western United States. For T > −35◦ C the
NC increases with supersaturation as more aerosols become active. However at cold temperatures
there is a step jump close to water saturation to much higher concentrations due to homogeneous
nucleation. . Homogeneous nucleation thus has the potential to create a much higher concen-
tration of cloud ice crystals - Q: Why? The homogeneous nucleation process creates more ice
particles since it is due to the freezing of aqueous solution droplets and thus related to the number
of CCN. The particles nucleated by homogeneous process are much smaller in size therefore.
Figure 4.46: Ice crystal categorization (source caltech.edu)
Figure 4.47: Nucleated ice crystal number in air samples as a function of water saturation at two
temperatures. (source: Demott et al., 2003).
Whereas in warm clouds the droplet concentration was related directly to the CCN number, in
ice clouds the relationship between IN and ice particle number concentration is more complicated
and nonlinear, falling into two distinct regimes, illustrated schematically in Fig 4.48. The figure
shows three cases of an lifted hypothetical parcel of air at cold temperature (T < −40◦ C). The left
case is for a clean sample of air with no IN. The parcel rises until the RH attains the threshold for
homogeneous nucleation, 160% in this example. Many ice particles ensue. These rapidly grow by
depositional growth as the air is highly ice supersaturated and the water vapour content returns
rapidly (in seconds to minutes) to the ice saturation vapour pressure.
The middle column shows a parcel of air with a low number of IN. These nucleate ice at lower
supersaturation thresholds (say 30%) but the number is too small to “use up” (by deposition)
much vapour and the homogeneous threshold is still reached. Nevertheless the “overshoot” over
this threshold and number of aqueous particle formed will be less due to the take-up of vapour
by the IN, and thus the final total number of ice crystals will be less. This also implies that the
return of the vapour content to the ice saturation value will be slower.
4.11. COMPETITION BETWEEN ICE NUCLEATION MECHANISMS 93
Finally the right hand column shows air with a high enough content of IN such that the ice
particles nucleated are sufficiently numerous to prevent the RH from even reaching the threshold
for homogeneous nucleation and the latter mechanism is effectively inhibited. The final number of
ice particles is orders of magnitude lower than in the clean air case and the return to ice saturation
is much slower and may take on the order of an hour or longer.
After this point further increases of IN will increase the ice particle concentrations as in warm
phase clouds, but always at much lower concentrations than produced by homogeneous nucleation.
Figure 4.48: Schematic of homogeneous and heterogeneous nucleation competition, see text and
lesson for details
The final number of ice crystals will be dependent on the both the IN concentration of the air
and the updraught velocity of the air parcel. As in warm clouds, parcel models are used to study
this, as in this example of Ren and Mackenzie (2005) in Fig.4.49. In a weather or climate model,
what are the issues involved in representing this relationship? The following points should be
Figure 4.49: Results from parcel model of Ren and Mackenzie (2005) using a single aerosol type
with a IN property threshold of S = 1.3
noted about the parcel model results in Fig.4.49:

• The overshoot past Scr for homogeneous nucleation determines the number of ice crystal
nucleated.
• The overshoot has a timescale of seconds, and thus is not temporarily resolved by numerical
models (parcel models use sub-second timesteps to resolve this),
• The overshoot depends on the updraught velocity and also the number of aerosols present
and their own nucleation (Scr ) properties.
• A numerical model needs to be able to resolve the updraught velocity of the parcel on the
scale of the cloud
• A numerical model needs to have accurate information concerning the aerosol quantities
present.
• In many cases the diffusional growth timescale is also less than the timestep of a typical
global model (O(1 hour)).
Another example of this work is Gierens (2003). To pick up on the point concerning the model
vertical velocity, Fig. 4.50 gives typical PDFs observed and also modelled using a global model.
It is seen that the coarse resolution of the model truncates the spectra and does not resolve the
highest velocities occurring on small spatial scales. Some success at reproducing the statistics of
the vertical velocity spectrum is achieved with a simple parametrization.
Figure 4.50: Results from Lohmann (citation needed) showing PDF of vertical velocity from (blue)
a global model (red) aircraft observations and (black) model results corrected with a simple tur-
bulence based parameterization for use in a ice-cloud model.
4.12 Aggregation
Aggregation
The ice particle that form from diffusional growth from an ice crystal are called pristine. However
pristine ice particles can clump together to form snowflakes in a process known as aggregation.
At temperatures as low as 200K, the surface of ice is highly disordered and water-like. As the
temperature approaches the freezing point, this region of disorder extends farther down from the
surface and acts as a lubricant.
Thus the efficiency of the aggregation process increases as the temperature exceeds -5◦ C, when
ice surfaces becomes sticky. There is also a secondary peak between -10◦ C and -16◦ C when dendrite
arms get entangled. Figures 4.51 and 4.52 show example aggregates of bullet rosettes and other
aggregates.
4.13 Riming
Riming to form graupel
If vapour exceeds the water saturation mixing ratio (in strong updraughts) then water can condense
on ice crystals, and then subsequently freeze to form graupel, which are round ice crystals of higher
densities.
4.14. ICE PARTICLE FALL-SPEEDS 95
Figure 4.51: Bullet Rosette aggregates aggregates (source Dr. Chris Westbrook, Reading Univer-
sity, UK)
Figure 4.52: Other aggregate example (source Eszter Barthazy).
Graupel and Hail are also formed by aggregating liquid drops in mixed phase clouds (see
schematic in Fig. 4.53).
Figure 4.53: Schematic of riming process (source unknown)
Some examples of lightly and heavily rimed ice are given in Fig. 4.54. If the latent heat
of condensation and fusion keeps temperature close to 273K then high density hail particles can
form, since the liquid water “spreads out” before freezing. Hail is dense and thus has high terminal
velocities (up to 40 m/s) implying that it only forms in convection with strong updraught able to
support the particle long enough for growth
4.14 Ice particle fall-speeds

The density and size of ice particles determines their respective fallspeeds and its dependence on
particle radius (Fig. 4.55).
There are implications for numerical models that may become clear examining radar animations
of cloud systems (see link). can you think what these might be?
Figure 4.54: Photos of lightly and heavily rimed ice (source caltech.edu)
Figure 4.55: Fallspeeds of bulk ice particle categories (source: Rogers and Yau, 1989).
4.15 Ice multiplication

Ice multiplication
In recent years it has become increasingly evident that concentrations of ice crystals in “real” clouds
are not always represented by the concentrations of IN measured or expected to be activated in
such environments. In particular, it has been found that at temperatures warmer than -10◦ C the
concentration of ice crystals can exceed the concentration of IN activated at cloud top temperature
by as much as three or four orders of magnitude
There have been a number of hypotheses which have been advocated to explain such discrepan-
cies. In particular, these hypotheses attempt to account for the unexpectedly high concentrations
at warm temperatures. Some of the leading hypotheses are:
1. Fragmentation of large drops during freezing.
2. Mechanical fracture of fragile ice crystals (i.e., dendrites and needles) caused by collision of
these crystals with graupel and other ice particles.
3. Splinter formation during riming of ice crystals.
4. Artificial measurement artifact of ice crystals shattering on impact with aircraft measurement
device inlet.
Aircraft observations of ice in mid-level clouds observed in aircraft campaigns by Fleishauer
et al. (2002) are shown in Fig. 4.56. Although many models make simple assumptions such that
ice particle size reduces with temperature, it is seen that this is not necessarily valid in clouds
with strong mixing. The small particles may only appear circular due to instrument resolution
limitations.
4.15. ICE MULTIPLICATION 97
Figure 4.56: Example ice crystals from aircraft observations in mid-level cloud systems by
Fleishauer et al. 2002 Fleishauer et al. (2002).
Snowflake size distributions

Most of the ice precipitation that reaches the ground does so as snowflakes and not pristine ice
crystals.
Since snowflakes are irregular aggregates of crystals there is no simple way to measure their
dimensions, thus they are usually measured in terms of their mass or, equivalently, the diameter
of the water drop that would form from their melting.
Gunn and Marshall (1958) found that the exponential approximation still fits the snowflake
diameter well with the following parameters:
− Λ(R) = 25.5R−0.48 , (4.34)
and
N0 = 3.8 × 10−2 R−0.87 (4.35)
where the units of R and Lambda are mm hour−1 and cm−1 (Fig 4.57). To summarize, the key
Figure 4.57: Snow flake size distributions (from Gunn and Marshall, 1958)
growth mechanisms are summarized in Fig. 4.58. and the location of some of these key processes
in a typical deep convective cloud are given in Fig. 4.59.
Reminder Questions
• explain at the board what it means for an aqueous aerosol solution droplet to become “acti-
vated”.
• What are the main mechanisms of ice nucleation that are relevant for ice cloud formation?
• How is a cloud called that consists only of ice, and one which consists of both liquid and
ice? In roughly which temperature regimes does each cloud type exist? In the mixed phase
explain what the Bergeron-Findeison effect is for ice growth
Figure 4.58: Crystal growth processes (a) condensation (b)deposition (c)collision-coalescence, (d)
aggregation, (e) riming, (f) Bergeron-Findeisen process. (from Lohmann, U.)
Figure 4.59: microphysical processes in a deep convective cloud (from Lohmann, U.)
• Explain why we can find air that is supersaturated with respect to ice but not with respect
to liquid water? What kind of cloud is an indication of supersaturated layers?
• Does the liquid cloud droplet radius increase or decrease with CCN number? Assuming that
in a forming cloud, condensing liquid water is equally distributed among CCN, write down
the relationship between droplet radius and cloud liquid water density in kgm−3
• Does the ice crystal radius increase or decrease with ice nuclei (IN) number? What is a
common source of IN?
• What are the two main mechanisms by which activated aerosols grow into raindrops, and
under which droplet size radius do they operate and why? We made a number of approxi-
mations in our mathematical descriptions of the two processes, can you name some?
• What factor directly determines the number of ice crystals nucleated by each homogeneous
nucleation, and what is this dependent on in turn?
Chapter 5
Radiation
Radiative processes
The sun (Fig. 5.1) is the fundamental energy source of the planet, and the uneven adsorption
of solar radiation leads to the large-scale weather patterns that redistribute heat. Perturbations to
radiative processes are responsible for climate variability through the greenhouse effect, volcanic
eruptions and solar variability. Understanding radiative processes is also important for remote
sensing of the atmosphere.
Figure 5.1: The sun! (http://z.about.com)
5.1 Definitions of the radiative field

