. 2
( 16)


15 °C and the temperature that would apply if greenhouse gases were absent.4
This blanketing is known as the natural greenhouse effect and the gases are

Figure 2.2 Schematic of the natural greenhouse effect.

known as greenhouse gases (Figure 2.2). It is called ˜natural™ because all the
atmospheric gases (apart from the chloro¬‚uorocarbons “ CFCs) were there
long before human beings came on the scene. Later on I will mention the
enhanced greenhouse effect: the added effect caused by the gases present in the
atmosphere due to human activities such as deforestation and the burning of
fossil fuels.
The basic science of the greenhouse effect has been known since early in the
nineteenth century (see box) when the similarity between the radiative prop-
erties of the Earth™s atmosphere and of the glass in a greenhouse (Figure 2.3)
was ¬rst pointed out “ hence the name ˜greenhouse effect™. In a greenhouse,
visible radiation from the Sun passes almost unimpeded through the glass and
is absorbed by the plants and the soil inside. The thermal radiation that is emit-
ted by the plants and soil is, however, absorbed by the glass that re-emits some

of it back into the greenhouse. The glass thus
Thermal radiation Radiation
emitted outwards from Sun
acts as a ˜radiation blanket™ helping to keep
by glass
the greenhouse warm.
However, the transfer of radiation is only
one of the ways heat is moved around in a
greenhouse. A more important means of heat
transfer is convection, in which less dense
warm air moves upwards and more dense cold
air moves downwards. A familiar example of
this process is the use of convective electric
heaters in the home, which heat a room by
stimulating convection in it. The situation in
the greenhouse is therefore more complicated
than would be the case if radiation were the
only process of heat transfer.
Mixing and convection are also present in
the atmosphere, although on a much larger
Thermal radiation
inside greenhouse
scale, and in order to achieve a proper under-
Figure 2.3 A greenhouse has a similar effect to the standing of the greenhouse effect, convect-
atmosphere on the incoming solar radiation and the
ive heat transfer processes in the atmosphere
emitted thermal radiation.
must be taken into account as well as radia-
tive ones.
Within the atmosphere itself (at least in the lowest three-quarters or so of
the atmosphere up to a height of about 10 km which is called the troposphere)
convection is, in fact, the dominant process for transferring heat. It acts as fol-
lows. The surface of the Earth is warmed by the sunlight it absorbs. Air close
to the surface is heated and rises because of its lower density. As the air rises it
expands and cools “ just as the air cools as it comes out of the valve of a tyre.
As some air masses rise, other air masses descend, so the air is continually
turning over as different movements balance each other out “ a situation of
convective equilibrium. Temperature in the troposphere falls with height at
a rate determined by these convective processes; the fall with height (called
the lapse rate) turns out on average to be about 6 °C per kilometre of height
(Figure 2.4).
A picture of the transfer of radiation in the atmosphere may be obtained
by looking at the thermal radiation emitted by the Earth and its atmosphere
as observed from instruments on satellites orbiting the Earth (Figure 2.5). At
some wavelengths in the infrared the atmosphere “ in the absence of clouds “
is largely transparent, just as it is in the visible part of the spectrum. If our
eyes were sensitive at these wavelengths we would be able to peer through the

Pioneers of the science of the greenhouse effect5
The warming effect of the greenhouse gases in the atmosphere was ¬rst recognised in 1827 by the French
scientist Jean-Baptiste Fourier, best known for his contributions to mathematics. He also pointed out the
similarity between what happens in the atmosphere and in the glass of a greenhouse, which led to the
name ˜greenhouse effect™. The next step was taken by a British scientist, John Tyndall, who, around 1860,
measured the absorption of infrared radiation by carbon dioxide
and water vapour; he also suggested that a cause of the ice ages
might be a decrease in the greenhouse effect of carbon dioxide.
It was a Swedish chemist, Svante Arrhenius, in 1896, who cal-
culated the effect of an increasing concentration of greenhouse
gases; he estimated that doubling the concentration of carbon
dioxide would increase the global average temperature by 5 to
6 °C, an estimate not too far from our present understanding.6
Nearly 50 years later, around 1940, G. S. Callendar, working
in England, was the ¬rst to calculate the warming due to the
increasing carbon dioxide from the burning of fossil fuels.
The ¬rst expression of concern about the climate change that
might be brought about by increasing greenhouse gases was in
1957, when Roger Revelle and Hans Suess of the Scripps Institute
of Oceanography in California published a paper which pointed
out that in the build-up of carbon dioxide in the atmosphere,
Svante August Arrhenius (19 February
human beings are carrying out a large-scale geophysical experi-
1859 “ 2 October 1927).
ment. In the same year, routine measurements of carbon dioxide
were started from the observatory on Mauna Kea in Hawaii. The rapidly increasing use of fossil fuels since
then, together with growing interest in the environment, has led to the topic of global warming moving up
the political agenda through the 1980s, and eventually to the Climate Convention signed in 1992 “ of which
more in later chapters.

atmosphere to the Sun, stars and Moon above, just as we can in the visible
spectrum. At these wavelengths all the radiation originating from the Earth™s
surface leaves the atmosphere.
At other wavelengths radiation from the surface is strongly absorbed by some
of the gases present in the atmosphere, in particular by water vapour and car-
bon dioxide.
Objects that are good absorbers of radiation are also good emitters of it. A
black surface is both a good absorber and a good emitter, while a highly re¬‚ect-
ing surface absorbs rather little and emits rather little too (which is why highly

Figure 2.4 The distribution of temperature in
Average height at which
a convective atmosphere (red line). The green
the outgoing radiation
line shows how the temperature increases when
originates. The temperature
of the atmosphere increases
the amount of carbon dioxide present in the
10 from the surface up
atmosphere is increased (in the diagram the
Height (km)

difference between the lines is exaggerated
“ for instance, for doubled carbon dioxide
5 in the absence of other effects the increase
in temperature is about 1.2 °C). Also shown
for the two cases are the average levels from
which thermal radiation leaving the atmosphere
originates (about 6 km for the unperturbed
“50 0
Temperature (°C) atmosphere).

Wavelength (cm “1)
1500 1000 500

H2O Window O3 Window CO2 H2O




“ 53

7 8 9 10 15 20
Wavelength ( m)

Figure 2.5 Thermal radiation in the infrared region (the visible part of the spectrum is between about
0.4 and 0.7 μm) emitted from the Earth™s surface and atmosphere as observed over the Mediterranean
Sea from a satellite instrument orbiting above the atmosphere, showing parts of the spectrum where
different gases contribute to the radiation. Between the wavelengths of about 8 and 14 μm, apart from
the ozone band, the atmosphere, in the absence of clouds, is substantially transparent; this is part of the
spectrum called a ˜window™ region. Superimposed on the spectrum are curves of radiation from a black
body at 7 °C, ’13 °C, ’33 °C and “53 °C. The units of radiance are watts per square metre per steradian
per wavenumber.
Ice, oceans, land surfaces and clouds all play a role in determining how much incoming solar radiation the
Earth re¬‚ects back into space.

re¬‚ecting foil is used to cover the surface of a vacuum ¬‚ask and why it is placed
above the insulation in the lofts of houses).
Absorbing gases in the atmosphere absorb some of the radiation emitted by
the Earth™s surface and in turn emit radiation out to space. The amount of ther-
mal radiation they emit is dependent on their temperature.
Radiation is emitted out to space by these gases from levels somewhere
near the top of the atmosphere “ typically from between 5 and 10 km high
(see Figure 2.5). Here, because of the convection processes mentioned earlier,
the temperature is much colder “ 30 to 50 °C or so colder “ than at the sur-
face. Because the gases are cold, they emit correspondingly less radiation. What
these gases have to do, therefore, is absorb some of the radiation emitted by the
Earth™s surface but then to emit much less radiation out to space. They, there-
fore, act as a radiation blanket over the surface (note that the outer surface of
a blanket is colder than inside the blanket) and help to keep it warmer than it
would otherwise be (Figure 2.6).

