Technical ReportPDF Available

Climate Change Impacts in the Mediterranean Resulting from a 2°C Global Temperature Rise

Authors:

Abstract and Figures

Our results show that a global temperature rise of 2°C is likely to lead to a corresponding warming of 1-3 °C in the Mediterranean region. The warming is likely to be higher inland than along the coast.The largest increase in temperature is expected to take place in the summer, when extremely hot days and heatwaves are expected to increase substantially, especially in inland and southern Mediterranean locations. Under the A2 scenario, a drop in precipitation seems to be the dominant feature of the future precipitation regime. Under the B2 scenario, rainfall increases in the northern Mediterranean, particularly in winter. However, under both scenarios precipitation decreases substantially in the summer in both the north and the south. Under both A2 and B2 scenarios, fire risk is shown to increase nearly everywhere in the Mediterranean region, especially in inland locations. The southern Mediterranean is at risk of forest fire all year round. In the Iberian Peninsula, northern Italy and over the Balkans, the period of extreme fire risk lengthens substantially. The only region that shows little change in fire risk is in the southeastern Mediterranean. Based on the same climatic data, we investigated the changes in agricultural crop yields using a well-established numerical model. Our results show a general reduction in crop yields (e.g. C3 and C4 summer crops, legumes, cereals, tuber crops). The southern Mediterranean is likely to experience an overall reduction of crop yields due to the change in climate. In some locations in the northern Mediterranean, the effects of climate change and its associated increase in carbon dioxide may have little or small positive impacts on yields, provided that additional water demands can be met. The adoption of specific cropmanagement options (e.g. changes in sowing dates or cultivars) may help in reducing thenegative responses of agricultural crops to climate change. However, such options could require up to 40% more water for irrigation, which may or may not be available in the future. We calculated heating degree days (HDD) and cooling degree days (CDD) in order to examine the change in heating and cooling requirements. Under both climate scenarios, HDD decreases substantiallyin the northern Mediterranean and CDD increases everywhere in the Mediterranean, especially in the south. This change can potentially shift the peak in energy demand to the summer season with implications for the need for additional energy capacity and increased stress on water resources. Changes to tourism in the Mediterranean were examined through discussions with experts and stakeholders. We expect that warmernorthern European summers would encourage northern Europeans to take domestic holidays and thus, not travel to the Mediterranean. In addition, more frequent and intense heat waves and drought are likely to discourage holidays in the Mediterranean in the summer. We expect that the Mediterranean holiday season may shift to spring and autumn.
Content may be subject to copyright.
Climate change impacts in the Mediterranean
resulting from a 2oC global temperature rise
A report for WWF
1 July 2005
C. Giannakopoulos, M. Bindi, M. Moriondo, P. LeSager and T. Tin
Published in July 2005 by WWF, the global conservation organization, Gland, Switzerland. Any
reproduction in full or in part of this publication must mention the title and credit the above-
mentioned publisher as the copyright owner. © text (2005) WWF. All rights reserved.
The geographical designations in this report do not imply the expression of any opinion
whatsoever on the part of WWF concerning the legal status of any country, territory, or area, or
concerning the delimination of its frontiers or boundaries.
B
Climate change impacts in the Mediterranean
resulting from a 2oC global temperature rise
Summary
Tina Tin, Christos Giannakopoulos, Marco Bindi
The goal of the present study is to provide the first piece of the puzzle in
understanding the impacts of a 2°C global temperature rise on the Mediterranean region,
using high temporal resolution climate model output that has been made newly available.
The analysis has been based on the temperature, precipitation and wind daily outputs of
the HadCM3 model using the IPCC SRES A2 and B2 emission scenarios. The study is
focussed on the thirty-year period (2031-2060) centred on the time that global
temperature is expected to reach 2oC above pre-industrial levels, as defined by an
earlier companion study. Changes in both the mean (temperature, precipitation) and the
extremes (heatwaves, drought) under the different scenarios were assessed. The
impacts of these climatic changes on energy demand, forest fire, tourism and agriculture
were subsequently investigated either using existing numerical models or an expert-
based approach. Based on recent studies, the impacts on biodiversity, water resources
and sea level rise in the region were also discussed.
Our results show that a global temperature rise of 2°C is likely to lead to a
corresponding warming of 1-3 °C in the Mediterranean region. The warming is likely to
be higher inland than along the coast. The largest increase in temperature is expected to
take place in the summer, when extremely hot days and heatwaves are expected to
increase substantially, especially in inland and southern Mediterranean locations.
Under the A2 scenario, a drop in precipitation seems to be the dominant feature of
the future precipitation regime. Under the B2 scenario, rainfall increases in the northern
Mediterranean, particularly in winter. However, under both scenarios precipitation
decreases substantially in the summer in both the north and the south. In the south, the
reduction in precipitation extends year round. Longer droughts are shown to be common,
and are accompanied by shifts in timing. In terms of extremes, the number of dry days is
shown to increase while the number of wet and very wet days remains unchanged. This
C
can imply that when it rains it will rain more intensely and strongly, especially at certain
locations in the northern Mediterranean.
