<<

. 4
( 13)



>>

water from the oceans provides a measure of the total volume of the
ice in the ice caps; it changes by about one part in 1000 between the
maximum ice extent of the ice ages and the warm periods in between.
Information about the 18 O content of ocean water at different times is
locked up in corals and in cores of sediment taken from the ocean bot-
tom, which contain carbonates from fossils of plankton and small sea
creatures from past centuries and millennia. Measurements of radioac-
tive isotopes, such as the carbon isotope 14 C, and correlations with other
signi¬cant past events enable the corals and sediment cores to be dated.
Since the separation between the oxygen isotopes which occurs as these
creatures are formed also depends on the temperature of the sea water
(although the dependence is weaker than the other dependencies consid-
ered above) information is also available about the distribution of ocean
surface temperature at different times in the past.6
68 Climates of the past



time at which the ice was formed “ gases such as carbon dioxide or
methane. Dust particles that may have come from volcanoes or from
the sea surface are also contained within the ice. Further information is
provided by analysis of the ice itself. Small quantities of different oxygen
isotopes and of the heavy isotope of hydrogen (deuterium) are contained
in the ice. The ratios of these isotopes that are present depend sensitively
on the temperatures at which evaporation and condensation took place
for the water in the clouds from which the ice originated (see box above).
These in turn are dependent on the average temperature near the surface
of the Earth. A temperature record for the polar regions can therefore
be constructed from analyses of the ice cores. The associated changes in
global average temperature are estimated to be about half the changes in
the polar regions.
Such a reconstruction from a Vostok core for the temperature and
the carbon dioxide content is shown in Figure 4.4 for the past 160 000
years, which includes the last major ice age that began about 120 000
years ago and began to come to an end about 20 000 years ago. It also
demonstrates the close connections that exist between temperature and
carbon dioxide concentrations. Similar close correlation is found with
the methane concentration. Note from Figure 4.4 the likely growth of
atmospheric carbon dioxide during the twenty-¬rst century, taking it to
levels that are unlikely to have been exceeded during the past twenty
million years.
Data from ice cores can take us back 400 000 years or so over four
ice age cycles during which the correlations between temperature and
carbon dioxide concentrations shown in Figure 4.4 are repeated.7 To
go further back, over the past million years, the composition of ocean
sediments can be investigated to yield information. Fossils of plankton
and other small sea creatures deposited in these sediments also contain
different isotopes of oxygen. In particular the amount of the heavier
isotope of oxygen (18 O) compared with the more abundant isotope (16 O)
is sensitive both to the temperature at which the fossils were formed and
to the total volume of ice in the world™s ice caps at the time of the fossils™
formation (see box above). For instance, from oxygen isotope and other
data we can deduce that the sea level at the last glacial maximum, 20 000
years ago, was about 120 m lower than today.
From the variety of paleoclimate data available, variations in the
volume of ice in the ice caps can be reconstructed over the greater part
of the last million years (Figure 4.5(c)). In this record six or seven major
ice ages can be identi¬ed with warmer periods in between, the period
between these major ice ages being approximately 100 000 years. Other
cycles are also evident in the record.
The past million years 69




Figure 4.4 Variations over the last 160 000 years of polar temperature and
atmospheric carbon dioxide concentrations derived from the Vostok ice core
from Antarctica. It is estimated that the variation of global average temperature
is about half that in the polar regions. Also shown is the current carbon dioxide
concentration of about 370 ppm and the likely rise during the twenty-¬rst
century under various projections of its growth.



The most obvious place to look for the cause of regular cycles in
climate is outside the Earth, in the Sun™s radiation. Has this varied in the
past in a cyclic way? So far as is known the output of the Sun itself has not
changed to any large extent over the last million years or so. But because
of variations in the Earth™s orbit, the distribution of solar radiation has
varied in a more or less regular way during the last millennium.
Three regular variations occur in the orbit of the Earth around the
Sun (Figure 4.5(a)). The Earth™s orbit, although nearly circular, is actually
an ellipse. The eccentricity of the ellipse (which is related to the ratio
between the greatest and the least diameters) varies with a period of
about 100 000 years; that is the slowest of the three variations. The Earth
also spins on its own axis, the axis of spin being tilted with respect
to the axis of the Earth™s orbit, the angle of tilt varying between 21.6—¦
and 24.5—¦ (currently it is 23.5—¦ ) with a period of about 41 000 years.
70 Climates of the past



The third variation is of the time of year when the Earth is closest to
the Sun (the Earth™s perihelion). The time of perihelion moves through
the months of the year with a period of about 23 000 years (see also
Figure 5.19); in the present con¬guration, the Earth is closest to the Sun in
January.
As the Earth™s orbit changes its relationship to the Sun, although the
total quantity of solar radiation reaching the Earth varies very little, the
distribution of that radiation with latitude and season over the Earth™s
surface changes considerably. The changes are especially large in polar
regions where the variations in summer sunshine, for instance, reach
about ten per cent (Figure 4.5(b)). James Croll, a British scientist, ¬rst
pointed out in 1867 that the major ice ages of the past might be linked
with these regular variations in the seasonal distribution of solar radi-
ation reaching the Earth. His ideas were developed in 1920 by Milutin
Milankovitch, a climatologist from Yugoslavia, whose name is usually
linked with the theory. Inspection by eye of the relationship between the
variations of polar summer sunshine and global ice volume shown in
Figure 4.5 suggests a signi¬cant connection. Careful study of the corre-
lation between the two curves con¬rms this and demonstrates that sixty
per cent of the variance in the climatic record of global ice volume falls
close to the three frequencies of regular variations in the Earth™s orbit,
thus providing support for the Milankovitch theory.
More careful study of the relationship between the ice ages and the
Earth™s orbital variations shows that the size of the climate changes is
larger than might be expected from forcing by the radiation changes
alone. Other processes that enhance the effect of the radiation changes
(in other words, positive feedback processes) have to be introduced to ex-
plain the climate variations. One such feedback arises from the changes in
carbon dioxide in¬‚uencing atmospheric temperature through the green-
house effect, illustrated by the strong correlation observed in the climatic
record between average atmospheric temperature and carbon dioxide
concentration (Figure 4.4). Such a correlation does not, of course, prove
the existence of the greenhouse feedback; in fact part of the correlation
arises because the atmospheric carbon dioxide concentration is itself in-
¬‚uenced, through biological feedbacks (see Chapter 3), by factors related
to the average global temperature. However, as we shall see in Chapter 5,
climates of the past cannot be modelled successfully without taking the
greenhouse feedback into account.8
An obvious question to ask is when, on the Milankovitch theory, is
the next ice age due? It so happens that we are currently in a period of
relatively small solar radiation variations and the best projections for the
long term are of a longer than normal interglacial period leading to the
beginning of a new ice age perhaps in 50 000 years™ time.9
How stable has past climate been? 71



Figure 4.5 Variations in
(a) 21.6 degrees
the Earth™s orbit (a), in its
eccentricity, the orientation
24.5 degrees
of its spin axis (between
Sun Earth
21.6—¦ and 24.5—¦ ) and the
longitude of perihelion (i.e.
the time of year when the
Earth is closest to the Sun,
see also Figure 5.19), cause
(b) (c)
changes in the average
0
amount of summer
sunshine (in millions of
joules per square metre per
day) near the poles (b).
100
These changes appear as
cycles in the climate record
in terms of the volume of
ice in the ice caps (c).
200
Thousands of years ago