There needs to be a balance between the radiation coming in and the radi-
ation leaving the top of the atmosphere “ as there was in the very simple
model with which this chapter started. Figure 2.7 shows the various compo-
nents of the radiation entering and leaving the top of the atmosphere for the
real atmosphere situation. On average, 235 watts per square metre of solar
radiation are absorbed by the atmosphere and the
surface; this is less than the 288 watts mentioned at
the beginning of the chapter, because now the effect
Greenhouse gases
of clouds is being taken into account. Clouds re¬‚ect
some of the incident radiation from the Sun back out
to space. However, they also absorb and emit thermal
radiation and have a blanketing effect similar to that of
Earth™s surface
the greenhouse gases. These two effects work in oppos-
Figure 2.6 The blanketing effect of ite senses: one (the re¬‚ection of solar radiation) tends
greenhouse gases.

Reflected solar 235 Outgoing
Incoming solar
107 342
radiation longwave
107 Wm “2 342 Wm “ 2 radiation
235 Wm “ 2

Reflected by clouds,
aerosol and
Emitted by
atmospheric 40
gases Atmospheric
77 window
Emitted by clouds
Absorbed by

78 heat

Reflected by 350 324
168 24 78
Absorbed by Thermals Evapo -
Absorbed by surface
surface transpiration

Figure 2.7 Components of the radiation (in watts per square metre) which on average enter and leave
the Earth™s atmosphere and make up the radiation budget for the atmosphere. About half of the incoming
solar radiation is absorbed by the Earth™s surface. This energy is transferred to the atmosphere by
warming the air in contact with the surface (thermals), by evapotranspiration and by longwave radiation
that is absorbed by clouds and greenhouse gases. The atmosphere in turn radiates longwave energy back
to Earth as well as out to space.

to cool the Earth™s surface and the other (the absorption of thermal radiation)
tends to warm it. Careful consideration of these two effects shows that on
average the net effect of clouds on the total budget of radiation results in a
slight cooling of the Earth™s surface.7
The numbers in Figure 2.7 demonstrate the required balance: 235 watts per
square metre on average coming in and 235 watts per square metre on aver-
age going out. The temperature of the surface and hence of the atmosphere
above adjusts itself to ensure that this balance is maintained. It is interest-
ing to note that the greenhouse effect can only operate if there are colder
temperatures in the higher atmosphere. Without the structure of decreasing
temperature with height, therefore, there would be no greenhouse effect on
the Earth.

Mars and Venus
Similar greenhouse effects also occur on our nearest planetary neighbours, Mars
and Venus. Mars is smaller than the Earth and possesses, by Earth™s standards,
a very thin atmosphere. A barometer on the surface of Mars would record an
atmospheric pressure less than 1% of that on the Earth. Its atmosphere, which
consists almost entirely of carbon dioxide, contributes a small but signi¬cant
greenhouse effect.
The planet Venus, which can often be seen fairly close to the Sun in the morn-
ing or evening sky, has a very different atmosphere to Mars. Venus is about the
same size as the Earth. A barometer for use on Venus would need to survive
very hostile conditions and would need to be able to measure a pressure about
100 times as great as that on the Earth. Within the Venus atmosphere, which
consists very largely of carbon dioxide, deep clouds consisting of droplets of
almost pure sulphuric acid completely cover the planet and prevent most of
the sunlight from reaching the surface. Some Russian space probes that have
landed there have recorded what would be dusk-like conditions on the Earth “
only 1% or 2% of the sunlight present above the clouds penetrates that far. One
might suppose, because of the small amount of solar energy available to keep
the surface warm, that it would be rather cool; on the contrary, measurements
from the same Russian space probes ¬nd a temperature there of about 525 °C “ a
dull red heat, in fact.
The reason for this very high temperature is the greenhouse effect. Because
of the very thick absorbing atmosphere of carbon dioxide, little of the thermal
radiation from the surface can get out. The atmosphere acts as such an effective
radiation blanket that, although there is not much solar energy to warm the
surface, the greenhouse effect amounts to nearly 500 °C.
The planets Mars, Earth and Venus have signi¬cant atmospheres. This diagram shows the approximate
relative sizes of the terrestrial planets.

The ˜runaway™ greenhouse effect
What occurs on Venus is an example of what has been called the ˜runaway™
greenhouse effect. It can be explained by imagining the early history of the
Venus atmosphere, which was formed by the release of gases from the interior
of the planet. To start with it would contain a lot of water vapour, a power-
ful greenhouse gas (Figure 2.8). The greenhouse effect of the water vapour
would cause the temperature at the surface to rise. The increased tempera-
ture would lead to more evaporation of water from the surface, giving more
atmospheric water vapour, a larger greenhouse effect and therefore a further
increased surface temperature. The process would continue until either the
atmosphere became saturated with water vapour or all the available water had
A runaway sequence something like this seems to have occurred on Venus.
Why, we may ask, has it not happened on the Earth, a planet of about the same
size as Venus and, so far as is known, of a similar initial chemical composition?
The reason is that Venus is closer to the Sun than the Earth; the amount of solar
energy per square metre falling on Venus is about twice that falling on the
Earth. The surface of Venus, when there was no atmosphere, would have started
off at a temperature of just over 50 °C (Figure 2.8). Throughout the sequence
described above for Venus, water on the surface would have been continuously
boiling. Because of the high temperature, the atmosphere would never have
become saturated with water vapour. The Earth, however, would have started
at a colder temperature; at each stage of the sequence it would have arrived at

Figure 2.8 The evolution of the 100
atmospheres of the Earth, Mars
and Venus. In this diagram, the
surface temperatures of the three
planets are plotted against the
vapour pressure of water in their

Temperature (°C)
atmospheres as they evolved.
Also on the diagram (dashed)
are the phase lines for water,
dividing the diagram into regions Earth
where vapour, liquid water or ice
are in equilibrium. For Mars and
the Earth the greenhouse effect Ice
is halted when water vapour is
in equilibrium with ice or liquid
water. For Venus no such halting
occurs and the diagram illustrates 0.01 1 100
the ˜runaway™ greenhouse effect. Vapour pressure of water (kPa)

an equilibrium between the surface and an atmosphere saturated with water
vapour. There is no possibility of such runaway greenhouse conditions occurring
on the Earth.

