Climate Change and the Mediterranean Region

EXECUTIVE SUMMARY

Water shortages and poor harvests during the droughts of the early 1990s
exposed the acute vulnerability of the Mediterranean region to climatic
extremes. Against this backdrop, the prospect of a major climate change
brought about by human activities is a source of growing concern, raising
serious questions over the sustainability of the region.

This report examines the potential implications of global climate change
for the Mediterranean region. Drawing on the results of recent studies, it
reviews possible changes in climate together with recent trends, the
potential impacts of climate change and the implications for sustainable
development.

One key finding is that future climate change could critically undermine
efforts for sustainable development in the Mediterranean region. In
particular, climate change may add to existing problems of desertification,
water scarcity and food production, while also introducing new threats to
human health, ecosystems and national economies of countries. The most
serious impacts are likely to be felt in North African and eastern
Mediterranean countries.

The report concludes that while there is some scope for adaptation,
ensuring the long-term sustainability of the region requires urgent action
to cut global emissions of greenhouse gases.

Specific findings are summarised below.


Hotter and drier times ahead?

If current trends in emissions of greenhouse gases continue, global
temperatures are expected to rise faster over the next century than over
any time during the last 10,000 years. Significant uncertainties surround
predictions of regional climate changes, but it is likely that the
Mediterranean region will also warm significantly.

The outlook for precipitation is much less certain, but most projections
point to more precipitation in winter and less in summer over the region as
a whole. A common feature of many projections is declining annual
precipitation over much of the Mediterranean region south of 40 or 45° N,
with increases to the north of this. Even areas receiving more
precipitation may get drier than today due to increased evaporation and
changes in the seasonal distribution of rainfall and its intensity.

As a consequence, the frequency and severity of droughts could increase
across the region. Changes in large-scale atmospheric circulation - as
represented by the El Niño-Southern Oscillation (ENSO) and the North
Atlantic Oscillation (NAO) - would further effect the occurrence of extreme
events.

An indication of the scale of possible changes is given by one scenario
based on the output from four climate models. This suggests that
temperatures could rise by over 4°C by 2100 over many inland areas and by
over half of this over the Mediterranean Sea. Over the same period, annual
precipitation is projected to decline by 10 to 40% over much of Africa and
southeastern Spain, with smaller - but potentially significant - changes
elsewhere.

Aerosol emissions may counter some of the effects of greenhouse gases in
some areas. But, in the long term prospect remains one of hotter, drier
conditions throughout the Mediterranean region as the relative influence of
greenhouse gases increases over time.


Coastal flooding and erosion

As the world warms, global sea levels will rise as oceans expand and
glaciers melt. Around much of the Mediterranean basin, sea levels could
rise by close to 1 m by 2100. As a consequence, some low-lying coastal
areas would be lost through flooding or erosion, while rivers and coastal
aquifers would become more salty. The worst affected areas will be the Nile
Delta, Venice and Thessaloniki where local subsidence means that sea levels
could rise by at least one-and-a-half times as much as elsewhere.


Climate shows possible signs of change

On a global scale, there is increasing evidence that climate is changing
and of a discernible human influence. The high natural variability of the
Mediterranean climate make both the detection of climate change and
attribution of its cause very difficult. Nevertheless, observations suggest
that climate may already be changing in the region.

Land records for the western Mediterranean show slight trends towards
warmer and drier conditions over the last century. In contrast, parts of
the eastern Mediterranean have experienced cooler, wetter conditions in
recent times than earlier this century.  Surface water temperature records
for the last 120 years show little overall trend but deep water records for
the western Mediterranean show a continuous warming trend since 1959.

During the period 1952 to 1992, the number and frequency of heat waves
affecting the region has increased. The early 1990s were notable for
recurrent droughts and for periods of intense rainfall in the western
Mediterranean and for extreme cold events and rainfall in the east. Recent
climatic extremes are linked with the exceptional behaviour of ENSO and of
the NAO. Record-breaking NAO values occurred in 1983, 1989 and 1990, while
the prolonged 1990 to 1995 El Niño event was the longest on record.

While all such trends and extremes could have occurred naturally, they are
broadly consistent with the potential effects of greenhouse gas emissions
and aerosol emissions to-date.


Increase in extent and severity of desertification

While much desertification is attributed to poor land use practices, hotter
and drier conditions would extend the area prone to desertification
northwards to encompass areas currently not at risk. In addition, the rate
of desertification would increase due to increases in erosion, salinisation
and fire hazard and reductions in soil quality. As a result, the process of
desertification is likely to become irreversible.

The economic and human costs of an increase in desertification would be
tremendous - even today, the annual costs of desertification in Tunisia and
Spain are US$100 million and US$200 million, respectively.


Increased frequency of water shortages and decline in water quality

It is likely that the first impacts of climate change will be felt in the
Mediterranean water resource system. Reductions in water availability would
hit southern Mediterranean countries the hardest. In Egypt, Libya, Tunisia,
Algeria, Morocco, Syria, Malta  and the Lebanon, water availability already
falls below, or approaches
1,000 m3 per person per year - the common benchmark for water scarcity.

Even relatively well-endowed countries, such as Spain, Greece and Italy,
could suffer ever-more frequent regional water shortages due to the twin
problems of climate change and rising demand. Crete, for example, could
experience serious water shortages in five out of six years by 2010.

Some water supplies could become unusable due to the penetration of salt
water into rivers and coastal aquifers as sea level rises. Water pollution
- already a major health hazard in the region - would become still worse as
pollutants become more concentrated with reductions in river flow.


Food security threatened by falls in production and world price rises

Livestock production would suffer due to a deterioration in the quality of
rangeland associated with higher concentrations of atmospheric carbon
dioxide and to changes in areas of rangeland as climate boundaries move
northwards. In the European Mediterranean, the area of unproductive
shrubland is expected to expand, while in North Africa and the Near East,
most of the steppe rangeland could give way to desert by 2050 or earlier.

Yields of grains and other crops could decrease substantially across the
Mediterranean region due to increased frequency of drought. While losses
may be partially offset by beneficial effects from carbon dioxide, crop
production would be further threatened by increases in competition for
water and the prevalence of pests and diseases and land losses through
desertification and sea level rise.

Climate change effects combined with wider socio-economic factors could
cause cereal production over much of southern Europe to become untenable.
At Kardista in central Greece, for example, the chance of obtaining current
yields of maize could drop to close to zero by 2050, while in Spain,
irrigation problems could force maize out of production.

In North Africa and the Near East, changes in average climate associated
with a doubling of carbon dioxide could cause yield losses of over 20% for
wheat, corn and other coarse grains - even before allowance is made for
losses through other causes. In coastal areas, large areas of productive
land may be lost through flooding, saline intrusion and waterlogging. In
Egypt, for example, agricultural production may cease altogether over an
area extending 20 km inland.

World prices for many key commodities such as wheat, maize, soybean meal
and poultry could rise significantly as a result of global climate changes.
Not only might Mediterranean countries loose substantially in economic
terms, but the combination of higher prices and crop losses would lead to a
deterioration in levels of food security in, particularly, southern
countries.


New, widespread risks to public health

Reductions in food security would increase the risks of malnutrition and
hunger for millions in the south. The combination of heat and pollution
would lead to an upsurge in respiratory illness among urban populations,
while extreme weather events could increase death and injury rates. Water
shortages and damaged infrastructure would increase the risk of cholera and
dysentery. Higher temperatures would increase the incidence and extent of
infectious diseases, such as malaria, dengue fever, schistosmaisis and
yellow fever.


Many valuable ecosystems would be lost

Many valuable ecosystems could be lost as species fail to keep up with the
shift in climate boundaries and/or find their migration paths blocked by
human activities. Wetland sites will face the dual threats of drying out
and sea level rise. Up to 85% of wetland sites in southern Europe could
disappear with a 3 to 4°C rise in temperatures. In Tunisia, for example,
rising temperatures could contribute to the loss of all food plants and
breeding waterfowl and the disappearance of nationally important fisheries.


Economic activity undermined in coastal zones

Industries, infrastructure and heritage sites in the coastal zone would be
threatened by inundation or erosion due to sea level rise. For example, a
rise in sea level of just 0.5 m would flood the western part of Kastala Bay
(Croatia) harbour and cause serious degradation to the historic cities of
Cres (Croatia) and Venice (Italy). Hydroelectric power output could be
constrained by water shortages, with potentially serious knock-on
implications for both domestic and industrial users.

Serious social disruption as the livelihood of millions is threatened and
international tensions over resources mounts.

Serious social disruption could occur as millions are forced from their
homelands as a result of desertification, poor harvests and sea level rise,
while international disputes over shared water resources could turn into
conflict in the face of declines in water availability and increased demand.


Losses to national economies

National economies would be adversely affected not only by the direct
impacts of climate change, but also through the cost of adaptive measures
and the knock-on implications of changes elsewhere. Quantitative estimates
of financial costs are unreliable but in general, developing countries are
expected to suffer larger relative economic damages than developed
countries.


Sustainable development hinges on international action to cut greenhouse
gas emissions

Future climate change could critically undermine efforts for sustainable
development in the Mediterranean region through its impacts on the
environment and social and economic well-being. While in many respects
climate change exacerbates existing problems rather than creates new ones,
the sheer magnitude of the potential problem means it cannot be ignored.

There is some scope for adaptation, but the fact that many measures would
be beneficial irrespective of climate change suggests that radical changes
in policies and practices will be needed. It is also vital that developed
countries meet their obligations to assist adaptation in developing
countries through access to know-how and financial assistance.

Ultimately, however, the long-term sustainability of the Mediterranean
region requires keeping climate change within tolerable bounds. Current
understanding of safe limits points to the need for prompt international
agreement - and action - to make the drastic cuts in emissions of
greenhouse gases required to stabilise atmospheric concentrations of these
gases.


1.	INTRODUCTION

In 1993, tourists from Malaga to Athens and from St. Tropez to Malta were
confronted by exhortations to "Save it" as the region was hit by its fifth
year of drought after winter rains failed to replenish the reservoirs and
aquifers (Pearce, 1993). Serious as such events were, they pale in
comparison with the potential impacts of a human-induced climate change.

Atmospheric concentrations of greenhouse gases1 are rising as a result of
human activities and, in particular fossil-fuel use, land-use changes and
agriculture. Greenhouse gases occur naturally in the atmosphere, where they
allow solar radiation to reach the Earth unhindered but trap a proportion
of outgoing radiation. In this way, greenhouse gases play a critical role
in maintaining the heat balance of the Earth. But as concentrations rise,
scientists believe the world will warm.

In 1986, the Scientific Committee of Problems of the Environment proclaimed
that global warming "should be considered one of today's most important
long-term problems" (Bolin and others, 1986). Research over the last 10
years has reaffirmed the magnitude of the looming threat. In 1996, the
Intergovernmental Panel on Climate Change (IPCC)2 reported for the first
time that past emissions appear to have had "a discernible influence on
global climate" (IPCC, 1996a). The IPCC further found that, if current
trends in emissions continue, this could cause a rate of warming over the
next century "probably greater than any seen in the last 10,000 years".

The magnitude, and possible immediacy, of a major change in climate has
alarming implications for countries world-wide as both ecological and human
systems are fundamentally dependant on climate. The Mediterranean region3
is particularly vulnerable to climate change as over much of the region,
summer rainfall is virtually zero. Water scarcity is endemic and changes in
the water balance would have substantial implications for, amongst other
things, agriculture and water supplies. This vulnerability is compounded by
the ongoing desertification of much of the region, together with population
growth and poverty in, particularly, the southern Basin.

Over the last five years, a number of studies have assessed how climate
change may affect the Mediterranean region. From these, it is clear that
while many uncertainties remain, climate change will have profound and
far-reaching implications for the 350 million or so people who live in the
Mediterranean region today - and for generations to come. This report draws
on recent work to examine, first, how climate may change and what recent
observations show. It then moves on to describe some of the potential
impacts of future climate change and their significance in light of recent
trends and, finally, discusses the implications for sustainable development
in the region.


Figure 1: 	The Mediterranean region.



2.	FUTURE CLIMATE CHANGE

Climate varies naturally on all timescales from decades to millennia due to
changes in atmospheric and ocean circulation, solar output and volcanic
activity. However, future climate change will be dominated by human
influences unless and until the composition of the atmosphere is
stabilised.

Stabilisation of concentrations of carbon dioxide - a key greenhouse gas -
requires cuts in emissions of between 50 and 70%. Emissions of other gases
would also have to be reduced significantly - or even stopped completely -
if atmospheric concentrations are to be stabilised and the risk of climate
change reduced. This section, examines the potential implications for both
global climate and climate in the Mediterranean region if cuts of this
magnitude are not achieved.


Global Changes in Climate

The magnitude and rate of future climate change will depend on the amount
of greenhouse gases emitted, the sensitivity of climate to these gases, and
the degree to which the effects are modified by aerosol emissions. The IPCC
present six scenarios of future emissions, based on widely differing
assumptions of future population and economic growth, energy consumption,
technological developments and land use. These all show that atmospheric
concentrations of greenhouse gases will continue to rise throughout the
21st century unless there is concerted action to curb emissions (Houghton
and others, 1996).