300




400




500




600

2.2 2.4 Ice volume
Summer sunshine




How stable has past climate been?
The major climate changes considered so far in this chapter have taken
place relatively slowly. The growth and recession of the large polar
ice-sheets between the ice ages and the intervening warmer interglacial
periods have taken on average many thousands of years. However, the ice
core records such as those in Figures 4.4 and 4.6 show evidence of large
and relatively rapid ¬‚uctuations. Ice cores from Greenland provide more
detailed evidence of these than those from Antarctica. This is because at
the summit of the Greenland ice cap, the rate of accumulation of snow
72 Climates of the past




Figure 4.6 Variations in Arctic temperature over the past 100 000 years as
deduced from oxygen isotope measurements (in terms of δ18 O) from the
˜Summit™ ice core in Greenland. The quantity δ18 O plotted in Figures 4.6 and 4.7
is the difference (in parts per thousand) between the 18 O/16 O ratio in the sample
and the same ratio in a laboratory standard. The overall shape of the record is
similar to that from the Vostok ice core shown in Figure 4.4 but much more
detail is apparent in the ˜Summit™ record™s stable period over the last 8000 years.
A change of ¬ve parts per thousand in δ18 O in the ice core corresponds to about
a 7 —¦ C change in temperature.


has been higher than that at the Antarctica drilling locations. For a given
period in the past, the relevant part of the Greenland ice core is longer
and more detail of variations over relatively short periods is therefore
available.
The data show that the last 8000 years have been unusually stable
compared with earlier epochs. In fact, as judged from the Vostok (Fig-
ure 4.4) and the Greenland records (Figure 4.6) this long stable period in
the Holocene is a unique feature of climate during the past 420 000 years.
It has been suggested that this had profound implications for the devel-
opment of civilisations.10 Model simulations (see Chapter 5) indicate
that the detail of long-term changes during the Holocene is consistent
with the in¬‚uence of orbital forcing (Figure 4.5).
It is also interesting to inspect the rate of temperature change during
the recovery period from the last glacial maximum about 20 000 years
ago and compare it with recent temperature changes. The data indicate
an average warming rate of about 0.2 —¦ C per century between 20 000
and 10 000 years before present (BP) over Greenland, with lower rates
How stable has past climate been? 73



for other regions. Compare this with a temperature rise during the twen-
tieth century of about 0.6 —¦ C and the rates of change of a few degrees
centigrade per century projected to occur during the twenty-¬rst century
because of human activities (see Chapter 6).
The ice core data (Figures 4.6) demonstrate that a series of rapid
warm and cold oscillations called Dansgaard“Oeschger events punctu-
ated the last glaciation. Comparison between the results from ice cores
drilled at different locations within the Greenland ice cap con¬rm the
details up to about 100 000 years ago. Comparison with data from Antarc-
tica suggests that the ¬‚uctuations of temperature over Greenland (perhaps
up to 16 —¦ C) have been larger than those over Antarctica. Similar large
and relatively rapid variations are evident from North Atlantic deep sea
sediment cores.
Another particularly interesting period of climatic history, more re-
cently, is the Younger Dryas event (so called because it was marked by
the spread of an arctic ¬‚ower, Dryas octopetala), which occurred over a
period of about 1500 years between about 12 000 and 10 700 years ago.
For 6000 years before the start of this event the Earth had been warming
up after the end of the last ice age. But then during the Younger Dryas
period, as demonstrated from many different sources of paleoclimatic
data, the climate swung back again into much colder conditions similar
to those at the end of the last ice age (Figure 4.7). The ice core record
shows that at the end of the event, 10 700 years ago, the warming in
the Arctic of about 7 —¦ C occurred over only about ¬fty years and was
associated with decreased storminess (shown by a dramatic fall in the
amount of dust in the ice core) and an increase of precipitation of about
¬fty per cent.
Two main reasons for these rapid variations in the past have been
suggested. One reason particularly applicable to ice age conditions is
that, as the ice-sheets over Greenland and eastern Canada have built
up, major break-ups have occurred from time to time, releasing massive
numbers of icebergs into the North Atlantic in what are called Heinrich
events. The second possibility is that the ocean circulation in the North
Atlantic region has been strongly affected by injections of fresh water
from the melting of ice. At present the ocean circulation in this region
is strongly in¬‚uenced by cold salty water sinking to deep ocean levels
because its saltiness makes it dense; this sinking process is part of the
˜conveyor belt™ which is the major feature of the circulation of deep ocean
water around the world (see Figure 5.18). Large quantities of fresh water
from the melting of ice would make the water less salty, preventing it
from sinking and thereby altering the whole Atlantic circulation.
This link between the melting of ice and the ocean circulation
is a key feature of the explanation put forward by Professor Wallace
74 Climates of the past



’35 ’30
’38 ’36 ’34 ’32 ’30 ’28 ° 1700
δ18O °
δ18O 8500 yr BP

DYE 3




Pre-Boreal
Deep ice core
9000 yr BP


1785
9.500 yr BP
Gerzensee 1750
Lake sediments
’10 ’8 ’6 °
∼50yr
1
Depth of sediment




1786
10.700 BP

Younger Dryas
2
Allerod 1800 7°C
Bolling
3
1787




Younger Dryas
m
4
m
Depth of
ice core 1850
m

Figure 4.7 Records of the variations of the oxygen isotope δ18 O from lake
sediments from Lake Gerzen in Switzerland and from the Greenland ice core
˜Dye 3™ showing the ˜Younger Dryas™ event and its rapid end about 10 700 years
ago. Dating of the ice core was by counting the annual layers down from the
surface; dating of the lake sediment was by the 14 C method. A change of ¬ve
parts in a thousand in δ18 O in the ice core corresponds to about a 7 —¦ C change
in temperature.



Broecker for the Younger Dryas event.11 As the great ice-sheet over
north America began to melt at the end of the last ice age, the melt water
at ¬rst drained through the Mississippi into the Gulf of Mexico. Even-
tually, however, the retreat of the ice opened up a channel for the water
in the region of the St Lawrence river. This in¬‚ux of fresh water into the
North Atlantic reduced its saltiness, thus, Broecker postulates, cutting
off the formation of deep water and that part of the ocean ˜conveyor
belt™.12 Warm water was therefore prevented from ¬‚owing northward,
resulting in a reversal to much colder conditions. The suggestion is
also that a reversal of this process with the starting up of the Atlantic
˜conveyor belt™ could lead to a sudden onset of warmer conditions.
Although debate continues regarding the details of the Younger
Dryas event, there is considerable evidence from paleodata, especially
those from ocean sediments, for the main elements of the Broecker
Notes 75



explanation which involve the deep ocean circulation. It is also clear
from paleodata that large changes have occurred at different times in the
past in the formation of deep water and in the deep ocean circulation.
Chapter 3 mentioned the possibility of such changes being induced by
global warming through the growth of greenhouse gas concentrations.
Our perspective regarding the possibilities of future climate change needs
to take into account the rapid climate changes that have occurred in the
past.
Having now in these early chapters set the scene, by describing the
basic science of global warming, the greenhouse gases and their origins
and the current state of knowledge regarding past climates, I move on in
the next chapter to describe how, through computer models of the climate,
predictions can be made about what climate change can be expected in
the future.