The enhanced greenhouse effect
After our excursion to Mars and Venus, let us return to Earth! The natural green-
house effect is due to the gases water vapour and carbon dioxide present in the
atmosphere in their natural abundances as now on Earth. The amount of water
vapour in our atmosphere depends mostly on the temperature of the surface of
the oceans; most of it originates through evaporation from the ocean surface
and is not in¬‚uenced directly by human activity. Carbon dioxide is different. Its
amount has changed substantially “ by nearly 40% so far “ since the Industrial
Revolution, due to human industry and also because of the removal of forests
(see Chapter 3). Future projections are that, in the absence of controlling factors,
the rate of increase in atmospheric carbon dioxide will accelerate and that its
atmospheric concentration will double from its pre-industrial value within the
next 100 years (Figure 6.2).
This increased amount of carbon dioxide is leading to global warming of the
Earth™s surface because of its enhanced greenhouse effect. Let us imagine, for
instance, that the amount of carbon dioxide in the atmosphere suddenly dou-
bled, everything else remaining the same (Figure 2.9). What would happen

to the numbers in the radi-
(a) (b) (c) (d)
ation budget presented earlier
(Figure 2.7). The solar radiation
240 240 240 236 240 240 240 240
budget would not be affected.
The greater amount of carbon
dioxide in the atmosphere
top of atmosphere
means that the thermal radi-
CO2 x 2
ation emitted from it will
CO2 x 2 CO2 x 2 + feedbacks
originate on average from a
higher and colder level than
Ts =15 °C Ts =15+1.2 °C Ts =15+3 °C
Ts =15°C before (Figure 2.4). The thermal
radiation budget will there-
Earth™s surface
fore be reduced, the amount
Figure 2.9 The enhanced greenhouse gas effect. Under natural
of reduction being about 4
conditions (a) the net solar radiation coming in (S = 240 watts per
watts per square metre (a more
square metre) is balanced by thermal radiation (L) leaving the top of
precise value is 3.7).
the atmosphere; average surface temperature (Ts) is 15°C. If the carbon
This causes a net imbalance
dioxide concentration is suddenly doubled (b), L is decreased by 4 watts
per square metre. Balance is restored if nothing else changes (c) apart in the overall budget of 4 watts
from the temperature of the surface and lower atmosphere, which per square metre. More energy
rises by 1.2 °C. If feedbacks are also taken into account (d), the average
is coming in than going out.
temperature of the surface rises by about 3 °C.
To restore the balance the sur-
face and lower atmosphere will warm up. If nothing changes apart from the
temperature “ in other words, the clouds, the water vapour, the ice and snow
cover and so on are all the same as before “ the temperature change turns out
to be about 1.2 °C.
In reality, of course, many of these other factors will change, some of them
in ways that add to the warming (these are called positive feedbacks), others
in ways that might reduce the warming (negative feedbacks). The situation is
therefore much more complicated than this simple calculation. These compli-
cations will be considered in more detail in Chapter 5. Suf¬ce it to say here
that the best estimate at the present time of the increased average tempera-
ture of the Earth™s surface if carbon dioxide levels were to be doubled is about
twice that of the simple calculation: 3.0 °C. As the last chapter explained, for
the global average temperature this is a large change. It is this global warming
expected to result from the enhanced greenhouse effect that is the cause of
current concern.
Having dealt with a doubling of the amount of carbon dioxide, it is interest-
ing to ask what would happen if all the carbon dioxide were removed from
the atmosphere. It is sometimes supposed that the outgoing radiation would be

changed by 4 watts per square metre in the other direction and that the Earth
would then cool by one or two degrees Celsius. In fact, that would happen if the
carbon dioxide amount were to be halved. If it were to be removed altogether, the
change in outgoing radiation would be around 25 watts per square metre “ six
times as big “ and the temperature change would be similarly increased. The
reason for this is that with the amount of carbon dioxide currently present in
the atmosphere there is maximum carbon dioxide absorption over much of the
region of the spectrum where it absorbs (Figure 2.5), so that a big change in gas
concentration leads to a relatively small change in the amount of radiation it
absorbs.8 This is like the situation in a pool of water: when it is clear, a small
amount of mud will make it appear muddy, but when it is muddy, adding more
mud only makes a small difference.
An obvious question to ask is: has evidence of the enhanced greenhouse effect
been seen in the recent climatic record? Chapter 4 will look at the record of
temperature on the Earth during the last century or so, during which the Earth
has warmed on average by about three-quarters of a degree Celsius. We shall
see in Chapters 4 and 5 that there are good reasons for attributing most of this
warming to the enhanced greenhouse effect, although because of the size of
natural climate variability the exact amount of that attribution remains subject
to some uncertainty.


• No one doubts the reality of the natural greenhouse effect, which keeps
us over 20 °C warmer than we would otherwise be. The science of it is well
understood; it is similar science that applies to the enhanced greenhouse
• Substantial greenhouse effects occur on our nearest planetary neighbours,
Mars and Venus. Given the conditions that exist on those planets, the sizes
of their greenhouse effects can be calculated, and good agreement has
been found with those measurements that are available.
• Study of climates of the past gives some clues about the greenhouse effect,
as Chapter 4 will show.

First, however, the greenhouse gases themselves must be considered. How
does carbon dioxide get into the atmosphere, and what other gases affect
global warming?

1 Carry out the calculation suggested in Note 4 (refer also to Note 2) to obtain
an equilibrium average temperature for an Earth partially covered with
clouds such that 30% of the incoming solar radiation is re¬‚ected. If clouds
are assumed to cover half the Earth and if the re¬‚ectivity of the clouds
increases by 1% what change will this make in the resulting equilibrium
average temperature?
2 It is sometimes argued that the greenhouse effect of carbon dioxide is
negligible because its absorption band in the infrared is so close to saturation
that there is very little additional absorption of radiation emitted from the
surface. What are the fallacies in this argument?
3 Use the information in Figure 2.5 to estimate approximately the surface
temperature that would result if carbon dioxide were completely removed
from the atmosphere. What is required is that the total energy radiated by
the Earth plus atmosphere should remain the same, i.e. the area under the
radiance curve in Figure 2.5 should be unaltered. On this basis construct a
new curve with the carbon dioxide band absent.9
4 Using information from books or articles on climatology or meteorology
describe why the presence of water vapour in the atmosphere is of such
importance in determining the atmosphere™s circulation.
5 Estimates of regional warming due to increased greenhouse gases are
generally larger over land areas than over ocean areas. What might be the
reasons for this?
6 (For students with a background in physics) What is meant by Local
Thermodynamic Equilibrium (LTE),10 a basic assumption underlying
calculations of radiative transfer in the lower atmosphere appropriate to
discussions of the greenhouse effect? Under what conditions does LTE