The climatic impact of this rise in greenhouse gases will be modified by
the influence of aerosol emissions (Box 1). Unlike greenhouse gases, the
effects of aerosols are localised and short-lived. Thus, their overall
effect is likely to be to mask - rather than to offset - the much more
fundamental, long-term influence of greenhouse gases on climate at some
locations. Nevertheless, aerosols could exert a strong influence on the
climate experienced at some locations and the IPCC make a range of
different assumptions over future aerosols in their various emissions
scenarios.


Box 1: The Aerosol Effect

Aerosols are microscopic airborne particles. Natural sources include dust
storms, fires and volcanic eruptions. Human sources include the combustion
of fossil fuels and the deliberate burning of forests and fields. The
climatic impact of aerosols depends on their size and composition. Some
aerosols, such as soot, may have a warming effect. Others, such as sulphate
aerosols, are believed to have a cooling effect. The dominant effect is,
however, currently thought to be one of cooling, although some uncertainty
surrounds even this because of uncertainties over the indirect effects of
aerosols.

In any case, the cooling effect is not, according to the IPCC, "...a simple
offset to the warming effect of greenhouse gases" (Houghton and others,
1996). Aerosols tend to have a much shorter lifetime in the atmosphere than
greenhouse gases. While greenhouse gases may stay in the atmosphere for
centuries, volcanic aerosols tend to stay in the atmosphere for months to
years. The atmospheric lifetime of most aerosols of human origin is still
shorter - only a few days.

The short lifetime of aerosols of human origin means that their effects
will be fairly localised and that, unlike greenhouse gases, aerosols do not
constantly build up in the atmosphere. This has two important implications.

*	The relative impact of aerosols is expected to become less as time
passes.

*	If all emissions from fossil fuel burning were stopped tomorrow,
the cooling from aerosols would end within a week, while greenhouse warming
would continue for decades to centuries.

Sulphate aerosols are also major contributors to acid rain and emissions
subject to controls. If, and when, emissions are reduced, their climate
effect will decrease also. Not controlling acid emissions to try and
mitigate the effects of greenhouse gases is not considered an option due to
their devastating effect on crops, ecosystems and materials.


Under the IPCC's "mid-range" emissions scenario (IS92a), concentrations of
greenhouse gases reach the equivalent of double pre-industrial levels of
carbon dioxide by 20304 - and continue to rise thereafter (Carter and
others, 1994). As a result of this, and allowing for potential increases in
aerosol emissions, the IPCC calculate that global temperatures will climb
by between 1.4 and 2.9°C by the year 2100, with a best-estimate of 2°C
(Kattenberg and others, 1996). This range in values reflects scientific
uncertainty over the sensitivity of the climate system to changes in
greenhouse gas levels. Even values at the low end of this scale suggest a
rate of warming greater that any seen in the past 10,000 years.

Even this may be an under-estimate of changes to come. The IPCC's emissions
scenario assumes that emissions of aerosols increase markedly, creating a
strong cooling effect. This is unrealistic. No account is taken of the
Second Sulphur Protocol or of amendments to US car emissions. Moreover, as
the World Energy Council point out, if emissions were to increase as
projected, then this "would cause deposition levels that exceeded the
'critical loads' for most ecosystems in South and East Asian regions" (WEC,
1996).
Given this, emissions of aerosols are likely to be controlled much earlier
than envisaged in the IPCC's main scenario and the global warming will be
consequently greater. Figure 2 compares projected changes in global mean
temperature under the IPCC's mid-range scenario with increasing aerosols
and using the same greenhouse gas scenario but with constant 1990 aerosols.
Assuming constant 1990 aerosols, global temperatures would increase by
between 1.6 and 3.5°C by 2100, with a best-estimate of 2.4°C. If greenhouse
gas emissions conformed to the highest of all the IPCC's scenarios then
temperatures would rise still further - by up to 4.5°C by 2100 (Kattenberg
and others, 1996).


Figure 2: 	Global mean temperature changes from 1990 to 2100 under the
IPCC's mid-range emission scenario (IS92a) and different climate
sensitivities (Kattenberg and others, 1996). The full line shows
temperature changes assuming aerosol emissions increase, while the dashed
line is for constant 1990 aerosols.

As the world warms, global precipitation is expected to increase on average
and other aspects of climate will change. The process of climate change
will, however, not be a smooth, gradual process. Rather, the IPCC state,
"[as] future climate extends beyond the boundaries of empirical knowledge,
the more likely outcomes will include surprises and unanticipated rapid
changes." (IPCC, 1996a).

Increasing temperatures will cause sea levels to rise as glaciers melt and
the water in the oceans expands. Under the IPCC's mid-range emissions
scenario, global sea levels could rise by 20 to 86 cm by 2100, with a best
estimate of 59 cm (Warrick and others, 1996). If aerosol emissions are held
constant at 1990 levels, then sea levels could rise more - between 23 and
96 cm by 2100, with a best-estimate of 55 cm. A rise of 50 cm is between
two and five times the rise in the past century. Time lags in the onset of
melting mean that sea levels would continue to rise many centuries after
2100, even if concentrations of greenhouse gases were stabilised by then.

Significant uncertainty surrounds the projections of sea level rise. This
is largely because of uncertainty over how much ice will melt. A particular
concern is the fate of the West Antarctic ice sheet - this would cause a
rise in sea level of up to 6 m. This threat is generally considered as
uncertain and remote in time compared with the more immediate threat posed
by the potential rise in sea levels over coming decades. However, the
relatively recent discovery that ice shelves in Antarctica may be melting
from beneath due the presence of warm water from the deep oceans adds to
concern over the stability of the continental ice sheets (Jenkins and
others, 1995).


Climate Change in the Mediterranean Region

As the world warms, climate will change in the Mediterranean region.
However, considerable uncertainty exists over just what form these changes
may take. This is primarily because of the acknowledged weaknesses of
global climate models (GCMs) in assessing regional climate changes (Box 2).

While such uncertainties are frustrating, the option of ignoring the
prospect of a major change in climate is no more acceptable. Scientists are
confident that global warming due to current trends in emissions will be
accompanied by significant changes in local climate. Decisions must be
taken on the basis of the best information available, taking account of the
uncertainties, and not on the simply-wrong assumption that future climate
will be the same as in the past.

The following sections draw on the results of variety of studies to give an
impression of how climate may change in the Mediterranean region as we move
into the 21st century. Where possible, the results from several studies are
compared to give a sense of a range of possible outcomes.


Box 2: Climate Models and their Limitations

Most regional studies of future climate change use output from global
circulation models (GCMs) of the atmosphere and ocean. In these models, the
physical laws and empirical relationships that describe atmospheric and
oceanic systems are represented by mathematical equations. The many complex
processes - such as the melting of sea ice and formation of water vapour -
that influence climate are taken into account.

By changing the inputs to GCMs, scientists can assess the effects of
increasing concentrations of greenhouse gases on the climate system. The
majority of experiments to-date are 'equilibrium response' experiments and
assess the ultimate impact of a sudden doubling of concentrations of carbon
dioxide. Recently, attention has focused on more realistic 'transient
response' experiments. These experiments measure real-time climate changes
in response to progressive increases (typically 1% per year) in carbon
dioxide concentrations.

All climate models have a number of limitations:

*	The coarse resolution of global climate models means they do not
adequately depict many geographic features and the interactions between the
atmosphere and the surface;

*	Natural variations in local climate are much greater than those in
climate averaged over continental or larger scales;

*	The uneven spatial impact of aerosols - not only have few model
experiments taken aerosols into account, but those that do include only
very simplified effects ; and

*	Land-use changes - such as deforestation and desertification - are
currently seldom allowed for, but will substantially affect local climates.

Despite these limitations, scientists are confident in GCM results for
large-scale changes in climate. Confidence in local and regional
predictions is, however, lower as most models do not represent current
climate well on this scale and projections vary widely. But, as climate
scientist Tom Wigley observes, "[in] spite of the problems that plague
current GCMs, they are the best tool we have for projecting future changes
in climate at a regional level." (Wigley, 1992). Nevertheless he cautions,
model results "should be treated strictly as scenarios of possible future
climate and not as predictions."


Changes in Temperature

Rising concentrations of greenhouse gases alone could cause warming over
the Mediterranean region similar in magnitude to the global increase.
Results from four equilibrium experiments indicate that temperatures over
the region as a whole could rise by about 3.5°C between now and the latter
half of the 21st century in response to a doubling of carbon dioxide (or
its equivalent) (Wigley, 1992). According to three transient model runs,
about half of this rise - between 1.4 and 2.6°C - could occur by the 2020s
(Rosenzweig and Tubiello, 1997). There is no evidence of marked seasonal
differences in response.

These results are towards the high end of expectations as the models used
have middle to high sensitivities5. An impression of the full range of
possible outcomes is given by an analysis of output from nine transient
models for southern Europe and Turkey6 (Kattenberg and others, 1996). This
points to temperature increases of 1 to 4.5°C (with a mid-point of about
2.5°C) during the winter and summer by the latter half of the 21st century.
Even if emissions of greenhouse gases were stabilised by then, temperatures
would continue to climb for several decades due to time lags in the
response of the oceans.

There will be marked regional differences in the rate of temperature
increase experienced at different locations - although there is wide
disagreement between the patterns of change projected by the various models
(Wigley, 1992 and Cubasch and others, 1996). A picture of possible changes
is given by an average of the output from four equilibrium experiments,
statistically down-scaled for further local details (Figure 3; Palutikof
and Wigley, 1996)7. The results show that temperatures across the region
could rise between 0.7 and 1.6°C for every degree rise in global mean
temperature.


Figure 3: 	Model average temperature changes (°C) over the
Mediterranean region for every °C rise in global mean temperature resulting
from rising concentrations of greenhouse gases (Palutikof and Wigley,
1996). The map values can be seen as broadly indicative of conditions which
may exist around 20308. Areas where temperatures are projected to rise less than the global mean are shaded. 

The greatest rates of temperature increase oc
cur over Africa, the Ukraine and eastern Turkey, while the lowest rates of
change occur over the Mediterranean Sea. The coastal zones are areas of
rapid transition. Between now and 2100, temperatures could have risen by up
to: 2.5 to 3°C over the Mediterranean Sea, 3 to 4°C over coastal areas and
4 to 4.5°C over most inland areas, with increases of up to 5.5°C over
Morocco9. This general pattern of change suggested by these results is
physically reasonable as warming over the sea is likely to lag behind that
over land areas. Also, these findings are broadly similar to those from
more detailed model experiments (Cubasch and others, 1996)10.

These results do not take account of possible increases in aerosol
emissions which could mask some of this warming. One transient experiment
suggests that aerosols may reduce warming over the Mediterranean region by
1-2°C over a period from 1795 to 2030-2050 (Mitchell and others, 1995). The
net effect may even be to give an impression of cooling over the central
Mediterranean in summer over the next few decades (Hasselmann and others,
1995). Given the likely exaggeration of aerosol effects discussed earlier,
such results probably over-estimate the potential for local cooling. But,
in any case, the long-term prospect remains one of warming throughout the
Mediterranean region as the relative influence of greenhouse gases
increases over time.


Changes in Precipitation

The prospects for precipitation over the Mediterranean region in a warmer
world are highly uncertain due to the general weakness of GCMs in
predicting regional precipitation. Models offer conflicting evidence over
how precipitation may change on average over the Mediterranean region. Two
out of three equilibrium experiments presented in one study suggest an
overall increase in precipitation across the region (Rosenzwieg and
Tubiello, 1997). However, recent transient model runs for the 2020s suggest
an overall decrease of between 1.5 and 7.3% (Rosenzwieg and Tubiello,
1997).

Most equilibrium and transient experiments show a widening in the seasonal
precipitation gradient with more precipitation in winter and less in
summer. An average of four equilibrium model results for the whole
Mediterranean region suggests an increase in winter precipitation of 10%
and a decrease in summer precipitation of 10% between now and 210011
(Palutikof and others, 1992). This finding is broadly supported by a more
recent comparison of nine transient model runs for southern Europe and
Turkey (Kattenberg and others, 1996). In this case, most models suggest
increases in winter precipitation of up to 10% and reductions in summer
precipitation of 5 to 15% by the latter half of the 21st century.

The patterns of precipitation produced by different model runs are so
divergent that it is difficult to have confidence in any single projection.
Nevertheless, a common feature of many model runs is decreasing annual
precipitation over much of the Mediterranean region south of 40 to 45oN,
and increasing precipitation north of this (see for example, Cubasch and
others 1996, Barrow and Hulme, 1995 and Palutikof and Wigley, 1996). This
is illustrated by a scenario based on the average results from four
equilibrium models, statistically down-scaled to give a sense of more
localised changes12 (Figure 4, Palutikof and Wigley, 1996).


Figure 4: 	Model average precipitation changes (%) over the
Mediterranean region for every °C rise in global mean temperature resulting
from rising concentrations of greenhouse gases (Palutikof and Wigley,
1996). The map values can be seen as broadly indicative of conditions which
may exist around 203013. Areas where precipitation is projected to decrease
are shaded.

