. 3
( 13)


CO2 concentration (ppm)

Figure 3.4 Partitioning of fossil fuel carbon dioxide uptake using oxygen
measurements. Shown is the relationship between changes in carbon dioxide
and oxygen concentrations. Observations are shown by solid circles and
triangles. The arrow labelled ˜fossil fuel burning™ denotes the effect of the
combustion of fossil fuels based on the O2 : CO2 stoichiometric relation of the
different fuel types. Uptake by land and ocean is constrained by the
stoichiometric ratio associated with these processes, de¬ning the slopes of the
respective arrows.
Future emissions of carbon dioxide 39

is more vegetation growth in the northern hemisphere than the south-
ern, a minimum in the annual cycle of carbon dioxide in the atmosphere
occurs in the northern summer. Estimates from carbon cycle models of
the uptake by the land biosphere are constrained by these observations
of the difference between the hemispheres.6
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 40).
Although scienti¬c knowledge cannot yet put precise ¬gures on them,
there are strong indications that some of the positive feedbacks 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. 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 3.8. A useful formula for the
radiative forcing R from atmospheric carbon dioxide when its atmos-
pheric 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
atmospheric 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 activities 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
concentration of carbon dioxide will therefore also increase more rapidly.
Later chapters (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 the knowledge of what changes
in carbon dioxide emissions there are likely to be. 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
40 The greenhouse gases

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 that arise from the
climate change that has been induced) in¬‚uence the rate of increase
of the atmospheric concentration of these gases. These processes will
either tend 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 men-
tioned in the text. Three other positive feedbacks are potentially impor-
tant, although our knowledge is currently insuf¬cient to quantify them
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
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 forty per cent change in global
average respiration rate7 “ 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 die-
back especially in forests because of the stress caused by climate change,
which may be particularly severe in Amazonia8 (see box in Chapter 7
on page 173). As with the last effect, this will increase as the amount of
climate change becomes larger. A number of carbon cycle models show
that, through these two effects, during the second half of the twenty-¬rst
century the residual terrestrial sink (Table 3.1) could change sign and
become a substantial net source (Figure 3.5).
The third positive feedback is the release of methane, as tempera-
tures 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 sediments this
latter feedback is unlikely to become operative to a signi¬cant extent
during the twenty-¬rst century. However, were global warming to con-
tinue to increase substantially for more than a hundred years, releases
from hydrates could make a large contribution to methane emissions into
the atmosphere and act as a large positive feedback on the climate.
Future emissions of carbon dioxide 41


Gt C






1900 1950 2000 2050 2100

Figure 3.5 Illustrating the possible effects of climate feedbacks on the carbon
cycle. Results are shown of the changing budgets of carbon (in gigatonnes of
carbon) in the atmosphere, land and ocean in an ocean“atmosphere model
coupled to an ocean carbon cycle model (which includes the transfer of carbon
dioxide to depth through both the solubility pump and the biological pump)
and a dynamic global vegetation model (which includes the exchange of carbon
with the soil and with ¬ve different types of plant). The model was run with the
fossil fuel carbon dioxide emissions from 1860 to the present and then projected
to 2100 assuming the IS 92a scenario shown in Figure 6.1. Note that because of
climate feedbacks, the terrestrial biosphere changes from being a net sink of
carbon to being a net source around the middle of the twenty-¬rst century. Note
also as this source becomes stronger, by 2100 the atmospheric carbon content is
increasing at about the same rate as the total emissions (i.e. the ˜airborne
fraction™, or the fraction of fossil fuel emissions that remains in the atmosphere,
has changed from being about a half in the year 2000 to being about unity in
2100). Note also that an atmospheric carbon content of 1500 Gt more than it
was in 1860 is equivalent to a concentration of nearly 1000 ppm.

estimating how human beings will behave and what their activities are
likely to be. For instance, assumptions need to be made about population
growth, economic 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
42 The greenhouse gases

unlikely to be ful¬lled accurately 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 are presented two sets of emission scenarios as devel-
oped respectively by the Intergovernmental Panel on Climate Change
(IPCC) and the World Energy Council (WEC). These emission scenar-
ios 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 that chapter, through the application of computer models of
the climate (see Chapter 5), projections of the resulting climate change
from different scenarios are also presented.

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 two thousand years before 1800 its concentration in the
atmosphere was about 700 ppb. Since then its concentration has more
than doubled (Figure 3.6). During the 1980s it was increasing at about
10 ppb year’1 but during the 1990s the average rate of increase fell to
around 5 ppb year’1 .9 Although the concentration of methane in the
atmosphere is much less than that of carbon dioxide (less than 2 ppm
compared with about 370 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.10

Figure 3.6 Change in
methane concentration (mole
fraction in ppb) over the last
1000 years determined from
ice cores, glacier ice and
whole air samples. Radiative
forcing since the pre-industrial
era due to the methane
increase is plotted on the right
Other greenhouse gases 43

Table 3.2 Estimated sources and sinks of methane in millions of tonnes
per year.a The ¬rst column of data shows the best estimate from each
source; the second column illustrates the uncertainty in the estimates
by giving a range of values

Best estimate Uncertainty

Wetlands 150 (90“240)
Termites 20 (10“50)
Ocean 15 (5“50)
Other (including hydrates) 15 (10“40)
Coal mining, natural gas, petroleum industry 100 (75“110)
Rice paddies 60 (30“90)
Enteric fermentation 90 (70“115)
Waste treatment 25 (15“70)
Land¬lls 40 (30“70)
Biomass burning 40 (20“60)
Atmospheric removal 545 (450“550)
Removal by soils 30 (15“45)
Atmospheric increase 22 (35“40)

From Prather, M., Ehhalt, D. et al. 2001. Atmospheric chemistry and greenhouse
gases. Chapter 4 in Houghton et al., Climate Change 2001. See also Prather,
M. et al. 1995. Other trace gases and atmospheric chemistry. In Climate Change
1994. Cambridge: Cambridge University Press. The ¬gure for atmospheric in-
crease is an average for the 1990s.