Electromagnetic radiation
Electromagnetic radiation (EMR) can be considered as an ensemble of waves propagating at the
speed of light, but also as photon particles travelling through space. The EMR wave has both
electric and magnetic field components and carry energy. EMR is created from other types of
energy, and converted to other types of energy when destroyed, usually when interacting with
matter. For example, the vibrational and rotational kinetic energy of molecules can be converted
into EMR by the emission of a photon, or can increase by the absorption of a photon.
As with other waves, we describe EMR in terms of its frequency and wavelength.
Electromagnetic spectrum
The electromagnetic (EM) spectrum is the range of all possible electromagnetic radiation frequen-
cies. The ”electromagnetic spectrum” of an object is the characteristic distribution of electromag-
netic radiation from that particular object.
Bands that are important for atmospheric radiative processes are the ultraviolet, visible (400
to 700nm), near/thermal infra-red (Fig. 5.2).
Electro-magnetic wave
Electro-magnetic waves have a speed of c = 3 × 108 m/s, (Fig. 5.3) with speed, frequency ν and
wavelength λ related by
c = νλ (5.1)
99
100 CHAPTER 5. RADIATION
Figure 5.2: Electromagnetic spectrum with visible bands highlighted (source see caption).
Figure 5.3: Propagating electromagnetic wave.
The Earth with radius1 re orbits the sun (rs ) at a radius of rd and intercepts a tiny fraction of
the emitted energy (see sketch in Fig. 5.4). We assume that the sun emits as a black body.
Figure 5.4: Schematic of sun earth system.
Black Body
A black body is an object that completely absorbs all incident electromagnetic radiation. No
electromagnetic radiation passes through it and none is reflected. It is also a perfect emitter of
radiation.
Irradiance and Radiant Emittance E

Irradiance and radiant emittance are radiometry terms for the power of electromagnetic radiation
arriving or leaving a flat-plate surface, per unit area, integrated over the hemisphere above the
plate (Fig. 5.5). Irradiance is used when the electromagnetic radiation is incident on the surface
while radiant emittance is used when the radiation is emerging from the surface.
The total energy per unit time is called the radiant flux and has units of J s−1 or W. As
the irradiance and radiant emittance are measured per unit area they have units W m−2 . Note
that there is nothing in the definition of irradiance that tells us from which direction the radiation
arrives or is emitted (Fig. 5.6), so we therefore introduce the solid angle.
1 In the radiation section r is used for radius, not to be confused with mixing ratio! Other duplications of notation
will also occur.

5.1. DEFINITIONS OF THE RADIATIVE FIELD 101
Figure 5.5: Sketch of radiation incident on a flat flat.
Figure 5.6: Sketch showing how the irradiance can be equivalent with the radiation arriving from
two different directions.
Solid Angle Ω
The solid angle, Ω is the angle in three-dimensional space that an object subtends at a point. It is
a measure of how big that object appears to an observer looking from that point. The solid angle
is proportional to the surface area, A, of a projection of that object onto a sphere centered at that
point, and inversely proportional to the square of the sphere’s radius, r (Fig. 5.7).
We define the solid angle
A
Ω= (5.2)
r2
and it has units of steradians or ster. An important special case for radiation is the integration
Figure 5.7: Definition of the solid angle.
over a hemisphere of unit radius in spherical coordinates (Fig. 5.8).

π
Z Z 2π Z 2
dΩ = sinθdθdφ = 2π ster (5.3)
hemisphere 0 0
The angle to the zenith is defined by θ, with θ = 0 marking the vertical axis normal to the surface. The
azimuth angle is defined by φ. We integrate elements of solid angle dΩ over concentric circles starting in
the horizontal direction. The radius of the circle element of width dθ from the vertical axis is rsinθ = sinθ.
Thus the azimuth dimension of the solid angle element dΩ is sinθdφ. The vertical dimension of dΩ is dθ.
Radiance
The radiance tells us the radiant flux density per unit solid angle coming from a direction (θ, φ).
For a surface area of ∆Anorm normal to the direction (θ, φ), the radiant flux received from the solid
angle ∆Ω will be L(θ, φ)∆Ω∆Anorm (Fig. 5.9).
L(θ, φ) is the radiance with units W m−2 ster−1
However, in general the receiving surface will not be normal to θ, φ. If we take the special case
where the receiving surface of dimension ∆A is horizontal, then the projection of this area on to
the surface normal to direction (θ, φ) is given by ∆Anorm = ∆Acosθ (Fig. 5.10).
Thus in general the contribution to the radiant flux incident on a horizontal surface is given
by: L(θ, φ)∆Ω∆Acosθ. If we take ∆A = 1 for a unit area, then the radiant flux per unit area
from solid angle (i.e. the contribution to the irradiance from that solid angle) is L(θ, φ)cosθ∆Ω
Figure 5.8: Integration of solid angle over a hemisphere
Figure 5.9: Definition of radiance sketch.
with units W m−2 . From now on we will assume the (θ, φ) angle dependence and simply write L
for radiance. With remote sensing we need a device capable of measuring radiation from different
directions with very small solid angles to achieve the best resolution possible (Fig. 5.11).
Irradiance E Definition
We are now able to return to the concept of irradiance, which is the total radiance integrated over
all solid angles over a hemisphere.
Thus π
Z Z 2π Z 2
E= LcosθdΩ = Lcosθsinθdθdφ (5.4)
hemisphere 0 0
Isotropic radiation
An important special case is isotropic radiation, namely that the radiative flux has the same
intensity regardless of the direction, and thus L is independent of direction. Q: can you think of
some examples? Examples include:
• Emission from a black body.
• Solar radiation beneath a cloud of moderate thickness due to multiple scattering of photons.
• Background microwave radiation from big bang.

5.1. DEFINITIONS OF THE RADIATIVE FIELD 103
Figure 5.10: Definition of radiance sketch.
Figure 5.11: Schematic of why small solid angles measurements are required for remote sensing
(source NASA website).
In this case we may write

π
Z 2π Z 2
E=L cosθsinθdθdφ = πL (5.5)
0 0
Exercise: show! Q: What are the units of π in (5.5)? The units of π are steradians.
Planck Function
Experiments lead to the derivation of Planck’s law, which describes the spectral radiance of
electromagnetic radiation at all wavelengths from a black body at temperature T . The Planck
Function giving the radiance for wavelength λ as a function of T is:
2hc2
Lλ (T ) = ch (5.6)
λ5 (e kλT − 1)
where h=6.63e−34 J s is the Planck constant and k=1.38e−23 J K−1 is the Boltzmann constant.
The emitted power per unit area of emitting surface, per unit solid angle, and per unit wavelength.
We can derive the frequency form of the Planck function. The change in the Planck function
for a finite change in wavelength dλ is equal to the change for an equivalent change in frequency
dν:
Lλ (T )dλ = −Lν (T )dν, (5.7)
with the minus sign indicating that an increase in wavelength corresponds to a decrease in fre-
−c
quency. Using the chainlaw on eqn. 5.1 we find that dλdν = ν 2 , substituted into eqn. 5.6 and using
again eqn. 5.1 gives:
2hν 3
Lν (T ) = hν (5.8)
c2 (e kT − 1)
At room temperature, black bodies emit mostly infrared light (λ > 4µm), but as the tempera-
ture increases past a few hundred degrees Celsius, black bodies start to emit visible wavelengths,
from red, through orange, yellow, and white before ending up at blue, beyond which the emission
includes increasing amounts of ultraviolet (e.g. Fig. 5.12). Examples of the Planck function for
Figure 5.12: The colours indicate the temperature of the emitting body
Figure 5.13: Black body emission at three tropospheric temperatures.
temperatures between 200 to 300 K are given in Fig. 5.13.

Note how the peak in Fig. 5.13 moves to shorter wavelengths as the temperature increases.
The position of the peak as a function of wavelength can be found solving dLλ /dλ = 0.
From eqn. 5.6 we see
hc −hckT hc

−hc2 λ5 e kλT (kλT )2 + (e kλT − 1)5λ4
dLλ /dλ = hc 2 =0 (5.9)

10
λ (e kλT − 1)
and with some cancellation and simplification we get
hc hc hc
− e kλT + (e kλT − 1)5 = 0 (5.10)
kλT
hc
We now substitute x = kλT to get:
5(ex − 1) − xex = 0. (5.11)

hc
This has to be solved numerically, which results in the solution x=4.965, and thus λmax = kxT .
Inserting the values of h,c,k and x we get
5.2. ENERGY BALANCE MODELS OF THE ATMOSPHERE 105
Wein’s Displacement Law

The relationship describing the wavelength at which emission is at a maximum is
2897µK
λmax = (5.12)
T
Integrating eqn.(5.6) across all wavelengths gives the total emitted energy per unit area per
unit solid angle, in other words the radiance L. Thus
∞ ∞
2hc2
Z Z
L≡ Lλ dλ = ch dλ. (5.13)
0 0 λ5 (e kλT − 1)
hc −hc
We again make the substitution of x = kλT , implying dx = kλ2 T dλ, to get
3 Z 0
−2hc2 kT x3

kT
L= dx, (5.14)
hc hc ∞ (ex − 1)
which simplifies to (dropping minus as we reverse the integration limits)

∞
2(kT )4 x3
Z
L= dx (5.15)
h3 c2 0 (ex − 1)
R∞ x3 π4
Using the solution 0 (ex −1) dx = 15 we get
2π 4 (kT )4
L= (5.16)
15h3 c2
We now want to integrate L over the hemisphere to give us the total radiant emittance E from
a unit area flat plate over all wavelengths. We calculated this in the derivation of eqn.5.5 which
tells us in the case of isotropic blackbody radiation E = πL. Thus we derive the
Stephan-Boltzmann Law
2π 5 k 4 4
E= T = σT 4 (5.17)
15h3 c2
where σ is the Stefan Boltzmann constant:
2π 5 k 4
σ= = 5.67 × 10−8 W m−2 K −4 (5.18)
15h3 c2
Grey Bodies and Kirchoff ’s Law

A grey body emittance is one which emits radiation according to
Egrey = σT 4 , (5.19)
where is the fractional emittance ( ≤ 1 ).