In this scenario, annual precipitation changes across the region range from
-12% to +13% per °C rise in global mean temperature. This translates into
annual precipitation decreases of between 10 and 40% over much of Africa
and southeast Spain, and of up to 10% over central Spain, southern France,
Greece and the Near East by 210014. There is also the suggestion of
possible increase in precipitation of up to 20% over central Italy.
However, as the authors stress, confidence in these scenarios is low
because of the uncertainty associated with GCM results for regional
precipitation.

In the short-term, aerosol effects may counter the effect of rising
concentrations of greenhouse gases in some areas. Results from transient
experiments for around the middle of the 21st century suggest that once
aerosol effects are allowed for precipitation over southern Europe and
Turkey as a whole may increase slightly (Kattenberg and others, 1996).
These changes are far from certain as they depend critically on both the
aerosol scenario used and how aerosols are represented in the models. In
any case, a long-term model run for the Mediterranean region suggests that
from 2050 onwards precipitation would decrease markedly as the relative
influence of greenhouse gases grows (Palutikof and others, 1996b).

Clearly, there remains considerable uncertainty over how precipitation will
change over the Mediterranean region in response to the changing
composition of the atmosphere. However, the balance of evidence seems to
suggest reductions in precipitation over much of the region, with a
possible transitional period for some areas due to aerosol effects.


Changes in Moisture Availability

In terms of the ecological and social impacts of climate change, changes in
moisture availability are more important than changes in precipitation or
temperature alone. Low levels of moisture availability are associated with
droughts.

Moisture availability is determined both by water gains from precipitation
and water losses through runoff and evapotranspiration15. As temperature
increases, evapotranspiration will also increase (all other things being
equal). This means that even where precipitation is projected to increase,
actual moisture availability could go down if the gains are outweighed by
losses. The projected widening of the seasonal precipitation gradient is
also likely to reduce water availability during the growing season
(Kattenberg and others, 1996; Wigley, 1992). This is because extra
precipitation in winter may not be stored in the soil, but lost as runoff.
The occurrence of precipitation in intense episodes has a similar effect
(Segal and others, 1994).

GCMs are particularly weak at determining moisture availability. This is
partly because potential evapotranspiration is not properly assessed by
GCMs due to crude treatment of the hydrological cycle (Rind and others,
1992) and partly because of the huge uncertainties over future
precipitation. Despite this, there is a high level of consistency in model
results for southern Europe and Turkey, with models showing an overall
reduction in summer moisture availability in response to rising
concentrations of greenhouse gases (Kattenberg and others, 1996). Results
from three equilibrium experiments for southern Europe and Turkey suggest
that soil moisture would decrease over the whole region by 15 to 25% during
the summer (IPCC, 1992). A preliminary assessment of changes in the water
balance over the eastern Mediterranean from Turkey through to Egypt also
found a tendency for a northwards shift of the desert line (Segal and
others, 1994).

Evidence of reductions in water availability over much of the Mediterranean
region during both winter and summer comes from a recent transient
experiment (Gordon and O'Farrell, 1996). This is supported by work for the
region using average temperature and precipitation output from four
equilibrium experiments (Palutikof and others, 1994 and 1996b)16. This
study indicates an unfavourable shift in the ratio of precipitation to
evapotranspiration throughout the whole Mediterranean region in every
season17. The greatest effects are over the north of the region, extending
over the Italian mainland, Sardinia and Corsica, in spring and autumn. The
impact on human activities may, however, be most acute in the south of the
region where water is in particularly short supply even now.

Again in the near-term, the effects of increased concentrations of
greenhouse gases may be mitigated in some areas by the effects of aerosols.
Two transient experiments show that if aerosol effects are included, then
soil moisture over southern Europe and Turkey as a whole could increase,
rather than decrease (Kattenberg and others, 1996). However, exaggeration
of aerosol effects and the localised nature of their impacts means that
some areas may still experience drier conditions. Moreover, these findings
are only relevant to around the middle of next century. Beyond this, the
relative influence of greenhouse gases is expected to grow and the
long-term prospect is one of a drying out of the whole Mediterranean region.


Changes in Extreme Events

As climate changes the frequency of extreme events in the Mediterranean
region will change in response to changes both in average climate and in
climate variability. Warmer conditions over the Mediterranean region should
lead to an increase in the occurrence of extremely high temperatures and a
decrease in extremely low temperature events. One study finds that by
around the middle of the next century, current maximum temperatures in
Athens could be exceeded in most months (Barrow and others, 1995).

Similarly, in areas experiencing a general decrease in precipitation,
droughts are likely to become more frequent as the probability of dry days
and the length of dry spells increases. The converse is true for areas
where precipitation increases. One study reports that the probability of a
dry spell lasting more than 30 days in summer in southern Europe would
increase by a factor of between two and five on a doubling of carbon
dioxide (Gregory, 1996). A study for Naxos (Greece) further suggests that a
10% reduction in winter precipitation could increase the length of dry
spells by up to 21 to 45%, while a 10% increase in summer precipitation
could increase the length of wet spells by 15% (Palutikof and others,
1992)18.

In general, scientists expect more heavy rain events in a warmer world due
to an intensification of the hydrological cycle. Most models suggest a
general increase in the intensity of precipitation of between 10 and 30% at
most latitudes for a doubling of carbon dioxide (Kattenberg and others,
1996). Storminess may also increase, although this is less certain.

On a wider scale , changes in climate variability will be influenced by
changes in general atmospheric circulation. A major source of year-to-year
variability world-wide is the El Niño-Southern Oscillation (ENSO)
phenomenon)19. ENSO is renowned for bringing climatic disruption world-wide
(Glantz and others, 1991). In the Mediterranean region, El Niño events have
been linked with low rainfall over much of the western and central basins
(Arkin and Xie, 1997, Lamb and Peppler, 1991 Rodó and others, 1997; Rodó
and Comins, 1996).

As yet, scientists are uncertain how ENSO will change in a warmer world -
models do not simulate the phenomena very well and under-estimate the
variability. Nevertheless, a number of models indicate that ENSO events
will continue to occur in a warmer world and there is some evidence that
precipitation anomalies will increase in tropical areas (Kattenberg and
others, 1996). However, a number of papers reviewed by the IPCC suggest
that "much of the effects of global warming may be modulated through a
change in the magnitude and regularity of the warm and cold phases of ENSO"
(Dickinson and others, 1996).

Of still greater significance to the Mediterranean region is the fate of
the North Atlantic Oscillation (NAO))20 although as yet little indication
has been given of likely changes in a warmer world. The state of the NAO
critically affects storm tracks, temperatures and precipitation across
Europe and eastern North America. High values have been linked with low
winter rainfall throughout much of the Mediterranean and cold conditions in
the east (Hurrell, 1995; Palutikof and others, 1996b; Trenberth and Shea,
1997).

Despite these uncertainties over exactly how climate variability and
extremes will change in the Mediterranean region, the overall picture does
suggest an increase in the frequency of extreme events and, in particular,
of droughts in the western Mediterranean.


Sea Level Rise

Locally, the apparent rise in sea level will critically depend on local
land movements. Most of the Mediterranean region currently appears to be
stable and is likely to experience a sea level rise comparable with the
global mean - up to about 96 cm by 210021 (Milliman, 1992; Warrick and
others, 1996). The Near East and Alexandria may, however, experience
slightly lower rates of sea level rise - up to 90 cm by 2100 - as the land
appears to be rising slightly.

The worst affected regions seem likely to be the larger river deltas of the
Nile, Thessaloniki and Venice, which are currently subsiding. In these
areas, sea levels could rise by up to 150 cm, 140 cm and 175 cm,
respectively by 2100. Rising sea levels would, in all areas, bring the risk
of inundation, higher rates of erosion and increased saline intrusion into
rivers and aquifers.



3.	OBSERVED CLIMATE CHANGES: SIGNS OF CHANGE

Past emissions of greenhouse gases have already affected the Earth's energy
balance and the effects on global and regional climates will become more
marked over time (Santer and others, 1996). This raises two key questions:
is climate changing? And, if so, can the observed changes be attributed to
the changing composition of the atmosphere?

Globally, at least, scientists appear to have detected the first signs of
climate change. Since 1860, mean global temperatures have risen by between
0.3°C and 0.6°C. Warming since the mid-1970s has been particularly rapid
with all eight of the warmest years on record occurring since 1983 (WMO,
1997; CRU, 1997). Early signs are that 1997 may also prove to be a
record-breaking year (Tiempo, 24 Jun 1997).

In 1996, the IPCC announced that the observed warming "is unlikely to be
entirely natural in origin" (IPCC, 1996a). On the basis of further detailed
assessments of patterns of atmospheric and oceanic temperatures and changes
in the hydrological cycle, the IPCC further concluded that the "balance of
evidence suggests a discernible human influence on global climate" .

As global climate appears to be changing, we would expect the Mediterranean
climate also to have changed. Detection of climate change on this scale is,
however, extremely difficult as the high variability in local climates
masks trends in the 'noise' of natural fluctuations. Moreover, the short
period of observations makes the identification of clear trends difficult
and creates uncertainty over the scale of natural variability.

Proving that any observed changes are the result of the changing
composition of the atmosphere is still more difficult due to the weakness
of models in predicting the regional effects of climate change. The picture
is further complicated by the influence of other human activities on
climate (Box 3) which may not only mask underlying trends but could either
accentuate or mitigate the effects of global warming.

For all this, observational records do suggest marked changes in the
climate of the Mediterranean region over recent years. While it is
impossible to be certain if these trends are "real" or if they can be
attributed to atmospheric pollution, a number of aspects of the observed
changes are consistent with a human influence. In either case, absolute
proof will only be available with hindsight - by which time significant
impacts will already be occurring.


Box 3: Human Influences on Regional Climates

Human activities can substantially affect regional climates. The cooling
effects of sulphate aerosols are discussed earlier, but other significant
impacts arise from urbanisation and other land use changes. These effects
complicate the detection of more fundamental climate changes.

*	Urbanisation and the associated pollution have the effect of
increasing both temperature and precipitation (Cotton and Pielke, 1995).
Warmer conditions result from a number of processes, including: the slowing
of winds by high buildings, heat released as energy is used and a reduction
in evaporation as rain runs off into drains rather than being retained in
soils. Precipitation increases as air rises and cools over what is
effectively a man-made hill.

The combined effects of urbanisation on local climates can be significant.
In Athens, urbanisation is held responsible for a 1°C rise in maximum
temperatures over the last 20 years which occurred despite a fall in
minimum temperatures (Metaxas and others, 1991). Similarly, rainfall over
the last 70 years has been higher than expected given trends in other
nearby regions (Amanatidis and others, 1993). Over the period 1970 to 1989,
the number of automobiles increased from about 200,000 to over a million,
but also many more, and higher, buildings were constructed between the
Athens National Observatory and the coast.

*	Desertification acts to increase maximum daily temperatures and
reduce precipitation. (Cotton and Pielke, 1995). While desertification is
in part a product of climate change, there are also important feedbacks on
local climate. Land degradation tends to reduce soil moisture and this in
turn reduces evaporation resulting in increased maximum temperatures and
lower rainfall. Reductions in vegetation have a similar effect as this
reduces the amount of water captured and then recycled through
evapotranspiration to create rain. Both processes also increases the
reflectively - or "albedo" - of the ground causing higher temperatures in
the day and reducing them at night.

	Analysis of temperature data for this century shows that warming
was nearly 0.2°C greater over dryland areas than over land areas as a whole
(Jones, 1994). It is unclear, however, how much of this difference is due
to recent desertification and how much is due to the existing arid state of
many dryland areas. Desertification is a major, long-term problem in the
Mediterranean region and it is possible that this accounts for, at least in
part, the observed decrease in rainfall in some areas.

*	Deforestation can increase maximum daily temperatures and reduce
precipitation in much the same way as desertification (Cotton and Pielke,
1995). Experiments in both the Amazon and in southern Nigeria reveal a much
greater range in temperature over cleared ground (Ghuman and Lal, 1986;
Salati and others, 1978). The role of forests in enhancing rainfall is also
well-established - an estimated 50% of rainfall in the Amazon is from local
evaporation and transpiration (Salati and others, 1978). In the
Mediterranean region, deforestation has occurred over many centuries and
the effects are unlikely to distort the recent record - although, of
course, the effects of past deforestation will be ever-present.

*	Irrigation and Man-made Lakes have the opposite effect on climate
than desertification as rainfall increases due an increase in local water
availability and day-time temperatures are lowered with an increase in
albedo (Cotton and Pielke, 1995). Few definitive studies of the scale of
these effects have been done but estimates of the possible effects of a
proposal to flood depressions in the Chott region in Algeria and Tunisia
suggest that, as a consequence, local precipitation could increase by up to
150 mm every year (Enger and Tjernström, 1991). The impact of existing
irrigation in, say, Egypt and Israel is unknown, but may have offset some
of the general decrease in precipitation observed locally.