The main natural source of methane is from wetlands. A variety
of other sources result directly or indirectly from human activities, for
instance from leakage 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 the best estimates of
the sizes of these sources 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 worldwide
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 37)
in atmospheric methane assist considerably in helping to tie down the
44 The greenhouse gases

proportion which comes from fossil fuel sources, such as leakage from
mines and from natural gas pipelines.
The main process for the removal of methane from the atmosphere
is through chemical destruction. It reacts with hydroxyl (OH) radicals,
which are present in the atmosphere because of processes involving sun-
light, oxygen, ozone and water vapour. The average lifetime of methane
in the atmosphere is determined by the rate of this loss process. At about
twelve years,11 it is much shorter than the lifetime of carbon dioxide.
Although most methane sources cannot be identi¬ed very precisely,
the largest sources apart from natural wetlands are closely associated
with human activities. It is interesting to note that the increase of atmo-
spheric methane (Figure 3.6) follows very closely the growth of human
population since the Industrial Revolution. However, even without the
introduction of deliberate measures to control human-related sources
of methane because of the impact on climate change, it is not likely
that this simple relationship with human population will continue. The
IPCC Special Report on Emission Scenarios (SRES) presented in Chap-
ter 6 include a wide range of estimates of the growth of human-related
methane emissions during the twenty-¬rst century “ from approximately
doubling over the century to reductions of about twenty-¬ve per cent.
In Chapter 10 (page 253) ways are suggested in which methane emis-
sions could be reduced and methane concentrations in the atmosphere

Nitrous oxide
Nitrous oxide, used as a common anaesthetic and known as laughing gas,
is another minor greenhouse gas. Its concentration in the atmosphere of
about 0.3 ppm is rising at about 0.25 per cent per year and is about
sixteen per cent greater than in pre-industrial times. It possesses a
relatively long atmospheric lifetime of about 115 years. The largest
emissions to the atmosphere are associated with natural and agricul-
tural ecosystems; those linked with human activities are probably due
to increasing fertiliser use. Biomass burning and the chemical industry
(for example, nylon production) also play some part. The sink of nitrous
oxide is photodissociation in the stratosphere and reaction with electron-
ically excited oxygen atoms, leading to an atmospheric lifetime of about
120 years.

Chloro¬‚uorocarbons (CFCs) and ozone
The CFCs are man-made chemicals which, because they vaporise
just below room temperature and because they are non-toxic and
Other greenhouse gases 45

non-¬‚ammable, appear to be ideal for use in refrigerators, the manufac-
ture of insulation and aerosol spray cans. Since they are so chemically
unreactive, once they are released into the atmosphere they remain for
a long time “ one or two hundred years “ before being destroyed. As
their use increased rapidly through the 1980s their concentration in the
atmosphere has been building up so that they are now present (adding
together all the different CFCs) in about l ppb (part per thousand mil-
lion “ or billion “ by volume). This may not sound very much, but it is
quite enough to cause two serious environmental problems.
The ¬rst problem is that they destroy ozone.12 Ozone (O3 ), a
molecule consisting of three atoms of oxygen, is an extremely reactive
gas present in small quantities in the stratosphere (a region of the atmo-
sphere between about 10 km and 50 km in altitude). Ozone molecules
are formed through the action of ultraviolet radiation from the Sun on
molecules of oxygen. They are in turn destroyed by a natural process as
they absorb solar ultraviolet radiation at slightly longer wavelengths “
radiation which would otherwise be harmful to us and to other forms
of life at the Earth™s surface. The amount of ozone in the stratosphere
is determined by the balance between these two processes, one forming
ozone and one destroying it. What happens when CFC molecules move
into the stratosphere is that some of the chlorine atoms they contain are
stripped off, also by the action of ultraviolet sunlight. These chlorine
atoms readily react with ozone, reducing it back to oxygen and adding
to the rate of destruction of ozone. This occurs in a catalytic cycle “ one
chlorine atom can destroy many molecules of ozone.
The problem of ozone destruction was brought to world attention
in 1985 when Joe Farman, Brian Gardiner and Jonathan Shanklin at
the British Antarctic Survey discovered a region of the atmosphere over
Antarctica where, during the southern spring, about half the ozone over-
head disappeared. The existence of the ˜ozone hole™ was a great sur-
prise to the scientists; it set off an intensive investigation into its causes.
The chemistry and dynamics of its formation turned out to be complex.
They have now been unravelled, at least as far as their main features
are concerned, leaving no doubt that chlorine atoms introduced into the
atmosphere by human activities are largely responsible. Not only is there
depletion of ozone in the spring over Antarctica (and to a lesser extent
over the Arctic) but also substantial reduction, of the order of ¬ve per
cent, of the total column of ozone “ the amount above one square me-
tre at a given point on the Earth™s surface “ at mid latitudes in both
Because of these serious consequences of the use of CFCs, inter-
national action has been taken. Many governments have signed the Mon-
treal Protocol set up in 1987 which, together with the Amendments
46 The greenhouse gases

agreed in London in 1991 and in Copenhagen in 1992, required that
manufacture of CFCs be phased out completely by the year 1996 in
industrialised countries and by 2006 in developing countries. Because
of this action the concentration of CFCs in the atmosphere is no longer
increasing. However, since they possess a long life in the atmosphere,
little decrease will be seen for some time and substantial quantities will
be present well over a hundred years from now.
So much for the problem of ozone destruction. The other problem
with CFCs and ozone, the one which concerns us here, is that they are
both greenhouse gases.13 They possess absorption bands in the region
known as the longwave atmospheric window (see Figure 2.4) where few
other gases absorb. Because, as we have seen, the CFCs destroy some
ozone, the greenhouse effect of the CFCs is partially compensated by
the reduced greenhouse effect of atmospheric ozone.
First considering the CFCs on their own, a CFC molecule added to
the atmosphere has a greenhouse effect ¬ve to ten thousand times greater
than an added molecule of carbon dioxide. Thus, despite their very small
concentration compared, for instance, with carbon dioxide, they have a
signi¬cant greenhouse effect. It is estimated that due to the CFCs now
present in the atmosphere the radiative forcing in the tropics (at higher
latitudes there is a compensating effect due to ozone reduction which
is explained below) is about 0.25 W m’2 “ or about twenty per cent of
the radiative forcing due to all greenhouse gases. This forcing will only
decrease very slowly next century.
Turning now to ozone, the effect from ozone depletion is complex
because the amount by which ozone greenhouse warming is reduced
depends critically on the height in the atmosphere at which it is being de-
stroyed. Further, ozone depletion is concentrated at high latitudes while
the greenhouse effect of the CFCs is uniformly spread over the globe.
In tropical regions there is virtually no ozone depletion so no change in
the ozone greenhouse effect. At mid latitudes, very approximately, the
greenhouse effects of ozone reduction and of the CFCs compensate for
each other. In polar regions the reduction in the greenhouse effect of
ozone more than compensates for the greenhouse warming effect of the
As CFCs are phased out, they are being replaced to some degree
by other halocarbons “ hydrochloro-¬‚uorocarbons (HCFCs) and hydro-
¬‚uorocarbons (HFCs). In Copenhagen in 1992, the international com-
munity decided that HCFCs would also be phased out by the year 2030.
While being less destructive to ozone than the CFCs, they are still green-
house gases. The HFCs contain no chlorine or bromine, so they do not
destroy ozone and are not covered by the Montreal Protocol. Because
of their shorter lifetime, typically tens rather than hundreds of years,
Gases with an indirect greenhouse effect 47