Questions
1 Given that the sea level at the end of the last glacial maximum was 120 m
lower than that today, estimate the volume of ice in the ice-sheets that covered
the northern parts of the American and Euroasian continents.
2 How much energy would be required to melt the volume of ice you have
calculated in question 1? Compare this with the extra summer sunshine
north of latitude 60—¦ which might have been available between 18 000 and
6000 years before the present according to the data in Figure 4.5. Does your
answer support the Milankovitch theory?
3 It is sometimes suggested that the large reserves of fossil fuels on Earth
should be preserved until the onset of the next ice age is closer so that some
of its impact can be postponed. From what you know of the greenhouse effect
and of the behaviour of carbon dioxide in the atmosphere and the oceans,
consider the in¬‚uences that human burning of the known reserves of fossil
fuels “ see Figure 11.2 “ could have on the onset of the next ice age.



Notes for Chapter 4
1 In the IPCC 2001 Report, expressions of certainty such as ˜very likely™ were
related so far as possible to quantitative statement of con¬dence, virtually
certain (greater than ninety-nine per cent chance that a result is true), very
likely (ninety to ninety-nine per cent chance), likely (sixty-six to ninety per
cent chance), medium likelihood (thirty-three to sixty-six per cent chance),
unlikely (ten to thirty-three per cent chance), very unlikely (one to ten per
cent chance), exceptionally unlikely (less than one per cent chance).
2 Further information in Folland, C. K. et al. (Chapter 2) and Mitchell, J. F. B.
et al. (Chapter 12). 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
76 Climates of the past



the Third Assessment Report of the Intergovernmental Panel on Climate
Change. Cambridge: Cambridge University Press.
3 From Summary for policymakers, Houghton et al., Climate Change 2001:
The Scienti¬c Basis.
4 See, for instance, Crowley, T. J. 2000. Causes of climate change over the
past 1000 years. Science, 289, 270“7.
5 More information in Mitchell, J. F. B. et al. (Chapter 12). In Houghton et al.
Climate Change 2001.
6 For more information about paleoclimates and how they are investigated
see, for instance, Crowley, T. L. and North, G. R. 1991. Paleoclimatology.
Oxford: Oxford University Press.
7 Folland, C. K. et al. (Chapter 2). Figure 2.22. In Houghton et al., Climate
Change 2001.
8 See also Raynaud, D. et al., 1993. The ice core record of greenhouse gases.
Science, 259, 926“34.
9 Berger, A. and Loutre, M. F. 2002. Science, 297, 1287“8.
10 Petit, J. R. et al. 1999. Nature, 399, 429“36.
11 Broecker, W. S. and Denton, G. H. 1990. What drives glacial cycles? Scien-
ti¬c American, 262, 43“50.
12 More information in Chapter 5, see especially Figure 5.18.
Chapter 5
Modelling the climate




Chapter 2 looked at the greenhouse effect in terms of a simple radiation
balance. That gave an estimate of the rise in the average temperature at
the surface of the Earth as greenhouse gases increase. But any change in
climate will not be distributed uniformly everywhere; the climate system
is much more complicated than that. More detail in climate change pre-
diction requires very much more elaborate calculations using computers.
The problem is so vast that the fastest and largest computers available
are needed. But before computers can be set to work on the calculation,
a model of the climate must be set up for them to use.1 A model of the
weather as used for weather forecasting will be used to explain what is
meant by a numerical model on a computer, followed by a description
of the increase in elaboration required to include all parts of the climate
system in the model.




Modelling the weather
An English mathematician, Lewis Fry Richardson, set up the ¬rst nu-
merical model of the weather. During his spare moments while working
for the Friends™ Ambulance Unit (he was a Quaker) in France during the
First World War he carried out the ¬rst numerical weather forecast. With
much painstaking calculation with his slide-rule, he solved the appropri-
ate equations and produced a six-hour forecast. It took him six months “
and then it was not a very good result. But his basic methods, described
in a book published in 1922,2 were correct. To apply his methods to real
forecasts, Richardson imagined the possibility of a very large concert
hall ¬lled with people, each person carrying out part of the calculation,

77
78 Modelling the climate




Figure 5.1 Illustrating the growth of computer power available at major
forecasting centres. The computers are those used by the UK Meteorological
Of¬ce for numerical weather prediction research, since 1965 for operational
weather forecasting and most recently for research into climate prediction.
Richardson™s computer refers to his dream of a large ˜human™ computer
mentioned at the beginning of the chapter.



so that the integration of the numerical model could keep up with the
weather. But he was many years before his time! It was not until some
forty years later that, essentially using Richardson™s methods, the ¬rst
operational weather forecast was produced on an electronic computer.
Computers more than one million times faster than the one used for that
¬rst forecast (Figure 5.1) now run the numerical models that are the basis
of all weather forecasts.
Numerical models of the weather and the climate are based on the
fundamental mathematical equations that describe the physics and dy-
namics of the movements and processes taking place in the atmosphere,
the ocean, the ice and on the land. Although they include empirical
information, they are not based to any large degree on empirical rela-
tionships “ unlike numerical models of many other systems, for instance
in the social sciences.
Setting up a model of the atmosphere for a weather forecast (see Fig-
ure 5.2) requires a mathematical description of the way in which energy
from the Sun enters the atmosphere from above, some being re¬‚ected
by the surface or by clouds and some being absorbed at the surface or
in the atmosphere (see Figure 2.6). The exchange of energy and water
vapour between the atmosphere and the surface must also be described.
Modelling the weather 79



Figure 5.2 Schematic
illustrating the parameters
and physical processes
involved in atmospheric
models.




Water vapour is important because of its associated latent heat (in
other words, it gives out heat when it condenses) and also because the
condensation of water vapour results in cloud formation, which modi¬es
substantially the interaction of the atmosphere with the incoming energy
from the Sun. Variations in both these energy inputs modify the atmo-
spheric temperature structure, causing changes in atmospheric density
(since warmed gases expand and are therefore less dense). It is these
density changes that drive atmospheric motions such as winds and air
currents, which in their turn alter and feed back on atmospheric density
and composition. More details of the model formulation are given in the
box below.
To forecast the weather for several days ahead a model covering the
whole globe is required; for example, the southern hemisphere circula-
tion today will affect northern hemisphere weather within a few days and
vica versa. In a global forecasting model, the parameters (i.e. pressure,
temperature, humidity, wind velocity and so on) that are needed to de-
scribe the dynamics and physics (listed in the box below) are speci¬ed
at a grid of points (Figure 5.3) covering the globe. A typical spacing
between points in the horizontal would be 100 km and about 1 km in
the vertical; typically there would be twenty levels or so in the model in
the vertical. The ¬neness of the spacing is limited by the power of the
computers currently available.
Having set up the model, to generate a forecast from the present, it is
started off from the atmosphere™s current state and then the equations are
integrated forward in time (see box below) to provide new descriptions of
the atmospheric circulation and structure up to six or more days ahead.
For a description of the atmosphere™s current state, data from a wide
variety of sources (see box below) have to be brought together and fed
into the model.
80 Modelling the climate