Kirchoff ’s Law states that, if at a given wavelength we know how good an emitter a body is,
then we also know how good an absorber it is. Kirchoff’s Law specifies:
• Absorptance a =
• Transmittance τ = 1 −
Figure 5.14: Inverse square law. source: http://jersey.uoregon.edu/ imamura/122/lecture-

1/lecture-1.html
5.2 Energy balance models of the atmosphere
Energy balance models of the atmosphere

We are now in a position to construct a simple energy balance model of the earth-atmosphere
system. If the sun is considered to be a black body (a good approximation) then its emittance is
4
given by σTsun where Tsun is the surface temperature of the sun. This is emitted over a surface
area of 4πrs giving a total energy of 4πrs2 σTsun
2 4
.
The energy of the sun is spread over the surface of a sphere of area 4πrd2 at a distance of rd ,
known as the inverse square law (Fig. 5.14). The irradiance E (W m−2 ) at a distance rd is
thus
2
4 rs
E = σTsun . (5.20)
rd
If rd corresponds to the distance of separation between the earth and the sun, then we define
S0 ≡ E, called the solar constant.
Solar Constant S0
S0 , the average irradiance at the top of the earth’s atmosphere, is known as the Solar Constant.
Measurements from satellites show that S0 ' 1370 W m−2 . The value varies by ≈ 1 W m−2 over
a solar cycle that has an average length of 11 years.
Knowing the solar constant we can invert (5.20) to get the surface temperature of the sun:
s
2
4 rd S0
Tsun = (5.21)
rs σ
giving Tsun as 5800K. The energy intercepted by the earth is (Fig.5.15) S0 πre2 , so that Ee , the
average incident solar radiation on the earth’s surface, which has a surface area of 4πre2 , is
S0 πre2 S0
Ee = = (5.22)
4πre2 4
which has a value of 340 W m−2 .

Planetary Albedo
A proportion of this radiation is scattered back to space by clouds, atmospheric gases and the
planet’s surface. We thus introduce the parameter αp as the planetary albedo. Which is defined as
Reflected solar irradiance (at top of atmosphere)

αp = (5.23)
Incident solar irradiance
Observations from Satellites put the earth albedo at αp ' 0.3. Thus the average irradiance
absorbed is S40 (1 − αp ) which is 240 W m−2 .
Effective Emitting temperature of the Earth
5.2. ENERGY BALANCE MODELS OF THE ATMOSPHERE 107
Figure 5.15: Shadow area of earth
The Earth/atmosphere system is assumed to emit to space as a black body at some effective
temperature Te . Thus the rate of emission is 4πre2 σTe4 . We will assume that there is a long term
balance (i.e. an equilibrium) between absorbed solar and emitted infrared radiation on an earth:
S0 πre2 (1 − αp ) = 4πre2 σTe4 (5.24)
Thus r
4 S0
Te = (1 − αp ) = 255K (5.25)
4σ
Neglecting this albedo issue, we have two contrasting temperature of the sun (5800K) and the
earth (255K). This is important since the peak wavelengths (λ) at which energy is emitted is a
function of temperature. The Planck function Lλ is shown graphically for the sun and earth
effective emitting temperatures in Fig. 5.16. It should be pointed out the solar radiation energy is
Figure 5.16: Planck function for sun and earth’s effective emitted temperatures of 5800K and
255K, respectively.
not restricted to the visible (0.4 to 0.7 µm), rather 35% of the solar energy received at the earth
is in this band, with 50% in the near-infrared and 15% in the ultra-violet.
Comparing the normalized curves for Lλ , depicted in Fig. 5.17, it is also clear that the result
of the different emitting temperatures is that the radiation from the sun and earth are in distinct
wavelength regions with little overlap. We found that Te = 255 K and yet surface measurements
indicate an average global temperature of roughly 288K. Q: Why are these so different? These are
different because Te is the effective emitting temperature of the whole earth-atmosphere system,
this is not the same as the surface temperature. Some gases in the atmosphere are not transparent
to infrared radiation emitted by earth, thus warming the surface relative to Te in a process known
as the greenhouse effect.
Figure 5.17: As Fig. 5.16 but normalised and plotted with a linear y-scale.
5.3 Sun and Earth Geometry

In the previous section we constructed a simple slab-atmosphere radiative balance model. We now
briefly discuss the sun-earth geometry in more detail as this determines the insolation of the earth
by the sun and the sun position in the sky needed for radiative transfer calculations.
Sun-Earth Geometry
The sun-earth geometry is shown in Fig.5.18. Note the
1. Declination angle of the earth ϑ
2. Eccentricity of the earth’s orbit
Figure 5.18: B
Declination Angle ϑ
• ϑ = +23.5◦ : at the summer solstice June 21st
• ϑ = -23.5◦ : at the winter solstice December 21st
• ϑ = 0◦ : at the spring and Autumn Equinoxes
Eccentricity of orbit
The Earth’s orbit is slightly elliptical with an eccentricity factor of e = 0.0167. Since the average
sun-earth separation is rd = 149.6 × 106 km, we can calculate:
5.4. RADIATION INTERACTIONS WITH A SLAB 109
• The Perihelion (distance of closest approach) is rd (1 − e) = 147.1 × 106 km.
• Thus the Aphelion (distance of furthest approach) is rd (1 + e) = 152.1 × 106 km.
Presently, the Perihelion occurs on January 3rd while the Aphelion July 4th, but the closeness
of these dates to the solstices is a present-day coincidence.
Solar Zenith Angle

The solar zenith angle (θ) is a function of latitude, time of year and time of day. The daily
averaged insolation is given in Fig. 5.19. Q: At midsummer, at which latitude do you think Qav is
greatest: Tropics, subtropics, midlatitudes or at the north pole?
Figure 5.19: Day averaged insolation Qav (W m−2 ) at the TOA as function of season and latitude.
The heavy dashed line marks the latitude of the subsolar point at noon ?.
In midsummer Qav is maximum at the north pole. θ ranges from 0◦ at a latitude of 23.5◦ N to
66.5◦ at the north pole. Q: If the sun is overhead at 23.5◦ N, why is Qav is maximum at the north
pole? It is due to the day length which more than offsets the lower value of S0 cosθ.
5.4 Radiation interactions with a slab
Radiation interaction with a slab

We now consider the interaction of radiation with a layer of the atmosphere in more detail,
considering the processes depicted in the schematic. We imagine a layer that is homogeneous in
the infinite horizontal direction, and has a finite vertical depth of ∆z (Fig. 5.20).
The radiance Lλ which we recall is a function of (θ, φ), can be divided into three distinct
components.
1. The direct solar component: radiative flux that passes unimpeded through the slab direction
(θ, φ).
2. The scattered component: radiative flux originating from a different solid angle, but undergoes
scattering within the slab to emerge in direction (θ, φ).
3. Emission: Radiation emitted from the slab in direction (θ, φ).

Figure 5.20: Schematic of radiation interaction with a medium of thickness dz.
5.4.1 Direct Radiation
Direct Radiation - Beer’s Law

Beer’s Law The law states that there is a logarithmic dependence between the transmission τ
of the a radiative flux through a substance and the product of the absorption coefficient of the
substance, a, and the thickness of the material (i.e. the path length).
To derive this, we note that the finite reduction in the direct component by the slab dLλ will
be directly proportional to (Fig. 5.21):
Figure 5.21: Schematic of radiation interaction with a medium of thickness dz.
• Lλ : The original intensity of the radiative beam
• secθdz: The path length through the slab of the radiation.
• ρ: The density of radiatively active material in the layer (e.g. cloud drops, gases).
• kλe : The mass extinction coefficient at wavelength λ
Thus
dLλ = −Lλ kλe ρsecθdz, (5.26)
where ρ is the density of the material (kg m−3 ).
Mass extinction, absorption and scatter coefficients
For a specific material, it is useful to define the mass extinction kλe . This specifies how strongly a
unit mass of a medium depletes a direct beam of radiation of a given wavelength λ (or equivalently
frequency ν) and has units m2 kg−1 .
The mass extinction coefficient is the sum of the absorption kλa and scattering kλs coefficients.
kλe = kλa + kλs (5.27)
Single scattering albedo

5.4. RADIATION INTERACTIONS WITH A SLAB 111
The single scattering albedo is defined as the ratio
kλs
ω= (5.28)
kλe
The two extreme regimes are
• ω = 1: Pure scatter (conservative scatter)
• ω = 0: Pure absorption
We now integrate (5.26) over a finite depth slab (Fig. 5.22 to get the direct component of
radiation as Z z2
z1 z2 e
Lλ = Lλ exp − kλ ρsecθdz . (5.29)
z1
Figure 5.22: Schematic of integral over slab depth dz.
Optical Thickness/Depth
If from (5.29) we set
Z z2
δλ = kλe ρsecθdz. (5.30)
z1
Then δλ is known as the optical thickness or optical depth of the layer between z1 and z2 ; the
integrated extinction coefficient.
Beer’s (or Lambert’s Law) then can be written
Lzλ1 = Lzλ2 e−δλ (5.31)
showing us that the transmittance τ is related to optical depth by
τλ = e−δλ (5.32)
We lastly emphasize that the transmittance, absorptance, extinction and optical depth are func-
tions of wavelength or equivalently frequency.
5.4.2 Emission from Slab
Emission from slab

The emission by a slab in direction (θ, φ) is equivalent to the absorption, and thus simply given by
Lemitted
λ = Lλ kλa ρsecθdz (5.33)
5.4.3 Scattering from other directions
Scattering from other directions

The scattering from other directions (θ0 , φ0 ) is defined as
π
Z 2π Z
ωλ 2
Lscatter
λ = Lλ (θ0 , φ0 )P (θ0 , φ0 , θ, φ)sinθdθdφ (5.34)
4π 0 −π
2
We note
• The two integrals are to integrate over the sphere.
• 4π is a normalization factor, the solid angle integrated over a sphere, while we recall that ωλ
is the single scattering albedo.
• Recall sinθdθdφ is the element of solid angle.
Phase Function
P (θ0 , φ0 , θ, φ) is called the Phase Function and gives the probability that a photon will be scattered
from direction (θ0 , φ0 ) into direction (θ, φ)
We note that for solar radiation (λ < 4µm) thermal emission is unimportant as Lλ is very
small at typical earth temperatures (recall Fig. 5.17). For thermal infrared (λ > 4µm) scattering
can generally be neglected, thus only the extinction and emission are important. We now examine
these in turn.
5.5 Absorption by atmospheric gases