While the effects of such activities on regional climates can clearly be
large, the effects on global climate are very small. Globally, urbanisation
accounts for only an estimated 0.05°C of the observed warming over land
areas this century (Jones and others, 1990). The global impact of
desertification is thought to be still smaller - only a few hundredths of a
degree (Nicholls and others, 1996).
Trends in Temperature

Sea surface temperature records for the Mediterranean region show clear
fluctuations in climate over the last 120 or so years, but little overall
trend (Figure 5). This record shows that temperatures were at a minimum
around 1910 and then rose sharply to a maximum around 1940 after which they
stabilised for around 20 years. After this, while global temperatures
continued to rise to unprecedented levels, the Mediterranean region
experienced a decade of rapid cooling. Warming resumed in the late 1970s,
but still temperatures remained below those experienced in the 1930s and
1940s up until 1989 at least.


Figure 5. 	Variations in annual sea surface temperatures across the
Mediterranean between 1873 and 1989, as represented by frequency
differences of warm minus cold months. The jagged line shows annual values
while the smooth line highlights variations over decadal and longer
timescales. (Source: Metaxas and others, 1991).

This basic pattern is also evident in sea surface temperature records for
both the eastern and western basins and in the seasonal records, but with
one potentially important difference. The cooling in the east of the region
during the 1970s was much more marked than in the west (Metaxas and others,
1991). As a result, temperatures remained substantially below average in
the east until at least the end of the 1980s. It is also notable that deep
water records for the western Mediterranean show a continuous warming trend
from 1959 (Bethoux and others, 1997).

Land records for the western and central Mediterranean do, however, suggest
a long-term warming trend. While all show a similar pattern of warming and
cooling, the 1970s minimum is much less pronounced at many locations, for
example, Cairo, Marseille, Perpignan and Athens (Metaxas and others, 1991;
Repapis and Philandras, 1988). While this may in part be attributable to
increasing urbanisation over the last 30 to 40 years, the overall impact is
an appearance of warming comparable with that seen in the global record.
This contrasts with Jerusalem (Israel) where annual temperatures in the
mid-1970s were lower than during any other time during the previous 100
years (Repapis and Philandras, 1988).

This east-west difference in temperature trends also shows up clearly when
average conditions for the period 1975 to 1994 are compared with average
conditions during the previous twenty years (Nicholls and others, 1996).
This shows that temperatures were, on average, higher during the recent
period over south-west Europe and north-west Africa. In contrast, average
temperatures in the eastern Mediterranean were lower than during the
previous 20 years. The area of colder conditions is centred on Turkey and
extends west as far as Italy in the north and Libya in the south.

Recent changes in temperature across the Mediterranean clearly fall within
the range of natural variability. But, the general pattern of change is
also broadly consistent with a GCM simulation of temperature changes over
the region associated with the combined effects of present-day carbon
dioxide levels and sulphur emissions (IPCC, 1996a). This being the case,
then it is possible that the warming over the last decade experienced over
much of the region may be a sign of things to come. Only time will tell.


Trends in Precipitation

Since 1900, precipitation decreased by over 5% over much of the land
bordering the Mediterranean Sea, with the exception of the stretch from
Tunisia through to Libya where it increased slightly (Nicholls and others,
1996a). Within these overall trends, regular alternations between wetter
and drier periods are discernible. Records for both the western
Mediterranean and the Balkans indicate major moist periods sometime during
the periods1900 to 1920, 1930 to 1956, and 1968 to 1980 with intervening
dry periods (Maheras, 1987; Maheras and Kolyva-Machera, 1990). Records for
the period 1951 onwards show a slight tendency towards decreasing rainfall
in almost all regions and in all seasons (Figure 6; Palutikof and others,
1996b). The only clear positive trend is in the eastern Mediterranean in
the autumn.


Figure 6: 	Yearly rainfall anomaly index for the northern
Mediterranean from Portugal to Syria, and including the islands north of
35°N (Palutikof and others, 1996b). The index is for the hydrological
(rainfall) year from September in one year to August in the following year.

Such regional trends underplay the scale of changes in precipitation
experienced locally. Over the period 1975 to 1994, precipitation was on
average more than 17% lower than during the preceding 20 years over much of
north-western Africa, Spain, Italy and Greece (Nicholls and others, 1996).
The recent dryness in the western Mediterranean contrasts with the
conditions elsewhere in northern Africa and the eastern basin. Here,
precipitation was generally higher over the last couple of decades compared
with the previous 20 years. (Nicholls and others, 1996).

Events took an abrupt about-turn in 1996, with drying areas suddenly
experiencing extreme wet conditions and vice versa (WMO, 1997). It is not
clear if this is just a temporary "blip" in overall trends, the start of a
new trend or a return to more "normal" conditions experienced earlier this
century. If the models are broadly correct about precipitation changes in
response to increases in greenhouse gases, the droughts over the western
Mediterranean could be symptomatic of a growing human influence on climate
in the region. Wetter conditions in the east might reflect the stronger
influence of aerosols in this area.


Occurrence of Extreme Events

During the period 1952 to 1992, the number and frequency of heat waves
affecting the Mediterranean region has increased (Geeson and Thornes,
1996), while the early 1990s were notable for a number of extreme events
(Box 4). It is impossible to gauge if the frequency and magnitude of
extremes has increased without a thorough analysis. Nevertheless, records
of ENSO and NAO - both of which are linked with the occurrence of extreme
events in the Mediterranean region - do show exceptional behaviour. This
may, in turn, suggest that the recent history of extremes in the
Mediterranean is unusual.


Box 4: Recent Climatic Extremes in the Mediterranean Region

*	The early 1990s were characterised by extreme drought over much of
this region. In 1995, precipitation was less than 75% of normal (1961-1990)
over much of the western Mediterranean (CRU, 1997). Over 1994 and 1995,
Spain received less than 50% of  normal  at some locations (CRU, 1997).

*	In the winters of 1991/2 and 1992/3, rarely seen snowfall fell in
many areas of north Africa and the eastern Mediterranean, while average
temperatures from December to March 1991/2 were the coldest on record in
Turkey (from 1930) and at Jerusalem (from 1865) (WMO/UNEP, 1994).

*	Between late September and early November 1993 large sections of
south-eastern France, western Spain, central Portugal, Corsica and northern
Morocco recorded 2 to 3 times the usual precipitation (WMO/UNEP, 1994). In
this period, Madrid had the highest amount of precipitation since records
began in 1854 while in mid-November 1993, Greece and Israel experienced
major floods (WMO/UNEP, 1994).

*	In 1995, some interior parts of Egypt saw rainfall for the first
time in nearly half a century. Similarly, conditions in Tunisia and Libya
were exceptionally wet (CRU, 1997).


Both the unusual coldness of over the eastern Med-iterranean over the last
decade and the dry conditions afflicting most of the region have been
linked with exceptionally high values in the NAO (Hurrell, 1995, Palutikof,
1996; Trenberth and Shea, 1997). From the 1940s to the early 1970s, NAO
values decreased markedly. This trend re-versed sharply 25 years ago,
resulting in largely un-precedented positive values of NAO values from 1980
onwards (with the notable exception of the 1995-6 winter). NAO values for
1983, 1989 and 1990 winters are the highest since records began in 1894.

Changes in parts of the western and central Med-iterranean have been
connected to the ENSO the phenomenon. The behaviour of ENSO has changed
mark-edly since 1976/1977, with the record being dominated by El Niño
events and showing only rare instances of La Niña events (Trenberth and
Shea, 1997). The prolonged 1990 to 1995 El Niño event is the longest on
record and would be expected to occur less than once every 2000 years
(Trenberth and Hoar, 1996). La Niña conditions returned abruptly in 1996
but, at the time writing, early signs of an imminent El Niño event have
been observed (Tiempo, 24 June 1997).

The extent to which observed changes in NAO and ENSO are due to increases
in greenhouse gases is not clear. Evidence exists that persistent ENSO
events, at least, may have occurred prior to the period of instrumental
data (Allan and D'Arrigo, 1996). But, as climate scientists Kevin Trenberth
and Dennis Shea point out: "the observational evidence is suggestive that
climate change, for whatever reason, is contributing to [these] changes in
circulation, which in turn alter the distribution of storm tracks and
rainfall." (Trenberth and Shea, 1997).


4.	IMPACTS OF CLIMATE CHANGE

Climate change will have diverse and far reaching consequences for the
Mediterranean region (Figure 7). An immediate concern is the potential to
exacerbate existing problems of desertification, water resources and food
production. But ultimately, the impacts will be much wider as the effects
cascade through the social and economic system. While all areas will be
affected, the type and extent of impacts experienced will vary markedly
depending on local circumstances.


Figure 7: 	Impact of climate change on environment and society
(Milliman and others, 1992).

While there has been an upsurge in impacts studies in recent years, it
remains difficult to be precise over the scale of impacts likely to occur.
This is partly due to fundamental uncertainties in modelling regional
climate change. Most studies focus on the possible impacts of hotter, drier
conditions. However, while current evidence suggests this is the most
likely response to increasing concentrations of greenhouse gases, it must
be noted that confidence in particularly the precipitation scenarios is low
(see Section 2). Also, other activities - and in particular aerosol
emissions - could have an important influence on climate in some areas, at
least in the short term.

Assessment of the impacts of climate change is further complicated by the
need to consider not only the nature of the climate change, but also the
sensitivity of ecological and social systems to change, the degree to which
adaptation is possible and the vulnerability of any given system (Box 5).
The extent to which existing studies take these factors into account
varies. But, common weaknesses include failures to consider: how systems
may evolve under progressive long-term climate change; the interactions
between different sectors; and/or the implications of multiple stress
factors. As a result, it seems likely that the potential impacts of climate
change are understated in many studies.

Nevertheless, there is clear evidence of potentially serious impacts
throughout the Mediterranean region, with the most acute impacts being felt
south of the socio-economic divide in Africa and the Near East. The
following sections, like most of the studies they draw on, focus on the
potential implications on hotter, drier conditions over much of the
Mediterranean region.


Box 5: Assessing the Impacts of Climate Change

The impacts in any particular area will depend on four key factors.

*	The magnitude and rate of climate change. This will critically
affect both the extent to which ecological and social systems can withstand
stress and their ability to adapt. The impacts of climate change will be
mediated through not only the direct effects of changes in temperature and
other climate variables, but also the associated rises in atmospheric
concentrations of carbon dioxide and in sea level. The rapid rate of change
anticipated under all but the lowest scenarios of climate change poses a
particular threat.

*	The sensitivity of ecological and social systems to climate change.
Low-lying coastal areas are obviously sensitive to changes in sea level,
while the droughts and floods of the 1990s clearly exposed the sensitivity
of, in particular, water supply and food production systems to climate
variations. Other key concerns such as desertification and the degradation
of natural ecosystems probably more immediately impacted by demographic
change and land use practices than by climate - although, even in these
areas, the fundamental change in underlying conditions suggested by climate
change has significant long-term implications.

*	The scope for adaptation. Both the rate of climate change and the
uncertainty over the nature of the expected changes makes adaptation
difficult, particularly in the many areas, such as infrastructure
development, where planning timescales are long in relation to the
timescales of the predicted changes.

*	The vulnerability of areas. This is determined both by the system's
sensitivity to change and by its ability to adapt. It is likely that
vulnerability will be dictated by as much by economic circumstances and
institutional infrastructure as by inherent sensitivity to climate change.

Finally, it is vital to assess potential impacts of climate change in the
context of other environmental and socio-economic trends. Many countries of
theMediterranean region are already under pressure due desertification,
population growth, tourism, pollution and (legitimate) aspirations to
improve economic well-being. Climate change is just one further stress.
But, the dependence of most human activities on the environment means that
changes can either enhance or undermine development in the region.


Desertification

In the Mediterranean region, future climate change is likely to aggravate
significantly the existing problem of desertification and critically
undermine the effectiveness of efforts to combat the problem.

The threat posed by desertification to human welfare is internationally
recognised and was the stimulus behind agreement on an International
Convention to Combat Desertification in 1992. UNEP define desertification
as "land degradation in arid, semi-arid and dry sub-humid areas resulting
from various factors, including climatic variations and human activities"
(ICCD, 1994). In the process of desertification, biologically and
economically productive land becomes less productive and less able to
support the communities that depend on it.

Desertification is considered one of the most serious problems facing the
Mediterranean region today (Table 1). The area affected extends across
northern Africa into the Near East and across large parts of Europe,
including Greece, southern Italy, Sicily, Corsica, and the Iberian
Peninsula (UNEP, 1992; Imeson and Emmer, 1992). Every year, Turkey, Tunisia
and Morocco lose around 54237, 18000 and 2200 hectares of land through
erosion, respectively (UNEP, 1987).


Table 1: 	Extent of desertification (%) in the early 1980s (Mabutt,
1984).



The economic and human costs of desertification are enormous. Tunisia alone
spends US$100 million on efforts to combat desertification (Kharrat, 1997).
Desertification in Spain causes an estimated 30,000 million pesetas (US$200
million) economic damages every year (La Mundo, 13 October 1993; La
Vanguardia, 7 January, 1994). Human costs often include malnutrition, the
risk of famine and dislocation of people who are forced abandon their lands.