Based on the above climatic variables, we calculated the Canadian Fire Weather
Index to provide an indication of the forest fire risk under the future climate scenarios.
Under both A2 and B2 scenarios, fire risk is shown to increase nearly everywhere in the
Mediterranean region, especially in inland locations. The southern Mediterranean is at
risk of forest fire all year round. In the Iberian Peninsula, northern Italy and over the
Balkans, the period of extreme fire risk lengthens substantially. The only region that
shows little change in fire risk is in the southeastern Mediterranean.
Based on the same climatic data, we investigated the changes in agricultural crop
yields using a well-established numerical model. Our results show a general reduction in
crop yields (e.g. C3 and C4 summer crops, legumes, cereals, tuber crops). The southern
Mediterranean is likely to experience an overall reduction of crop yields due to the
change in climate. In some locations in the northern Mediterranean, the effects of
climate change and its associated increase in carbon dioxide may have little or small
positive impacts on yields, provided that additional water demands can be met. The
adoption of specific crop management options (e.g. changes in sowing dates or cultivars)
may help in reducing the negative responses of agricultural crops to climate change.
However, such options could require up to 40% more water for irrigation, which may or
may not be available in the future.
We calculated heating degree days (HDD) and cooling degree days (CDD) in order
to examine the change in heating and cooling requirements. Under both climate
scenarios, HDD decreases substantially in the northern Mediterranean and CDD
increases everywhere in the Mediterranean, especially in the south. This change can
potentially shift the peak in energy demand to the summer season with implications for
the need for additional energy capacity and increased stress on water resources.
Changes to tourism in the Mediterranean were examined through discussions with
experts and stakeholders. We expect that warmer northern European summers would
encourage northern Europeans to take domestic holidays and thus, not travel to the
Mediterranean. In addition, more frequent and intense heat waves and drought are likely
D
E
to discourage holidays in the Mediterranean in the summer. We expect that the
Mediterranean holiday season may shift to spring and autumn.
Based on results from existing studies, a global warming of 2°C and its associated
reduction in precipitation are expected to reduce surface runoff and water yields in the
Mediterranean region. In some countries, this could result in water demand exceeding
available water supply. In terms of biodiversity, climate change is likely to lead to shifts in
the distributions and abundances of species, potentially increasing the risks of extinction.
In addition, forest fires are expected to encourage the spread of invasive species which
in turn, have been shown to fuel more frequent and more intense forest fires.
Acknowledgments
The authors would like to thank Clare Goodess, Bob Bunce, Rafael Navarro, Antonio
Navarra, Riccardo Valentini, Michael Case and Lara Hansen for their comments on
earlier drafts of this report. Special thanks goes to Clare Goodess for her help during the
initial phase of the project, and to Mark New, Daniel Scott and Jacqueline Hamilton for
their helpful discussions.
F
Contents::
Page
C Summary
1 Climate change impacts on the Mediterranean resulting from a 2°C
temperature rise.
54 Impact of a 2° C global temperature rise on the Mediterranean
region: Agriculture analysis assessment.
G
H
Climate change impacts in the Mediterranean
resulting from a 2oC global temperature rise
Dr. Christos Giannakopoulos,
Dr. Philippe LeSager,
National Observatory of Athens, Athens, Greece
2
1. The Mediterranean region and climate: Basic issues
The Mediterranean Region has many morphologic, geographical, historical and
societal characteristics, which make its climate scientifically interesting. In general, the
Mediterranean climate is characterised by mild wet winters and by warm to hot, dry
summers and may occur on the West Side of continents between about 30° and 40°
latitude.
The Mediterranean Sea, a marginal and semi-enclosed sea, is located on the
western side of a large continental area and is surrounded by Europe to the north, Africa
to the south, and Asia to the east. Its area, excluding the Black Sea, is about 2.5 million
km 2; its extent is about 3700 km in longitude, 1600 km in latitude and surrounded by 21
African, Asian and European countries. The average depth is 1500 m. with a maximum
depth of 5150 m in the Ionian Sea. The Mediterranean Sea is an almost completely
closed basin, being connected to the Atlantic Ocean through the narrow Gibraltar strait
(14.5 km wide, less than 300m deep at the sill). These morphologic characteristics are
rather unique. In fact, most of the other marginal basins have much smaller extent and
depth or they are connected through much wider openings to the open ocean. Moreover,
high mountain ridges surround the Mediterranean Sea on almost every side.
Furthermore, strong albedo differences exist in south-north directions (Bolle, 2003).
These characteristics have important consequences on air masses and atmospheric
circulation at the regional scale (e.g. Xoplaki 2002). The Mediterranean sea is an
important heat reservoir and source of moisture for surrounding land areas. It represents
an important source of energy and moisture for cyclone development and its complex
land topography plays a crucial role in steering air flow, so that energetic meso-scale
features are present in the atmospheric circulation.