Historical overview of climate change science. Chapter 1, in Solomon, S., Qin, D.,
Manning, M., Chen, Z., Marquis, M., Averyt, K. B., Tignor, M., Miller, H. L. (eds.)
2007. Climate Change 2007: The Physical Science Basis. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental Panel on
Climate Change. Cambridge: Cambridge University Press.
Houghton J. 2002. The Physics of Atmospheres, third edition. Cambridge: Cambridge
University Press, Chapters 1 and 14.
N OT E S F O R C H A P T E R 2

the different re¬‚ectivity of ice compared with
1 It is about one-quarter because the area of the Earth™s
the present surface but also ignored the presence
surface is four times the area of the disc which is the
of clouds. Depending on the assumptions made
projection of the Earth facing the Sun; see Figure 2.1.
regarding clouds and other factors, values ranging
2 The radiation by a black body is the Stefan“
Boltzmann constant (5.67 — 10 ’8 J m’2 K’4 s’1) multi- between 20 and 30 °C are quoted for the difference
in surface temperature with and without green-
plied by the fourth power of the body™s absolute
house gases present.
temperature in Kelvin. The absolute temperature
5 Further details can be found in Mudge, F. B. 1997. The
is the temperature in degrees Celsius plus 273
development of greenhouse theory of global climate
(1 K = 1 °C).
change from Victorian times. Weather, 52, 13“16.
3 These calculations using a simple model of an atmos-
6 A range of 2 to 4.5 °C is quoted in Chapter 6,
phere containing nitrogen and oxygen only have
page 143.
been carried out to illustrate the effect of the other
7 More detail of the radiative effects of clouds is given
gases, especially water vapour and carbon dioxide.
in Chapter 5; see Figures 5.14 and 5.15.
It is not, of course, a model that can exist in reality.
8 The dependence of the absorption on the concentra-
All the water vapour could not be removed from the
tion of gas is approximately logarithmic.
atmosphere above a water or ice surface. Further,
9 For some helpful diagrams and more information
with an average surface temperature of ’6 °C, in a
about the infrared spectrum of different green-
real situation the surface would have much more ice
house gases, see Harries, J. E. 1996. The greenhouse
cover. The additional ice would re¬‚ect more solar
Earth: a view from space. Quarterly Journal of the Royal
energy out to space leading to a further lowering of
Meteorological Society, 122, 799“818.
the surface temperature.
10 For information about LTE see, for instance,
4 The calculation I made giving a temperature of ’6 °C
Houghton, J. T. 2002. The Physics of Atmospheres, third
for the average temperature of the Earth™s surface if
edition. Cambridge: Cambridge University Press.
greenhouse gases are not present not only ignored
3 The greenhouse gases

Industrial activity: a source of carbon dioxide and other gaseous and particulate pollution.

T HE GREENHOUSE gases are those gases in the atmosphere which, by absorbing thermal
radiation emitted by the Earth™s surface, have a blanketing effect upon it. The most important
of the greenhouse gases is water vapour, but its amount in the atmosphere is not changing directly
because of human activities. The important greenhouse gases that are directly in¬‚uenced by human
activities are carbon dioxide, methane, nitrous oxide, the chloro¬‚uorocarbons (CFCs) and ozone.
This chapter will describe what is known about the origin of these gases, how their concentration in
the atmosphere is changing and how it is controlled. Also considered will be particles in the
atmosphere of anthropogenic origin, some of which can act to cool the surface.

Which are the most important greenhouse gases?
Figure 2.5 illustrated the regions of the infrared spectrum where the greenhouse
gases absorb. Their importance as greenhouse gases depends both on their con-
centration in the atmosphere (Table 2.1) and on the strength of their absorption
of infrared radiation. Both these quantities differ greatly for various gases.
Carbon dioxide is the most important of the greenhouse gases that are increas-
ing in atmospheric concentration because of human activities. If, for the moment,
we ignore the effects of the CFCs and of changes in ozone, which vary considerably
over the globe and which are therefore more dif¬cult to quantify, the increase in
carbon dioxide (CO2) has contributed about 72% of the enhanced greenhouse effect
to date, methane (CH4) about 21% and nitrous oxide (N2O) about 7% (Figure 3.11).

Radiative forcing
In this chapter we shall use the concept of radiative forcing to compare the rela-
tive greenhouse effects of different atmospheric constituents. It is necessary
therefore ¬rst to de¬ne radiative forcing.
In Chapter 2 we noted that, if the carbon dioxide in the atmosphere were
suddenly doubled, everything else remaining the same, a net radiation imbal-
ance near the top of the atmosphere of 3.7 W m’2 would result. This radiation
imbalance is an example of radiation forcing, which is de¬ned as the change
in average net radiation at the top of the troposphere1 (the lower atmosphere;
for de¬nition see Glossary) which occurs because of a change in the concentra-
tion of a greenhouse gas or because of some other change in the overall climate
system; for instance, a change in the incoming solar radiation would constitute
a radiative forcing. As we saw in the discussion in Chapter 2, over time the cli-
mate responds to restore the radiative balance between incoming and outgoing
radiation. A positive radiative forcing tends on average to warm the surface and
a negative radiative forcing tends on average to cool the surface.

Carbon dioxide and the carbon cycle
Carbon dioxide provides the dominant means through which carbon is trans-
ferred in nature between a number of natural carbon reservoirs “ a process
known as the carbon cycle. We contribute to this cycle every time we breathe.
Using the oxygen we take in from the atmosphere, carbon from our food is
burnt and turned into carbon dioxide that we then exhale; in this way we are
provided with the energy we need to maintain our life. Animals contribute
to atmospheric carbon dioxide in the same way; so do ¬res, rotting wood and
decomposition of organic material in the soil and elsewhere. To offset these

Atmosphere = 760
Accumulation 3.3 ± 0.2

Fossil fuels and Net terrestrial Net ocean
cement production uptake uptake
6.3 ± 0.6 0.7 ± 1.0 2.3 ± 0.8

Global net primary Air/sea
productivity, respiration exchange = 90
and fire = 60

Vegetation = 500
Run-off = 0.8
Soils and detritus =2000
Ocean = 39 000

Fossil organic carbon
Sedimentation = 0.2
carbonate minerals

Figure 3.1 The global carbon cycle, showing the approximate carbon stocks in reservoirs (in Gt) and carbon
¬‚ows (in Gt year “1) relevant to the anthropogenic perturbation as annual averages over the decade from
1989 to 1998. Net ocean uptake of the anthropogenic perturbation equals the net air/sea input plus
run-off minus sediment. The units are thousand millions of tonnes or gigatonnes (Gt). (More detail in
Fig. 7.3 in Chapter 7 of IPCC AR4 WGI 2007.)