Much desertification is attributed to human activities going back over
millennia. Human impacts arise from overstocking, over-cultivation and
deforestation and, to a lesser degree, irrigation and urbanisation. Past
degradation is held responsible for decline ancient civilisations within
the region and elsewhere (Box 6). However, drylands are inherently
vulnerable to water stress and drought and as the United Nations Framework
Convention on Climate Change (UNFCCC) points out, "countries in arid and
semi-arid areas or areas liable to floods, drought and desertification ...
are particularly vulnerable to the adverse effects of climate change"
(preamble).


Box 6: Lessons of the Past - Desertification and the Decline of Civilisations

The devastating consequences of desertification are evidenced by the
destruction of ancient civilisations. Six thousand years ago in Mesopotamia
(the Middle East), food production and Sumerian culture declined as poor
irrigation practices led to huge tracts of land becoming salinised
(WRI/IIED/UNEP, 1989). The political, military and economic decline of
Assyria and Babylonia some three thousand years ago coincided with a
notable period of warming and drying in the region (Neumann and Parpola,
1987). Climate change could explain this aridity - but so could
desertification. Elsewhere, in Rajastan in north-west India, archaeological
evidence tells of a well-developed civilisation existing at around the same
time in an area that is now tropical desert. (Cotton and Pielke, 1995) This
transformation is thought to be as much the consequence of overpopulation
and denudation of the local vegetation as of natural climate change. Around
the Mediterranean region, areas covered today by Maquis vegetation (scrub,
oak, heath and so on) were once productive forests (WRI/IIED/UNEP, 1989).


Climate change would both affect the extent of areas prone to
desertification and the severity of desertification in existing drylands.
At a fundamental level, the conditions for desertification are dictated by
climatic factors since the process occurs mainly in arid, semi-arid and dry
sub-humid zones22. The reduction in moisture availability projected under
climate change would both increase the aridity of existing drylands and
progressively shift the boundaries of areas susceptible to desertification
northwards in the Mediterranean region to encompass areas currently not at
risk.

The particular vulnerability of ecosystems in semi-arid and arid areas to
climate change is highlighted by the IPCC which points out that "[whereas]
most terrestrial ecosystems have some built-in ability to buffer the
effects of climate variability, this is not so true of those in arid and
semi-arid lands - where even small changes in climate can intensify the
already high natural variability and lead to permanent degradation of the
productive potential of such lands." (Bullock and Le Houérou, 1996).

The potential for desertification is still further enhanced through the
direct effects of climate change on erosion, soil quality, salinisation and
fire hazard. Key processes include the following (Rosenzweig and Hillel,
1993; Imeson and Emmer, 1996).

*	As soils dry they become more susceptible to wind erosion,
especially where there is no vegetation cover or the area is cultivated -
this could ultimately generate "dust bowl" conditions in some areas. The
hazard of water erosion would also be made worse by any accompanying
increase in rainfall intensity.

*	Higher temperatures could result in a reduction in soil fertility
due to higher rates of decomposition and losses of organic matter, and
could affect nutrient cycling.

*	A general decrease in precipitation or increase in evaporation will
cause an increase in the area affected by saline conditions. Spain and
Italy, in particular, could experience an increase in the area affected by
low permeability, shrinking and swelling and waterlogging. Coastal areas
could be more directly affected by salinisation due to the increased
penetration of salt water into the groundwater.

*	Increasing temperatures and drier conditions could give rise to
more forest fires, although the extent to which land degradation results
will vary. In Sardinia (Italy), forest fires are the most important cause
of severe soil erosion (Aru, 1984). A Spanish newspaper in October 1994
suggested that soil erosion resulting from forest fires and torrential rain
cost Spain more than 55 000 million pesetas (US$360 million) per year
(López-Bermúdez, 1995).

There could also be important feedback effects. For example, a reduction in
surface moisture or vegetative cover would increase temperatures and reduce
rainfall as less energy is used in evapo-transipiration and less water is
recycled. Where vegetation cover is lost completely, soil surfaces become
sealed and encrusted, reducing water intake and resulting in a still drier
environment (Bullock and Le Houérou, 1996).

For all this, it is impossible to quantify the combined impact of climate
change on desertification. Much will depend not just on the climate change
but also of coincident pressures arising from other human activities and
the effectiveness of responses to the problem of desertification in
general. However, as the IPCC warns: "[desertification] is likely to become
irreversible if the environment becomes drier and the soil becomes further
degraded through erosion and compaction" (Bullock and Le Houérou, 1996).
Thus, in a very fundamental way climate change threatens to undermine all
current efforts to reduce desertification in the Mediterranean region.


Water Resources

Climate change could further exacerbate existing problems over water
scarcity in many Mediterranean countries and cause a decline in water
quality through increased concentrations of pollutants, salinisation and
increased salt water intrusion in coastal aquifers.

Water scarcity is endemic in many parts of the Mediterranean, making these
countries particularly vulnerable to any reduction in supplies. The common
benchmark for water scarcity is 1,000m3 per person per year. Water
availability either already falls below this level or is expected to within
two to three decades in southern Mediterranean countries including Egypt,
Libya, Tunisia, Algeria, Morocco, Syria, Malta and the Lebanon
(WRI/IIED/UNEP, 1997). Even in the generally well-endowed countries of the
northern Mediterranean such as Spain, Italy and Greece, regional water
problems exist.

The reduction of moisture availability anticipated in the climate change
scenarios would inevitably add to the problem of water scarcity throughout
the Mediterranean region. The changes in precipitation combined with
increased evaporation would directly reduce runoff and ground water levels
throughout the Mediterranean region. The acute sensitivity of water
resources to rainfall reductions is illustrated by estimates that a 20%
decrease in rainfall in Acheloos basin (Greece) would increase the risk of
water system failure (inability to provide targeted supplies) from less
than 1% at present to 38% (Mimikou and others, 1991)

Other processes would further damage water supplies. Poorer infiltration
due to soil degradation would reduce aquifer recharge, while reservoirs
could be seriously affected by an increase in sedimentation due to erosion.
Such sedimentation problems are already anticipated in the Apollakia
reservoir in Rhodes (Perissoratis and others, 1996).

The problem of reduced water availability will be compounded by increases
in demand driven by both socio-economic factors and climate change itself.
In Crete, it has been estimated that increases in urban demand alone could
increase the likelihood of water shortages from 20% in 1980 to 85% in 2010
(MEDALUS II, 1996). As climate changes, demand for irrigation water is
likely to increase. This is significant, as irrigation accounts for about
72% of all water consumption in the Mediterranean region (Blue Plan, 1988).
Experiments for Lesotho point to an increase in demand of 7% with a 10%
decrease in runoff and of more than 20% with a 2°C rise in temperature
(Arnell and Piper, 1995).

Water quality could also be adversely affected by climate change. Higher
temperatures and evaporation would cause rises in the salinity of lakes and
reservoirs, while in coastal areas, rises in sea level would increase
saline intrusion into aquifers and estuaries. In Malta, a 1 m rise in sea
level could reduce water from the main reservoir by 40% (Attard and others,
1996), while in France, the Vaccares and lower lakes of the Camargue are
anticipated to become hyper-saline (Corre, 1996). Problems of saline
intrusion would be further exacerbated by reductions in runoff and by
increased withdrawals in response to higher demand. Excessive demand
already contributes to saline intrusion problems in many coastal areas of
Italy, Spain, Greece and North Africa (Aru, 1996).

Pollution concentrations would also increase with a reduction in river and
lake levels. Dumping of a whole variety of wastes in seasonal channels
around the Mediterranean is already a major health hazard and the problem
could become more acute with climate change (MEDALUS II, 1996). Similarly,
the situation could become critical in Venice where there are already
massive inflows of urban, industrial and agricultural waste (Sestini, 1992).

There is no doubt that many opportunities exist to improve supply through
demand side management and increasing the efficiency of water use, for
example through improved irrigation systems, changes in crops and so on
(Kaczmarek, 1996). In the Syrian coastal area alone, it has been estimated
that 40% of water supplied is lost through poorly constructed and
maintained mains and illegal tapping (Al-Shalberi and others, 1996).
Increasing the amount of fresh water available through desalinisation
processes is less attractive as it is both very expensive and highly
intensive in the use of fossil fuels (although renewable energy sources can
be used).

However, as the IPCC point out, "[it] remains uncertain whether water
supply systems will evolve substantially enough in the future to compensate
forthe anticipated negative impacts of climate change on water resources
and for potential increases in demand." (IPCC, 1996b). It is therefore not
surprising that, as early as 1990, UNEP warned "[it] is likely that the
impact of climate change will first be felt in the Mediterranean water
resource system." (UNEP, 1990).


Agriculture

Future climate change would affect food production in the Mediterranean
region in a number of ways. Direct effects would be incurred through the
change in climate itself and the associated increase in carbon dioxide
levels and sea level rise. But, in many areas, food production would be
further affected by other climate impacts such as desertification,
increased fire risk, spread of pests and diseases and changes in the global
markets.

The full impact of climate change on food production remains uncertain. As
yet, no fully integrated studies of the overall impact of these various
changes have been undertaken. Moreover, most studies focus on a limited
range of foodstuffs and consider changes in yield under present-day
cultivation conditions and even then often only double carbon dioxide
conditions. Nevertheless, the available evidence suggests that climate
change will have a deleterious impact on food production throughout the
region, increase prices and add to food insecurity in the southern basin.


Rangelands

Future climate change is likely to both reduce the productivity of
rangelands and change the areas amenable to livestock production. The most
serious impacts on livestock production would be in the southern
Mediterranean, where the rangelands are already under pressure from land
use changes and population growth.

Rangelands sustain a large number of people in the Mediterranean region
through their support for livestock and forage crops. They are currently
receding across the region, but particularly in the south where some 50% of
arid steppe rangeland has been cleared for crops over the last 30 years (Le
Houérou, 1992). This ever-increasing clearance is also a major cause of
desertification in northern Africa and the Near East (Le Houérou, 1988).
This fragmentation of rangeland may add to their vulnerability to climate
change (Archer, 1994). As the IPCC point out, "[with] the addition of
climate change to existing impacts, rangelands may be more susceptible to
extreme events such as drought, 100-year floods, and insect outbreaks"
(Allen Diaz, 1996).

Higher levels of carbon dioxide will worsen conditions for grazing across
the region as this will increase the carbon to nitrogen ratio in forage,
thus reducing its food value (Allen Diaz, 1996). Moreover, a reduction in
moisture availability would change the species composition in favour of
woody, less palatable, plants. A further effect of a shift of carbon
storage from soil to biomass is that it is likely to adversely affect soil
stability and increase erosion.

As climate changes, the areas suitable for rangeland will change in
response to changes in the water balance. A doubling of carbon dioxide (or
its equivalent) would shift the climatic limits for grass and dwarf shrub
steppes 300 to 500 km polewards at the expense of shrubland (Allen-Diaz,
1996). In European Mediterranean countries, a northwards expansion of
unproductive shrubland is expected due to the abandonment of agricultural
land for climatic and wider socio-economic reasons, increases in wildfires
and a decrease in livestock grazing (Allen-Diaz, 1996).

The potential effects of climate change on rangelands in North Africa and
the Near East are much more serious. The projected reduction in moisture
availability alone would mean the encroachment of desert further north.
Decreases in plant cover may also increase erosion and lead to a nearly
irreversible loss of productive potential (Parton and others, 1993). It is
likely that this "natural response" will be compounded by desertification
due to over-grazing and the current trend towards conversion of rangeland
to agriculture. The net result is that by 2050 (or earlier), most of the
steppe rangeland is expected to have given way to desert (Le Houérou, 1992;
Allen-Diaz, 1996).

Desertification of the North African and Near Eastern steppes would have
significant implications for the livestock industry and the wider economy
in these areas. Over 50% of the sheep industry is in the arid-steppe zone
and increasingly this would have to depend on imported feed and the
international cereal market to survive (Le Houérou, 1992). Thus, national
economies could be hit twice-over: first, by a decline in the contribution
of the livestock industry to the economy; and second, by the cost of
importing more feed. As the IPCC point out, the losses to the national
economy would "have serious implications for the food policies of many
underdeveloped countries and on the lives of thousands of pastoral people"
(Allen-Diaz, 1996).


Crop Production

Future changes in climate would significantly affect both crop yields and
what it is possible to grow. The overall impact on crop production will,
however, be determined by the ability of farmers to adapt as existing
problems over land and water use are ex-acerbated. There will be marked
north-south differences in both the impacts and the re-sponses, with the
most serious problems being faced in the south.

Crop production remains important for both domestic food con-sumption and
the gen-eration of export income in the Mediterranean region. Currently,
Med-iterranean agriculture accounts for nearly all the olive oil production
world-wide, 60% of wine production, 45% of grape production, 20% of citrus
production and about 12% of total cereal production (FAO, 1993). Today, the
northern Mediterranean is self sufficient in food production, while the
southern Mediterranean countries produce less than 60% of their food
(Rosenzweig and Tubiello, 1997). Even in the absence of climate change,
basic food security in southern countries is likely to deteriorate due to a
combination of population growth, land use changes and water problems.
Water use has also become a problem in Spain, Greece and southern Italy.