Because of its latitude, the Mediterranean Sea is located in a transitional zone where
both mid-latitude and tropical variability is important and competes against each other.
The Mediterranean climate is exposed to the South Asian Monsoon in summer and the
Siberian high- pressure system in winter. The southern part of the region is mostly under
the influence of the descending branch of the Hadley cell, while the northern part is more
linked to the mid-latitude variability.
A further important characteristic of the Mediterranean Sea is the emergence of the
first highly populated and technologically advanced societies since, at least, 2000BC.
3
Because of the demographic pressure and exploitation of land for agriculture, the region
presents since ancient times important patterns of land-use change and important
anthropogenic effects on the environment, which are themselves interesting research
topics.
Nowadays, about 400 million people live in the countries around the Mediterranean
Sea. This densely populated area has large economic, cultural and demographic
contrasts. There are approximately 10-fold differences in GDP between the largest
economies of the European Union countries and small Middle East nations, and a 3 to 6-
fold difference in the GDP per-capita between Western European countries and the
other nations. Demographic trends are also quite different. European countries (also
including non EU nations) are close to a null growth and expected to stabilise or even
decrease their population, while North African and Asian countries are growing and are
expected to double their population by mid 21st century. In contrast to European
Countries, urbanisation for most African Nations is an ongoing process that is changing
the socio-economic structures of these regions. All these different trends are likely to
produce contrasts and conflicts in a condition of limited available resources. Moreover,
different level of services, of readiness to emergencies, technological and economical
resources, are likely to result in very different adaptation capabilities to environmental
and climate changes. Poorer societies with recently increased urbanisation are likely to
be critically vulnerable to weather extremes and incapable of adapting to changing
climate patterns. Hence the need is paramount for the best possible prediction of future
climate scenarios and descriptions of possible impacts and adaptation strategies.
2. Present trends of the Mediterranean climate
2.1 Introduction
Instrumental data reveal significant trends of Mediterranean temperature and
precipitation at different time and space scales. For instance, during the last 50 years of
the 20th century large parts of the Mediterranean experienced winter and summer
warming. For the same period, precipitation over the Mediterranean decreased.
However, the statistical significance is low due to the large interannual variability. These
trends, however, differ across regions and periods under consideration showing
variability at a range of scales in response to changes in the direct radiative forcing and
variations in internal modes of the climate system. It is one of the main challenges for
future research to understand the physical processes and causes responsible for these
trends. They seem to be hemispheric to global (such as external forcings and changes in
the large-scale atmospheric circulation), anthropogenic as well as local/regional (such as
changes in earth surface and land use, orography).
2.2 Observed Temperature Trends over the Mediterranean
Giorgi (2002) analysed the surface air temperature variability and trends over the
larger Mediterranean land-area for the 20th century based on gridded data of New et al.
(2000). He found a significant warming trend of 0.75°C per century, mostly from
contributions in the early and late decades of the century. Slightly higher values were
observed for winter and summer. Based on the same data, Jacobeit et al. (2003) found a
distinct summer warming for the 1969-1998 period. The structure of climate series can
differ considerably across regions showing variability at a range of scales in response to
changes in the direct radiative forcing and variations in internal modes of the climate
system (New et al. 2001; Hansen et al. 2001; Giorgi 2002). Figure 1 (right) presents the
linear trends of summer surface air temperatures (°C/50yr) for the period 1950-1999. It
also shows the stations, which experienced a significant trend.
A clear east-west differentiation in Mediterranean summer air temperature trends is
visible. Cooling, though mostly not significant, was experienced over the Balkans, and
parts of the eastern basin. In the other areas, there is a significant warming trend of up
to 3°C/50yr. However, the warming in these regions did not occur in a steady or
monotonic fashion. Over most of western Mediterranean for instance, it has been mainly
registered in two phases: from the mid-1920s to 1950 and from the mid-1970s onwards
(e.g. Brunet et al. 2001a, 2002, Galan et al. 2001). A glance at summer air temperature
trends for the 1900-1949 period reveals that warming, though less extreme than that in
1950-1999, was experienced in the western basin. A cooling trend over 1900-1949 was
only prevalent over Libya and Egypt. The trend of winter temperature over 1900-1949
indicates a general cooling in the central basin but a warming in the east and west. For
the 1950-1999 period, except for the eastern part, there was warming experienced.
Xoplaki et al. (2003) found a significant cooling trend of Mediterranean winter Sea
Surface Temperatures (SSTs) east of 20°E over the period 1950-1999, while the
western basin experienced warming.
4
5
Figure 1. (left) Linear trends of winter (NDJF) station precipitation (mm/50yr) and (right)
summer (JJAS) surface air temperatures (°C/50yr) for the 1950-1999 period. Stations with a
significant trend (90% confidence level, based on the Mann-Kendall test) are encircled (adapted
from Xoplaki 2002).