Gases are not like black bodies that absorb equally and completely at all wavelengths. Rather,
they absorb only at specific, often narrow ranges of wavelengths.
Absorption by atmospheric gases
In general molecules posses three kinds of quantized energy
• Electronic Energy (Visible and UV)
• Vibrational Energy (Near IR, Thermal IR 0.7-20 µm)
• Rotational Energy (Thermal IR)
To absorb or emit radiation the molecule must possess a permanent dipole, as it vibrates or
rotates the electric field varies, allowing the interaction with the incoming/outgoing radiation.
N2 and O2
The major species of N2 and O2 are simple molecules that possess no permanent dipole (Fig. 5.23),
and thus possess almost no vibrational or rotational absorption.
Figure 5.23: N2 molecule
CO2
Carbon dioxide CO2 is a linear molecule, and thus from its symmetry possesses no dipole and thus
no rotational spectrum. However, there are certain vibrational modes shown in (Fig. 5.24).
H2 O and O3
5.5. ABSORPTION BY ATMOSPHERIC GASES 113
Figure 5.24: CO2 molecule with its main vibrational modes
Figure 5.25: H2 O molecule and its charge structure
H2 O and O3 do possess a permanent dipole as the molecular structure is “bent” and hence asym-
metric2 . They therefore also have a rotational spectrum. (Fig. 5.25),
Around the turn of the century Max Planck theorized that radiation energy is quantized in
order to address the shortcomings of the classical theory. As radiation is quantized, it implies that
for a given frequency of radiation, there can be only one value of quantum energy for the photons
of that radiation. Planck suggested the relation
E = hν (5.35)
The energy levels of atoms and molecules can have only certain quantized values and transitions
between these quantized states occur by absorption and emission.
The photon energy given by the Planck relationship is equal to the energy separation of the
participating pair of quantum energy states (Fig. ??)
Figure 5.26: Schematic of state transitions (source: http://hyperphysics.phy-astr.gsu.edu)
Different rotational and vibrational states can be combined to give a variety of possible energy
transitions. The excited states do not persist: after some random amount of time, the molecules
revert back to their original, lower energy state. The vibrational or rotational mode decays,
emitting a photon. The energy associated with with a transition between two states is
∆E = hν (5.36)
where h is the Planck Constant
Due to the existence of various rotational and vibrational discrete levels, molecules can absorb
or emit at many different wavelengths or wavenumbers, with the strength of the absorption related
2 see the atmospheric thermodynamics lecture notes from term 1, specifically Fig. 32
Figure 5.27: Quantized vibrational energy levels.
to the probability of the transition between the relevant energy levels. This leads to the unique
Absorption Spectrum of gases
Line broadening
Due to the discretised nature of the molecular energy levels one might expect absorption and
emission to occur at distinct set delta function wavelengths as in Fig. 5.28a. In reality the energy
levels are slightly “fuzzy” due to line broadening mechanisms so that absorption is spread over a
finite width of wavelengths (Fig. 5.28b).
Figure 5.28: (a) Hypothetical discrete absorption lines (b) actual absorption due to line broadening
mechanisms
Three mechanisms for line broadening are
• Natural Broadening: a fundamental limit to the narrowness of the spectral lines due to
Heisenberg’s uncertainty principle, which relates the uncertainty of the time spent in excited
states to the energy of the state transition
• Pressure Broadening: The energy levels are perturbed by the presence of other particles
nearby the emitting one, for example, collisions of other particles with the emitting particle
interrupts the emission process.
• Doppler Broadening: the emitting particles will have a variety of motions towards and away
from the observer leading to a Doppler shift in the radiation.
Natural Broadening is overwhelmed by the other two processes. Q: Where in the atmosphere
will pressure broadening dominate? Pressure broadening dominates in the troposphere where p
is relatively high and Doppler broadening is important in the upper stratosphere (z > 30km).
We recall from earlier (5.32) that the transmittance is related to optical depth as τλ = e−δλ =
5.6. SCATTERING 115
Rz
exp(− z12 kλe ρsecθdz). If we assume a homogeneous material (kλe is constant) and a solar zenith
angle of zero, then this simplifies to
τλ = exp(−kλe ρ(z2 − z1 )) (5.37)
The absorption in a band depends on the product kλe ρ(z2 − z1 ). If a particular gas is an efficient
absorber (high kλe ) and/or is present is high amounts (large ρ(z2 − z1 )) then the incoming/outgoing
radiation of a band of wavelengths around the line centre may be completely absorbed. We thus
find that we can have two regimes of absorption lines (Fig. 5.29):
• Weak Limit: Absorption is linear with absorber amount
• Strong Limit: Radiation is totally absorbed at line centre, adding more absorber only in-
creases absorption at band edges. Absorption increases proportionally to the square root/log
of the absorber amount with the constant of proportionally depending on pressure.
Figure 5.29: Schematic of integral over slab depth dz.
This is important when considering the greenhouse effect of gases. Carbon dioxide CO2 for
example is present in substantial amounts and is in the strong limit. Doubling CO2 does not
double its greenhouse affect. On the other hand CFCs are not present naturally and absorb in a
part of the spectrum where other atmospheric gases are not active (referred to as an absorption
window), and therefore their greenhouse effect increases linearly with concentration.
The major absorption bands are shown in Fig. 5.30 with the contribution by the main active
gases. It is seen that nearly all of the outgoing infra-red radiation from the earth’s surface is
absorbed by water vapour except for the window between 8 and approximately 15 µm. However,
CO2 is strongly active in this window with an absorption band between 14 and 16µm. We saw
in Fig. 5.30 the existence of the water vapour window in the IR. Q: In remote sensing what would
this be useful for? The window region of 8-13 µm is useful for remote sensing of the surface (e.g.
SST).
In contrast to the earth’s IR radiation, the majority of the solar radiation at the top of atmo-
sphere reaches the earth’s surface, with the atmosphere on average absorbing around 17% of solar
radiation. Fig. 5.31 shows the measured incoming solar radiation spectra and marks the specific
gas absorption bands marked in addition to a reference blackbody curve.
5.6 Scattering
The scattering of radiation by a particle, depends strongly on the ratio of the wavelength of the
radiation to the radius of the particle. The theory that governs this is called Mie theory, involving
the internal and external refraction of the radiation by the particle and the full description is
beyond the scope of this course. (Fig.5.32)
The scattering regime has a large impact on the phase function P (θ0 , φ0 , θ, φ) (Fig. 5.33, see
also Maxwell animations online). The relevant parameter to determine the scattering regime is the
ratio of the particle radius r to the radiation wavelength 2πr/λ.
One can see that in the regime 2πr/λ << 1 the scattering is more regular in direction, with
the phase function loosing its directional dependence in the limit 2πr/λ → 0.
Rayleigh Scattering
Figure 5.30: Upper panel shows the transmitted radiation in comparison to theoretical blackbody
curves. Lower panels show the percentage absorptance due to the atmosphere and the contribution
to this by the main active gases (source Wikipedia).
Figure 5.31: Solar incoming TOA and surface radiation spectra with the gases responsible for the
major bands marked.
If we assume in the regime 2πr/λ << 1 (the limit is about 5 × 10−3 ) that the phase function P is
independent of direction and thus the scattering is a function of λ only, then this approximation
is called Rayleigh Scattering. The scattering extinction coefficient is proportional to the inverse
fourth power of the wavelength, kλs ∝ λ14 . In the Rayleigh regime the scatterer can be treated as
a single dipole.
The regimes where Rayleigh scattering approximation is valid are shown in Fig 5.34. In general
the radiatively “active” species are the trace gases such as CO2 , H2 O, O3 . Rayleigh Scattering is
one of the few examples where the major species such as N2 and O2 are “active”. Q: why is the
sky blue?
We recall that the scattering coefficient in the Rayleigh regime is proportional to the inverse
fourth power of the wavelength. Red light hasa λ = 0.7µm while blue light has λ = 0.4µm,
indicating an order of magnitude increase in scattering at blue wavelengths.
Contribution of Rayleigh scatter to planetary albedo
Rayleigh scattering’s contribution to albedo depends on the solar zenith angle. Q: why?. The solar
zenith angle determines the pathlength of the radiation and the likelihood of a scattering event.
When the sun is overhead Rayleigh scattering contributes 0.037 to the albedo, while the figure is
0.13 for a solar zenith angle of 80◦ . The global average value is about 0.06, thus 0.06/0.30= 20%
of the planetary albedo is due to Rayleigh scatter.
Q: How does a greenhouse work?
5.6. SCATTERING 117
Figure 5.32: Schematic of mie scattering (source www.malvern.co.uk)
Figure 5.33: Scatter regimes as a function of 2πr/λ (a) 2πr/λ << 1, (b)2πr/λ ∼ 1 (c) 2πr/λ > 1
A simple model for the greenhouse effect

To illustrate the Greenhouse Effect we will now construct a simple radiative balance model for
the earth surface with an overlying atmosphere which will be considered a single radiatively active
homogeneous slab.
We make the following simplifying assumptions:
• The earth surface is a black body.
• The atmosphere does not absorb solar radiation.
• Atmosphere can be represented as a grey body.
We can now write down the radiative energy balance of each component (according to the
sketch in Fig. 5.35), assuming a state of radiative equilibrium.
Q: Derive these components.
Top of atmosphere
S0
(1 − αp ) = σTs4 (1 − ) + σTa4 (5.38)
4
Atmosphere
σTs4 = 2σTa4 (5.39)
Q: Why is there a factor 2 on the RHS?
Surface
S0
σTa4 + (1 − αp ) = σTs4 (5.40)
4
Q: Why isn’t there an factor in the solar term?
We thus have 3 equations with 2 unknowns Ts and Ta , and can use any two to calculate these
unknowns. For example, from (5.39) we have
Ts
Ta = √
4
, (5.41)
2
Figure 5.34: Scattering regimes as a function of wavelength and particle radius.
Figure 5.35: Sketch of the greenhouse slab model (source: me!)
which if we substitute into (5.38) gives