Increasing levels of atmospheric carbon dioxide may increase the yield of
major food crops as a result of higher rates of photosynthesis and
improvements in water use efficiency. Under controlled conditions, yields
of C3 plants - such as wheat, rice, soybean and barley - increase by about
30% with a doubling of present-day carbon dioxide levels (Cure and Acock,
1986; Rogers and Dahlman, 1993). The measured response for C4 plants - such
as maize, sugarcane, millet and sorghum - is much less. There is, however,
considerable debate about whether or not these benefits would occur in the
real world.

Climate change is likely to result in a northward shift of crop growing
zones, with some areas becoming progressively less suitable for growing
certain crops and more suitable for others. Crop response models are
commonly used to assess the potential effects of both increased carbon
dioxide and changes in mean climate. Result from these models generally
show yields of grains and other crops could decrease substantially in
Mediterranean areas due to an increased frequency of drought (Reilly, 1996).

Regional results from a global crop-response modelling study suggest that
yields of a number of major crops could decrease markedly in North Africa
and the Near East, under three equilibrium climate scenarios (Table 2,
Rosenzweig and others, 1993; Reilly and others, 1993)23. Estimated yield
losses are over 20% for wheat, corn and other coarse grains in some
scenarios, even allowing for effects of carbon dioxide. The largest losses
- of up to 51% in the case of wheat - are anticipated in Egypt. The
potential for large reductions in Egyptian wheat and maize yields is also
supported by more recent modelling work (El-Shear and others, 1997). Even
allowing for some adaptation, yield losses may occur throughout the
southern and eastern Mediterranean region24.


Table 2: 	Estimated yield changes of major crops by region. The range
takes account of results under three equilibrium GCM scenarios for a
doubling of carbon dioxide and includes thedirect carbon dioxide effects.
Estimates derived by Rosenzweig and others, 1993 and extracted from Reilly
and others, 1993.

 * Yields decrease under at least one climate scenario if no account is
taken of carbon dioxide fertilisation effects.
** Yields decrease under at least one climate scenario even allowing for
carbon dioxide fertilisation effects and some adaptation.
The European Community region includes: Portugal, Spain, France, Italy and
Greece
The Eastern Europe region includes: Former Yugoslavia, Albania
The Middle East/North Africa - oil producers region includes: Syria,
Algeria, Tunisia and Libya
The Middle East/North Africa - other countries region includes: Turkey,
Cyprus, Lebanon, Israel, Morocco


The global crop modelling study suggests more promising prospects for
European countries as a whole, but detailed country studies for southern
European suggest the outlook for cereal production is poor in these
countries. Experiments for Spain, southern Italy and Greece all show a
decline in maize yields, even allowing for the potential benefits of carbon
dioxide fertilisation (see review by Rosenzweig and Tubiello, 1997).

*	One study found that under the IPCC's mid-range reference scenario,
yields at Kardista (central Greece) may drop by about 8% by 2030 and that
by 2050 the probability of obtaining current yields could be close to zero
(Kapetanaki and Rosenzweig, 1997). Yield reductions of almost twice this
amount were calculated for a site in northern Greece.

*	A study of the implications of climate change for maize production
in Spain suggests that "the yield reductions and the exacerbated problems
of irrigation water availability ... may force the crop out of production
in some regions" (Iglesais and Minguez, 1997). The same situation could
arise in many other parts of the Mediterranean region.

The prospects for wheat in southern Europe are less certain. Crop-climate
experiments for a number of different climate scenarios suggest that wheat
yields may increase in northern Mediterranean countries (Harrison and
Butterfield, 1995; Igelais and Minguez, 1997). This may, however, be
over-optimistic as the yield increases are largely attributable to higher
carbon dioxide levels and thus must be open to question. Indeed, results of
another experiment in which carbon dioxide effects are not included suggest
that yields in northern and central Italy would decline (Bindi and others,
1993).

Overall, the outlook for cereal production is probably significantly worse
than suggested by the crop-climate models. This is because the models have
a number of inherent limitations which suggest that they may be
over-optimistic in the yields predicted (Box 7). Uncertainties surrounding
the effects of carbon dioxide have already been described. Aside from this,
their failure to consider the impact of extreme events, soil conditions,
competition for water supplies, changes in the prevalence and distribution
of pests and diseases and sea level rise could be particularly important.


Box 7: False Optimism? Crop-Climate Experiments

Crop-climate modelling experiments make a number of highly optimistic
assumptions, including:

*	The CO2 fertilisation effect works to the full - the effects of
carbon dioxide on yields may be overestimated as the experimental results
used to calibrate the models may not be replicated in variable, windy,
pest-infected field conditions (Körner, 1990).

*	Climate variability will not alter - changes in the frequency and
magnitude of extreme events (for example, extended periods of high
temperatures, droughts, and so on) could have a greater impact than changes
in mean climate alone (Semenov and Porter, 1994).

*	Nutrients are not limiting and diseases and insect pests are
controlled - ensuring these conditions are met could require the use of
expensive agrochemicals and may make farming uneconomic.

*	Irrigation water is available in unlimited supply (sometimes, but
not always, assumed) - in reality, reductions in water availability would
increase competition between sectors for water supplies.

*	There are no problem soil conditions - there are.

*	No account is taken of land degradation or sea level rise on
production - this could be particularly important in the southern
Mediterranean.

Taken together, these limitations suggest that crop-climate experiments may
give substantially higher yields than would be achieved in reality. Thus,
while such experiments do provide a useful idea of the sensitivity of
particular crops to mean climate change, their results should be treated
with caution.


Crop production in all areas could be adversely impacted by changes in the
incidence of plant diseases and insect pests. Milder winters could increase
the incidence of outbreaks of, for example, powdery mildew, brown leaf rust
of barley, strip rust of cereals (Meier, 1985), while also reducing
incidence of most fungal diseases. Higher temperatures could also cause a
proliferation of insect pests as warmer and longer growing seasons provide
time for pests to reproduce more often (Pimentel and Pimentel, 1978).
Ironically, responses to such outbreaks, could actually damage agricultural
production in the long term. For example, an increased use of pesticides
will enhance soil degradation and thus soil erosion (Baric and Gasparovic,
1992).

The anticipated increase in the frequency of high temperature and drought
events could further undermine agricultural production in many regions.
This effects of just a short period of extreme conditions is illustrated by
the drought in 1993 when more than half the harvest was lost to drought in
some areas of Spain. In the Segura river basin (southeast Spain), 10,000
hectares could not be irrigated and 400,000 tons of green products were
lost (Revista del MOPT, January, 1994). Ultimately, the viability of crop
production would be jeopardised by an increase in the frequency of such
events. One study suggests that the anticipated reduction in moisture
availability would simply make soils unsuitable for cereal growing
throughout large parts of Spain, southern Italy and Greece (Le Houérou,
1992).

As conditions deteriorate for cereals, opportunities for other crops may
open up. Warmer climates and a longer growing season would extend the scope
for olive and citrus throughout much of the northern Mediterranean region.
One study suggests that a 3°C warming could extend the area of citrus crops
in Europe by a factor of three and effectively close the European market
for this product (Le Houérou, 1992). However, a study for the Kastela Bay
region (Croatia) is more pessimistic finding that "citrus and kiwi fruits
are already impossible to grow without irrigation" (Baric and others,
1996). Similarly, another study points out that warmer winters could lead
to the loss of adequate winter chilling for crops such as peaches,
nectarines and kiwi fruits (Reilly, 1996).

In the southern Mediterranean, the scope for olive production may also
increase in some areas. Moreover, increases in temperature may, however,
open the way for the more tropical species such as avocado, mango, banana,
paw paw and sugar cane (Le Houérou, 1992) - assuming there is an adequate
supply of water.

Crop production of all types in coastal areas will be vulnerable to sea
level rise. While the extent of inundation may be limited, problems of
increased saline intrusion and waterlogging due to the higher water table
are expected. The most vulnerable, and most studied, area at risk is the
Nile Delta (Egypt). If sea level rises by 1m, then 12 to 15% of existing
agricultural land in the Delta may be lost (Nicholls and Leatherman, 1995).
Even a 0.2 to 0.4 m rise in sea level could eventually cause agricultural
land in Egypt to be withdrawn from production over an area extending 20 km
inland (El-Shaer and others, 1997).

The actual impact of climate change on the agricultural production will
depend, not just on the nature of the change, but also on demographic and
socio-economic factors. These factors mean there are likely to be large
north-south differences in the impacts of climate change on production.
Cereal production is already marginal over much of southern Europe and
could according to one study "become untenable with a worsened water
balance, particularly if, as expected the price of cereals in the EEC
countries progressively declines to world market prices." (Le Houérou,
1992). As a result the land devoted to cereals in the northern
Mediterranean seems likely to shrink as climate changes, with marginal
lands being converted to other uses.

This contrasts markedly with the situation in North Africa and the Near
East where it is anticipated that cereal production may continue despite
ever-increasing risks (Le Houérou, 1992). Reductions in moisture
availability could cause crop expectancy in semi-arid areas to drop by 20%
to around 50 or 60%. In arid areas - where crop expectancy is only
one-fifth or less today, cereal growing would become more of a gamble than
it is today. Despite this, demographic pressures mean that cereal
production in semi-arid and arid areas is likely to continue and the area
devoted to it may even expand despite the crop losses (Le Houérou, 1992).
Such an expansion would bring its own problems, as the IPCC note, "[in]
Afro-Asian Mediterranean countries, extension of agriculture and
overgrazing in marginal areas, ... will probably lead to further
degradation of plant cover and soil loss" (Allen-Diaz, 1996).

It is unclear whether or not countries on both sides of the socio-economic
divide would be able to take advantage of the opportunities for new crops
that may arise. One study suggests that in the northern Mediterranean
"olive cultivation will inevitably shrink in spite of more favourable
conditions, because of increasing costs of labour and the difficulty in
developing mechanised harvesting" (Le Houérou, 1992). While, a study for
Tunisia suggests, "[a] great increase in olive cultivation could be
anticipated in the plain of Mateur except that the expected EEC olive
products import quotas will limit this change" (Hollis, 1992).

Clearly, radical changes in agricultural practice and policies may be
required as climate changes. The key question is: could these adaptations
be made? The IPCC is far from optimistic, highlighting firstly "there are
significant uncertainties about the capacity of different regions to adapt
successfully to the projected climate change" (IPCC, 1996a). The IPCC goes
on to suggest that "[during] water shortages, allocations to agriculture
will most likely decline before allocations to other uses". For the
developing countries of the south, the extent to which they can adapt will
also depend critically on the affordability of appropriate measures and
access to both know-how and technology.


Food Security

More important than food production is the question, will people have
enough to eat?  And if so, at what cost? Vulnerability to hunger or
malnutrition depends on a range of factors, of which local food production
is only one. These factors include: changes in the world market for
agricultural products, population trends, national economic well being and
income distribution. Experience tells us it is not enough just to produce
enough food globally, people must be able to access it.

Even today, there are significant differences in the vulnerability of
countries in the Mediterranean region to hunger. Most European countries
currently enjoy high levels of food security while Morocco and Algeria have
only moderate food security (Downing, 1992)25. Even in the absence of
climate change, it seems probable that food security in southern countries
will go down due to population growth, their low level of economic
development and ongoing land degradation.

Assessing the impacts of climate change on food security is immensely
complex. To do so, requires not just climate and crop modelling, but also
economic modelling of the world's trading system - thus adding yet another
level of uncertainty. Nevertheless, results from initial studies suggest
potentially serious implications for southern Mediterranean countries.
World trade models have been used to examine the implications of three
equilibrium GCM climate scenariosand associated crop yield changes (with-
or without-carbon dioxide effects and with- or without-adaptation) for food
prices and national welfare (Reilly and others, 1993)26.

The results show virtually all commodities could become more expensive, if
no allowance is made for carbon dioxide fertilisation effects and
adaptation. Under all climate scenarios the price of wheat, rice,
groundnuts, cotton, sugar and tobacco at least doubles, with one scenario
suggesting price hikes of over 500% for rice and tobacco. Even if allowance
is made both carbon dioxide effects and adaptation, then prices of key
commodities such as wheat, maize, sorghum, rice, soybean meal, sugar, beef,
pork and poultry rise in at least two out of three climate scenarios

The effect of the price rises means that agricultural exporters may gain
even though their supplies fall, but some food-importing countries may
suffer economic losses even if they manage to increase food production.
Assessments of net welfare change for regions encompassing Mediterranean
countries suggest that all regions may loose substantially in economic
terms, although there is a chance of a small gain under more optimistic
scenarios (Table 3).


Table 3: 	Economic welfare implications of the impact of climate
change on global agricultural production. The range takes account of
results under three equilibrium GCM scenarios for a doubling of carbon
dioxide and includes direct carbon dioxide effects. Data extracted from
Reilly and others, 1993.


