Loading ...
Sorry, an error occurred while loading the content.

Climate Change and the Mediterranean Region

Expand Messages
  • souscayrous
    EXECUTIVE SUMMARY Water shortages and poor harvests during the droughts of the early 1990s exposed the acute vulnerability of the Mediterranean region to
    Message 1 of 1 , Oct 28, 2005

      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

      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

      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

      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

      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

      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


      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.


      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

      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

      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

      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

      * 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,
      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

      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,

      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

      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

      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,

      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,

      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.


      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).


      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.


      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,

      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 o<br/><br/>(Message over 64 KB, truncated)
    Your message has been successfully submitted and would be delivered to recipients shortly.