the concentration in the atmosphere of both the HCFCs and the HFCs,
and therefore their contribution to global warming for a given rate of
emission, will be less than for the CFCs. However, since their rate of
manufacture could increase substantially their potential contribution to
greenhouse warming is being included alongside other greenhouse gases
(see Chapter 10 page 247).
Concern has also extended to some other related compounds which
are greenhouse gases, the per¬‚uorocarbons (e.g. CF4 , C2 F6 ) and sulphur
hexa¬‚uoride (SF6 ), which are produced in some industrial processes.
Because they possess very long atmospheric lifetimes, probably more
than 1000 years, all emissions of these gases accumulate in the atmo-
sphere and will continue to in¬‚uence climate for thousands of years. They
are also therefore being included as potentially important greenhouse
Ozone is also present in the lower atmosphere or troposphere, where
some of it is transferred downwards from the stratosphere and where
some is generated by chemical action, particularly as a result of the action
of sunlight on the oxides of nitrogen. It is especially noticeable in polluted
atmospheres near the surface; if present in high enough concentration,
it can become a health hazard. In the northern hemisphere the limited
observations available together with model simulations of the chemical
reactions leading to ozone formation suggest that ozone concentrations
in the troposphere have doubled since pre-industrial times “ an increase
which is estimated to have led to a global average radiative forcing of
between 0.2 and 0.6 W m’2 (Figure 3.8). Ozone is also generated at
levels in the upper troposphere as a result of the nitrogen oxides emitted
from aircraft exhausts; nitrogen oxides emitted from aircraft are more
effective at producing ozone in the upper troposphere than are equivalent
emissions at the surface. The radiative forcing in northern mid latitudes
from aircraft due to this additional ozone15 is of similar magnitude to that
from the carbon dioxide emitted from the combustion of aviation fuel
which is about three per cent of current global fossil fuel consumption.

Gases with an indirect greenhouse effect
I have described all the gases present in the atmosphere which have
a direct greenhouse effect. There are also gases which through their
chemical action on greenhouse gases, for instance on methane or on
lower atmospheric ozone, have an in¬‚uence on the overall size of green-
house warming. Carbon monoxide (CO) and the nitrogen oxides (NO and
NO2 ) emitted, for instance, by motor vehicles are some of these. Carbon
monoxide has no direct greenhouse effect of its own but, as a result of
chemical reactions, it forms carbon dioxide. These reactions also affect
48 The greenhouse gases

the amount of the hydroxyl radical (OH) which in turn affects the con-
centration of methane. Substantial research has been carried out on the
chemical processes in the atmosphere that lead to these indirect effects
on greenhouse gases.16 It is of course important to take them properly
into account, but it is also important to recognise that their combined ef-
fect is much less than that of the major contributors to human-generated
greenhouse warming, namely carbon dioxide and methane.

Particles in the atmosphere17
Small particles suspended in the atmosphere (often known as aerosol)
affect its energy balance because they both absorb radiation from the Sun
and scatter it back to space. We can easily see the effect of this on a bright
day in the summer with a light wind when downwind of an industrial
area. Although no cloud appears to be present, the sun appears hazy. We
call it ˜industrial haze™. Under these conditions a signi¬cant proportion
of the sunlight incident at the top of the atmosphere is being lost as it
is scattered back and out of the atmosphere by the millions of small
particles (typically between 0.001 and 0.01 mm in diameter) in the haze.
Atmospheric particles come from a variety of sources. They arise
partially from natural causes; they are blown off the land surface, espe-
cially in desert areas; they result from forest ¬res and they come from sea
spray. From time to time large quantities of particles are injected into the
upper atmosphere from volcanoes “ the Pinatubo volcano which erupted
in 1991 provides a good example (see Chapter 5). Some particles are
also formed in the atmosphere itself, for instance sulphate particles from
the sulphur-containing gases emitted from volcanoes.
Other particles arise from human activities “ from biomass burning
(e.g. the burning of forests) and the sulphates and soot resulting from the
burning of fossil fuels. The sulphate particles are particularly important.
They are formed as a result of chemical action on sulphur dioxide, a gas
that is produced in large quantities by power stations and other industries
in which coal and oil (both of which contain sulphur in varying quan-
tities) are burnt. Because these particles remain in the atmosphere only
for about ¬ve days on average, their effect is mainly con¬ned to regions
near the sources of the particles, i.e. the major industrial regions of the
northern hemisphere (Figure 3.7). Over limited regions of the north-
ern hemisphere the radiative effect of these particles is comparable in
size, although opposite in effect, to that of human-generated greenhouse
gases up to the present time. Estimates of the direct radiative forcing, av-
eraged over the globe, due to the particles from various human-generated
sources are shown in Figure 3.8. It will be seen that there are substantial
uncertainties associated with these estimates.
Particles in the atmosphere 49

Figure 3.7 Modelled geographic distribution of estimates of the annual mean
direct radiative forcing (in watts per square metre) from anthropogenic sulphate
aerosols in the troposphere. The radiative forcing, which is negative, is largest in
regions close to industrial activity.

So far for aerosol we have been describing direct radiative forcing.
There is a further way by which particles in the atmosphere could in-
¬‚uence the climate; that is through their effect on cloud formation that
is described as indirect radiative forcing. Two mechanisms of indirect
forcing have been proposed. The ¬rst is the in¬‚uence of the number of
particles and their size on cloud radiative properties. It arises as follows.
If particles are present in large numbers when clouds are forming, the
resulting cloud consists of a large number of smaller drops “ smaller
than would otherwise be the case “ this is similar to what happens as
polluted fogs form in cities. Such a cloud will be more highly re¬‚ecting
to sunlight than one consisting of larger particles, thus further increasing
the energy loss resulting from the presence of the particles. The second
mechanism arises because of the in¬‚uence of droplet size and number on
precipitation ef¬ciency, the lifetime of clouds and hence the geographic
extent of cloudiness. There is observational evidence for both of these
mechanisms but the processes involved are dif¬cult to model and will
vary a great deal with the particular situation. Estimates of their magni-
tude as shown in Figure 3.8 therefore remain very uncertain. To re¬ne
these estimates, more studies need to be made especially by making
careful measurements on suitable clouds.
The estimates for the radiative effects of particles as in Figure 3.8
can be compared with the global average radiative forcing to date due
50 The greenhouse gases

Global and annual mean radiative forcing (1750 to present)

2 N2O
Radiative forcing (W m’2)


1 Mineral
Trop. O3 Aviation-induced

Contrails Cirrus
Strat. O3 FF Land-
(oc) BB use
’1 (albedo)
Trop. aerosol
indirect effect
(1st type)
Level of scientific understanding (LOSU)