2.3 Observed Precipitation Trends over the Mediterranean
Recent studies revealed that the 20th century was characterized by significant
precipitation trends at different time and space scales (e.g. New et al. 2001, Folland et
al. 2001). Giorgi (2002) found negative winter precipitation trends over the larger
Mediterranean land-area for the 20th century.
Using the same data, Jacobeit et al. (2003) showed for the last three decades some
rainfall increases in autumn (western Iberia and southern Turkey), but dominating
decreases in winter and spring.
A glance at the Mediterranean regional precipitation trends reveals a more detailed
picture of the general findings. Sub regional variability is high, particularly in areas with
contrasted topography near coastland where also significant trend in variability and
monthly totals have been observed (e.g. Turkes 1996, 1998). The evaluation of regional
data series (Figure 1, left) indicate, that trends in many regions are not statistically
significant in view of the large interannual variability (Xoplaki, 2002). However, significant
decreases are prevalent in western, central and the eastern Mediterranean.
For the Mediterranean Sea, precipitation variability has been investigated using
gauge-satellite merged products and atmospheric re-analyses (Mariotti and Struglia,
2002). NCEP re-analyses show that during the last 50 years of the 20th century
Mediterranean averaged winter precipitation has decreased by about 20%, with the
decrease mostly occurring during the period late 1970s to early 1990s.
2.4 Observed Daily Rainfall and Temperature Trends over the
Mediterranean
Only few areas have been studied on a daily basis in the Mediterranean because
high quality data are rather scarce (e.g. De Luis et al. 2000). Difficulties exist in
determining trends of very rare events (e.g. Frei and Schär, 2001). One exception is the
dense daily rainfall data-base for the east of the Iberian Peninsula (Romero et al. 1998,
1999) which show successive drying in western Catalonia and central and western
Andalusia for the period 1964-1993. Brunetti et al. (2001ab) have found a negative trend
for the number of wet days and annual rainfall in Italy, while the heaviest events class
interval show a positive trends. Alpert et al. (2002), Brunetti et al. (2001ab), Goodess
and Jones (2002) also report on a tendency to more intense concentration of rainfall to
have occurred along some Mediterranean coastal areas, essentially Italy and Spain.
Similar results were found in two long observations in the north-eastern inland of Spain
(Ramos, 2001).
Over the western Mediterranean little change or even an increase of the day/night
temperature differences has been highlighted for the last 130 years (Brunet et al.
2001bc) and for the 20th century (Brunet et al. 1999, Abaurrea et al. 2001, Horcas et al.
2001). Maximum temperature increased at larger rates than minimum temperature. This
diurnal differential rate of warming, opposite to the observed on larger spatial scales, has
been mainly intensified during the second half of the 20th century.
3. Study overview
In this study, we conduct a “first-order” investigation on the impacts of a 2oC global
temperature rise in the Mediterranean basin. The analysis is based on the temperature,
precipitation and wind daily outputs of the HadCM3 model using two emission scenarios.
The study period is a thirty-year period (2031-2060) centred on the time of the 2oC global
temperature rise. Changes in both the mean (temperature, precipitation) and the
extremes (heatwaves, drought) under different future climate scenarios are assessed.
Subsequently, the impacts of these climatic changes on energy demand, forest fire,
tourism and agriculture are investigated. Impacts are examined using impact models
where such models exist in the literature (such as agriculture, forest fire, energy) or
using an expert-based approach when such models have not yet been developed (such
as for tourism). The likely magnitudes of uncertainties and the sensitivity of HadCM3
6
7
relative to other climate models are discussed. Based on recent work in the published
literature, we also examine the consequences of such a change in climate on water
availability, biodiversity and sea level rise in the region. The goal of the present study is
to provide the first piece of the puzzle in understanding the impacts of a 2°C global
temperature rise on the Mediterranean region, using high temporal resolution climate
model output that has been made newly available. Although only a single climate model
has been used in this study, we believe that our results are representative of average or
conservative values within the range of similar estimates from currently available climate
models. This arises from the use of a model of average sensitivity to investigate a period
during which there is reasonable agreement among models (Section 4.3). Before larger
scale and more comprehensive multi-model, multi-year studies are undertaken, such a
first-order study can provide invaluable information in the planning of future research or
policy directions.
Within the context of ongoing studies, the present study is an extension of the 3-year
European project MICE (Modeling the Impacts of Climate Extremes), which has involved
8 institutes since 2002 in analyzing the occurrence of extremes in climate models and
quantifying the impacts of climate extremes on selected European environments, using
the same climate model output from HadCM3 (Giannakopoulos and Palutikof, 2005;
Palutikof, 2004). MICE has now been superseded by the Integrated EU project
ENSEMBLES (involving 72 European research Institutes) , which compliments MICE by
providing probabilistic estimates of climatic risk and by characterising the level of
confidence in future climate scenarios through ensemble integrations of climate models
(Hewitt and Griggs, 2004).