Setting up a numerical atmospheric model
A numerical model of the atmosphere contains descriptions, in appro-
priate computer form and with necessary approximations, of the basic
dynamics and physics of the different components of the atmosphere and
their interactions.3 When a physical process is described in terms of an
algorithm (a process of step-by-step calculation) and simple parameters
(the quantities that are included in a mathematical equation), the process
is said to have been parameterised.
The dynamical equations are:
r The horizontal momentum equations (Newton™s Second Law of
Motion). In these, the horizontal acceleration of a volume of air
is balanced by the horizontal pressure gradient and the friction.
Because the Earth is rotating, this acceleration includes the Cori-
olis acceleration. The ˜friction™ in the model mainly arises from
motions smaller than the grid spacing, which have to be parame-
terised.
r The hydrostatic equation. The pressure at a point is given by the
mass of the atmosphere above that point. Vertical accelerations are
neglected.
r The continuity equation. This ensures conservation of mass.

The model™s physics consists of:
r The equation of state. This connects the quantities of pressure,
volume and temperature for the atmosphere.
r The thermodynamic equation (the law of conservation of energy).
r Parameterisation of moist processes (such as evaporation, conden-
sation, formation and dispersal of clouds).
r Parameterisation of absorption, emission and re¬‚ection of solar
radiation and of thermal radiation.
r Parameterisation of convective processes.
r Parameterisation of exchange of momentum (in other words, fric-
tion), heat and water vapour at the surface.
Most of the equations in the model are differential equations, which
means they describe the way in which quantities such as pressure and
wind velocity change with time and with location. If the rate of change of
a quantity such as wind velocity and its value at a given time are known,
then its value at a later time can be calculated. Constant repetition of
this procedure is called integration. Integration of the equations is the
process whereby new values of all necessary quantities are calculated at
later times, providing the model™s predictive powers.
Modelling the weather 81



Figure 5.3 Illustration of
a model grid. The levels in
the vertical are not equally
spaced; the top level is
typically at about 30 km in
altitude.




Since computer models for weather forecasting were ¬rst introduced,
their forecast skill has improved to an extent beyond any envisaged by
those involved in the development of the early models. As improvements
have been made in the model formulation, in the accuracy or coverage
of the data used for initialisation (see box above) or in the resolution of
the model (the distance between grid points), the resulting forecast skill
has increased. For instance, for the British Isles, three-day forecasts of
surface pressure today are as skilful on average as two-day forecasts of
ten years ago, as can be seen from Figure 5.5.
When looking at the continued improvement in weather forecasts, the
question obviously arises as to whether the improvement will continue or
whether there is a limit to the predictability we can expect. Because the
atmosphere is a partially chaotic system (see box below), even if perfect
observations of the atmospheric state and circulation could be provided,
there would be a limit to our ability to forecast the detailed state of the
atmosphere at some time in the future. In Figure 5.6 current forecast skill
is compared with the best estimate of the limit of the forecast skill for
the British Isles (similar results would be obtained with any other mid-
latitude situation) with a perfect model and near perfect data. According
to that estimate, the limit of signi¬cant future skill is about 20 days
ahead.
82 Modelling the climate




Data to initialise the model
At a major global weather forecasting centre, data receipt of data from around the world a dedi-
from many sources are collected and fed into cated communication network has been set up, used
the model. This process is called initialisation. solely for this purpose. Great care needs to be taken
Figure 5.4 illustrates some of the sources of data with the methods for assimilation of the data into
for the forecast beginning at 1200 hours Universal the model as well as with the data™s quality and
Time (UT) on 1 July 1990. To ensure the timely accuracy.




Figure 5.4 Illustrating some of the sources of data for input into the UK Meteorological Of¬ce global
weather forecasting model on a typical day. Surface observations are from land observing stations
(manned and unmanned), from ships and from buoys. Radiosonde balloons make observations at up to
30 km altitude from land and from ship-borne stations. Satellite soundings are of temperature and
humidity at different atmospheric levels deduced from observations of infrared or microwave radiation.
Satellite cloud-track winds are derived from observing the motion of clouds in images from geostationary
satellites.



Forecast skill varies considerably between different weather situa-
tions or patterns. In other words some situations are more ˜chaotic™ (in
the technical sense in which that word is used “ see box below) than
others. One way of identifying the skill that might be achieved in a
Figure 5.5 Errors (root mean square differences of forecasts of surface pressure
compared with analyses) of UK Meteorological Of¬ce forecasting models for the
north Atlantic and Western Europe since 1966 for 24-h, 48-h and 72-h forecasts.
Note that 1 hPa = 1 mbar.




Figure 5.6 Illustrating potential improvements in forecast skill if there were
better data or a better model. The ordinate (vertical axis) is a measure of the
error of model forecasts (it is the root mean square differences of forecasts of the
500 hPa height ¬eld compared with analyses). Curve (a) is the error of 1990 UK
Meteorological Of¬ce forecasts as a function of forecast range. Curve (b) is an
estimate showing how, with the same initial data, the error would be reduced if
a perfect model could be used. Curve (c) is an estimate showing the further
improvement which might be expected if near-perfect data could be provided
for the initial state. After a suf¬ciently long period, all the curves approach a
saturation value of the average root mean square difference between any
forecasts chosen at random.
Weather forecasting and chaos4

The science of chaos has developed rapidly since only moves from about six days to about twenty
the 1960s (when a meteorologist Edward Lorenz days, because the atmosphere is a chaotic system.
was one of its pioneers) along with the power of For the simple pendulum not all situations are
electronic computers. In this context, chaos is a chaotic. Not surprisingly, therefore, in a system
term with a particular technical meaning. A chaotic as complex as the atmosphere, some occasions
system is one whose behaviour is so highly sensi- are more predictable than others. A good illustra-
tive to the initial conditions from which it started tion of an occasion with particular sensitivity to
that precise future prediction is not possible. Even the initial data and to the way in which the data
quite simple systems can exhibit chaos under some were assimilated into the model is provided by
conditions. For instance, the motion of a simple the Meteorological Of¬ce forecasts for the storm
pendulum (Figure 5.7) can be ˜chaotic™ under some which hit southeast England in the early hours of
circumstances, and, because of its extreme sensitiv- Friday, 16 October 1987. During this storm gusts of
over 90 knots (170 km h’1 ) were recorded and ap-
ity to small disturbances, its detailed motion is not
then predictable. proximately ¬fteen million trees were blown down.
A condition for chaotic behaviour is that the Although as early as the previous Sunday forecasts
relationship between the quantities which govern had given good early warning of a storm of un-
the motion of the system be non-linear; in other usual severity, the model forecasts available during
words, a description of the relationship on a graph 15 October gave much poorer guidance than earlier
would be a curve rather than a straight line.5 Since forecasts and failed to predict the intensity or the
the appropriate relationships for the atmosphere correct track of the storm. The question was raised
are non-linear it can be expected to show chaotic at the time as to whether the numerical models used
behaviour. This is illustrated in Figure 5.6, which were capable of the accurate prediction of such an
shows the improvement in predictability that can be exceptional event. Figure 5.8 shows that, using all
expected if the data describing the initial state are the data which could have been available and bet-
improved. However, even with virtually perfect ini- ter assimilation procedures in the model, a good
tial data, the predictability in terms of days ahead forecast of the event can be achieved.