Figure 3.8 Global, annual mean radiative forcings (W m’2 ) due to a number of
agents for the period from pre-industrial (1750) to 2000. The height of the
rectangular bar denotes a best estimate value while its absence denotes no best
estimate is possible because of large uncertainties. The vertical lines with ˜x™ or
˜o™ delimiters indicates estimates of the uncertainty ranges. A ˜level of scienti¬c
understanding (LOSU)™ index is accorded to each forcing, with H, M, L and VL
denoting high, medium, low and very low levels respectively. This represents a
judgement about the reliability of the forcing estimate involving factors such as
the assumptions necessary to evaluate the forcing, the degree of knowledge of
the mechanisms determining the forcing and the uncertainties surrounding the
quantitative estimate of the forcing. The well-mixed greenhouse gases are
grouped together into a single rectangular bar with the individual contributions
shown. The second and third bars apply to stratospheric and tropospheric
ozone. The next bars denote the direct effect of aerosols. FF denotes aerosol
from fossil fuel burning and is separated into black carbon (bc) and organic
carbon (oc) components. BB denotes aerosols from biomass burning. The sign
of the effects due to mineral dust is itself an uncertainty. Only the ¬rst indirect
aerosol effect is estimated as little quantitative evidence exists regarding the
second. All the forcings have distinct spatial and seasonal variations (see Figure
3.7) so that they cannot be added up and viewed a priori as providing offsets in
terms of complete global climate impact.
Questions 51

to the increase in greenhouse gases of about 2.6 W m’2 . Comparing
global average forcings, however, is not the whole story. Because of
the large regional variation of particles in the atmosphere (Figure 3.7),
any effect they have on the climate can be expected to be substantially
different from the effect of increases in greenhouse gases, which is much
more uniform over the globe. This will be illustrated in Chapter 5 where
we consider the likely pattern of anthropogenic climate change to date.
More consideration of it will also be given in Chapter 6 when discussing
projections of climate change on regional scales that can depend critically
on assumptions about the likely concentrations of atmospheric particles
in the future.
An important factor that will in¬‚uence the future concentrations of
sulphate particles is ˜acid rain™ pollution, caused mainly by sulphur diox-
ide emissions. This leads to the degradation of forests and ¬sh stocks in
lakes especially in regions downwind of major industrial areas. Serious
efforts are therefore under way, especially in Europe and North America,
to curb these emissions to a substantial degree. Although the amount
of sulphur-rich coal being burnt elsewhere in the world, for instance
in Asia, is increasing rapidly, the damaging effects of sulphur pollution
are such that tight controls on sulphur emissions are being extended to
these regions also. For the globe as a whole therefore, sulphur emissions
are likely to rise much less rapidly than emissions of carbon dioxide;
the IPCC SRES scenarios of future emissions as presented in Chapter 6
anticipate these changes. The climate change resulting from an increase
in sulphate particles therefore will become increasingly less by compar-
ison with that from the likely increase of greenhouse gases.

Estimates of radiative forcing
This chapter has summarised current scienti¬c knowledge about the
sources and sinks of the main greenhouse gases and the exchanges which
occur between the components of the climate system “ the atmosphere,
the ocean and the land surface “ including the close balances that are
maintained between the different components and the way in which these
balances are being disturbed by human-generated emissions. Different
assumptions about future emissions have been used to generate emission
scenarios. From these scenarios estimates have been made (for carbon
dioxide, for instance, using a computer model of the carbon cycle) of
likely increases in greenhouse gas concentrations in the future.
Given information about the possible increases in greenhouse gases,
the next step is to calculate the effect of these increases on the amounts
of thermal (infrared) radiation absorbed and emitted by the atmosphere.
52 The greenhouse gases

This is done using information about how the different gases absorb
radiation in the infrared part of the spectrum, as mentioned in Chapter 2.
The radiative forcing associated with the increases in each of the gases
can then be calculated. In Figure 3.8 are brought together estimates of
global average radiative forcing for the period from 1750 to 2000 for
the different greenhouse gases and for tropospheric aerosols of different
origins which we have been considering in this chapter.
It is useful to be able to compare the radiative forcing generated by
the different greenhouse gases. Because of their different life-times, the
future pro¬le of radiative forcing due to releases of greenhouse gases
varies very much from gas to gas. An index called the global warm-
ing potential (GWP) has been de¬ned for greenhouse gases that takes
the ratio of the time-integrated radiative forcing from the instantaneous
release of 1 kg of a given gas to that from the release of 1 kg of car-
bon dioxide. A time horizon has also to be speci¬ed for the period over
which the integration is carried out. The GWPs of the six greenhouse
gases included in the Kyoto Protocol are listed in Table 10.2. Applying
the GWPs to the emissions from a mixture of greenhouse gases enables
the mixture to be considered in terms of an equivalent amount of car-
bon dioxide. However, because the GWPs for different time horizons are
very different, GWPs are of limited application and must be used with
When considering radiative forcing of climate, the question is bound
to be asked as to whether variations have occurred or are likely to occur
in the amount of energy from the Sun that is incident on the Earth and
that could, therefore, affect the climate. We shall see, for instance, in the
next chapter that ice ages in the past have been triggered by variations in
the geometry of the Earth™s orbit. It is considered possible that the Sun™s
output itself could vary by small amounts over time (see box in Chapter 6
page 138). Figure 3.8 (see also Figure 6.12) indicates the range of es-
timates of solar variability that may have occurred since 1850 showing
that its in¬‚uence is much less than that of the increase in greenhouse
Also included in Figure 3.8 are the possible effects of aviation on
radiative forcing and effects due to land-use changes that arise because
of changes in the albedo (see Glossary) of the surface. The effects of
aviation that are in addition to the source of carbon dioxide emissions it
provides arise from its in¬‚uence on high cloud cover through its emis-
sions of water vapour. As we shall see in Chapter 5 (page 91) high
cloud provides a blanketing effect on the Earth™s surface similar to that
of greenhouse gases and therefore leads to positive radiative forcing.
Many examples exist of extensive contrail formation over regions where
many aircraft ¬‚ights regularly occur. Aircraft may also in¬‚uence cirrus
Questions 53

cloud formation through the effect of the particles in aircraft emissions.
The overall greenhouse effect of aircraft has been estimated as up to
the equivalent of two or three times the effect of their carbon dioxide
Details of radiative forcings as projected for the twenty-¬rst century
are presented in Chapter 6 (page 120). Chapters 5 and 6 will explain how
estimates of radiative forcing can be incorporated into computer climate
models so as to predict the climate change that is likely to occur because
of human activities. However, before considering predictions of future
climate change, it is helpful to gain perspective by looking at some of
the climate changes that have occurred in the past.


1 The lifetime of a carbon dioxide molecule in the atmosphere before it is
exchanged with the ocean is typically less than a year, while the time taken
for an increase in carbon dioxide concentration from fossil fuel burning to
diminish substantially is typically many years. Explain the reasons for this
2 Estimate how much carbon dioxide you emit each year through breathing.
3 Estimate the size of your share of carbon dioxide emissions from the burning
of fossil fuels.
4 A typical city in the developed world with a population of about one million
produces about half a million tonnes of municipal waste each year. Suppose
the waste is buried in a land¬ll site where the waste decays producing equal
quantities of carbon dioxide and methane. Making assumptions about the
likely carbon content of the waste and the proportion that eventually decays,
estimate the annual production of methane. If all the methane leaks away,
using the information in note 10, compare the greenhouse effect of the carbon
dioxide and methane produced from the land¬ll site with that of the carbon
dioxide produced if the waste were incinerated instead.
5 A new forest is planted containing a million trees which will mature in forty
years. Estimate the amount of carbon sequestered per year by the forest.
6 Find ¬gures for the amount of fuel used by a typical aircraft and the size of
¬‚eets of the world™s airlines and airforces and estimate the carbon dioxide
emitted globally each year by the world™s aircraft.
7 Search for information about the ozone hole and explain why it occurs mainly
in the Antarctic.
8 What are the main uses of CFCs? Suggest ways in which the emissions of
CFCs to the atmosphere could be reduced more rapidly.
9 Evidence is sometimes presented suggesting that the variations in global av-
erage temperature over the last century or more can all be explained as due
to variations in the energy output of the Sun. There is therefore nothing left
to attribute to the increase in greenhouse gases. What is the fallacy in this
54 The greenhouse gases

10 With the use of the formula given in the text, calculate the radiative forcing
due to carbon dioxide for atmospheric concentrations of 150, 280, 450, 560
and 1000 ppm.