4. Data and methods
4.1 Time of 2oC global temperature rise
Monthly data from six coupled ocean-atmosphere GCMs, each driven by several
forcing scenarios, were downloaded from the IPCC Data Distribution Centre with the aim
to determine the time of the 2oC global temperature rise. These models (seen in Figure
2) exhibit a range of sensitivity to greenhouse gas forcing. Following New (2005), for
each model, control-run surface temperature data were used to calculate a “pre-
industrial” mean temperature climatology, and these were spatially averaged to calculate
8
a global mean pre-industrial surface temperature. For each climate change simulation,
the global temperature fields were spatially averaged to calculate time-series of global
mean annual temperature, which were then differenced from the “pre-industrial” global
mean temperature. The resulting global mean temperature-anomaly series were then
smoothed with a 21-year moving average, and the date at which the 21-year mean
global temperature anomaly exceeded 2°C above pre-industrial levels was taken as the
time of 2°C global temperature change.
The time at which the simulated global mean temperature exceeds the control run
global mean by 2°C ranges from between 2026 and 2060. The inter-model spread for a
single scenario (e.g. B2) is nearly as large as the total spread; however, there is a
tendency for the scenarios with greater accumulated radiative forcing (e.g. A2) to exhibit
a greater rate of warming, and an earlier year of 2oC global rise.
Figure 2. Global mean annual temperature anomalies relative to control climatology, smoothed with a 21-
year moving average. Vertical lines indicate the range in time at which the 21-year global mean
temperature anomaly exceeds +2°C. Figures on the right show the time at which the 21-year mean global
temperature anomaly exceeds +2°C for each GCM-scenario combination (adapted from New, 2005).
9
300
400
500
600
700
800
900
1000
1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
CO
2
Concetration (ppm)
A1B
A1T
A1FI
A2
B1
B2
IS92a
1.0
3.0
5.0
7.0
9.0
1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Radiative Forcing (W/m
2
)
A1B
A1T
A1FI
A2
B1
B2
IS92ags
IS92agg
Figure 3. Estimated concentrations of CO2 and globally averaged increase in radiative forcing from all
greenhouse gases and aerosols (relative to preindustrial levels) arising from various IPCC emissions
scenarios. Scenarios used in this study are in bold. Data from Appendix II of IPCC (2001b). Adapted from
New (2005).
We focused on the results from 2 IPCC forcing scenarios, namely A2 and B2. As
described in Nakicenovic et al. (2000), the A2 storyline and scenario family describes a
very heterogeneous world. The underlying theme is self-reliance and preservation of
local identities. Fertility patterns across regions converge very slowly, which results in
continuously increasing global population. Economic development is primarily regionally
oriented and per capita economic growth and technological change are more
10
fragmented and slower than in other storylines. The B2 storyline and scenario family
describes a world in which the emphasis is on local solutions to economic, social, and
environmental sustainability. It is a world with continuously increasing global population
at a rate lower than A2 and intermediate levels of economic development. While the
scenario is also oriented toward environmental protection and social equity, it focuses on
local and regional levels. The B2 scenario family is based on the long-term UN Medium
1998 population projection of 10.4 billion by 2100. The A2 scenario family is based on a
high population growth scenario of 15 billion by 2100 that assumes a significant decline
in fertility for most regions and stabilization at above replacement levels. It falls below
the long-term 1998 UN High projection of 18 billion.
The radiative forcing under the two scenarios are less divergent for the expected
period of 2°C global temperature rise (2026-2060, Fig. 2) than for the latter part of the
century (Fig. 3).
4.2 The HadCM3 ocean-atmosphere coupled GCM
We have used daily output data from a coupled atmosphere-ocean general
circulation model (GCM) HadCM3, consisting of an atmospheric GCM coupled to an
ocean GCM. HadCM3 is a coupled atmosphere-ocean GCM, developed at the Hadley
Centre and described by Gordon et al (2000) and Pope et al (2000). Unlike earlier
atmosphere-ocean GCMs at the Hadley Centre and elsewhere, HadCM3 does not need
flux adjustment (additional "artificial" heat and freshwater fluxes at the ocean surface) to
produce a good simulation. The higher ocean resolution of HadCM3 is a major factor in
this. HadCM3 has been run for over a thousand years, showing little drift in its surface
climate. The control run is basically the GCM being run for 240 years at 1961-1990
atmospheric concentrations. Any variation in the control run is, hopefully, due solely to
natural variability. The last 30 years of the control run, 1961-1990, are used here as
reference for comparison with future predictions. The control run is unforced and thus
common to any scenario that may apply after 1990.
The atmospheric component of HadCM3 has 19 levels with a horizontal resolution of
2.5° of latitude by 3.75° of longitude, which produces a global grid of 96 x 73 grid cells.
This is equivalent to a surface resolution of about 417 km x 278 km at the Equator,
reducing to 295 km x 278 km at 45° of latitude (comparable to a spectral resolution of
T42). Thus the model geography is much simpler than the real-world geography. As an
example, only five grid boxes cover the UK and one land grid box represents continental
11
Greece (Fig. 4). This has to be kept in mind when analysing results. The oceanic
component of HadCM3 has 20 levels with a horizontal resolution of 1.25 x 1.25°. At this
resolution it is possible to represent important details in oceanic current structures.