* Yields decrease in at least one climate scenario if no account is taken
of carbon dioxide fertilisation effects.
** Yields decrease in at least one climate scenario even allowing for
carbon dioxide fertilisation effects and some adaptation.
For country coverage, see table 2.


For developed countries, such losses may be simply financial, but in poorer
developing countries, there is a risk of malnutrition, starvation or
hunger. In the Mediterranean region, the most at risk seem likely to be
Albania, Algeria, Egypt, Lebanon, Morocco, Syria and Tunisia - all of which
has a Gross Domestic Product below US$2,000 per person, closely followed by
Croatia and Turkey (1993 data from WRI/IIED/UNEP, 1997).


Wider Socio-Economic Implications

Almost all aspects of human society are dependant on climate and most would
be affected by climate change either directly or as the effects cascade
through the socio-economic system. The combined impacts could be a much
more serious risk to human well-being than the effects on any individual
sector.

As yet, there are few fully integrated studies of the impacts of climate
change in the Mediterranean region and this makes it is impossible to gauge
the full ramifications of climate change. Nevertheless, even a brief
assessment of four areas: human health, natural ecosystems, industry and
infrastructure and social disruption illustrates the potential seriousness
and diversity of threats posed by climate change.


Human Health

In a 1995 WHO/UNEP report, Anthony McMicheal declares, "[in] destabilising
the world's climate and its dependent ecosystems, we are posing new and
widespread risks to public health" (WHO/UNEP, 1995). This finding is
further supported by the work of the IPCC (McMichael, 1996). Direct effects
will arise from increases in the frequency and severity of heat waves and
other extreme weather events. But, the most significant impacts are likely
to come indirectly through the effects of climate change on, for example,
the distribution and prevalence of infectious diseases, water quality, food
security and sea level rise.

Quantifying the full impact of climate change on health is extremely
difficult. This is partly because many modelling techniques are still in
their infancy, but partly because impacts will depend on numerous
interacting factors including other environmental trends, social resources
and pre-existing health status. Nevertheless, sufficient understanding
exists to identify some of the major pathways by which the health of the
population of the Mediterranean region may be at put atrisk by climate
change.

An increase in the frequency and severity of heat waves would both increase
illness and death rates. Some of the most serious effects may be felt in
large cities, where extreme heat could aggravate local pollution problems
and increase the incidence of respiratory illness. The sensitivity of
populations in the Mediterranean region to such an outcome is highlighted
by conditions in Athens in June to August 1993 where a combination of
pollution and high temperatures during a heat wave caused more than a
thousand people to seek medical attention in Athens (WMO/UNEP, 1994).

Increases in extreme weather events, such as storms or intense rainfall
would increase death and injury rates. Torrential rain and the floods
and/or destruction associated with it is already responsible for deaths in
Mediterranean countries. For example, torrential rain in September 1991 led
to one of the deadliest floods on record in southern France as 32 people
drowned (WMO/UNEP, 1994). Events such as this could become more common if
rainfall intensifies as climate changes.

Warmer conditions could increase the incidence and extent of infectious
diseases such as malaria, dengue fever, schistosomiasis and yellow fever.
In recent years, the prevalence of the schistosome-spreading snails has
increased world-wide due to the spread of irrigation. Climate changes
could, however, increase the incidence of schistosomiasis in Egypt at least
(Gillet, 1974; WH0, 1990). Currently, snails lose their infections in the
winter, but with higher temperatures, they could cause infection throughout
the year. Climate change will also increase the extent of potential
malarial areas.

Climate change is already playing a part in the recent resurgence of
infectious diseases world-wide. Even with the minor increase in temperature
experienced to-date, malaria and dengue are now being reported at higher
latitudes in Africa, Asia and Latin America than at any time this century
(WHO/UNEP, 1995). How great an impact climate change has on human health
will ultimately depend, however, on finding ways to tackle the ongoing
growth in pesticide and drug resistance.

Some of the most important impacts are likely to come, however, through the
deteoriation of social and economic circumstances in areas affected by
desertification and sea level rise. Falls in food production and higher
prices could significantly increase the risk of widespread malnutrition and
hunger in countries with little, or inadequate welfare support mechanisms.
Water shortages and increased concentrations of pollutants could, together
with the higher temperatures, increase the risk of cholera, salmonella and
dysentery. This problem could be compounded by damages to drainage and
sewerage infrastructure through sea level rise (Attard and others, 1996).

The health impacts on populations in the Mediterranean region will vary
significantly both in terms of the risks individual countries are exposed
to and still more critically, health services available and economic
well-being. The most severe problems are likely to be faced by those
countries which already have the biggest problems today - notably those in
North Africa and the near East. In these areas, child mortality rates among
under-fives are already average six times higher than rates in southern
European countries (WRI/IIED/UNP, 1989). Climate change could further
increase this divide.


Natural Ecosystems

Future climate change and the associated environmental changes would
directly threaten natural ecosystems in the Mediterranean region, including
biosphere reserves and wetland sites of recognised international importance
(Figure 8; Milliman and others, 1992). Ecosystems have both an intrinsic
value and a value to humanity as they provide a variety of goods and
services - from fire wood to soil retention and from filtering pollution to
recreation.


Figure 8: 	Wetland sites of international importance, MAB biosphere
reserves and UNESCO World heritage sites (UNEP, 1987).

Despite some legal protection, natural ecosystems throughout the Mediterranean region are being lost or damaged as a result of, for example, land co
nversion, pollution and infrastructure developments. Climate change will,
however, pose a still more fundamental threat to the integrity of
ecosystems and the activities they support.

Climate change will directly affect ecosystems though the associated
increase in carbon dioxide levels, sea level rise and changes in
temperature and other climate variables. But there will also be secondary
impacts, for instance, an increase frequency of drought might decrease the
ability of trees to resist pests.

The ability of natural ecosystems to respond to changes in climate
boundaries will highly dependent on the rate of change itself. While
animals and birds can respond fairly swiftly it is, as the IPCC point out,
"unlikely that future rates of [tree] species migration could match those
required by the currently expected rates of climate change" (Kirschbaum and
others 1996). In theory, some other plant species could keep up with the
rate of change, but in practice their ability to establish themselves
elsewhere will be constrained by the extent of desertification and by human
land uses. As a result many valuable species and habitats may be lost.

Wetland sites would be affected both by the drier conditions and by the
effects of sea level rise. The acute sensitivity of wetland sites to higher
temperatures is illustrated by the results of one study that suggests that
an increase of 3 to 4°C would decrease the aerial extent of hydrophyte
wetland in southern Europe by 70 to 85% (Brock and van Vierssen, 1992). The
IPCC observe that sea level rise in combination with human activities could
threaten half the world's coastal wetland sites, including those around the
Mediterranean sea (Warrick and others, 1996).

The potentially devastating impact of climate change when combined with
other pressures is illustrated by a study on the Ichkeul National Park
(Tunisia) (Hollis, 1992). This shows that the combination of rise in
temperature and a dam scheme would cause the loss of all food plants for
wintering and breeding waterfowl and the disappearance of nationally
important fisheries. Some groups of birds, notably flamingos and waders,
may benefit but only marginally.

In the case of natural ecosystems, the scope for adaptation through human
intervention is limited as yet there is no way of maintaining species
diversity and ecosystem integrity artificially in the face of long-term
environmental change.


Industry and Infrastructure

Impacts on industry will vary, with some benefiting - and others losing -
from climate change. Food processing industries could clearly suffer if
food production goes down, while one soft drinks industry could be a
potential winner - during the first six months of 1993, Seville inhabitants
drank 11,000 million pesetas (almost US$80 million) in bottled water
(Revista del MOPT, January 1994). However, the soft drinks and other
industries could all be adversely affected by irregularities in both water
and energy supplies.

Supplies of hydroelectric power would be significantly reduced if the
frequency of droughts increases and/or there is an overall reduction in
runoff and increase in evaporation. This risk has already been identified
in Greece, where one study points to a very large increase in the
probability of being unable to generate the design power from hydroelectric
reservoirs (Mimikou and others, 1991). The extent, and possibly the
immediacy, of this problem was highlighted in 1993 when restrictions on
irrigation and hydroelectric output were imposed in Morocco during a
prolonged dry spell (WMO/UNEP, 1994).

The anticipated increase in rainfall intensity could also cause havoc. In
July 1991, exceptionally heavy showers and thunderstorms in Turkey led to
widespread flooding, several deaths and extensive material damage from
Istanbul to Samsum, including the demolition of several bridges and
toppling of power lines (WMO/UNEP, 1994).

The most serious impacts on both industry and infrastructure are likely to
be caused by rising seal levels. The coastal zone is host to both thermal
and nuclear power-plants (Figure 9) and host of other industries: including
73 petrochemical plants, 28 metallurgic plants, and 56 chemical industry
plants (Baric and Gasporavic, 1992). Located in the coastal zone for easy
access to transportation and water supplies, they are particularly
vulnerable to sea level rise. The extent to which any particular facility
is affected will depend on local conditions and as yet no full assessment
has been undertaken, although a number of case studies have highlighted
some industries which may be adversely affected.


Figure 9: 	Thermal power-plants located in the coastal zones (Baric
and Gasporavic, 1992).

In the Inner Thermaikos Gulf (Greece) region, low-lying areas such as the
developing industrial zone of Sindos and the Micra airport have been
identified as totally exposed to even a minor rise in sea level (Georgas
and Perissoratis, 1992). In the Kastela Bay region (Croatia), sea level
rise would directly affect the ship yards, chemical plants and ferro-alloy
plants, while a rise of just 0.5 m would flood the western part of the city
harbour (Baric and others, 1996). In the Cres-Losinj Islands, 5-10% of the
total water supply network is close to the shore and would risk inundation
at high tides (Randic and others, 1996)

Rising sea levels would also have serious implications for many heritage
sites of national and/or international importance (Figure 8) and popular
beach resorts. For example, a rise in sea level of just 0.5 m would cause
serious degradation to Venice (Italy) and threaten neighbouring towns of
artistic and historical significance (Sestini, 1996). Similarly, in the
Cres Losinj Islands (Croatia), the 4000 year old city of Osor and the
historic city of Cres would be directly at risk from rising sea level
(Randic and others, 1996). Beach resorts throughout the region would be
affected by loss of their key attraction - the beach. In Crete, for
example, many regions could lose at least half their current extent of
beach with only a 50 cm rise in sea level (Georgas and others, 1996).

The loss of some of the regions natural and historical heritage, combined
with enhanced beach erosion and growing water scarcity could have serious
implications for the tourist industry. Tourism is a major industry in
Mediterranean countries - in 1984, there were 100 million tourists in
coastal areas. In the absence of climate change, this number is projected
to grow to 170-340 million by 2025 (UNEP, 1987). But, with the loss of some
of the attractions this growth may not be realised, despite possible
benefits arising from an extension of the tourist season due to warmer
weather. In any event, climate change combined with a growth in tourism
would add to water scarcity problems in the region.

Social Disruption

Serious social disruption could arise due to tensions arising from
reductions in water availability and both internal and external migrations
of displaced people. Examples of minor disruptions exist even today. For
instance, protesters blocked Istanbul's international airport to highlight
the failure of the government to deal with the countries worst water
problem in 40 years (Toronto Star, 5 August, 1993). Desertification is also
known to exacerbate political instability as it contributes significantly
to water scarcity, famine, displacement of people and social breakdown.
This is, according to UNEP, a recipe for political instability and tensions
between countries (UNEP Fact Sheet 10).

International conflicts could arise over rights to shared river and aquifer
resources as a result of both climate change and population and development
pressures (Gleick, 1993). There are already disputes between, for example,
Syria and Turkey over shares in the waters of the Euphrates and between
Egypt, Sudan and Ethiopia over the Nile. Existing regional agreements will
become still more strained, if water availability declines or as demand
increases as a result of climate change or regional developments.

Environmental refugees could become a still greater problem as millions are
uprooted by shoreline erosion, coastal flooding and agricultural
disruptions" (IPCC, 1990, Myers, 1993). Desertification inevitably leads to
migration to cities as the land cannot support the original inhabitants
(El-Karouri, 1996), while in Egypt alone it has been estimated that a sea
level rise of only 0.5 m would permenantly displace 16% of population
(Broadus and others, 1986). Northward migration flows across the
Mediterranean Sea are already increasing as people are attracted into the
newly prosperous southern European countries (King, 1996). As climate
changes, European countries are likely to come under still greater pressure
from hungry multitudes to the south.


National Economy

Adverse impacts on national economies will arise from a number of
directions, including:

*	Direct impacts of climate change. These could range from possible
negative impacts on the balance of payments due to changes in agricultural
production and relative food prices, increased health problems and to
opportunity costs arising from irreversible damage.

*	Responses required to mitigate the impacts of climate change. These
could range from increases in the cost of efforts to reduce desertification
to welfare payments for displaced people. These can be expensive, for
example, the cost of combating desertification, even a decade ago, was
estimated to range from US$25 per hectare for rangeland to around US$750
per hectare of irrigated land (Dregne, 1983).