Notes for Chapter 3
1 It is convenient to de¬ne radiative forcing as the radiative imbalance at the
top of the troposphere rather than at the top of the whole atmosphere.
2 Reference sources for Table 3.1 are as follows:
Prentice, I. C. et al. 2001. The carbon cycle and atmospheric carbon dioxide.
Chapter 3 in Houghton, J. T., Ding, Y., Griggs, D. J., Noguer, M., van der
Linden, P. J., Dai, X., Maskell, K., Johnson, C. A. (eds.) Climate Change
2001: The Scienti¬c Basis. Contribution of Working Group I to the Third As-
sessment Report of the Intergovernmental Panel on Climate Change. Cam-
bridge: Cambridge University Press.
House, J. I. et al. 2003. Reconciling apparent inconsistencies in estimates
of terrestrial CO2 sources and sinks. Tellus, 55B, pp. 345“63.
See also Table 1.2 in Watson, R.T., Noble, I. R., Bolin, B., Ravindranath,
N. H., Verardo, D. J., Dokken, D. J. (eds.) Land Use, Land-use Change, and
Forestry, Chapter 1. Cambridge: Cambridge University Press.
3 See Prentice et al., Climate Change 2001.
4 Woods, J., Barkmann, W. 1993. The Plankton multiplier “ positive feedback
in the greenhouse. Journal of Plankton Research, 15, pp. 1053“74.
5 Keeling, R. F., Piper, S. C., Heimann, M. 1996. Global and hemispheric
sinks deduced from changes in atmospheric O2 concentration. Nature, 381,
pp. 218“21.
6 For more details see House, J. I. et al. 2003. Reconciling apparent incon-
sistencies in estimates of terrestrial CO2 sources and sinks. Tellus, 55B,
pp. 345“63.
7 See Jones, C. D. and Cox, P. M. 2001. Atmospheric science letters.
doi.1006/asle. 2001.0041; the respiration rate varies approximately with
a factor 2(T ’10)/10 .
8 Cox, P. M. et al. 2004. Amazon die-back under climate-carbon cycle pro-
jections for the 21st century. Theoretical and Applied Climatology, in press.
9 In 1992 the increase slowed to almost zero. The reason for this is not known
but one suggestion is that, because of the collapse of the Russian economy,
the leakage from Siberian natural gas pipelines was much reduced.
10 The ratio of the enhanced greenhouse effect of a molecule of methane com-
pared to a molecule of carbon dioxide is known as its global warming po-
tential (GWP); a de¬nition of GWP is given later in the chapter. The ¬gure
of about 8 given here for the GWP of methane is for a time horizon of
100 years “ see Lelieveld, J., Crutzen, P. J. Nature, 355, 1992, pp. 339“41;
see also Prather et al., Chapter 4, and Ramaswamy et al., Chapter 6 in,
Houghton et al., Climate Change 2001. The GWP is also often expressed as
the ratio of the effect for unit mass of each gas in which case the GWP for
methane (whose molecular mass is 0.36 of that of carbon dioxide) becomes
Notes 55

about 23 for the 100 year time horizon. About seventy-¬ve per cent of the
contribution of methane to the greenhouse effect is because of its direct
effect on the outgoing thermal radiation. The other twenty-¬ve per cent
arises because of its in¬‚uence on the overall chemistry of the atmosphere.
Increased methane eventually results in small increases in water vapour in
the upper atmosphere, in tropospheric ozone and in carbon dioxide, all of
which in turn add to the greenhouse effect.
11 Taking into account the loss processes due to reaction with OH in the tropo-
sphere, chemical reactions and soil loss lead to a lifetime of about ten years.
However, the effective lifetime of methane against a perturbation in con-
centration in the atmosphere (the number quoted here) is complex because
it depends on the methane concentration. This is because the concentration
of the radical OH (interaction with which is the main cause of methane
destruction), due to chemical feedbacks, is itself dependent on the methane
concentration (more details in Prather et al., Chapter 4, in Houghton et al.,
Climate Change 2001).
12 For more detail, see Scienti¬c Assessment of Ozone Depletion: 1998. Global
Ozone Research and Monitoring Project “ Report No 44. Geneva: World
Meteorological Organization, 732 pp.
13 Prather et al., Chapter 4, in Houghton et al., Climate Change 2001.
14 More detail on this and the radiative effects of minor gases and particles
can be found in Ramaswamy et al., Chapter 6, in Houghton et al. Climate
Change 2001.
15 More detail in Penner, J. E. et al. (eds.) 1999. Aviation and the Global
Atmosphere. An IPCC Special Report. Cambridge: Cambridge University
16 Prather et al., Chapter 4, in Houghton et al., Climate Change 2001.
17 Penner et al., Chapter 5, in Houghton et al., Climate Change 2001.
18 For more complete information of the effects of aviation see Penner, Aviation
and the Global Atmosphere.
Chapter 4
Climates of the past

To obtain some perspective against which to view future climate
change, it is helpful to look at some of the climate changes that have
occurred in the past. This chapter will brie¬‚y consider climatic records
and climate changes in three periods: the last hundred years, then the
last thousand years and ¬nally the last million years. At the end of the
chapter some interesting recent evidence for the existence of relatively
rapid climate change at various times during the past one or two hundred
thousand years will be presented.