Mediterranean water is partially mixed with Atlantic water across the Strait of Gibraltar as
a simple representation of water mass exchange since the channel is not resolved in the
model.
Figure 4. Land-sea mask of HadCM3 over the Mediterranean. The squares represent the size of
the HadCM3 grid cell. Grey colour denotes land and white denotes sea as represented in the
model.
The transient climate response (TCR) is often used to compare differences in model
response to the same standardised forcing. The TCR of HadCM3 is 2°C, which occupies
an average position within the range of TCRs (1.4°C – 3.1°C) of the 19 GCMs assessed
in the IPCC Third Assessment Report (IPCC, 2001a). More specifically for this study,
HadCM3 exhibits an average response to the A2 scenario and the most conservative
response to the B2 scenario, when compared with the five other models used to derive
the time of 2°C global temperature rise (Figure 2). Therefore, we believe that our results
represent average or conservative estimates of the impacts of a 2°C global temperature
rise.
The choice of HadCM3 over the other GCMs was made mainly because of the
availability of daily data in HadCM3 (which is fundamental for the study of extremes in a
region such as the Mediterranean) over the desired time period of the 2oC global
12
temperature rise. Output from regional models would provide higher spatial resolution
but are not available for this period. A study that uses one model, with one ensemble,
only gives one picture of the climate response to a given forcing scenario. In theory,
using more models and more ensembles should give more statistically reliable results. In
practice, this can be misleading because the models may all use the same forcing
model, which can influence the projected climate response more than the emissions
scenario. Moreover, there may be insufficient models/ensembles to give a representative
mean projected climate response so that the climate response can be biased towards
the least representative (probably the worst) model. Under these cases, it is arguably
better to use a single model that generally gives a good overall representation of current
climate and extremes and where you understand its weaknesses, which is what was
done in the present study. In addition, model comparison studies have shown that
results from different models for the study period agree with one another fairly well, while
most of the divergence takes place in the latter part of the century (2070-2100) (IPCC,
2001a).
4.3 Methodology
In order to determine the changes in the Mediterranean climate and their impacts as
a result of a 2°C global temperature rise, our study focused on the period of 2031-2060
(Fig. 2).
The meteorological data from HadCM3 model have been processed to produce
yearly characteristics such as the maximum length of the drought, the number of
summer days, or percentile values. These parameters have been averaged over the
1961-1990 (reference or control period) and the 2031-2060 periods. The results for the
two periods are then compared.
Both scenarios (A2 and B2) give similar results, except in few cases specifically
noted hereafter. The two scenarios bring more differences in the 2070-2100 period,
which is not the period of interest here.
Precipitation, and maximum (daytime), minimum (night- time), and mean
temperatures have been examined. Wind (max and mean) do not show any significant
changes and will not be discussed here.
13
4.4 Uncertainties
Most of our findings in this study will be subject to uncertainties corresponding to
more than one of the classes below:
1. Incomplete or imperfect observation. This is a joint property of the system being
studied (e.g., Earth’s climate, crop responses to climate) and our ability to measure it.
2. Incomplete conceptual frameworks, e.g., models that do not include all relevant
processes, etc.
3. Inaccurate prescriptions of known processes, e.g., poor parameterizations etc.
4. Chaos. This is a property of the system (e.g., Earth’s climate, crop responses to
climate) being studied.
5. Lack of predictability. Some aspects of societal prediction are much less amenable to
prediction than others. For example, in considering new technologies, uncertainties
associated with rates of market penetration of new technologies are smaller than those
associated with rates of onset of the new technologies themselves.
In seeking to characterize uncertainty, the concept of the ‘uncertainty cascade’,
shown below, has been developed (IPCC, 2001b). This starts with different socio-
economic assumptions that affect projections of GHG emissions, and flows through
differing potential emission scenarios and ranges of GHG concentrations, radiative
forcing, and climate system responses and feedbacks. These in turn affect the
estimation of the range of potential impacts, and the consideration of adaptation and
mitigation responses and policies.
Figure 5. The concept of “uncertainty cascade” in climate change impact analyses. Adapted from
IPCC (2001b).
14
From the schematic, in this study, we are working in the region where the range of
uncertainty is potentially large, because of the combined effects of the sources of
uncertainty higher up the cascade. It, therefore follows that, even if we had used more
than one models to perform the simulations and assess the impacts, our results would
still be subject to uncertainty. For example, in discussing uncertainties associated with
predictions of crop yields under climate change, the sources of uncertainty will come
both from the crop-climate impacts models and from the climate models which
provide the climate scenarios. For this, it is important to treat the results in this report
with caution. They do provide an estimate of the future climate and impacts in the
Mediterranean, but under no circumstances should we use them in a formal quantitative
way. A qualitative discussion, as has been done in this report, is only possible.