s
4 S0 (1 − αp )
Ts = (5.42)
2σ(2 − )
Exercise: Check this derivation and also try other equation combinations using (5.40)
If we assume that = 0 then we get the same result we achieved earlier, namely Ts = 255K, Q:
Why?.A more reasonable value is = 0.6, which results in Ts = 278K and Ta = 235K. Thus we can
see that introducing an atmosphere that can absorb in the thermal IR causes a surface warming.
This is known as the greenhouse effect. We will later consider what determines and its variation
in a continuous media.
What happens to Ta as increases? From eqn. 5.41 we see that Ta increases. In fact the
warming is a function of height 3 , but in contrast temperatures in the stratosphere will reduce
since they receive less IR radiation from below.
We could take this approach further and divide the atmosphere up into many vertical slabs, each
with its own emittance according to the absorbing gases present in the layer. This would provide
a vertical atmospheric temperature profile of the earth-atmosphere system in a state of radiative
equilibrium (sketch). However this would not be a good model of the atmosphere temperature
structure, Q: Can you think why this would be the case?
If one calculates the vertical temperature structure that would arise from radiative equilibrium
it turns out that it is absolutely unstable to both dry and moist deep convection. Thus convection
can (and does) occur to transport heat from near the earth’s surface to higher in the atmosphere.
One of the earliest studies to examine this using a column model of the atmosphere was conducted
by Manabe and Strickler (1964) (Fig. 5.37). Q: in a state of radiative equilibrium, what would be
the net radiative heating everywhere in the atmosphere? In a radiative equilibrium the net heating
3 in the tropics due to the role of convection we expect a slightly larger increase in the upper troposphere
5.7. RADIATION BUDGET OF CLOUDS 119
Figure 5.36: Observed and modelled temperature trends from Seidel et al. (2011) (?)
Figure 5.37: Temperature profiles resulting from a pure radiative equilibrium, or a radiative-
convective equilibrium assuming a simple parametrization of convection that adjust the profile
to a dry (9.8 C km−1 ) or mean moist (6.5 C km−1 ) temperature lapse rate. From Manabe and
Strickler (1964).
rate would be zero everywhere by definition. Instead in the troposphere there is a net cooling rate
of O(1 K day−1 ) which is balanced by convective heating processes.
5.7 Radiation budget of clouds

Scattering and absorption by clouds
In clouds Rayleigh scattering approximation no longer holds and 2πr/λ ≈ 100, and the full (com-
plicated) Mie theory is required.
We saw in Fig. 5.33 that in this case there was a strong “forward lobe” where the forward scatter
was much stronger than the backwards scatter. For liquid water droplets the forward scattering
lobe can be many orders of magnitudes greater than the back-scattered radiation. This leads to
an apparent paradox as clouds reflect up to 70% of the incident solar radiation back to space. Q:
how can this occur? The solution is that we have so far considered only single scattering events.
In optically thin clouds photons may undergo only a single scattering events and thus the vast
majority pass through unimpeded. In optically thick clouds though multiple scattering may occur
(Fig. 5.38). Thus the possibility of light reemerging at the cloud top is much higher. This is why
thick clouds appear dark when viewed from the surface. Cloud albedo asymptotes to around 0.8
for thick clouds. We recall that ω tells us according to (5.28) what proportion of the direct
beam extinction is due to scatter or absorption. For solar radiation ω > 0.9 meaning most of the
radiation is scattered and indeed for visible wavelengths ω > 0.999; the conservative scattering
regime. However, in the thermal infra-red generally ω < 0.5 implying that there is little chance of
Figure 5.38: Schematic of multiple scattering with a cloud.
a photon at these wavelength undergoing multiple scattering events and escaping from the cloud.
Thus clouds are essentially black bodies in the infra-red. We can see from the above that clouds
will have large effect on the earth’s radiation budget depending on their height and optical depth.
Low clouds
• SW: Low clouds will sufficient optical depth will reflect sunlight back to space efficiently,
thereby reducing the amount of solar radiation absorbed at the surface.
• LW: In the long wave infra red, cloud absorb the surface emitted radiation efficiently as they
are effectively black bodies. They emit infrared radiation back to earth at the approximate
temperature of the cloud lower surface, and to space the cloud top temperature. Q: Do low
clouds have a strong impact in the IR therefore?
The answer depends on the level you are concerned with! The TOA budget is not impacted
since the cloud top temperature will be similar to the surface. But low cloud will change the
surface budget witnessed by the fact that winter nights are warmer with low cloud cover.
High clouds
High clouds of both reflect light back to space, but they can also have a strong IR signature since
they are very cold (temperatures close to that of the tropopause). The balance can be in favour
of warming or cooling depending on the crystal type4 and the optical depth.
An intriguing result in the tropics is that on the average it appears that high cloud radiative
forcing is almost exactly zero, i.e. the solar and the infra-red effects balance each other. This
may be a coincidence of the present climate and not hold in past or future climates, but some
authors have suggested that this result is no accident, and have put forward mechanisms involved
negative feedbacks between radiation and convection that maintain this balance (see Hartmann
et al., 2001).
5.8 Summary of Earth’s radiation budget

We now examine the radiative budget of Earth (Fig. 5.39). This emphasizes the fact that once
clouds are accounted for, very little of the surface emitted radiation escapes directly to space.
Fig. 5.40 shows the net TOA shortwave radiation budget.
• Up to 300 Wm−2 in the tropics and below 100 Wm−2 in polar regions (why?)
• the highest values are over the oceans (why?)
• the lowest vaues are over the deserts (why?)
Fig. 5.40 shows the net TOA longwave radiation budget.
• Strong gradient from equator to poles (why?),
4 Remember that these are ice clouds with complex crystal geometries affecting scattering properties. See Macke
et al. (1996) for a seminal analysis of ice crystal scattering.

5.8. SUMMARY OF EARTH’S RADIATION BUDGET 121
Figure 5.39: Radiative budget from Kiehl and Trenberth(2008)
Figure 5.40: top of atmosphere (TOA) SW radiation budget
• But with low values in some tropical regions (why?)
• highest values over the deserts (why?)
If the net (SW+IR) effect is shown (Fig. ??) we see
• Surplus of incoming energy in tropics, deficit at poles.
• There is also a deficit over some desert regions
The effects of clouds can be isolated by calculating the cloud radiative forcing in the SW, IR
and net (SW+IR) (Fig. 5.43). These are calculated at each location by comparing the mean TOA
outgoing radiative flux at all times with the mean averaged for only those times when clouds are
present. The difference between these two values gives a measure of the cloud radiative forcing.
The TOA radiative forcing of clouds is shown in Fig. 5.43, and it should be possible to identify
which panel is for SW and which for LW.
Calculating heating rates
To calculate the atmospheric heating rate a range of models are used that divide the spectrum up
into a number of bands, with the spectrum resolution depending on available computing time.
• Line by Line: The most complicated models will divide the spectrum of 0-2500cm−1 into
well over 100,000 spectral intervals. These models are very expensive to run but provide the
benchmarks by which to test the cheaper models below.
Figure 5.41: TOA OLR budget from ERBE
Figure 5.42: Net TOA radiation budget
• Narrow Band: These models will divide the infra-red spectrum into ∼ 250 spectral intervals,
but these models are still too expensive to use in weather prediction or climate models.
• Broad Band: These may employ 5 to 20 spectral intervals, although numerical tricks can be
used to improve representation of gas absorption bands.
5.9 Climate change

Climate change is often view as synonymous with human induced climate change. But climates
have always been changing in the past. Around 2.6 million years ago climate variability increased,
with marked cold periods characterized by extensive coverage of ice sheets, known as ice ages
(Fig. 5.44). The key cyclicity was 41Kyr until circa 1 million years ago, when the 100kyr period
dominated.
Many factors are involved in the temperature variations. One factor is the movement of the
earth land masses. It is thought that when land masses are found predominately at high latitudes,
glaciation is more readily favoured (Fig. 5.45).
Biogeochemical cycles though play a strong role in feedbacks through the radiation budget.
The temperature is strongly correlated with fluctuations in CO2 and methane. The CO2 leads
the temperature record by 800 years or so. CO2 is a positive feedback in the climate system (Fig.
5.46).
5.9. CLIMATE CHANGE 123
Figure 5.43: Cloud radiative forcing calculated from ERBE in the (top) LW (middle) SW and
lower net (LW+SW)
Figure 5.44: Temperature record (source Nasa.gov)
Variations in Earth’s orbit (Milankovitch cycles)

There are variations in the Earth’s orbit parameters, that affect the eccentricity of the earth orbit,
the Axial tilt (obliquity) and Precession (wobble).
The variations in the solar insolation can strongly modulate climate (Fig. 5.48).
These variations are not thought to be the only drivers of ice ages as the forcing variability
occurs at a higher frequency than the major glaciations (Fig. 5.49), again emphasizing the key
role of biogeophysical cycles which alter the temperature response.
Recent reconstructions emphasize the present period of warming (Fig. 5.50).
Anthropogenic global warming
From the above discussion it is clear that increasing the concentrations of radiatively active gases
that absorb in the infra-red will tend to warm the surface. Thus burning fossil fuels increases
concentrations of CO2 which to date are estimated to have had an equivalent TOA radiative
forcing of around 1.5 W m−2 (Fig. 5.51).
Likewise, humans can have other climate impacts such as changing surface albedo through land-
use changes or changing atmospheric albedo through the emission of aerosols that reflect sunlight
(Fig. 5.52). Indeed it is the later effect that is now thought to be partly the cause of the cooling in
northern latitudes during the 1950s and 1960s, although natural decadal variability due to ocean
circulation is also key (and the reason given for the recent hiatus in warming from 2000-2014).
Fig. 5.53 gives the estimate of aerosol forcing of the climate along with other key anthropogenic
forcing. Note the large uncertainty of this and some other terms. The calculation of the projected
man-induced warming in future is greatly complicated by the existence of natural feedbacks, that
may amplify or reduce the eventual warming. Two of the major positive feedbacks are:
Figure 5.45: Land masses (source Met Today)
Figure 5.46: Land masses (source Met Today)
• Ice Albedo Feedback: Increased temperatures → ice melting at poles → lower albedo →
increased solar absorption → increased temperatures
• Water Vapour Feedback: Increased temperatures → more water vapour → enhanced green-
house effect → increased temperatures
• cloud feedback
One of the major uncertainties though is the future role played by clouds. This is complicated
as clouds could get thicker, thinner, higher, change phases (ice,liquid), change radiative properties,
etc. Global climate models do not agree on the role of clouds and many models show clouds
amplifying the future warming, while other show clouds offsetting future possible warming.
Note: Just because clouds tend to cool the present-day climate does not imply that they will
offset future warming. It is interesting to note that the uncertainty due to clouds has not reduced
with increasing knowledge of cloud physics over the past three decades! The uncertainty is
emphasized in this intercomparison of CMIP3 (IPCC 4th assessment report) models in Fig. 5.54.
Despite these uncertainties, estimates of future warming are made and must also take into account
uncertainties in emission scenarios. Projections from IPCC AR4 and AR5 are shown in Fig. 5.55.
Figure 5.47: Schematic (Left) Axial tilt (obliquity), (middle)Precession and (right) eccentricity
(source Wikipedia)
Figure 5.48: Solar flux variation as a result of earth orbit variations
In terms of the evolution, the system is fairly linear, Fig. 5.56:

The courses in the second term will aim to go into to more details concerns the components of
earth system models for NWP and climate and highlight so of the reasons for the uncertainty in
the modelling of these important processes.
Figure 5.49: Solar flux variations as a result of earth orbit variations and the associated recon-
structed temperature records (source Wikipedia)
Figure 5.50: The past 1000 years temperature on the northern hemisphere and the modern record
(source Nasa)
Figure 5.51: Greenhouse gas emissions and their equivalent top of atmosphere (TOA) radiative
forcing (source IPCC AR4)
Figure 5.52: NOAA land and ocean global mean temperature 1880 to 2013, with a 5 year running
mean applied. (source NOAA climate monitor: http://www.ncdc.noaa.gov/cag/time-series/global
)
Figure 5.53: Human impacts and their assessed radiative forcing equivalent. (source IPCC AR4)
Figure 5.54: (a) net and (b) SW/LW cloud feedback in the GCMs that contributed to the 4th
assessment IPCC report, from Cess et al. JGR 1996
Figure 5.55: Future projected warming from many different climate models under different emission
scenarios from the 3rd and 5th CMIP intercomparison exercise.
Figure 5.56: Future projected warming from many different climate models relative to the CO2
emission for all four RCPs from AR5.
Chapter 6
Exercises
6.1 Thermodynamics Exercises

1. A parcel of air at 800 hPa has a temperature of 280K is surrounded by air with a temperature
of 278K and is lifted to a pressure of 700 hPa where the air temperature is 272K. Is the parcel
warmer or cooler than its environmental when arriving at 700 hPa? What have you assumed
to make this calculation?
2. Use the equation of state for dry air to calculate the air density at the summit of Mt Everest on
an occasion where the pressure and temperature were 313 hPa and -38.5o C there, respectively.
3. In the subtropical free (above the boundary layer) troposphere there is an approximate
balance between radiative cooling and dry adiabatic motion of the air. Assuming a constant
net radiative diabatic cooling of Qrad =1.5 K day−1 , what is the velocity w of the air to
maintain temperature equilibrium? Is the motion upwards or downwards? How long would
it take an air parcel to move through the depth of the troposphere (assume a depth of 10km)
In general, let’s assume that we can neglect the impact of radiation if the heating/cooling
associated with vertical motion of an air parcel is at least two orders of magnitude (hundred
times) greater. What would be the minimum vertical velocity of an air parcel for this to be
the case? Do you know how this compares to typical rates of ascent in cumulus clouds?
4. Starting from eqn. 36, derive eqn. 46 for the definition of entropy.
5. Find the density of water vapour and dry air at STP (1000hPa and 0C). Which density is
lower? At a given pressure, will a moist parcel of air with unit volume therefore have a lower
or higher density than a unit volume of dry air?
6. Find the density of moist air with a specific humidity of 15 g kg−1 , temperature 15o C and
pressure 1010 hPa, and compare it with the density of dry air at the same temperature and
pressure. Find the temperature at which dry air would have the same density as the moist
air at the same total pressure. This temperature was introduced as the virtual temperature
Tv
7. A parcel of air passes through a deep convective cloud and reaches the LNB in the upper
troposphere. We assume it is dry (rv = 0) and θ=350K. Since the parcel arrives at the LNB
with upward momentum it overshoots the LNB. The environmental profile is stable with a
potential temperature (θenv ) lapse rate that is constant in height at dθdz
env
= 5 × 10−3 Km−1 ,
so that the parcel will return to the LNB and oscillate. Neglecting virtual temperature effect,
θ−θenv
the parcel acceleration is related to the buoyancy as Dw Dt = g θenv . Assuming that all the
humidity has condensed and fallen out of the parcel as precipitation, and thus its motion is
dry adiabatic, derive the period of the parcel oscillation.
8. (3pt) An air parcel of moist (but non-cloudy) air has a temperature of 24.9o C. It is neutrally
buoyant sitting in an environment that has a temperature of 25.0 o C and a mixing ratio of
10 g kg−1 . What is the mixing ratio in the air parcel?
129
130 CHAPTER 6. EXERCISES
9. The environmental air with temperature of 25.0 o C and a mixing ratio of 10 g kg−1 is located
near the surface at a pressure of 1000hPa. What is the relative humidity (assume RH = rrvs .)
?
The airstarts to cool isobarically at nightnight until eventually a fog forms. Assuming the
pressure is 1000hPa. Which is the temperature at which the fog occurs? (Use the tephigram
to answer this)
What is the common name given to this temperature?
10. name of all the following quantities that are conserved in pseudoadiabatic moist saturated
ascent: mixing ratio rv , pressure p, wet bulb potential temperature θw , (dry bulb) temper-
ature T , potential temperature θ, dew point temperature Td , vapour pressure e, wet bulb
temperature Tw , adiabatic equivalent temperature Te , equivalent potential temperature θe ?
11. Explain in words why you can sometimes see your breath in winter when there are cold,
damp conditions?
Answers
1. You need eqn. 35 to give you the parcel’s final temperature. It is not necessary to calculate
Rc d
θ, simply take the reference level at the parcel origin of 800 hPa, and then T = T0 pp0
p
,
which is 269.5 K. Thus the parcel is cooler than its surroundings on arrival.
2. ans: 0.465 kg m−3
3. The rate of change of temperature due to vertical motion of an air parcel is w dT dz , where
the lapse rate is that of the parcel and is given by the adiabatic lapse rate (eqn. 32).
Thus, if the vertical motion balances radiative cooling rate given −w cgp = Qrad , this gives
w = 1.7 × 10−3 ms−1 . The motion is downwards (watch the signs!) and to move through
10000 metres would take 68 days, around 2 months! In order to neglect radiation the vertical
velocity would have a velocity of 0.14 ms−1 , which is less that observed in shallow convection
( 1ms−1 ) or deep convection thunderstorms ( 10ms−1 ).
4. To start, we take the log of eqn. 46

Rd
lnθ = ln(T pp0 cp
)
Rd
lnθ = lnT + cp (lnp0 − lnp)
then differentiate
Rd dT Rd dp
dlnθ = dlnT − cp dlnp = T − cp p
thus using pv = Rd T we get

cp dT −vdp dq
cp dlnθ = T = T = dφ
5. Using the different gas constants for water vapour and dry air and the gas law, ρv =0.79 kg
m−3 and ρd =1.28 kg m−3 . Thus a moist parcel will be less dense than the equivalent unit
volume of dry air at STP!
6. Tv =290.7K cf T=288.2K
7. We write the momentum equation in terms of the parcel displacement dz about the LNB:
D2 z θ−θenv
Dt2 = g θenv . As θ is conserved (dry adiabatic) then θ = θenv at the LNB. We can write
dθ dθ
θenv (δz) = θenv (0) + dz )δz. Thus θ − θenv = θenv (0) − θenv = − dz δz.
D2
This gives Dt2 = − θenvg (0) dz
dθ
dz = −N 2 dz
where N = θenvg (0) dz

dθ
is usually referred to as the buoyancy frequency. We look for solutions
of the form dz = Aexp(iN t), which gives a period of 2π/N . Thus the period is 531 seconds.
6.2. EXERCISES: DEEP CONVECTION 131
8. The virtual temperature has to be the same

Tv−parcel = Tv−env = Tenv (1 + Krenv )
1−
where K = = 0.608
(Tenv /Tparcel )(1+Krenv )−1
Thus rparcel = / K
plugging in the numbers we find rv =0.01056 or 10.56 g kg−1
9. You need to insert p and T into Teton’s formula to get rs =19.8 g/kg. thus rv /rs =50.4% or
0.54.
The fog forms at approximately 14C and it is known as the wet bulb temperature.
10. θw and θe
11. Due to the nonlinearity of the saturation vapour pressure curve and the linearity of temper-
ature and mixing ratio mixing (recall diagram).
6.2 Exercises: Deep Convection

Tephigram Exercise I Using the correct tephigram shown in Fig. (40) plot the sounding data
given in table 6.1.
p(hP a) T (◦ C) r (g kg−1 )
1000 10.0 5.0
950 6.5 5.0
900 3.1 5.0
850 2.5 3.2
800 0.0 2.8
700 -5.1 1.4
600 -12.0 1.3
500 -17.6 0.9
Table 6.1: An arbitrary atmospheric sounding
1. How deep is the planetary boundary layer (PBL) in hPa?
2. Is the PBL neutral, stable or unstable ?
3. What is the RH at 1000 hPa and 900 hPa?
4. What is Td at 1000 hPa?
5. Is the PBL “well mixed” in humidity?
6. What is the potential temperature θ of an air parcel at p=800hPa?
7. If this air parcel undergoes adiabatic ascent until it reaches a pressures of p=700hPa. What
is its new T and θ?
Tephigram Exercise II
Return to the sounding you plotted using the data in Table 6.1.
1. For a parcel of air at 1000 hPa, what is the condensation temperature Tc and pressure ?
2. For a parcel of air at 1000 hPa, what is the dry bulb or dew point temperature Td ?
3. If rainfall is evaporated into this parcel, at what temperature will it become saturated and
what is this temperature called?
4. What will be the parcel’s mixing ratio at this point?
5. What is the parcel’s equivalent potential energy θe ?
6. What is θe for a parcel at 900 and 700 hPa?