processes of respiration whereby carbon is turned into carbon dioxide, there are
processes involving photosynthesis in plants and trees which work the opposite
way; in the presence of light, they take in carbon dioxide, use the carbon for
growth and return the oxygen back to the atmosphere. Both respiration and
photosynthesis also occur in the ocean.
Figure 3.1 is a simple diagram of the way carbon cycles between the vari-
ous reservoirs “ the atmosphere, the oceans (including the ocean biota), the
soil and the land biota (biota is a word that covers all living things “ plants,
trees, animals and so on “ on land and in the ocean, which make up a whole
known as the biosphere). The diagram shows that the movements of carbon
(in the form of carbon dioxide) into and out of the atmosphere are quite large;
about one-¬fth of the total amount in the atmosphere is cycled in and out each

year, part with the land biota and part through physical and chemical processes
across the ocean surface. The land and ocean reservoirs are much larger than
the amount in the atmosphere; small changes in these larger reservoirs could
therefore have a large effect on the atmospheric concentration; the release of
just 2% of the carbon stored in the oceans would double the amount of atmos-
pheric carbon dioxide.
It is important to realise that on the timescales with which we are con-
cerned anthropogenic carbon emitted into the atmosphere as carbon dioxide
is not destroyed but redistributed among the various carbon reservoirs. Carbon
dioxide is therefore different from other greenhouse gases that are destroyed
by chemical action in the atmosphere. The carbon reservoirs exchange car-
bon between themselves on a wide range of timescales determined by their
respective turnover times “ which range from less than a year to decades (for
exchange with the top layers of the ocean and the land biosphere) to millennia
(for exchange with the deep ocean or long-lived soil pools). These timescales
are generally much longer than the average time a particular carbon dioxide
molecule spends in the atmosphere, which is only about four years. The large
range of turnover times means that the time taken for a perturbation in the
atmospheric carbon dioxide concentration to relax back to an equilibrium can-
not be described by a single time constant. About 50% of an increase in atmos-
pheric carbon dioxide will be removed within 30 years, a further 30% within a
few centuries and the remaining 20% may remain in the atmosphere for many
thousands of years.2 Although a lifetime of about 100 years is often quoted for
atmospheric carbon dioxide so as to provide some guide, use of a single lifetime
can be very misleading.
Before human activities became a signi¬cant disturbance, and over periods
short compared with geological timescales, the exchanges between the reser-
voirs were remarkably constant. For thousands of years before the beginning of
industrialisation around 1750, a steady balance was maintained, such that the
mixing ratio (or mole fraction; for de¬nition see Glossary) of carbon dioxide in
the atmosphere as measured from ice cores (see Chapter 4) kept within about 20
parts per million (ppm) of a mean value of about 280 ppm (Figure 3.2a).
The Industrial Revolution disturbed this balance and since its beginning over
600 thousand million tonnes (or gigatonnes, Gt) of carbon have been emitted
into the atmosphere from fossil fuel burning. This has resulted in a concentra-
tion of carbon dioxide in the atmosphere that has increased by about 36%, from
280 ppm around 1700 to a value of over 380 ppm at the present day (Figure 3.2a),
a greater concentration than for at least 650 000 years. Accurate measurements,
which have been made since 1959 from an observatory near the summit of
Mauna Loa in Hawaii, show that from 1995 to 2005 carbon dioxide increased
Didcot power station, near Oxford, UK.

on average each year by about 1.9 ppm (an increase from the average for the
1990s of about 1.5 ppm, although there are large variations from year to year
(Figure 3.2b)). This increase spread through the atmosphere adds about 3.8 Gt to
the atmospheric carbon reservoir each year.
It is easy to establish how much coal, oil and gas is being burnt worldwide
each year. Most of it is to provide energy for human needs: for heating and
domestic appliances, for industry and for transport (considered in detail in
Chapter 11). The burning of these fossil fuels has increased rapidly since the
Industrial Revolution (Table 3.1). Over the 1990s emissions rose about 0.7% per
year; from 1999 to 2005 annual emissions rose systematically from 6.5 to 7.8 Gt
of carbon (an annual increase averaging about 3%), nearly all of which enters
the atmosphere as carbon dioxide. Another contribution to atmospheric carbon
dioxide due to human activities comes from land-use change, in particular from
tropical deforestation balanced in part by afforestation or forest regrowth. This
contribution is not easy to quantify but some estimates are given in Table 3.1. For
the 1990s (see Table 3.1), annual anthropogenic emissions from fossil fuel burn-
ing, cement manufacture (about 3% of the total) and land-use change amounted
to about 8.0 Gt; over three-quarters of these resulted from fossil fuel burning.
Since the annual net increase in the atmosphere was about 3.2 Gt, about 40%
of the 8 Gt of new carbon remained to increase the atmospheric concentration

Figure 3.2 Atmospheric carbon (a)
dioxide concentration. (a) Over 400
the last 10 000 years (inset
since 1750) from various ice
cores (symbols with different
colours for different studies)

Radiative forcing (Wm“2)
Carbon dioxide (ppm)
and atmospheric samples (red 300

lines). Corresponding radiative 1

forcings shown on right-hand 1900
1800 2000
axis. (b) Annual changes in
global mean and their ¬ve-year
means from two different 300
measurement networks (red
and black stepped lines).
The ¬ve-year means smooth
out short-term perturbations
associated with strong El Ni±o
Southern Oscillation (ENSO) 250
events in 1972, 1982, 1987 and
1997. The upper dark green 10 000 5000 0
Time (before 2005)
line shows the annual increases
that would occur if all fossil
fuel emissions stayed in the
atmosphere and there were no
other emissions.

Carbon dioxide annual change (ppm)






1960 1970 1980 1990 2000

Table 3.1 Components of annual average global carbon budget for 1980s and
1990s in Gt of carbon per year (positive values are ¬‚uxes to the atmosphere,
negative values represent uptake from the atmosphere)

1980s 1990s 2000“2005

5.4 ± 0.3 6.4 ± 0.4 7.2 ± 0.3
Emissions (fossil fuel, cement)
3.3 ± 0.1 3.2 ± 0.1 4.1 ± 0.1
Atmospheric increase
“1.8 ± 0.8 “2.2 ± 0.4 “2.2 ± 0.5
Ocean“atmosphere ¬‚ux
“0.3 ± 0.9 “1.0 ± 0.6 “0.9 ± 0.6
Land“atmosphere ¬‚ux*
partitioned as follows
Land-use change 1.4 (0.6 to 2.3) 1.6 (0.5 to 2.7) not available
Residual terrestrial sink “1.7 (“3.4 to 0.2) “2.6 (“4.3 to “0.9) not available

(Figure 3.2b). The other 60% was taken up between the other two reservoirs: the
oceans and the land biota. Figure 3.5 shows that, as global average temperatures
increase the fractions taken up by both land and ocean are likely to reduce.
About 95% of fossil fuel burning occurs in the northern hemisphere, so there
is more carbon dioxide there than in the southern hemisphere. The difference
is currently about 2 ppm (Figure 3.3) and, over the years, has grown in parallel
with fossil fuel emissions, thus adding further compelling evidence that the
atmospheric increase in carbon dioxide levels results from these emissions.
We now turn to what happens in the oceans. We know that carbon dioxide
dissolves in water; carbonated drinks make use of that fact. Carbon dioxide is
continually being exchanged with the air above the ocean across the whole
ocean surface (about 90 Gt per year is so exchanged “ Figure 3.1), particularly
as waves break. An equilibrium is established between the concentration of
carbon dioxide dissolved in the surface waters and the concentration in the air
above the surface. The chemical laws governing this equilibrium are such that
if the atmospheric concentration changes by 10% the concentration in solution
in the water changes by only one-tenth of this: 1%.
This change will occur quite rapidly in the upper waters of the ocean, the
top 100 m or so, thus enabling part of the anthropogenic (i.e. human-generated)
carbon dioxide added to the atmosphere (most of the ocean™s share of the 60%
mentioned above) to be taken up quite rapidly. Absorption in the lower levels in
the ocean takes longer; mixing of surface water with water at lower levels takes
up to several hundred years or for the deep ocean over a thousand years. This
process whereby carbon dioxide is gradually drawn from the atmosphere into
the ocean™s lower levels is sometimes known as the solubility pump.