Figure 5.7 (a) Illustrating a simple pendulum consisting of a bob at the end of a string of length 10 cm
attached to a point of suspension which is moved with a linear oscillatory forcing motion at frequencies
near the pendulum™s resonance frequency f 0 . (b) and (c) show plots of the bob™s motion on a horizontal
plane, the scale being in centimetres. (b) For a forcing frequency just above f 0 the motion of the bob
settles down to a simple, regular pattern. (c) For a forcing frequency just below f 0 the bob shows
˜chaotic™ motion (although contained within a given region) which varies randomly and discontinuously
as a function of the initial conditions.
Seasonal forecasting 85




Weather forecasting and chaos continued




Figure 5.8 Analyses and forecasts of the surface pressure in millibars (1 mbar = 100 Pa = 1hPa) for the
storm which passed over southern England on 15/16 October 1987. The fully developed storm is shown in
(b) about 4 hours before it reached the London area; (a) is the situation twenty-four hours earlier from
which the storm developed. (c) and (d) are twenty-four-hour forecasts with the Meteorological Of¬ce
operational ¬ne-mesh model, (c) being the one available at the time and (d) being one produced after the
event using more complete data and better assimilation procedures.



given situation is to employ ensemble forecasting in which an ensemble
of forecasts is run from a cluster of initial states that are generated by
adding to an initial state small perturbations that are within the range of
observational or analysis errors. The forecasts provided from the means
of such ensembles show signi¬cant improvement compared with indi-
vidual forecasts. Further, ensemble forecasts where the spread amongst
the ensemble is low possess more skill that those where the spread in the
ensemble is comparatively high.


Seasonal forecasting
So far short-term forecasts of detailed weather have been considered.
After twenty days or so they run out of skill. What about further into the
86 Modelling the climate



future? Although we cannot expect to forecast the weather in detail, is
there any possibility of predicting the average weather, say, a few months
ahead? As this section shows, it is possible for some parts of the world,
because of the in¬‚uence of the distribution of ocean surface temperatures
on the atmosphere™s behaviour. For seasonal forecasting it is no longer
the initial state of the atmosphere about which detailed knowledge is
required. Rather, we need to know the conditions at the surface and how
they might be changing.
In the tropics, the atmosphere is particularly sensitive to sea surface
temperature. This is not surprising because the largest contribution to the
heat input to the atmosphere is due to evaporation of water vapour from
the ocean surface and its subsequent condensation in the atmosphere,
releasing its latent heat. Because the saturation water vapour pressure
increases rapidly with temperature “ at higher temperatures more water
can be evaporated before the atmosphere is saturated “ evaporation from
Figure 5.9 Changes in sea the surface and hence the heat input to the atmosphere is particularly
surface temperature
large in the tropics.
1871“2002, relative to the
It is during El Ni˜ o events in the east tropical Paci¬c (see Figure 5.9)
n
1961“90 average, for the
that the largest variations are found in ocean temperature. Anomalies in
eastern tropical Paci¬c off
the circulation and rainfall in all tropical regions and to a lesser extent at
Peru.
Forecasting for the African Sahel region

The Sahel region of Africa forms a band about trates the difference in pattern of the mean world-
500 km wide along the southern edge of the Sahara wide SST between the ¬ve wettest and ¬ve driest
Desert that gets most of its rainfall during northern Sahel years since 1901. This and other patterns form
hemisphere summer (particularly July to Septem- the basis of the statistical and atmosphere global
ber). This region is known for its prolonged periods circulation model seasonal forecasts that have been
of drought, which can have a devastating impact on made by the UK Meteorological Of¬ce for the Sahel
since 1986.8 Long-lead forecasts are made in May
the local economy, an example of which occurred
in the 1970s and 1980s (Figure 5.10). The droughts for the July“September rainfall season and updated
have been linked to ¬‚uctuations in several patterns in early July.
of sea surface temperature (SST). Figure 5.11 illus-

Figure 5.10 Annual rainfall
Anomoly (standardised units)




differences from a long-term
1
average (anomalies) for the
Sahel in standardised units
0 Wetter
(the standard deviation from
the average). The smoothed
-1
curve is approximately a
decadal average.
drier
1900 1910 1920 1930 1940 1950 1960 1970 1980 1980 2000


A common dif¬culty with the forecasts has been (correlation between observed and modelled mean
that changes in the SST patterns occur between the rainfall is about 0.37), in particular major changes
to drier conditions are being picked up.9 Such mod-
time of the forecast and the main rainfall season. To
overcome this problem, an atmosphere“ocean cou- els, additionally coupled to variations in the char-
pled model (see ˜The climate system™, next section) acteristics of land surface vegetation and soil which
have also been shown to be important,10 provide the
is being used to forecast changes in SST in addi-
tion to changes in atmospheric circulation. Fore- greatest promise for seasonal forecasts in the Sahel
cast skill has improved although remaining modest and in many other regions of the world.

Figure 5.11 Mean difference
in July“September SST (in —¦ C)
observed during the ¬ve driest
years in the Sahel since 1900
(1972, 1982, 1983, 1984,
1987) and during the ¬ve
wettest years (1927, 1950,
1953, 1954, 1958). Shaded
areas show the most
consistently (statistically
signi¬cant) different regions.
88 Modelling the climate



mid latitudes are associated with these El Ni˜ o events (see Figure 1.4).
n
A good test of the atmospheric models described above is to run them
with an El Ni˜ o sequence of sea surface temperatures and see whether
n
they are able to simulate these climate anomalies. This has now been
done with a number of different atmospheric models; they have shown
considerable skill in the simulation of many of the observed anomalies,
especially those in the tropics and sub-tropics.6
Because of the large heat capacity of the oceans, anomalies of ocean
surface temperature tend to persist for some months. The possibility
therefore exists, for regions where there is a strong correlation between
weather and patterns of ocean surface temperature, of making forecasts
of climate (or average weather) some weeks or months in advance. Such
seasonal forecasts have been attempted especially for regions with low
rainfall; for instance, for north east Brazil and for the Sahel region of
sub-Saharan Africa, a region where human survival is very dependent
on the marginal rainfall (see box below).
The ability to make accurate seasonal forecasts signi¬cantly in ad-
vance depends critically on being able to forecast changes in ocean
surface temperature. To do that requires understanding of, and the abil-
ity to model, the ocean circulation and the way it is coupled to the
atmospheric circulation. Because the largest changes in ocean surface
temperature occur in the tropics and because there are reasons to suppose
that the ocean may be more predictable in the tropics than elsewhere,
most emphasis on the prediction of ocean surface temperature has been
placed in tropical regions, in particular on the prediction of the El Ni˜ o
n
events themselves (see box below).
Later on in this chapter the coupling of atmospheric models and
ocean models is described. For the moment it will suf¬ce to say that, using
coupled models together with detailed observations of both atmosphere
and ocean in the Paci¬c region, signi¬cant skill in the prediction of
El Ni˜ o events up to a year in advance has been achieved7 (see also
n
Chapter 7).