*	Knock-on implications from changes elsewhere. The impacts of
climate change will be felt world-wide and this could significantly affect
both import and export markets. On another level, the cost of accommodating
environmental refugees could be substantial.

Quantifying such costs is extremely problematic, however, studies do point
to substantial economic costs for Mediterranean countries. A study for
Greece, for example, found that the total cost of climate change could be
an estimated US$2.6 to US$5.9 billion per year (Dalianis and Petassis,
1993). This includes estimated annual losses equivalent to: US$2.3 to
US$3.9 billion due to lost agricultural production; US$35.6 to US$71.2
million to ameliorate the impacts of sea level rise; a minimum of US$8.4
million for preservation of biodiversity at risk from climate change.

Past efforts to cost the global impacts of climate change have generally
met with controversy owing to several factors, including: the relative
economic value assigned to lives in developed and developing countries and
the discounting of future impacts. Economists assign a much lower monetary
value to life in developing countries, which is inappropriate when dealing
with a global problem. Even so, these estimates show that developing
countries will suffer larger relative damages in economic terms than
developed countries.

While economic assessments may provide some insights into the nature of
economic impacts, they are ultimately misleading. Many aspects of human and
ecological welfare simply cannot be measured in monetary terms. Moreover, a
perverse characteristic of economic studies is that some attempts to
mitigate adverse impacts of climate change could increase Gross Domestic
Product (GDP) as they increase cash flow through the economy. To simply
count this as a plus is, of course, too simplistic since money spent on
tackling problems produces only marginal and short-term benefits and, in
the absence of climate change, could be more constructively used to benefit
human well-being.



5.	IMPLICATIONS FOR SUSTAINABLE DEVELOPMENT

The UN Conference on Environment and Development - the "The Earth Summit" -
in 1992 signalled international recognition of both the inherent
unsustainability of current development paths and the pressing need to
reorientate social and economic policies accordingly. Climate change and
its consequences threatens to undermine all such attempts to do so in the
Mediterranean region unless action is taken to both reduce adverse impacts
and emissions. Achieving this will require fundamental changes to social
and economic policies and significant international support for developing
countries.


Linking Climate Change and Sustainable Development

The most widely accepted definition of sustainable development is that
adopted by the World Commission on Environment and Development:
"development which meets the needs of the present generation without
compromising the ability of future generations to meet their own needs"
(WCED, 1987). Three basic principles underlie this objective: environmental
sustainability, development and equity (Karas, 1995). Climate change and
its consequences may cut against all three.

*	Environmental sustainability - recognises that sufficient resources
need to be retained or renewed to allow both current and future generations
to meet their needs. Climate change in the Mediterranean region may reduce
the capacity of the environment to support human activity in future by
exacerbating problems of desertification and water scarcity in the region.

*	Development - is a broad concept that encompasses improvements in
not merely economic terms, but also in social and environmental welfare.
The sheer magnitude of the potential problems created by climate change
seriously undermine the ability of countries to improve conditions for
their people.

*	Equity - with a particular stress on meeting the needs of the poor
today and of future generations. The fact that poorer southern
Mediterranean regions are likely to be hardest hit by climate change goes
directly against this principle, as does the prospect of permanent
degradation of the environment.

Even without climate change, current developments in the Mediterranean
region are wholly unsustainable. The most fundamental problem is that human
activities are already seriously degrading the environment as a life
support system, through their key role in desertification, over-extraction
of water from aquifers and pollution. Problems are further exacerbated in
the south where population growth presents the ever-more challenging
prospect of improving quality of life while also accommodating more people.

In this sense, climate change will largely exacerbate existing problems
rather than create new ones. But, there are three key dimensions to the
problem of climate change which mean that it cannot be ignored. First, the
near-inevitability of impacts over the next few decades due to time lags in
the climate systems response to past emissions; second, the irreversibility
on human timescales of, in particular, the impacts on desertification,
water resources and sea level rise; and third, the need for international
action to cut emissions of greenhouse gases if the threat of climate change
is to be reduced.

In view of these factors, ensuring sustainable development in the
Mediterranean region will require action both to adapt to the impacts of
climate change and to minimise emissions of greenhouse gases.


Tackling the Impacts

The commonalties between the possible adverse impacts of climate change and
existing problems of environmental degradation, food and water insecurity,
and so on suggest a variety of common solutions. These range from
improvements in land use and water management, through to technical fixes
and better social support systems.

Mediterranean countries are already signed up to a number of international
and regional agreements which effectively establish a policy framework for
actions that would mitigate many of the impacts of climate change. These
include: the Agenda 21, the Framework Convention on Climate Change, the
International Convention to Combat Desertification, the Convention on
Wetlands of International Importance and the Biodiversity Convention, the
Mediterranean Action Plan, and the European Union's (EU) Fifth Action
Programme on the Environment "Towards Sustainability".

Climate change adds to the need for the effective implementation of these
agreements. However, at the moment, the prospects do not look good. While
the agreements have stimulated significant research and a range of
small-scale innovative projects, the pace of environmental degradation
continues to increase.

The root of the problem is that the principles of sustainable development
have yet to be reflected in wider social and economic policies. The
consequences of this have been noted by the IPCC: "[many] current
agricultural and resource policies - already a source of land degradation
and misuse ... will discourage effective adaptation measures" (IPCC,
1996b). For example, the EU's Common Agricultural Policy has been shown to
actively contribute to the ongoing problem of desertification (MEDALUS II,
1996). Similarly, there seems little prospect of improving water security
in Syria while free water is handed out to industry (Al-Shalabei and
others, 1996).

Clearly, radical changes in both policies and practices are needed if
countries are going to successfully reduce existing problems, let alone the
adverse impacts of climate change. Furthermore, it is widely recognised
that climate change will add to the costs of appropriate responses and, as
the IPCC note in the case of agriculture, "[the] incremental costs of
adaptation strategies could create a serious burden for developing
countries (Watson and others,1996). In view of this, it will be vital that
developed countries meet their obligations under the Climate Convention to,
"assist the developing country Parties that are particularly vulnerable to
the adverse effects of climate change in meeting costs of adaptations to
those adverse effects" (Article 4, paragraph 4).


Reducing the Problem

Ultimately, sustainable development in the Mediterranean region could hinge
critically on getting international agreement - and the action to back it
up - to curb emissions of greenhouse gases. The Framework Convention on
Climate Change is clear in its objective, which is:

"...stabilisation of greenhouse gas concentrations in the atmosphere at a
level that would prevent dangerous anthropogenic interference with the
climate system. Such a level should be achieved within a time frame
sufficient to allow ecosystems to adapt naturally to climate change, to
ensure that food production is not threatened and to enable economic
development to proceed in a sustainable manner. (Article 4)

Achieving this objective will require stopping and then reversing the
current global growth in emissions by deep cuts (Alcamo and Kreileman,
1996; Karas, 1991). But, while the signing of the Convention at the Earth
Summit in June 1992 was an important landmark, emissions continue to
escalate.

Clearly, it is in the interests of Mediterranean Countries to push for
international action to cut emissions of greenhouse gases particularly by
countries with the greatest per-capita emissions. However, the deep cuts in
emissions required to stabilise atmospheric concentrations of greenhouse
gases means that all countries need to find ways of meeting their
development requirements while also minimising greenhouse gases emissions.

The IPCC highlights the feasibility achieving this goal, stating, "[the]
consequences of different technological choices and strategies to date
demonstrate that different industrial development paths - with
substantially lower emissions are possible" (Kashiwagi, 1996). The IPCC
further points out, A[if] carried out with care ..... responses would ...
enhance the prospects for sustainable economic development for all peoples
and nations" (Watson and others, 1996).

Reorientating development along more a more sustainable path requires
fundamental changes in sectoral and fiscal policies, particularly with
regard to transport and energy. A central plank of any response strategy in
Mediterranean countries, as elsewhere, must be to find ways of meeting
growing demand for energy while minimising greenhouse gas emissions. While
the possible reduction of the viability of hydroelectricity is unfortunate,
other renewable sources of energy - solar, biomass and wind - are of proven
value. Of these, solar power is particularly promising in the Mediterranean
region.

As in the case of adaptation strategies, major obstacles to the adoption of
appropriate strategies include limited technical knowledge and financial
resources. Here again, it will be vital that developed countries meet their
obligations under the climate convention to "take all practical steps to
promote, facilitate and finance, as appropriate, the transfer of, or access
to, environmentally sound technologies and know-how" (Article 4, paragraph
5).


6.	CONCLUSIONS

Scientists are confident that if current emissions of greenhouse gases
continue, the world will warm, sea levels will rise and regional climates
will change. However, large uncertainties remain over how climate in the
Mediterranean region will change and over its likely impacts. Nevertheless,
after reviewing recent work in the area, a number of conclusions can be
drawn.

*	The Mediterranean region is likely to warm significantly over the
next century and beyond in response to rising concentrations of greenhouse
gases. It is impossible to be certain over the precise pattern or scale of
warming, but it is likely that warming rates over some inland areas will be
much greater than the global average, while rates elsewhere may be slightly
lower than average.

*	Warming will be accompanied by changes in precipitation, moisture
availability and the frequency and severity of extreme events. Significant
uncertainties remain over future precipitation patterns in the region, but
the balance of current evidence suggests annual precipitation may decline
over much of the Mediterranean region. Moisture availability may go down
even in areas where precipitation goes up due to higher evaporation and
changes in the seasonal distribution of rainfall and its intensity. As a
consequence, the frequency and severity of droughts could increase.

*	Aerosols and other human influences on climate may mask more
fundamental changes in climate. In some areas, the effects of these other
influences may be sufficient to give an illusion either of no underlying
trend, or an opposite tendency to that expected. But, in the longer-term
the effects of greenhouse gases are expected to dominate future climate
changes in the Mediterranean as their influence grows over time.

*	It is likely that the changing composition of the atmosphere is
already influencing climate in the Mediterranean region. Climate records
for the region suggest a number of unusual characteristics to recent
climate behaviour. These fluctuations could all have occurred naturally.
But, it is notable that some aspects of recent climate variations are
consistent with changes that climate models suggest could occur.

*	Sea level rise and a reduction in moisture availability would
exacerbate existing problems of desertification and water scarcity and
substantially increase the risks associated with food production. Coastal
areas are directly threatened by rising sea levels, but the risks arising
from changes in moisture availability and the intensity of rainfall remain
difficult to quantify because of the large scientific certainties and the
concurrence of ongoing trends in land degradation.

*	Climate change would have far-reaching consequences affecting, for
example, peoples health, the integrity of ecosystems and the services they
support, industry, the risk of social disruption and national economies.
Again the most greatest adverse impacts would arise from rising sea levels
and the possible reduction in moisture availability.

*	The most serious impacts are likely to be experienced North African
and the eastern Mediterranean countries where ongoing desertification,
poverty and demographic change are already major problems. Adaptation could
be impeded in many areas by anticipated rate of climate change, limited
access to technical expertise and wider social and economic circumstances
and policies.

*	The impacts of climate change could seriously undermine efforts to
reorientate societies towards sustainable development. The unsustainability
of existing trends in the Mediterranean region is already widely
recognised. However, the irreversible character of some of the possible
impacts of climate change threatens to undermine all efforts to reorientate
societies along more sustainable paths.

Finally, Mediterranean countries have little choice but to try and adapt as
best they can to the almost inevitable changes in climate arising from past
emissions of greenhouse gases. But, the long-term sustainability of the
region could hinge critically on prompt international agreement, backed up
by action, to substantially cut emissions of greenhouse gases. Developing
countries of the southern and eastern Mediterranean are likely to need
international support to enable them both to adapt and to keep their own
emissions of greenhouse gases to a minimum.



GLOSSARY

Acronyms

CO2	Carbon dioxide
EEC	European Economic Community
ENSO	El-Niño-Southern Oscillation
EU	European Union
GCM	Global Circulation Model
ICCD	International Convention on Combating Desertification
IPCC	Intergovernmental Panel on Climate Change
IPCC92a	The IPCC's "baseline" or "reference" scenario of future emissions
of 			greenhouse gases
FCCC	Framework Convention on Climate Change
MEDALUS	Mediterranean Desertification and Land Use
NAO	North Atlantic Oscillation
UNEP	United Nations Environment Programme
US	United States of America
WHO	World Health Organisation


Units

cm	centimetres
m	metres
mm	millimetres
°C	degrees centigrade/Celsius
km	kilometres
m	cubic metres
ppbV	parts per billion by volume
ppmV	parts per million by volume
%	percent

REFERENCES

Alcamo, J. and Kreileman, E., 1996. The Global Climate System: Near Term
Action for Long Term Protection. RIVM Report no. 481508001. Bilthoven, The
Netherlands; Rijkinstitut voor Volkesgezonheid en Milieu.

Allan, R. J., and D'Arrigo, R. D., 1996. Persistent ENSO sequences: how
unusual is the recent El Niño? Holocene.  Submitted. Cited in: Nicholls and
others, 1996.

Allen-Diaz, B., 1996. Rangelands in a changing climate: impacts,
adaptations, and mitigation. In: Watson, R.T., and