5. Climate change analysis
5.1 Mean temperatures
We focus here on the annual and seasonal mean temperature changes averaged
over the two 30year periods: the control (1961-1990) and the study periods (2031-2060).
This will provide a summary of how much the temperature will be increased above
present values, as a result of an increase of 2oC in global mean temperature over pre-
industrial levels. Spatial variations along the Mediterranean region are discussed.
5.1.1.Annual changes
Fig. 6 presents the differences between the daily mean temperature averaged over
2031-2060 and 1961-1990 for scenario A2. It is clear that the average rise in
temperature (daily mean averaged over 30 full years) is between 1-2 °C along the coast
and France, and between 2-3 °C inland. The thermal inertia of the Mediterranean Sea is
obvious. Under scenario B2 (fig.7), the rise in temperature is somewhat larger than in A2
(especially in France), but in general, the patterns of changes are broadly similar.
15
Figure 6. Difference between the daily mean temperature averaged over 2030-2060 and over 1961-1990,
for scenario A2.
Figure 7. As Fig. 6 but for scenario B2.
Figs. 8 and 9 show also the same pattern as Figure 6 but for the daily maximum (Tmax)
and minimum temperatures (Tmin) respectively. It is worth noting that the average rise is
slightly larger for Tmax than for Tmin. In the figures, this is particularly evident in the
Iberian Peninsula.
16
Figure 8. As Fig. 6 but for the daily maximum temperature.
Figure 9. As Fig. 6 but for the daily minimum temperature.
17
5.1.2 Seasonal changes
Figs. 10, 11 & 12 present the variations in mean (Tmean), maximum (Tmax), and
minimum (Tmin) temperatures, respectively, for each of the four seasons, as projected
under the A2 scenario.
Figure 10. As Fig. 6 but for each season (Winter=DJF, Spring=MAM, Summer=JJA, Fall=SON).
For Tmean (Fig. 10), the rise occurs mainly in summer, when it reaches 4 °C inland
on average. Fall is the second season to get warmer, with temp rises above the 2 °C
average. Winter is likely to be uniformly warmer by 1-2 °C. Spring experiences the
average 2 °C increase, except in the north-western part of the region, where the
warming is less.
Tmin (Fig. 11) features the same seasonal variation, with a slightly smaller increase
in summer than Tmean.
18
Figure 11. As Fig. 10 but for the daily minimum temperature.
Tmax (Fig. 12) features the same seasonal variation, with a rise notably larger than
Tmean in summer and slightly larger in fall.
19
Figure 12. As Fig. 10 but for the daily maximum temperature.
5.1.3 Summary
The global 2 °C temperature rise is expected to be seasonally and spatially
translated in the Mediterranean region by:
Largest increase in summer, and inland: Tmean by 4 °C and Tmax by 5 °C, on
average.
Second largest increase in fall: 2-3 °C everywhere.
Spring temperatures could rise by about 2 °C.
Winter and spring temperatures could rise less than 2 °C.
Although less pronounced, thanks to the sea, the rise in coastal region temperature
is expected to be in the 1-2 °C range on average, and a bit more than 2 °C in
summer for Tmax.
Tmax is expected to rise more than Tmin.
20
5.2 High temperatures
5.2.1 Summer days
The increase in the number of summer days, defined as the number of days when
Tmax exceeds 25oC, is from 2-to-6 weeks (Fig. 13 top). This is translated to about one
additional month of summer days on average.
Large increases are found in Central Mediterranean Region (i.e. Crete,
Peloponnese, South Greece, Sicily), North Adriatic, and inland (within Maghreb,
Spain, Turkey, South of France and the Balkans). In this group, the largest increase
seems to occur in Crete with an additional 7 weeks of summer and the smallest in
the Maghreb with an additional 3-5 weeks of summer.
On the other hand, the coastal regions of the western Mediterranean, the Black Sea
and the Middle East is expected to have the smallest increases with only 2-3
additional weeks of summer.
It is expected that there will be, on average, one additional summer month
everywhere inland, as well as in the central Mediterranean coastal region, and about
half a month on all other coastal regions except the central Mediterranean Region.
5.2.2 Hot days
The pattern in the number of hot days, defined as the number of days with Tmax> 30 °C
(Fig. 13 bottom) is somewhat different from the pattern of the number of summer days.
The increase is from 2 weeks along the coast to 5-6 weeks inland (within Spain, Turkey,
South of France, the Balkans and in the Maghreb) indicating the role the Mediterranean
Sea exerts in preventing too hot days.
5.2.3 Summary
According to our study, there exist four types of regions:
Inland: on average one additional month of hot days, and also one additional month
of summer days.
Along the coasts outside Central Mediterranean region: 1-3 weeks of additional hot
days, and 2-3 weeks of additional summer days.