7. Is θe increasing or decreasing with height?
Tephigram Exercises III

Return to the sounding you plotted using the data in Table Tab:sounding1.
1. What is the LCL of a surface parcel of air?
2. What is the LFC?
3. What is the convective trigger temperature?
4. Do you think deep moist convection is likely to occur here?
Tephigram Exercises IV
A 50hPa thick layer of air residing between has the values T = 0◦ C and rv = 4.8 g kg−1 at its
lower boundary at 700 hPa, and T = −2.5◦ C and rv = 3.0 g kg−1 at the upper boundary at 650
hPa.
1. Is the layer absolutely stable/unstable, or conditionally unstable?
2. The layer is lifted by 100 hPa such that the lower boundary is at 600 hPa, what are the new
temperatures at its boundaries?
3. Will the lifted layer be unstable to overturning?
4. Can you think of a criterion for this layer instability ?
A parcel lifted from the boundary layer forms deep convection since it has negligibly small CIN,
and between the LCL at 900 hPa and LNB at 200 the cloudy has an average temperature excess of
1K. Assuming that the temperature excess does not vary much throughout the depth of the cloud,
estimate the CAP E.
(Hint, recall that CAPE is given by
LN B
Tu − Tenv
Z
CAP Ei = g dz, (6.1)
i Tenv
but you may find it useful to assume the atmosphere is in hydrostatic balance and rewrite the
definition in terms of pressure). (b) Apart from the constant temperature excess, what are the
approximations made in this calculation (c) Estimate the CAPE from the GATE atmospheric
sounding profile you plotted earlier in the course.
Tephigram Exercises VI
Plot this (slightly doctored) GATE profile on the tephigram.
1. Determine the Lifting condensation level (LCL) for the surface parcel
2. Determine the Trigger temperature for the surface parcel
3. Draw the temperature profile for a moist adiabatically (no mixing) ascending parcel
4. Determine the level of neutral buoyancy (LNB)

6.3. RADIATION EXERCISES 133
An updraught air parcel mixes with the cloud environment in a single entrainment event at
p = 700hP a, such that the mixture is composed of equal amounts of cloudy and environmental
air. Precipitation then passes through the parcel and evaporates, bringing the mixed parcel to
saturation.
1. What is the final temperature of the mixed parcel?
2. Can the parcel form a convective downdraught?
3. Recalculate this for a mixing event between updraught at 600 hPa.
4. If a downdraught can form, we assume it is keep at saturation during its descent by further
precipitation evaporation. What temperature does it have at the surface?
5. Finally compute the CAPE of the surface parcel (you will probably find it easier to use
pressure coordinates for this, assume hydrostatic balance)
6. Explain with a sketch diagram what the LCL, LFC and LNB are, and graphically what
CAPE and CIN are. Explain with a diagram how one graphically determines the wet-bulb
temperature and with which process is it associated? How is the wet bulb temperature
related to the equivalent potential temperature?
7. Explain what the criteria are for absolute convective stability and instability, and also con-
ditional instability.
6.3 Radiation Exercises

1. Rederive the energy balance equations with a slight modification to the assumptions. In the
IR part of the spectrum, the atmosphere is still a grey body with emissivity of IR , but now we
assume that the atmosphere is also a grey body in solar part of the spectrum with SW = 0.1.
In order to derive the equilibrium temperature Ts and Ta we need one further simplifying
assumption concerning the fraction β of the total emitted energy by the atmosphere that lies
in the short-wave part of the spectrum. What would be a good assumption for β do you
think?
2. Extinction in the atmosphere
i A parallel beam radiation of wavelength λ is passing vertically through a layer of 1

km in thickness, containing a homogeneous gas with a density of 0.01 kg m−3 with an
absorption coefficient kλe =0.1 m2 kg−1 and a zero scattering coefficient.
What is the optical thickness of the layer?
ii What is the transmissivity τ of the layer?
iii How thick would the layer have to be to absorb half of the incidence radiation beam?
Answer:
R z2
i optical thickness δλ = z1
kλe ρsecθdz thus 1000 x 0.01 x 0.1 = 1 (no units)
ii transmission = exp(-d) = 0.367
iii exp(-k.rho.dz)=0.5 thus dz=693 m
3. What is special characteristic of the scattering phase function in the Rayleigh scattering
approximation?
4. Can the scattering of solar radiation by clouds be described by the Rayleigh scattering
approximation and why (or why not)?
6.4 Cloud Physics Exercises

1. (a) Calculate the time taken for a cloud droplet of radius 5µm to avaporate completely
in an environment of ambient relative humidity of 90% and when the temperature of the
droplet and environment is 11o C. You may use Bolton’s empirical formula given
in (74)
or
17.67T
the saturation vapour pressure as a function of T in o C: es (T ) = 611.2exp T +243.5 (b)
What are the approximations/assumptions that are made with this calculation? (c) Repeat
for a temperature of -15o C
2. (a) Calculate the time taken for a droplet to grow from 20µm to 500 µm by the collision and
coallescence in a cloud with mean liquid water density of L = 1.2 × 10−3 g m−3 . (b) How
much longer is required to grow the 500 µm to 1000 µm?
3. In two clouds 5 × 10−4 kgm−3 water condenses. In the first cloud the CCN concentration is
100 cm−3 and in the second the concentration is 1000 cm−3 . (a) Assuming the liquid water
is equally distributed among the CCN, what is the droplet size in each cloud? (b) What
might the different concentrations of CCN indicate about the locations of the clouds? (c)
What are the consequences for the subsequent development of the clouds?
4. Explain why we never observe significant (more than 1%) supersaturations with respect to
liquid water, but supersaturations with respect to ice are common
5. What is the Bergeron-Findeison Effect, and explain how it leads to accelerated ice crystals
growth rates than might otherwise be expected.
6. Which ice nucleation mechanism leads to the highest ice crystal number concentrations,
homogeneous or heterogeneous, and why?
Answers:
1. (a) Use 4.26, setting S=0.9 to get 0.56 s (b) The approximations are numerous: neglect
of latent heat diffusion away from droplet, that the ambient relative humidity is fixed and
unaffected by droplet evaporation, the curvature and solute effects when radius falls below 1
micron.
2. (a) Use 4.28 and substitute the relevant forms for the terminal fall speeds to find that 462s
are required to grow the drop to 30 microns (where Vt = X1 r2 ), and then a further 1170s
are required to grow this 30µm droplet to 500 µm (where Vt = X2 r). (b) 289 seconds
3. The mass of N droplets is N ρL 43 πr3 . This is equal to the liquid water water content of the
cloud in kg m−3 , thus (a) r=12.8 µm and (b) 5.96 µm (c) in the first case the distribution
is likely to have drops greater than 20 µm for coalescence to take place, so the cloud may
produce rainfall.
Index
Absorption terminal fall velocity, 82

by atmospheric gases, 118 Dry static energy, 13
heating rate, 127
mass absorption coefficient, 116 Effective Emitting temperature, 113
spectrum, 120 Electro-magnetic wave, 105
Adiabatic Electromagnetic spectrum, 105
definition, 12 Enthalpy, 9
dry lapse rate, 12, 13 Entrainment zone, 47
motion, 13 Entropy, 14
observations, 12 Equation of state, 5
process, 12 Evaporation, 25
Adiabatic:super, 47
Advection, 11 First law of thermodynamics, 7, 9
Aerosols
Gibbs Energy, 26
as CCN, 77
Gibbs Function, 26
Air parcel theory, 11
Greenhouse effect, 123
Anthropogenic global warming, 129
Grey bodies, 111
Avoggadro’s hypothesis, 5
Haze particles, 79
Beer’s Law, 116 Heat Capacity, 8
Black body at constant pressure, 8
definition, 106 at constant volume, 8
Boltzmann constant, 109 Heterogeneous nucleation
Boundary layer, 18, 44 liquid drops, 77
Boundary layer, 45 Homogeneous nucleation
Bowen ratio, 46 liquid drops, 77
Buoyancy, 15 Hydrostatic balance, 10, 12, 13
Cloud Ice
particle modes, 71 molecular structure, 88
processes, 71 Ice ages, 128
Coldpool, 64 Ideal gas, 8
Condensation, 25 ideal gas, 5
Conduction, 11 Intensive and extensive variables, 5
Conserved quantities, 13 Inverse square law, 112
Convection, 11, 17, 42 Irradiance, 106, 108
convective available potential energy, 54 Isentropic process, 14
inhibition energy, 54 Isobaric process, 9
level of free convection, 55 Isotropic radiation, 108
non precipitating, 49
Rayleigh Bénard, 17, 43 Köhler curve, 79
trigger temperature, 58 Kirchoff’s Law, 111
wake, 64 Kirchoff’s law, 111
Dalton’s Law, 6 Laminar layer, 47

Diffusion, 79 Latent Heat
Droplet of Fusion, 27
Activation, 79 of Vaporization, 26
growth by coalescence, 85 Latent heating, 12
growth by diffusion, 80, 81 Level of neutral buoyancy, 54
135
136 INDEX
Line broadening mechanisms, 120 Temperature