Figure 3.3 Carbon dioxide 380
concentrations (monthly
averages) observed from Mauna
Loa, Hawaii, 19 °N, green and
from Baring Head, New Zealand,

CO2 mixing ratio (ppm)
41°S, red. Also shown are
measurements of deviations in “100
the O2/N2 ratio from an arbitrary 340
reference multiplied by 106 from

O2 (per meg)
samples from Alert, Canada,
82° N, blue and from Cape Grim,
320 “300
Australia, 41° S, dark blue (after
Manning and Keeling).
1970 1975 1980 1985 1990 1995 2000 2005

So the oceans do not provide as immediate a sink for increased atmospheric
carbon dioxide as might be suggested by the size of the exchanges with the
large ocean reservoir. For short-term changes only the surface layers of water
play a large part in the carbon cycle. Further, it is likely that a warmer regime
will be associated on average with weaker overturning in the ocean and there-
fore with reduced carbon dioxide uptake.
Biological activity in the oceans also plays an important role. It may not be
immediately apparent, but the oceans are literally teeming with life. Although
the total mass of living matter within the oceans is not large, it has a high
rate of turnover. Living material in the oceans is produced at some 30“40%
of the rate of production on land. Most of this production is of plant and ani-
mal plankton which go through a rapid series of life cycles. As they die and
decay some of the carbon they contain is carried downwards into lower levels
of the ocean adding to the carbon content of those levels. Some is carried
to the very deep water or to the ocean bottom where, so far as the carbon
cycle is concerned, it is out of circulation for hundreds or thousands of years.
This process, whose contribution to the carbon cycle is known as the biological
pump (see box), was important in determining the changes of carbon dioxide
concentration in both the atmosphere and the ocean during the ice ages (see
Chapter 4).
Computer models “ which calculate solutions for the mathematical equa-
tions describing a given physical situation, in order to predict its behaviour
(see Chapter 5) “ have been set up to describe in detail the exchanges of carbon
between the atmosphere and different parts of the ocean. To test the validity of
these models, they have also been applied to the dispersal in the ocean of the
carbon isotope 14C that entered the ocean after the nuclear tests of the 1950s;

A large aquamarine-coloured plankton bloom is shown stretching across the length of Ireland in the North
Atlantic Ocean in this image, captured on 6 June 2006 by Envisat™s MERIS satellite, a dedicated ocean
colour sensor able to identify phytoplankton concentrations.

the models simulate this dispersal quite well. From the model results, it is esti-
mated that about 2 Gt (± 0.8 Gt) of the carbon dioxide added to the atmosphere
each year ends up in the oceans (see Table 3.1). Observations of the relative dis-
tribution of the other isotopes of carbon in the atmosphere and in the oceans
also con¬rm this estimate (see box).

The biological pump in the oceans3
In temperate and high latitudes there is a peak each spring in ocean biological activity. During the winter,
water rich in nutrients is transferred from deep water to levels near the surface. As sunlight increases in
the spring an explosive growth of the plankton population occurs, known as the ˜spring bloom™. Pictures
of ocean colour taken from satellites demonstrate dramatically where this is happening.
Plankton are small plants (phytoplankton) and animals (zooplankton) that live in the surface waters
of the ocean; they range in size between about 0.001mm across and the size of typical insects on
land. Herbivorous zooplankton graze on phytoplankton; carnivorous zooplankton eat herbivorous zoo-
plankton. Plant and animal debris from these living systems sinks in the ocean. While sinking, some
decomposes and returns to the water as nutrients, some (perhaps about 1%) reaches the deep ocean or
the ocean ¬‚oor, where it is lost to the carbon cycle for hundreds, thousands or even millions of years.
The net effect of the ˜biological pump™ is to move carbon from the surface waters to lower levels in the
ocean. As the amount of carbon in the surface waters is reduced, more carbon dioxide from the atmos-
phere can be drawn down to restore the surface equilibrium. It is thought that the ˜biological pump™
has remained substantially constant in its operation during the last century unaffected by the increase
in carbon dioxide.
Evidence of the importance of the ˜biological pump™ comes from the palaeoclimate record from ice cores
(see Chapter 4). One of the constituents from the atmosphere trapped in bubbles in the ice is the gas
methyl sulphonic acid, which originates from decaying ocean plankton; its concentration is therefore an
indicator of plankton activity. As the global temperature began to increase when the last ice age receded
nearly 20 000 years ago and as the carbon dioxide in the atmosphere began to increase (Figure 4.4), the
methyl sulphonic acid concentration decreased. An interesting link is thereby provided between the carbon
dioxide in the atmosphere and marine biological activity. During the cold periods of the ice ages, enhanced
biological activity in the ocean could have been responsible for maintaining the atmospheric carbon diox-
ide at a lower level of concentration “ the ˜biological pump™ was having an effect.
There is some evidence from the palaeo record of the biological activity in the ocean being stimulated
by the presence of iron-containing dust blown over the oceans from the land surface. This has led to some
proposals in recent years to enhance the ˜biological pump™ through arti¬cially introducing iron over suitable
parts of the ocean. While an interesting idea, it seems from careful studies that even a very large-scale
operation would not have a large practical effect.
The question then remains as to why the ice ages were periods of greater marine biological activity
than the warm periods in between. One possible contributing process is suggested by considering what
happens in winter as nutrients are fed into the upper ocean ready for the spring bloom of biological activ-
ity. When there is less atmospheric carbon dioxide, cooling by radiation from the surface of the ocean
increases. Since convection in the upper layers of the ocean is driven by cooling at the surface, increased
cooling results in a greater depth of the mixed layer near the top of the ocean where all the biological
activity occurs. This is an example of a positive biological feedback; a greater depth of layer means more
plankton growth.4