The last hundred years
The 1980s and early 1990s have brought unusually warm years for the
globe as a whole (see Chapter 1) as is illustrated in Figure 4.1, which
shows the global average temperature since 1860, the period for which
the instrumental record is available with good accuracy and coverage.
An increase over this period has taken place of about 0.6 —¦ C (ninety-¬ve
per cent con¬dence limits of 0.4 to 0.8 —¦ C). The year 1998 is very likely1
to have been the warmest year during this period. An even more striking
statistic is that each of the ¬rst eight months of 1998 was very likely the
warmest of those months in the record. Although there is a distinct trend
in the record, the increase is by no means a uniform one. In fact, some
periods of cooling as well as warming have occurred and an obvious
feature of the record is the degree of variability from year to year and
from decade to decade.
A sceptic may wonder how a diagram like Figure 4.1 can be prepared
and whether any reliance can be placed upon it. After all, temperature

The last hundred years 57

Departures in temperature(°C)

from the 1961“1990 average





Data from thermometers
1880 1900 1920 1940 1960 1980 2000
Figure 4.1 Variations of the globally averaged Earth™s surface temperature over
the last 140 years. The dark bars are the year by year averaged values; the grey
line is a smoothed annual curve to illustrate decadal variations. Uncertainties in
the data are also shown; the thin whiskers represent the 95% con¬dence range.
The graph is based on an improved analysis of all years since the original
publication in the Third IPCC Scienti¬c Assessment. The changes are small.

varies from place to place, from season to season and from day to day by
many tens of degrees. But here we are not considering changes in local
temperature but in the average over the whole globe. A change of a few
tenths of a degree in that average is a large change.
First of all, just how is a change in global average temperature es-
timated from a combination of records of changes in the near-surface
temperature over land and changes in the temperature of the sea surface?
To estimate the changes over land, weather stations are chosen where
consistent observations have been taken from the same location over a
substantial proportion of the whole 130-year period. Changes in sea sur-
face temperature have been estimated by processing over sixty million
observations from ships “ mostly merchant ships “ over the same period.
All the observations, from land stations and from ships, are then located
within a grid of squares, say 1—¦ of latitude by 1—¦ of longitude, covering
the Earth™s surface. Observations within each square are averaged; the
global average is obtained by averaging (after weighting them by area)
over the averages for each of the squares.
58 Climates of the past

A number of research groups in different countries have made careful
and independent analyses of these observations. In somewhat different
ways they have made allowances for factors that could have introduced
arti¬cial changes in the records. For instance, the record at some land
stations could have been affected by changes in their surroundings as
these have become more urban. In the case of ships, the standard method
of observation used to be to insert a thermometer into a bucket of water
taken from the sea. Small changes of temperature have been shown to
occur during this process; the size of the changes varies between day and
night and is also dependent on several other factors including the material
from which the bucket is made “ over the years wooden, canvas and metal
buckets have been variously employed. Nowadays, a large proportion of
the observations are made by measuring the temperature of the water
entering the engine cooling system. Careful analysis of the effects of
these details on observations both on land and from ships has enabled
appropriate corrections to be made to the record, and good agreement
has been achieved between analyses carried out at different centres.
Con¬dence that the observed variations are real is increased by notic-
ing that the trend and the shape of the changes are similar when different
selections of the total observations are made. For instance, the separate
records from the land and sea surface and from the northern and south-
ern hemispheres are closely in accord. Further indirect indicators such
as changes in borehole temperatures and sub-surface ocean tempera-
tures, decrease in snow cover and glacier shrinkage provide independent
support for the observed warming.
During the last thirty years or so observations have been available
from satellites orbiting around the Earth. Their great advantage is that
they automatically provide data with global coverage, which are often
lacking in other data sets. The length of the record from satellites, how-
ever, is generally less than twenty years, a comparatively short period in
climate terms. It has been suggested that satellite measurements of lower
atmospheric temperature since 1979 are not consistent with the trend of
rising temperatures in surface observations. The satellite observations
do not cast doubt on the accuracy of the surface measurements “ the two
measurements are of different quantities. However, it is expected that
trends in lower atmosphere measurements and surface measurements
should be related and much careful work has been carried out studying
the two records and interpreting the differences (see box).
The most obvious feature of the climate record illustrated in
Figure 4.1 is that of considerable variability, not just from year to year,
but from decade to decade. Some of this variability will have arisen
through causes external to the atmosphere and the oceans, for instance
as a result of volcanic eruptions such as those of Krakatoa in 1883 or of
The last hundred years 59

Atmospheric temperature observed by satellites
Since 1979 meteorological satellites ¬‚own by the National Oceanic and
Atmospheric Administration (NOAA) of the United States have carried
a microwave instrument, the Microwave Sounding Unit (MSU), for the
remote observation of the average temperature of the lower part of the
Figure. 4.2(a) shows the record of global average temperature de-
duced from the MSU and compares it with data from sounding instru-
ments carried on balloons for the same region of the atmosphere, showing
very good agreement for the period of overlap. The record of temperature
at the surface is also added for comparison over the period from 1960
to 2000. All three measurements show similar variability, the variations
at the surface tracking well with those in the lower troposphere. The
plots also illustrate the dif¬culty of deriving accurate trends from a short
period of record where there is also substantial variability. Since 1979 the
trends from the three measurement sources have been carefully studied
and compared. They are 0.04 ± 0.11 —¦ C per decade and 0.03 ± 0.10 —¦ C
per decade for the satellite and balloon data respectively compared with
0.16 ± 0.06 —¦ C per decade for the surface data. The trend in the differ-
ence of the surface and lower troposphere of 0.13 ± 0.06 —¦ C per decade
is statistically signi¬cant. This is in contrast to near-zero surface tem-
perature trends from 1958 to 1978 when the global lower troposphere
temperature warmed by about 0.03 —¦ C per decade relative to the surface.
There are substantial regional variations in the differences between sur-
face and lower tropospheric temperature trends since 1979. For instance,
the differences are particularly apparent in many parts of the tropics and
sub-tropics where the surface has warmed faster than the lower tropo-
sphere, while over some other regions, e.g. North America, Europe and
Australia, the trends are very similar.
Why the surface and lower temperature trends show signi¬cant dif-
ferences, especially over the tropical and sub-tropical oceans, is not com-
pletely understood although there are a number of possible reasons for
the differences.2 It is of course well known that the presence of increased
concentrations of greenhouse gases leads to cooling of the atmosphere
at higher levels (see Chapter 2). In the stratosphere therefore the tem-
perature trends are reversed (Figure 4.2(b)) ranging from a decrease of
about 0.5 —¦ C per decade in the lower stratosphere to 2.5 —¦ C per decade in
the upper stratosphere.

Pinatubo in the Philippines in 1991 (the low global average temperature
in 1992 and 1993, compared with neighbouring years, is almost certainly
due to the Pinatubo volcano). But there is no need to invoke volcanoes or
other external causes to explain all the variations in the record. Many of
60 Climates of the past


Anomaly (°C)


’0.5 Balloons
1960 1970 1980 1990 2000


Anomaly (°C)


Agung El Chichon Pinatubo
1960 1970 1980 1990 2000
Figure 4.2 Time series of observations of global average temperature (—¦ C)
(relative to average for 1979“90) (a) for the lower troposphere based on
measurements from balloons and satellites compared with measurements at the
surface; (b) for the lower stratosphere from balloons and satellites.