definition, 5
Mass Extinction Coefficient, 116 inversion, 47
Mesoscale convective systems, 66 Tephigram, 14, 30
Milankovitch cycles, 129
Mixed layer, 47 Vapour pressure, 25
Virtual temperature, 30
Nucleation, 74
homogeneous from liquid to ice, 90 Water vapour
IR absorption, 121
Optical molecular structure, 23
depth, 117 Wein’s displacement law, 111
thickness, 117
Orbit
Eccentricity, 114
Partial pressure, 6
Pathlength, 116
Planck constant, 109
Planck function, 109
Planetary albedo, 112
Poisson’s Equation, 13
Potential temperature, 13
Radiance, 107
Radiant Emittance, 106
Radiation, 12
Radiative transfer, 115
Radiative balance models, 112
Relative humidity, 29
Saturation
over curved surface, 75
over ice, 89
solution term, 78
Bolton’s formula, 27
Clausius Clapeyon Equation, 26
definition, 25
saturation mixing ratio, 29
vapour pressure, 26
Scattering, 108, 115, 116, 121, 122, 124–126
clouds, 125
mass scattering coefficient, 116
phase function, 118
Rayleigh, 121, 122
Solar constant, 112
Solar declination angle, 114
Solar Zenith Angle, 115
Solid angle, 107
specific gas constant, 6
Specific humidity, 29
saturation, 29
Specific volume, 6
Stability, 18, 44
Stephan-Boltzmann law, 111
Sun-Earth geometry, 114
Surface heat balance, 46
Surface layer, 47
Surface tension, 73
Bibliography
Banta, R. M., and C. Barker Schaaf, 1987: Thunderstorm genesis zones in the Colorado Rocky
Mountains as determined by traceback of geosynchronous satellite images. Mon. Wea. Rev.,
115 (2), 463–476.
Berry, E. X., 1967: Cloud droplet growth by collection. J. Atmos. Sci., 24, 688–701.
Berry, E. X., and R. L. Reinhardt, 1974: An analysis of cloud drop growth by collection Part II.
Single initial distributions. J. Atmos. Sci., 31, 1825–1831.
Bretherton, C. S., P. N. Blossey, and M. Khairoutdinov, 2005: An Energy-Balance Analysis of
Deep Convective Self-Aggregation above Uniform SST. J. Atmos. Sci., 62, 4273–4292.
Byers, H. R., and R. R. Braham Jr., 1948: Thunderstorm structure and circulation. J. Meteorology,
5, 71–86.
Demott, P. J., D. J. Cziczo, A. J. Prenni, D. M. Murphy, S. M. Kreidenweis, and D. S. Thomson,
2003: Measurements of the concentrations and composition of nuclei for cirrus formation. Proc.
Nat. Acad. Sci., 100, 14 655–14 660.
Emanuel, K. A., 1994: Atmospheric Convection. Oxford University Press, 580 pp.
Fischer, E. M., S. I. Seneviratne, P. L. Vidale, D. Lüthi, and C. Schär, 2007: Soil Moisture–
Atmosphere Interactions during the 2003 European Summer Heat Wave. J. Climate, 20, 5081–
5099.
Fleishauer, R. P., V. E. Larson, and T. H. Vonder Haar, 2002: Observed microphysical structure
of midlevel, mixed-phase clouds. J. Atmos. Sci., 59, 1779–1804.
Fletcher, N. H., 1962: The Physics of Rainclouds. Cambridge Univ. Press, New York, 386pp.
Gierens, K., 2003: On the transition between heterogeneous and homogeneous freezing. Atmos.
Chem. Phys., 3, 437–446.
Grabowski, W. W., and M. W. Moncrieff, 2002: Large-scale organization of tropical convection
in two-dimensional explicit numerical simulations: Effects of interactive radiation. Q. J. R.
Meteorol. Soc., 128, 2349–2375.
Grabowski, W. W., and M. W. Moncrieff, 2004: Moisture-convection feedback in the Tropics. Q.
J. R. Meteorol. Soc., 130, 3081–3104.
Grabowski, W. W., J.-I. Yano, and M. W. Moncrieff, 2000: Cloud resolving modeling of tropical
circulations driven by large-scale SST gradients. J. Atmos. Sci., 57, 2022–2040.
Gray, W. M., and R. W. Jacobson Jr, 1977: Diurnal-variation of deep cumulus convection. Mon.
Wea. Rev., 105, 1171–1188.
Gunn, K. L. S., and J. S. Marshall, 1958: The distribution with size of aggregate snowflakes. J.
Atmos. Sci., 15, 452–461.
Hartmann, D., L. Moy, and Q. Fu, 2001: Tropical Convection and the Energy Balance at the Top
of the Atmosphere. J. Climate, 14, 4495–4511.
Held, I. M., R. S. Hemler, and V. Ramaswamy, 1993: Radiative-convective equilibrium with explicit
two-dimensional moist convection. J. Atmos. Sci., 50, 3909–3927.
137
138 BIBLIOGRAPHY
Hoose, C., and O. Möhler, 2012: Heterogeneous ice nucleation on atmospheric aerosols: a review of
results from laboratory experiments. Atmospheric Chemistry and Physics, 12 (20), 9817–9854.
Johnson, R. H., T. M. Rickenbach, S. A. Rutledge, P. E. Ciesielski, and W. H. Shubert, 1999:

Trimodal characteristics of tropical convection. J. Climate, 12, 2397–2418.
Kashchiev, D., 2000: Nucleation. Butterworth-Heinemann.
Klett, J. D., and M. H. Davis, 1973: J. Atmos. Sci., 30, 107–117.
Koop, T., B. P. Luo, A. Tsias, and T. Peter, 2000: Water activity as the determinant for homoge-
neous ice nucleation in aqueous solutions. Nature, 406, 611–614.
Macke, A., J. Mueller, and E. Raschke, 1996: Single scattering properties of atmospheric ice
crystals. J. Atmos. Sci., 53, 2813–2825.
Manabe, S., and R. F. Strickler, 1964: Thermal equilibrium of the atmosphere with a convective
adjustment. J. Atmos. Sci., 21, 361–385.
Marshall, J. S., and W. M. K. Palmer, 1948: The distribution of raindrops with size. J. Atmos.
Sci., 5, 165–166.
Mason, B. J., 1978: Physics of a raindrop. Physics Education, 13, 414–419.
McIlveen, R., 1991: Fundamentals of weather and climate. Psychology Press.
Noilhan, J., and S. Planton, 1989: A simple parametrization of land surface processes for meteo-
rological models. Mon. Wea. Rev., 117 (3), 536–549.
Paluch, I. R., 1979: The Entrainment Mechanism in Colorado Cumuli. J. Atmos. Sci., 36, 2467–
2478.
Pruppacher, H. R., and J. D. Klett, 1997: The Microphysics of Clouds and Precipitation. Kluwer
Academic Publishers, pp. 954.
Quante, M., 2004: The role of clouds in the climate system. Journal de Physique IV (Proceedings),
EDP sciences, Vol. 121, 61–86.
Ren, C., and A. R. Mackenzie, 2005: Cirrus parametrization and the role of ice nuclei. Q. J. R.
Meteorol. Soc., 131, 1585–1605.
Rogers, R. R., and M. K. Yau, 1989: A short course in cloud physics. Pergamon Press, 290pp pp.
Rotunno, R., J. B. Klemp, and M. L. Weisman, 1988: A theory for long-lived squall lines. J.
Atmos. Sci., 45, 463–485.
Seigel, R. B., and S. C. van den Heever, 2012: Dust lofting and ingestion by supercell storms. J.
Atmos. Sci., 69 (5), 1453–1473.
Spichtinger, P., K. Gierens, and W. Read, 2003: The global distribution of ice-supersaturated
regions as seen by the Microwave Limb Sounder. Q. J. R. Meteorol. Soc., 129, 3391–3410.
Taylor, C. M., F. Said, and T. Lebel, 1997: Interactions between the land surface and mesoscale
rainfall variability during HAPEX-Sahel. Mon. Wea. Rev., 125, 2211–2227.
Telford, J. W., and J. Warner, 1962: On the measurement from an aircraft of buoyancy and vertical
air velocity in cloud. J. Atmos. Sci., 19, 415–423.
Tompkins, A. M., 2001a: On the relationship between tropical convection and sea surface temper-
ature. J. Climate, 14, 633–637.
Tompkins, A. M., 2001b: Organization of tropical convection in low vertical wind shears: The role
of coldpools. J. Atmos. Sci., 58, 1650–1672.
Tompkins, A. M., 2001c: Organization of tropical convection in low vertical wind shears: The role
of water vapor. J. Atmos. Sci., 58, 529–545.
BIBLIOGRAPHY 139
Tompkins, A. M., and G. C. Craig, 1998: Radiative-convective equilibrium in a three-dimensional

cloud ensemble model. Q. J. R. Meteorol. Soc., 124, 2073–2097.
Turner, J., 1963: The motion of buoyant elements in turbulent surroundings. Journal of Fluid
Mechanics, 16 (01), 1–16.
Warner, J., 1955: The water content of cumuliform cloud. Tellus, 7, 449–457.
WARNER, J., and T. D. Newnham, 1952: new method of measurement of cloud water content.
Q. J. R. Meteorol. Soc., 78, 46–52.
Willis, P. T., 1984: Functional fits to some observed drop size distributions and parameterization
of rain. J. Atmos. Sci., 41, 1648–1661.

Thompkins Atmos Physics Handout

Uploaded by

Copyright:

Available Formats

Thompkins Atmos Physics Handout

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Thompkins Atmos Physics Handout

Uploaded by

Copyright:

Available Formats

1

Adrian Tompkins, ICTP, Trieste, Italy

4.7 Ice saturation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

1.1 Kinetic theory of heat

1.2 Equation of state: The ideal gas law

In general, it is useful to formulate the thermodynamic relations in terms of intensive variables.

1. What is the definition of a Ideal Gas?

2. Can dry air be considered an ideal gas?

3. What is Dalton’s law?

Gas Molecular Molar Mass Specific (Mi /Mtot )Ri

1.3 The first law of thermodynamics

first law of thermodynamics

1. Heat is a form of energy that can be transferred

and for a unit mass of gas:

The rate of work (visualize a piston, as in Fig. 10) is given by

such that (8) can be expressed as

1.4 Rules for differentiating

p = f1 (v, T ) , v = f2 (p, T ) or T = f3 (p, v). (1.12)

In other words, there are only two independent state variables,

∂f1 ∂f1 ∂f2 ∂f2 ∂f3 ∂f3

1.5 Enthalpy and specific heat

dq = cv dT + RdT − vdp, (1.22)

dq = (cv + R)dT − vdp, (1.23)

dq = du + pdv = dh − d(pv) + pdv = dh − vdp. (1.27)

The enthalpy gives a measure of total potential energy of the atmosphere.

1.6 Hydrostatic balance

(p(z) − p(z + dz))A = ρAdzg (1.28)

which can be integrated to give

1.7 Adiabatic Processes

Convection (and advection)

Thus if the pressure changes, a corresponding temperature change will result.

1.8 Potential Temperature

and using pv = Rd T we get

1.10 Thermodynamic charts

Figure 1.9: The tephigram basic structure

Figure 1.10: An orthogonal chart

Figure 1.11: A non-orthogonal chart

Figure 1.13: Reminder tephigram Questions

1.11 Buoyancy force

F = ρenv g∆z∆x∆y − ρgV = gV (ρenv − ρ) (1.50)

Figure 1.14: Reminder tephigram Questions

the fluid it displaces.

1.11.2 Buoyancy of an air parcel

Equation 50 can be rewritten

1.12 Introduction to convection

Figure 1.17: Examples of bénard cells (http://www.catea.gatech.edu)

1.13 Atmospheric Stability

as positively (negatively) buoyant. Equal temperatures are referred to as neutral buoyancy.

Lenticular or lee-wave cloud indicating atmospheric stable

Figure 2.2: The structure of the water vapour molecule(www.chem1.com)

Stephen Lower. The full link is http://www.chem1.com/acad/webtext/virtualtextbook.html with an abbreviated

Figure 2.3: The charge dipole (www.chem1.com)

Equation of state for water vapour

Figure 2.5: Schematic of molecules in moist air (Meteorology today)

The vapour vapour pressure increases as a result, which will be denoted e.

1. es depends on T and T only.

2. Saturation is independent of the pressure of other gases

However, since temperature is also constant we can write

Equating (63) and (64) shows

Gibbs Energy or Gibbs Function

The is defined by replacing e with es . At all normal atmospheric conditions e p, implying

Similarly to Tv , if we assume rl 1 and ri 1 and ignore second order terms we get