What we can learn from carbon isotopes
Isotopes are chemically identical forms of the same element but with different atomic weights. Three isotopes
of carbon are important in studies of the carbon cycle: the most abundant isotope 12C which makes up 98.9%
of ordinary carbon, 13C present at about 1.1% and the radioactive isotope 14C which is present only in very
small quantities. About 10 kg of 14C is produced in the atmosphere each year by the action of particle radiation
from the Sun; half of this will decay into nitrogen over a period of 5730 years (the ˜half-life™ of 14C).
When carbon in carbon dioxide is taken up by plants and other living things, less 13C is taken up in pro-
portion than 12C. Fossil fuel such as coal and oil was originally living matter so also contains less 13C (by
about 18 parts per 1000) than the carbon dioxide in ordinary air in the atmosphere today. Adding carbon
to the atmosphere from burning forests, decaying vegetation or fossil fuel will therefore tend to reduce
the proportion of 13C.
Because fossil fuel has been stored in the Earth for much longer than 5730 years (the half-life of 14C), it
contains no 14C at all. Therefore, carbon from fossil fuel added to the atmosphere reduces the proportion
of 14C the atmosphere contains.
By studying the ratio of the different isotopes of carbon in the atmosphere, in the oceans, in gas trapped
in ice cores and in tree rings, it is possible to ¬nd out where the additional carbon dioxide in the atmos-
phere has come from and also what amount has been transferred to the ocean. For instance, it has been
possible to estimate for different times how much carbon dioxide has entered the atmosphere from the
burning or decay of forests and other vegetation and how much from fossil fuels.
Similar isotopic measurements on the carbon in atmospheric methane provide information about how
much methane from fossil fuel sources has entered the atmosphere at different times.

Further information regarding the broad partitioning of added atmospheric
carbon dioxide between the atmosphere, the oceans and the land biota as pre-
sented in Table 3.1 comes from comparing the trends in atmospheric carbon
dioxide concentration with the trends in very accurate measurements of the
atmospheric oxygen/nitrogen ratio (Figures 3.3 and 3.4). This possibility arises
because the relation between the exchanges of carbon dioxide and oxygen with
the atmosphere over land is different from that over the ocean. On land, living
organisms through photosynthesis take in carbon dioxide from the atmosphere
and build up carbohydrates, returning the oxygen to the atmosphere. In the pro-
cess of respiration they also take in oxygen from the atmosphere and convert
it to carbon dioxide. In the ocean, by contrast, carbon dioxide taken from the
atmosphere is dissolved, both the carbon and the oxygen in the molecules being
removed. How such measurements can be interpreted for the period 1990“4 is
shown in Figure 3.4. These data are consistent with budget for the 1990s shown
in Table 3.1.

Figure 3.4 Partitioning of fossil 1990
fuel carbon dioxide uptake using
oxygen measurements. Shown
is the relationship between 1991
changes in carbon dioxide
and oxygen concentrations. 1992
Observations are shown by 1993
solid circles. The arrow labelled

O2 concentration, difference from standard (ppm)
˜fossil fuel burning™ denotes 1994
the effect of the combustion of “30
fossil fuels based on the O2 : CO2
stoichiometric relation of the
different fuel types. Uptake by Fossil
land and ocean is constrained 1996
by the stoichiometric ratio
associated with these processes, 1997
de¬ning the slopes of the
respective arrows.


The global land“atmosphere 2000
¬‚ux in Table 3.1 represents
the balance of a net ¬‚ux due Outgassing
to land-use changes which “55

has generally been positive
or a source of carbon to the Atmospheric increase Land uptake Ocean uptake
atmosphere and a residual
350 355 360 365 370 375 380 385
component that is, by infer- CO 2 concentration (ppm)
ence, a negative ¬‚ux or carbon
sink. The estimates of land-use
changes (Table 3.1) are dominated by deforestation in tropical regions although
some uptake of carbon has occurred through forest regrowth in temperate
regions of the northern hemisphere and other changes in land management.
The main processes that contribute to the residual carbon sink are believed
to be the carbon dioxide ˜fertilisation™ effect (increased carbon dioxide in the
atmosphere leads to increased growth in some plants “ see box in Chapter 7
on page 199), the effects of increased use of nitrogen fertilisers and of some
changes in climate. The magnitudes of these contributions (Table 3.1) are dif-
¬cult to estimate directly and are subject to much more uncertainty than their
total, which can be inferred from the requirement to balance the overall carbon
cycle budget.

A clue to the uptake of carbon by the land biosphere is provided from obser-
vations of the atmospheric concentration of carbon dioxide which, each year,
show a regular cycle; the seasonal variation, for instance, at the observatory site
at Mauna Loa in Hawaii approaches about 10 ppm (Figure 3.3). Carbon dioxide
is removed from the atmosphere during the growing season and is returned as
the vegetation dies away in the winter. In the northern hemisphere therefore
a minimum in the annual cycle of carbon dioxide occurs in the northern sum-
mer. Since there is a larger amount of the terrestrial biosphere in the northern
hemisphere the annual cycle there has a much greater amplitude than in the
southern hemisphere. Estimates from carbon cycle models of the uptake by the
land biosphere are constrained by these observations of the seasonal cycle and
the difference between the hemispheres.5
The carbon dioxide fertilisation effect is an example of a biological feedback
process. It is a negative feedback because, as carbon dioxide increases, it tends
to increase the take-up of carbon dioxide by plants and therefore reduce the
amount in the atmosphere, decreasing the rate of global warming. Positive
feedback processes, which would tend to accelerate the rate of global warming,
also exist; in fact there are more potentially positive processes than negative
ones (see box on page 48“9). Although scienti¬c knowledge cannot yet put pre-
cise ¬gures on them, there are strong indications that some of the positive feed-
backs could be large, especially if carbon dioxide were to continue to increase,
with its associated global warming, through the twenty-¬rst century into the
twenty-second. These feedbacks are often called climate/carbon-cycle feedbacks
as they all result from changes in the climate affecting the performance of the
carbon cycle. In later chapters, it will be seen that these feedbacks can assert
large in¬‚uence on future concentrations of carbon dioxide and hence on future
Carbon dioxide provides the largest single contribution to anthropogenic
radiative forcing. Its radiative forcing from pre-industrial times to the present
is shown in Figure 3.11. A useful formula for the radiative forcing R from atmos-
pheric carbon dioxide when its atmospheric concentration is C ppm is: R = 5.3
ln(C/C0) where C0 is its pre-industrial concentration of 280 ppm.

Future emissions of carbon dioxide
To obtain information about future climate we need to estimate the future atmos-
pheric concentrations of carbon dioxide, which depend on future anthropogenic

emissions. In these estimates, the long time constants associated with the
response of atmospheric carbon dioxide to change have important implications.
Suppose, for instance, that all emissions into the atmosphere from human activi-
ties were suddenly halted. No sudden change would occur in the atmospheric
concentration, which would decline only slowly. We could not expect it to
approach its pre-industrial value for several hundred years.
But emissions of carbon dioxide are not halting, nor are they slowing; their
increase is, in fact, becoming larger each year. The atmospheric concentra-
tion of carbon dioxide will therefore also increase more rapidly. Later chap-
ters (especially Chapter 6) will present estimates of climate change during the
twenty-¬rst century due to the increase in greenhouse gases. A prerequisite
for such estimates is knowledge of likely changes in carbon dioxide emissions.
Estimating what will happen in the future is, of course, not easy. Because
nearly everything we do has an in¬‚uence on the emissions of carbon dioxide, it
means estimating how human beings will behave and what their activities will
be. For instance, assumptions have to be made about population growth, eco-
nomic growth, energy use, the development of energy sources and the likely
in¬‚uence of pressures to preserve the environment. These assumptions are
required for all countries of the world, both developing as well as developed
ones. Further, since any assumptions made are unlikely to be ful¬lled accu-
rately in practice, it is necessary to make a variety of different assumptions,
so that we can get some idea of the range of possibilities. Such possible futures
are called scenarios.
In Chapter 6 and Chapter 11 are presented two sets of emission scenarios as
developed respectively by the Intergovernmental Panel on Climate Change (IPCC)
and the International Energy Agency (IEA). These emission scenarios are then
turned into future projections of atmospheric carbon dioxide concentrations
through the application of a computer model of the carbon cycle that includes
descriptions of all the exchanges already mentioned. Further in Chapter 6, through
the application of computer models of the climate (see Chapter 5), projections of
the resulting climate change from different scenarios are also presented.