them result from internal variations within the total climate system, for
instance between different parts of the ocean. (See Chapter 5 for more
The warming during the twentieth century has not been uniform
over the globe. For instance, the recent warming has been greatest over
Northern Hemisphere continents at mid to high latitudes. There have
also been areas of cooling, for instance over some parts of the
North Atlantic ocean associated with changes in ocean circulation (see
Chapter 5). Some of the recent regional patterns of temperature change
are related to different phases of atmosphere-ocean oscillations, such as
the North Atlantic Oscillation (NAO). The positive phase of the NAO,
The last hundred years 61

with high pressure over the sub-tropical Atlantic and southern Europe
and mild winters over northwest Europe, has tended to be dominant since
the mid 1980s.
An interesting feature of the increasing temperature during the last
few decades has been that, in the daily cycle of temperature, minimum
temperatures over land have increased about twice as much as maximum
temperatures. A likely explanation for this, in addition to the effects of
enhanced greenhouse gases, is an increase in cloud cover which has
been observed in many of the areas with reduced temperature range. An
increase in cloud tends to obstruct daytime sunshine and tends also to
reduce the escape of terrestrial radiation at night.
As might be expected the increases in temperature have led on
average to increases in precipitation, although precipitation shows even
more variability in both space and time than temperature. The increases
have been particularly noticeable in the northern hemisphere in mid to
high latitudes, often appearing particularly as increases in heavy rainfall
events (see Table 4.1).
The broad features of these changes in temperature and precipitation
are consistent with what is expected because of the in¬‚uence of increased
greenhouse gases (see Chapter 5), although there is much variability in
the record that arises for reasons not associated with human activities.
For instance, the particular increase from 1910 to 1940 (Figure 4.1)
is too rapid to have been due to the rather small increase in greenhouse
gases during that period. The particular reasons for this will be discussed
in the next chapter where comparisons of observed temperatures with
simulations from climate models for the whole of the twentieth century
will be presented, not just as they concern the global mean but also
the regional patterns of change. We conclude there that, although the
expected signal is still emerging from the noise of natural variability,
most of the observed warming over the last ¬fty years is likely to have
been due to the increase in greenhouse gas concentrations.3
Signi¬cant cooling of the lower stratosphere (the region at altitudes
between about 10 and 30 km) has been observed over the last two decades
(Figure 4.2). This is to be expected both because of the decrease in the
concentration of ozone (which absorbs solar radiation) and because of the
increased carbon dioxide concentration which leads to increased cooling
at these levels (see Chapter 2).
A further source of information regarding climate change comes
from measurements of change in sea level. Over the last hundred years sea
level has risen by between 10 and 20 cm. The best known contributions
to this rise are from the thermal expansion of ocean water because of
the global average temperature rise (estimated as up to about 7 cm)
62 Climates of the past

Table 4.1 Twentieth-century changes in the Earth™s atmosphere, climate and biophysical system

Indicator Observed changes

Concentration indicators
280 ppm for the period 1000“1750 to 368 ppm in year 2000 (31 ± 4%
Atmospheric concentration
of CO2 increase)
Terrestrial biospheric CO2 Cumulative source of about 30 GtC between the years 1800 and 2000;
but during the 1990s a net sink of about 14 ± 7 GtC
700 ppb for the period 1000“1750 to 1750 ppb in year 2000 (151 ± 25%
Atmospheric concentration
of CH4 increase)
270 ppb for the period 1000“1750 to 316 ppb in the year 2000 (17 ± 5%
Atmospheric concentration
of N2 O increase)
Increased by 35 ± 15% from the years 1750 to 2000, varies with region
Tropospheric concentration
of O3
Stratospheric concentration Decreased over the years 1970 to 2000, varies with altitude and latitude
of O3
Atmospheric Increased globally over the last ¬fty years
concentrations of
HFCs, PFCs and SF6

Weather indicators
Increased by 0.6 ± 0.2 —¦ C over the twentieth century; land areas warned
Global mean surface
temperature more than the oceans (very likely )
Northern hemisphere Increase over the twentieth century greater than during any other
surface temperature century in the last 1000 years; 1990s warmest decade of the
millennium (likely )
Diurnal surface temperature Decreased over the years 1950 to 2000 over land; night-time minimum
range temperatures increased at twice the rate of daytime maximum
temperatures (likely )
Hot days/heat index Increased (likely )
Cold/frost days Decreased for nearly all land areas during the twentieth century (very
likely )
Continental precipitation Increased by ¬ve to ten per cent over the twentieth century in the
Northern Hemisphere (very likely ), although decreased in some
regions (e.g. north and west Africa and parts of the Mediterranean)
Heavy precipitation events Increased at mid and high northern latitudes (likely )
Frequency and severity of Increased summer drying and associated incidence of drought in a few
drought areas (likely ). In some regions, such as parts of Asia and Africa, the
frequency and intensity of droughts have been observed to increase in
recent decades
The last hundred years 63

Table 4.1 (cont.)

Indicator Observed changes

Biological and physical indicators
Global mean sea level Increased at an average annual rate of 1“2 mm during the twentieth
Duration of ice cover of Decreased by about two weeks over the twentieth century in mid and
rivers and lakes high latitudes of the Northern Hemisphere (very likely )
Arctic sea-ice extent and Thinned by forty per cent in recent decades in late summer to early
thickness autumn (likely ) and decreased in extent by ten to ¬fteen per cent since
the 1950s in spring and summer
Non-polar glaciers Widespread retreat during the twentieth century
Snow cover Decreased in area by ten per cent since global observations became
available from satellites in the 1960s (very likely )
Permafrost Thawed, warmed and degraded in parts of the polar, sub-polar and
mountainous regions
El Ni˜ o events
n Became more frequent, persistent and intense during the last twenty to
thirty years compared to the previous 100 years
Growing season Lengthened by about one to four days per decade during the last forty
years in the Northern Hemisphere, especially at higher latitudes
Plant and animal ranges Shifted poleward and up in elevation for plants, insects, birds and ¬sh
Breeding, ¬‚owering and Earlier plant ¬‚owering, earlier bird arrival, earlier dates of breeding
migration season and earlier emergence of insects in the Northern Hemisphere
Coral reef bleaching Increased frequency, especially during El Ni˜ o events

Economic indicators
Weather-related economic Global in¬‚ation-adjusted losses rose an order of magnitude over the last
losses forty years. Part of the observed upward trend is linked to
socio-economic factors and part is linked to climatic factors

This table provides examples of key observed changes and is not an exhaustive list. It includes both changes
attributable to anthropogenic climate change and those that may be caused by natural variations or anthropogenic
climate change. Con¬dence levels (for explanation see note 1) are reported where they are explicitly assessed by
the relevant Working Group of the IPCC.
Source: Table SPM-1 from IPCC 2001 Synthesis Report.

and from glaciers which have generally been retreating over the last
century (estimated as up to about 4 cm). The net contribution from the
Greenland and Antarctic ice caps is more uncertain but is believed to be
In Chapter 1, we mentioned the increasing vulnerability of hu-
man populations to climate extremes, which has brought about more
64 Climates of the past

awareness of recent extremes in the forms of ¬‚oods, droughts, tropical
cyclones and windstorms. It is therefore of great importance to know
whether there is evidence of an increase in the frequency or severity of
these and other extreme events. The available evidence regarding how
these and other relevant parameters have changed during the twentieth
century is summarised in Table 4.1 in terms of different indicators: con-
centrations of greenhouse gases; temperature, hydrological and storm-
related indicators and biological and physical indicators. To what extent
these changes are expected to continue or to intensify during the twenty-
¬rst century will be addressed in Chapter 6.
Eventually, as greenhouse gases increase further, the amount of
warming is expected to become suf¬ciently large that it will swamp
the natural variations in climate. In the meantime, the global average
temperature may continue to increase or it could, because of natural
variability, show periods of decrease. Over the next few years, scientists
will be inspecting climate changes and climate events most carefully
as they occur to see how far actual events can be related to scienti¬c
predictions especially those associated with increasing greenhouse gas
emissions. Some details of these predictions will be discussed in later
chapters especially Chapter 6.