The climate system
So far the forecasting of detailed weather over a few days and of average
weather for a month or so, up to perhaps a season ahead, has been de-
scribed in order to introduce the science and technology of modelling,
and also because some of the scienti¬c con¬dence in the more elabo-
rate climate models arises from their ability to describe and forecast the
processes involved in day-to-day weather.
Climate is concerned with substantially longer periods of time, from
a few years to perhaps a decade or longer. A description of the climate
The climate system 89




A simple model of the El Ni˜ o
n

El Ni˜ o events are good examples of the strong
n boundary it is re¬‚ected as a different sort of wave,
coupling which occurs between the circulations of known as a Kelvin wave, which travels eastward.
ocean and atmosphere. The stress exerted by atmo- This Kelvin wave cancels and reverses the sign of
spheric circulation “ the wind “ on the ocean surface the original warm anomaly, so triggering a cold
is a main driver for the ocean circulation. Also, as event. The time taken for this half-cycle of the whole
we have seen, the heat input to the atmosphere from El Ni˜ o process is determined by the speed with
n
the ocean, especially that arising from evaporation, which the waves propagate in the ocean; it takes
has a big in¬‚uence on the atmospheric circulation. about two years. It is essentially driven by ocean dy-
A simple model of an El Ni˜ o event that shows
n namics, the associated atmospheric changes being
the effect of different kinds of wave motions that determined by the patterns of ocean surface temper-
can propagate within the ocean is illustrated in Fig- ature (and in turn reinforcing those patterns) which
ure 5.12. In this model a wave in the ocean, known result from the ocean dynamics. Expressed in terms
as a Rossby wave, propagates westwards from a of this simple model, some of the characteristics of
warm anomaly in ocean surface temperature near the El Ni˜ o process appear to be essentially pre-
n
the equator. When it reaches the ocean™s western dictable.




Figure 5.12 Schematic to illustrate El Ni˜ o oscillation
n




over a period involves the averages of appropriate components of the
weather (for example, temperature and rainfall) over that period together
with the statistical variations of those components. In considering the
effect of human activities such as the burning of fossil fuels, changes in
climate over periods of decades up to a century or two ahead must be
predicted .
90 Modelling the climate




Since we live in the atmosphere the variables commonly used to
Figure 5.13 Schematic of
the climate system. describe climate are mainly concerned with the atmosphere. But cli-
mate cannot be described in terms of atmosphere alone. Atmospheric
processes are strongly coupled to the oceans (see above); they are also
coupled to the land surface. There is also strong coupling to those parts
of the Earth covered with ice (the cryosphere) and to the vegetation and
other living systems on the land and in the ocean (the biosphere). These
¬ve components “ atmosphere, ocean, land, ice and biosphere “ together
make up the climate system (Figure 5.13).



Feedbacks in the climate system
Chapter 2 considered the rise in global average temperature which would
result from the doubling of the concentration of atmospheric carbon
dioxide assuming that no other changes occurred apart from the increased
temperature at the surface and in the lower atmosphere. The rise in
temperature was found to be 1.2 —¦ C. However, it was also established that,
because of feedbacks (which may be positive or negative) associated with
the temperature increase, the actual rise in global average temperature
was likely to be approximately doubled to about 2.5 —¦ C. This section lists
the most important of these feedbacks.
Feedbacks in the climate system 91




Figure 5.14 Schematic of the physical processes associated with clouds.




Water vapour feedback
This is the most important.11 With a warmer atmosphere more evapo-
ration occurs from the ocean and from wet land surfaces. On average,
therefore, a warmer atmosphere will be a wetter one; it will possess a
higher water vapour content. Since water vapour is a powerful green-
house gas, on average a positive feedback results of a magnitude that
models estimate to approximately double the increase in the global
average temperature that would arise with ¬xed water vapour.12



Cloud-radiation feedback
This is more complicated as several processes are involved. Clouds in-
terfere with the transfer of radiation in the atmosphere in two ways
(Figure 5.14). Firstly, they re¬‚ect a certain proportion of solar radia-
tion back to space, so reducing the total energy available to the sys-
tem. Secondly, they act as blankets to thermal radiation from the Earth™s
surface in a similar way to greenhouse gases. By absorbing thermal
radiation emitted by the Earth™s surface below, and by themselves
emitting thermal radiation, they act to reduce the heat loss to space by the
surface.
The effect that dominates for any particular cloud depends on the
cloud temperature (and hence on the cloud height) and on its detailed
optical properties (those properties which determine its re¬‚ectivity to
solar radiation and its interaction with thermal radiation). The latter
depend on whether the cloud is of water or ice, on its liquid or solid
92 Modelling the climate




Cloud radiative forcing
A concept helpful in distinguishing between the two effects of clouds
mentioned in the text is that of cloud radiative forcing. Take the radi-
ation leaving the top of the atmosphere above a cloud; suppose it has
a value R. Now imagine the cloud to be removed, leaving everything
else the same; suppose the radiation leaving the top of the atmosphere
is now R . The difference R “R is the cloud radiative forcing. It can be
separated into solar radiation and thermal radiation. Values of the cloud
radiative forcing deduced from satellite observations are illustrated in
Figure 5.15. On average, it is found that clouds tend slightly to cool the
Earth-atmosphere system.




Figure 5.15 The cloud radiative forcing is made up of a solar radiation
and a thermal radiation component, which generally act in opposite
senses, each typically of magnitude between 50 and 100 W.m’2 . The
average net forcing is shown here for the period January to July as a
function of latitude as observed from satellites in two years (1985“6) of
the Earth Radiative Budget Experiment (ERBE) and also as simulated in a
climate model with different schemes of cloud formulation (a simple
threshold relative humidity scheme (RH); a scheme that includes cloud
water as a separate variable (CW); as CW but also with cloud radiative
properties dependent on the cloud water content “ thick clouds with a
large water content are more re¬‚ective than thin clouds (CWRP)).There
is encouraging agreement between the models™ results and
observations, but also differences that need to be understood. It is
through comparisons of this kind that further elucidation of cloud
radiation feedback will be achieved.
Feedbacks in the climate system 93



Table 5.1 Estimates of global average temperature changes under
different assumptions about changes in greenhouse gases and clouds

Change (in —¦ C) from current
average global surface
temperature of 15 —¦ C
Greenhouse gases Clouds

As now As now 0
’32
None As now
’21
None None
As now None 4
As now but +3% high
As now 0.3
cloud
As now but +3% low ’1.0
As now
cloud
Doubled CO2 As now (no additional 1.2
concentration cloud feedback)
otherwise as now
Doubled CO2 Cloud feedback 2.5
concentration + included
best estimate of
feedbacks




water content (how thick or thin it is) and on the average size of the cloud
particles. In general for low clouds the re¬‚ectivity effect wins so they
tend to cool the Earth-atmosphere system; for high clouds, by contrast,
the blanketing effect is dominant and they tend to warm the system. The
overall feedback effect of clouds, therefore, can be either positive or
negative (see box below).
Climate is very sensitive to possible changes in cloud amount or
structure, as can be seen from the results of models discussed in later
chapters. To illustrate this, Table 5.1 shows that the hypothetical effect
on the climate of a small percentage change in cloud cover is comparable
with the expected changes due to a doubling of the carbon dioxide con-
centration.