Feedbacks in the biosphere
As the greenhouse gases carbon dioxide and methane are added to the atmosphere because of human
activities, biological or other feedback processes occurring in the biosphere (such as those arising from the
climate change that has been induced) in¬‚uence the rate of increase of the atmospheric concentration of
these gases. These processes will tend either to add to the anthropogenic increase (positive feedbacks) or
to subtract from it (negative feedbacks).
Two feedbacks, one positive (the plankton multiplier in the ocean) and one negative (carbon dioxide
fertilisation), have already been mentioned in the text. Four other positive feedbacks are potentially impor-
tant, although our knowledge is currently insuf¬cient to quantify them precisely.
One is the effect of higher temperatures on respiration, especially through microbes in soils, leading to
increased carbon dioxide emissions. Evidence regarding the magnitude of this effect has come from studies

(a) (b)
Emissions Emissions
CO2 changes CO2 changes
Land uptake Land uptake
Ocean uptake Ocean uptake
Gigatonnes carbon
Gigatonnes carbon




1850 1900 1950 2000 2050 2100 1850 1900 1950 2000 2050 2100
Year Year
Figure 3.5 The possible effects of climate feedbacks on the carbon cycle. Shown are (1) accumulated fossil
fuel CO2 emissions from 1860 to the present and then projected to 2100 assuming the A2 SRES scenario
(Figure 6.1) “ in red (2) CO2 from (1) absorbed into the ocean “ in blue, (3) CO2 taken up by the land (of the
same sign as the other contributions but plotted below the axis for clarity) “ in orange and (4) the residual CO2
from (1) added to the atmosphere “ in green. The nine different ocean and land budgets resulted from a study
with nine coupled atmosphere“ocean general circulation climate models (AOGCMs “ see chapter 5) organised
internationally as part of a climate model intercomparison project (CMIP). (a) Shows results assuming no
feedbacks from climate change into elements of the carbon cycle. (b) Shows results when climate feedbacks
into the carbon cycle are included.

of the short-term variations of atmospheric carbon dioxide that have occurred during El Ni±o events and
during the cooler period following the Pinatubo volcanic eruption in 1991. These studies, which covered
variations over a few years, indicate a relation such that a change of 5 °C in average temperature leads to a
40% change in global average respiration rate6 “ a substantial effect. A question that needs to be resolved
is whether this relation still holds over longer-term changes of the order of several decades to a century.
A second positive feedback is the reduction of growth or the dieback especially in forests because of the
stress caused by climate change, which may be particularly severe in Amazonia (see box in Chapter 7 on
page 208).7 As with the last effect, this will increase as the amount of climate change becomes larger. The
combined result of these two feedbacks is that less carbon is taken up by the biosphere and more remains
in the atmosphere.8
Figure 3.5 shows this combined result as estimated for the twenty-¬rst century by nine different climate
models that incorporate the relevant processes in both the ocean and the land biospheres. Note that one of
the models in the ensemble, from the Hadley Centre, predicts the strongest values for the climate/carbon-
cycle feedbacks mentioned above and projects the highest atmospheric carbon dioxide level by 2100.9 The
curve in Figure 3.5 for land uptake relating to this model begins to curve upwards from the middle of the
century at which time the terrestrial biosphere changes from being a net sink of carbon (as in Table 3.1) to
being a net source.
Taking the average of the nine models under the A2 scenario, about 50 Gt more carbon remains in
the atmosphere in 2050 and 150 Gt in 2100 compared with what would occur in the absence of climate/
carbon-cycle feedback. For the Hadley Centre model the numbers are about 50 Gt and 350 Gt respectively
for 2050 and 2100. In terms of carbon dioxide concentration an additional 100 Gt means an additional
50 ppm.
A third positive feedback occurs through the release of greenhouse gases into the atmosphere due to
the increase of ¬res in forested areas because of the drier conditions as climate warms or because of the
dieback due to climate stress mentioned above10.
The fourth positive feedback is the release of methane, as temperatures increase “ from wetlands and
from very large reservoirs of methane trapped in sediments in a hydrate form (tied to water molecules
when under pressure) “ mostly at high latitudes. Methane has been generated from the decomposition of
organic matter present in these sediments over many millions of years. Because of the depth of the sedi-
ments this latter feedback is unlikely to become operative to a signi¬cant extent in the near future. However,
were global warming to continue to increase unchecked for many decades, releases from hydrates could
make a large contribution to methane emissions into the atmosphere and act as a large positive feedback
on the climate.

Other greenhouse gases
Methane is the main component of natural gas. Its common name used to be
marsh gas because it can be seen bubbling up from marshy areas where organic
material is decomposing. Data from ice cores show that for at least 2000 years
before 1800 its concentration in the atmosphere was about 700 ppb. Since then
its concentration has more than doubled (Figure 3.6a) to a value that the ice core
record shows is unprecedented over at least the last 650 000 years. During the
1980s it was increasing at about 10 ppb per year but during the 1990s the aver-
age rate of increase fell to around 5 ppb per year11 and close to zero from 1999
to 2005. Although the concentration of methane in the atmosphere is much
less than that of carbon dioxide (only 1.775 ppm in 2005 compared with about
380 ppm for carbon dioxide), its greenhouse effect is far from negligible. That
is because the enhanced greenhouse effect caused by a molecule of methane is
about eight times that of a molecule of carbon dioxide.12
The main natural source of methane is from wetlands. A variety of other
sources result directly or indirectly from human activities, for instance from leak-
age from natural gas pipelines and from oil wells, from generation in rice paddy
¬elds, from enteric fermentation (belching) from cattle and other livestock, from
the decay of rubbish in land¬ll sites and from wood and peat burning. Details of
estimates of the sizes of these sources during the 1990s are shown in Table 3.2.
Attached to many of the numbers is a wide range of uncertainty. It is, for instance,
dif¬cult to estimate the amount produced in paddy ¬elds averaged on a world-
wide basis. The amount varies enormously during the rice growing season and
also very widely from region to region. Similar problems arise when trying to
estimate the amount produced by animals. Measurements of the proportions of
the different isotopes of carbon (see box on page 44) in atmospheric methane


. 2
( 16)