The last thousand years
The detailed systematic record of weather parameters such as tem-
perature, rainfall, cloudiness and the like presented above for the last
140 years and which covers a good proportion of the globe is not avail-
able for earlier periods. Further back, the record is more sparse and doubt
arises over the consistency of the instruments used for observation. Most
thermometers in use 200 years ago were not well calibrated or carefully
exposed. However, many diarists and writers kept records at different
times; from a wide variety of sources weather and climate information
can be pieced together. Indirect sources, such as are provided by ice cores,
tree rings and records of lake levels, of glacier advance and retreat, and
of pollen distribution, can also yield information to assist in building
up the whole climatic story. From a variety of sources, for instance, it
has been possible to put together for China a systematic atlas of weather
patterns covering the last 500 years.
Similarly, from direct and indirect sources, it has been possible to
deduce the average temperature over the Northern Hemisphere for the
last millennium (Figure 4.3). Suf¬cient data are not available for the
same reconstruction to be carried out over the Southern Hemisphere.
In Figure 4.3 it is just possible to identify the ˜Medieval Warm Period™
associated with the eleventh to fourteenth centuries and a relatively cool
The last thousand years 65

Instrumental data (AD 1902 to 1999)
Reconstruction (AD 1000 to 1980)
Reconstruction (40 year smoothed)
1998 instrumental value
Northern Hemisphere anomaly (°C)

relative to 1961 to 1990



1000 1200 1400 1600 1800 2000
Figure 4.3 Millennial Northern Hemisphere temperature reconstruction from
tree rings, corals, ice cores and historical records (for the period 1000“1980) and
from instrumental data (for 1902“1999). A smoothed version of the proxy
record is also shown, as are the 95% con¬dence limits (grey shaded).

period the ˜Little Ice Age™ associated with the ¬fteenth to nineteenth
centuries. These only affected part of the Northern Hemisphere and
are therefore more prominent in local records, for instance those from
central England. The increase in temperature over the twentieth century
is particularly striking. The 1990s are likely to have been the warmest
decade of the millennium in the Northern Hemisphere and 1998 is likely
to have been the warmest year.
Although there is as yet no certain explanation for the variations that
occurred between 1000 and 1900, it is clear that greenhouse gases such
as carbon dioxide and methane cannot have been the cause of change.
For the millennium before 1800 their concentration in the atmosphere
was rather stable, the carbon dioxide concentration, for instance, vary-
ing by less than three per cent. However, the combined in¬‚uences of
variations in volcanic activity and variations in the output of energy
from the Sun can provide some part of the explanation.4 The effect of
individual volcanic eruptions can be very noticeable. For instance, one of
the largest eruptions during the period was that of Tambora in Indonesia
66 Climates of the past

in April 1815, which was followed in many places by two exception-
ally cold years: 1816 was described in New England and Canada as the
˜year without a summer™. Although the effect on the climate even of
an eruption of the magnitude of Tambora only lasts a few years, varia-
tions in average volcanic activity have a longer term effect. Regarding
the Sun™s output, although accurate direct measurements are not avail-
able (apart from those made during the last two decades from satellite
instruments), other evidence suggests that the solar output could have
varied signi¬cantly in the past. For instance, compared with its value
today it may have been somewhat lower (by a few tenths of a watt per
square metre) during the Maunder Minimum in the seventeenth century
(a period when almost no sunspots were recorded; see also box on p. 138).
There is no need, however, to invoke volcanoes or variations in solar
output as the cause of all the climate variations over this period. As
with the shorter-term changes mentioned earlier, such variations of cli-
mate can arise naturally from internal variations within the atmosphere
and the ocean and in the two-way relationship “ coupling “ between
The millennial record of Figure 4.3 is particularly important be-
cause it provides an indication of the range and character of climate
variability that arises from natural causes. As we shall see in the next
chapter, climate models also provide some information on natural cli-
mate variability. Careful assessments of these observational and model
results con¬rm that natural variability (the combination of internal vari-
ability and naturally forced, e.g. by volcanoes or change in solar output)
is unlikely to explain the warming in the latter half of the twentieth

The past million years
To go back before recorded human history, scientists have to rely on
indirect methods to unravel much of the story of the past climate. A
particularly valuable information source is the record stored in the ice that
caps Greenland and the Antarctic continent. These ice caps are several
thousands of metres thick. Snow deposited on their surface gradually
becomes compacted as further snow falls, becoming solid ice. The ice
moves steadily downwards, eventually ¬‚owing outwards at the bottom of
the ice-sheet. Ice near the top of the layer will have been deposited fairly
recently; ice near the bottom will have fallen on the surface many tens
or hundreds of thousands of years ago. Analysis of the ice at different
levels can, therefore, provide information about the conditions prevailing
at different times in the past.
The past million years 67

Deep cores have been drilled out of the ice at several locations in both
Greenland and Antarctica. At Russia™s Vostok station in east Antarctica,
for instance, drilling has been carried out for over twenty years. The
longest and most recent core reached a depth of over 3.5 km; the ice
at the bottom of the hole fell as snow on the surface of the Antarctic
continent over 400 000 years ago.
Small bubbles of air are trapped within the ice. Analysis of the com-
position of that air shows what was present in the atmosphere for the

Paleoclimate reconstruction from isotope data
The isotope 18 O is present in natural oxygen at a concentration of about 1
part in 500 compared with the more abundant isotope 16 O. When water
evaporates, water containing the lighter isotope is more easily vapor-
ised, so that water vapour in the atmosphere contains less 18 O compared
with sea water. Similar separation occurs in the process of condensation
when ice crystals form in clouds. The amount of separation between the
two oxygen isotopes in these processes depends on the temperatures at
which evaporation and condensation occur. Measurements on snowfall
in different places can be used to calibrate the method; it is found that the
concentration of 18 O varies by about 0.7 of a part per thousand for each
degree of change in average temperature at the surface. Information is
therefore available in the ice cores taken from polar ice caps concern-
ing the variation in atmospheric temperature in polar regions during the
whole period when the ice core was laid down.
Since the ice caps are formed from accumulated snowfall which
contains less 18 O compared with sea water, the concentration of 18 O in


. 3
( 13)