Ocean-circulation feedback
The oceans play a large part in determining the existing climate of the
Earth; they are likely therefore to have an important in¬‚uence on climate
change due to human activities.
94 Modelling the climate




Figure 5.16 Estimates of transport of heat by the oceans. Units are terawatts
(1012 W or million million watts). Note the linkages between the oceans and
that some of the heat transported by the north Atlantic originates in the Paci¬c.



The oceans act on the climate in four important ways. Firstly, there are
close interactions between the ocean and the atmosphere; they behave
as a strongly coupled system. As we have already noted, evaporation
from the oceans provides the main source of atmospheric water vapour
which, through its latent heat of condensation in clouds, provides the
largest single heat source for the atmosphere. The atmosphere in its turn
acts through wind stress on the ocean surface as the main driver of the
ocean circulation.
Secondly, they possess a large heat capacity compared with the at-
mosphere, in other words a large quantity of heat is needed to raise the
temperature of the oceans only slightly. In comparison, the entire heat
capacity of the atmosphere is equivalent to less than three metres depth
of water. That means that in a world that is warming, the oceans warm
much more slowly than the atmosphere. We experience this effect of the
oceans as they tend to reduce the extremes of atmospheric temperature.
For instance, the range of temperature change both during the day and
seasonally is much less at places near the coast than at places far in-
land. The oceans therefore exert a dominant control on the rate at which
atmospheric changes occur.
Thirdly, through their internal circulation the oceans redistribute heat
throughout the climate system. The total amount of heat transported
from the equator to the polar regions by the oceans is similar to that
Models for climate prediction 95



transported by the atmosphere. However, the regional distribution of
that transport is very different (Figure 5.16). Even small changes in the
regional heat transport by the oceans could have large implications for
climate change. For instance, the amount of heat transported by the north
Atlantic Ocean is over 1000 terawatts (1 terawatt = 1 million million
watts = 1012 watts). To give an idea of how large this is, we can note
that a large power station puts out about 1000 million (109 ) watts and the
total amount of commercial energy produced globally is about twelve
terawatts. To put it further in context, considering the region of the north
Atlantic Ocean between north west Europe and Iceland, the heat input
(Figure 5.16) carried by the ocean circulation is of similar magnitude to
that reaching the ocean surface there from the incident solar radiation.
Any accurate simulation of likely climate change, therefore, especially
of its regional variations, must include a description of ocean structure
and dynamics.



Ice-albedo feedback
An ice or snow surface is a powerful re¬‚ector of solar radiation (the
albedo is a measure of its re¬‚ectivity). As some ice melts, therefore, at
the warmer surface, solar radiation which had previously been re¬‚ected
back to space by the ice or snow is absorbed, leading to further increased
warming. This is another positive feedback which on its own would in-
crease the global average temperature rise due to doubled carbon dioxide
by about twenty per cent.
Four feedbacks have been identi¬ed, all of which play a large part
in the determination of climate, especially its regional distribution. It is
therefore necessary to introduce them into climate models. Because the
global models allow for regional variation and also include the important
non-linear processes in their formulation, they are able in principle to
provide a full description of the effect of these feedbacks. They are,
in fact, the only tools available with this potential capability. It is to a
description of climate prediction models that we now turn.



Models for climate prediction
For models to be successful they need to include an adequate description
of the feedbacks we have listed. The water vapour feedback and its
regional distribution depend on the detailed processes of evaporation,
condensation and advection (the transfer of heat by horizontal air ¬‚ow) of
water vapour, and on the way in which convection processes (responsible
96 Modelling the climate



for showers and thunderstorms) are affected by higher surface tem-
peratures. All these processes are already well included in weather
forecasting models, and water vapour feedback has been very thoroughly
studied.13 The most important of the others are cloud radiation feedback
and ocean-circulation feedback. How are these incorporated into the
models?
For modelling purposes, clouds divide into two types “ layer clouds
present on scales larger than the grid size and convective clouds generally
on smaller scales than a grid box. For the introduction of layer clouds,
early weather-forecasting and climate models employed comparatively
simple schemes. A typical scheme would generate cloud at speci¬ed
levels whenever the relative humidity exceeded a critical value, chosen
for broad agreement between model-generated cloud cover and that ob-
served from climatological records. More recent models parameterise the
processes of condensation, freezing, precipitation and cloud formation
much more completely. They also take into account detailed cloud prop-
erties (e.g. water droplets or ice crystals and droplet number and size)
that enable their radiative properties (e.g. their re¬‚ectivity and transmis-
sivity) to be speci¬ed suf¬ciently well for the in¬‚uence of clouds on the
atmosphere™s overall energy budget to be properly described. The most
sophisticated models also include allowance for the effect of aerosols
on cloud properties “ denoted the indirect aerosol effect in Figure 3.8.
The effects of convective clouds are incorporated as part of the model™s
scheme for the parametrisation of convection.
The amount and sign (positive or negative) of the average cloud-
radiation feedback in a particular climate model is dependent on many
aspects of the model™s formulation as well as on the particular scheme
used for the description of cloud formation. Different climate models,
therefore, can show average cloud-radiation feedback that can be either
positive or negative; further, the feedback can show substantial regional
variation. For instance, models differ in their treatment of low cloud such
that in some models the amount of low cloud increases with increased
greenhouse gases, with other models it decreases. Uncertainty regarding
cloud-radiation feedback is the main reason for the wide uncertainty
range in what is called climate sensitivity (see Chapter 6) or the change
in global average surface temperature due a doubling of carbon dioxide
concentration.
The second important feedback is that due to the effects of the ocean
circulation. Compared with a global atmospheric model for weather fore-
casting, the most important elaboration of a climate model is the inclu-
sion of the effects of the ocean. Early climate models only included the
ocean very crudely; they represented it by a simple slab some ¬fty or
one hundred metres deep, the approximate depth of the ˜mixed layer™ of
Models for climate prediction 97



Figure 5.17 Component
elements and parameters
of a coupled
atmosphere“ocean model
including the exchanges at
the atmosphere“ocean
interface.




ocean which responds to the seasonal heating and cooling at the Earth™s
surface. In such models, adjustments had to be made to allow for the
transport of heat by ocean currents. When running the model with a per-
turbation such as increased carbon dioxide, it was not possible to make
allowance for any changes in that transport which might occur. Such
models therefore possess severe limitations.
For an adequate description of the in¬‚uence of the ocean it is neces-
sary to model the ocean circulation and its coupling to the atmospheric
circulation. Figure 5.17 shows the ingredients of such a model. For the
atmospheric part of the model, in order to accommodate long runs on
available computers, the size of the grid has to be substantially larger,
typically 300 km in the horizontal. Otherwise it is essentially the same
as the global model for weather forecasting described earlier. The for-
mulation of the dynamics and physics of the ocean part of the model
is similar to that of its atmospheric counterpart. The effects of water
vapour are of course peculiar to the atmosphere, but the salinity (the
salt content) of the oceans has to be included as a parameter together
with its considerable effects on the water density. Because dynamical
systems, e.g. largescale eddies in the oceans, are of smaller scale than
their atmospheric counterparts, the grid size of the ocean component is
typically about half that of the atmospheric component. On the other
hand, because changes in the ocean are slower, the time step for model
integration can be greater for the ocean component.
At the ocean“atmosphere interface there is exchange of heat, water
and momentum (exchange of momentum leads to friction) between the
two ¬‚uids. The importance of water in the atmosphere and its in¬‚uence on
the atmospheric circulation have already been shown. The distribution
of fresh water precipitated from the atmosphere as rain or snow also
has a large in¬‚uence on the ocean™s circulation through its effect on the
98 Modelling the climate



distribution of salt in the ocean, which in turn affects the ocean density.
It is not surprising, therefore, to ¬nd that the ˜climate™ described by the
model is quite sensitive to the size and the distribution of water exchanges
at the interface.
Before the model can be used for prediction it has to be run for a
considerable time until it reaches a steady ˜climate™. The ˜climate™ of the
model, when it is run unperturbed by increasing greenhouse gases, should
be as close as possible to the current actual climate. If the exchanges are
not correctly described, this will not be the case. Getting the exchanges
right in the model has proved to be dif¬cult. Until recently, many coupled
models introduced arti¬cial adjustments to the ¬‚uxes (known as ¬‚ux ad-
justments) so as to ensure that the model™s ˜climate™ is as identical as
possible to the current climate. However, the ocean component of the
model has been improved especially through introducing higher resolu-
tion (150 km or less), so the need for such adjustments has diminished
and many models are now able to provide an adequate description of the
climate with no such adjustments.
Before leaving the oceans, there is a particular feedback which
should be mentioned between the hydrological cycle and the deep ocean
circulation (see box below). Changes in rainfall, by altering the ocean
salinity, can interact with the ocean circulation. This could affect the
climate, particularly of the North Atlantic region; it may also have
been responsible for some dramatic climate changes in the past (see
Chapter 4).
The most important feedbacks belong to the atmospheric and the
ocean components of the model. They are the largest components, and,
because they are both ¬‚uids and have to be dynamically coupled together,
their incorporation into the model is highly demanding. However, another
feedback to be modelled is the ice-albedo feedback, which arises from
the variations of sea-ice and of snow.
Sea-ice covers a large part of the polar regions in the winter. It is
moved about by both the surface wind and the ocean circulation. So that
the ice-albedo feedback can be properly described, the growth, decay and
dynamics of sea-ice have to be included in the model. Land ice is also
included, essentially as a boundary condition “ a ¬xed quantity “ because
its coverage changes little from year to year. However, the model needs to
show whether there are likely to be changes in ice volume, even though
these are small, in order to ¬nd out their effect on sea level (Chapter 7
considers the impacts of sea-level change).
Interactions with the land surface must also be adequately described.
The most important properties for the model are land surface wetness or,
more precisely, soil moisture content (which will determine the amount
Models for climate prediction 99




The ocean™s deep circulation

west of Greenland) and in the region of Antarctica.
For climate change over periods up to a decade, only
Salt-laden deep water formed in this way con-
the upper layers of the ocean have any substantial
tributes to a deep ocean circulation that involves
interaction with the atmosphere. For longer periods,
all the oceans (Figure 5.18) and is known as the
however, links with the deep ocean circulation be-
thermohaline circulation (THC).
come important. The effects of changes in the deep
circulation are of particular importance In Chapter 4 we mentioned the link between
Experiments using chemical tracers, for in- the THC and the melting of ice. Increases in the ice
stance those illustrated in Figure 5.20 (see next melt can lead to the ocean surface water becoming
box), have been helpful in indicating the regions less salty and therefore less dense. It will not sink so
where strong coupling to the deep ocean occurs. To readily, the deep water formation will be inhibited
sink to the deep ocean, water needs to be partic- and the THC is weakened. In Chapter 6, the link be-
ularly dense, in other words both cold and salty. tween the THC and the hydrological (water) cycle
There are two main regions where such dense in the atmosphere is mentioned. Increased precipi-
water sinks down to the deep ocean, namely in the tation in the north Atlantic region, for instance, can
north Atlantic Ocean (in the Greenland sea between lead to a weakening of the THC.
Scandinavia and Greenland and the Labrador sea




RRENT
SALTY CU
DEEP




Figure 5.18 Deep water formation and circulation. The deep salty current originates in the north
Atlantic. Northward ¬‚owing water near the surface that is unusually salty becomes cooler and even saltier
through evaporation, so increasing its density and causing it to sink.
100 Modelling the climate



of evaporation) and albedo (re¬‚ectivity to solar radiation). The models
keep track of the changes in soil moisture through evaporation and pre-
cipitation. The albedo depends on soil type, vegetation, snow cover and
surface wetness.


Validation of the model
In discussing various aspects of modelling we have already indicated
how some validation of the components of climate models may be carried
out. The successful predictions of weather-forecasting models provide
validation of important aspects of the atmospheric component, as do
the simulations mentioned earlier in the chapter of the connections be-
tween sea surface temperature anomalies and precipitation patterns in
some parts of the world. Various tests have also been carried out of the
ocean component of climate models; for instance, through comparisons
between the simulation and observation of the movement of chemical
tracers (see box below).
Once a comprehensive climate model has been formulated it can be
tested in three main ways. Firstly, it can be run for a number of years
of simulated time and the climate generated by the model compared in
detail to the current climate. For the model to be seen as a valid one, the
average distribution and the seasonal variations of appropriate parame-
ters such as surface pressure, temperature and rainfall have to compare
well with observation. In the same way, the variability of the model™s
climate must be similar to the observed variability. Climate models that
are currently employed for climate prediction stand up well to such
comparisons.14
Secondly, models can be compared against simulations of past cli-
mates when the distribution of key variables was substantially different
than at present; for example, the period around 9000 years ago when
the con¬guration of the Earth™s orbit around the Sun was different (see
Figure 5.19). The perihelion (minimum Earth“Sun distance) was in July
rather than in January as it is now; also the tilt of the Earth™s axis was
slightly different from its current value (24—¦ rather than 23.5—¦ ). Result-
ing from these orbital differences (see Chapter 4), there were signi¬cant
differences in the distribution of solar radiation throughout the year.
The incoming solar energy when averaged over the northern hemisphere
was about seven per cent greater in July and correspondingly less in
January.
When these altered parameters are incorporated into a model, a dif-
ferent climate results. For instance, northern continents are warmer in
summer and colder in winter. In summer a signi¬cantly expanded low
Validation of the model 101




Figure 5.19 Changes in the Earth™s elliptical orbit from the present
con¬guration to 9000 years ago and (right hand side) changes in the average
solar radiation during the year over the northern hemisphere.




pressure region develops over north Africa and south Asia because of
the increased land“ocean temperature contrast. The summer monsoons
in these regions are strengthened and there is increased rainfall. These
simulated changes are in qualitative agreement with paleoclimate data;
for example, these data provide evidence for that period (around 9000
years ago) of lakes and vegetation in the southern Sahara about 1000 km
north of the present limits of vegetation.

<<

. 4
( 13)



>>