The coastal regions in Central and Eastern Mediterranean region are expected to
have only few additional hot days (like the other coastal regions), but one additional
month of summer days, as in the continental part of the Mediterranean. Crete is the
21
perfect example: no change in the number of hot days, but an average of 7 additional
weeks of summer days.
It looks as if the Mediterranean Sea moderates temperature increases in the coastal
areas so while more summer days are forecast, few of these days will be on the hot
side.
Figure 13. Differences in the number of summer (top) and hot (bottom) days between control and future
period for scenario A2.
5.2.4 Tropical nights
The number of tropical nights, defined as the number of nights with Tmin > 20 °C
(fig.14) increases by about a month almost everywhere. Only regions well within land are
22
expected to keep their night fresh, with only 1-2 weeks per year of additional tropical
nights. A larger increase in tropical nights than in summer days is expected to be seen in
the South Eastern Mediterranean Region (SEMR). The SEMR extends from the
Israeli/Syrian coast to Egypt. The opposite is likely to occur inland (South of France,
Spanish interior and Turkey, the Balkans and North Adriatic).
Figure 14. Difference in the number of tropical nights (scenario A2).
5.2.5 Extremely hot days
Fig. 15 shows the percentage of days with temperatures in the future period (2031-
2060) above the value that is exceeded by 10% of the days in the reference period
(1961-1990). This value is also called the 90th percentile of the reference period, and
changes with location. In other words, Fig. 15 (top and bottom) gives the percentage of
days in the future period that fall within the range of the 10% hottest days of the
reference period. It can be seen that the temperature range of the 10% hottest reference
days and nights will be reached or exceeded by 20% of the future days and nights on
average. This is a doubling, which is associated with the increase in the number of
summer and hot days.Note that the lowest increases are found near the Atlantic Ocean,
and in Cyprus (in contrast to Crete), while the sharpest increase is in the SEMR hot spot.
Figure 16 shows the changes in the number of heatwaves in the future period,
defined as the change in the number of weeks per year with temperatures exceeding
35oC. The changes are about the same for both scenarios. Continental areas in Spain,
the Middle East, Turkey, the Balkans, North Africa and North Italy are expected to
experience an increase of 3-5 weeks of heatwave days. Areas under the moderating
23
influence of the Mediterranean sea are likely to see very small or no changes. Such
areas are all the islands, South Italy, and Peloponese. The sole exception occurs in the
islands in the North Aegean Sea, which are expected to experience more heatwaves (4
weeks more) in the future period.
Figure 15. Percentage of days (top) and nights (bottom) nights with Tmin above the 90th percentile of the
reference period (scenario A2).
24
Figure 16. Difference in the number of heatwaves (i.e. weeks with Tmax>35oC) under A2 scenario
between the two periods.
5.3 Low temperatures
Fig. 17 reveals that the number of frost nights, defined as the number of nights
with Tmin < 0 °C, falls by 1-2 weeks along the coast, and up to a month inland. The
number of very cold nights (Tmin < -5 °C) is not shown but it also has a decreasing
trend (though not as strong).
Figure 17. Difference in the number of frost nights (scenario A2).
25
Finally, in symmetry with the 10% hotter days and nights examined above, we
examine the 10th percentile of the coldest days and nights in relation to the reference
period. The percentage of days and nights reaching these ranges is shown in Fig. 18 for
both scenarios. The ranges of the reference period are expected to be reached by only
5% of the future days and nights. A 50% decrease in the number of coldest days is
therefore evident. The drop is larger in the SEMR, both during night and during day, and
a little bit more under B2 scenario.
Figure 18. Percentage of days (top row) and nights (bottom row) with temperature below the 10th
percentile of the control years, under scenario A2 (left column) and B2 (right column).
5.4 Precipitation
5.4.1 Annual rainfall changes
Figs. 19 & 20 give the variations in the mean total yearly rainfall for both scenarios.
Under scenario A2, a drop in precipitation seems to be the dominant feature of the future
precipitation regime. Under B2, some rainfall increases are expected in the northern part
of the region. Compared to temperature, precipitation exhibits larger differences between
the two scenarios .
Under A2, we can roughly identify two main areas:
0-10% drop in precipitation relative to the 1961-1990 values in the northern part
10-20% drop in precipitation relative to the reference period values in the southern
part including Spain.
26
Under B2, the northern part is expected to experience both increases and decreases
in total yearly rainfall while a 0-20% drop is evident in the southern part.
Figure 19. As Fig. 6, but for precipitation under A2 scenario.
Figure 20. As Fig. 6, but for precipitation under B2 scenario.
27
5.4.2 Seasonal rainfall changes
Figs. 21 and 22 represent the relative changes in precipitation between the two
periods and for scenarios A2 and B2 respectively. From these two figures it becomes
evident that:
Spring and summer rainfall regime exhibits no differences between the two
scenarios.
Fall seems to experience a little bit more rain under B2 scenario.
Winter exhibits the main differences between the two scenarios, with more rainfall
under B2, especially in France and Spain.
Figure 21. As Fig. 10, but for percentage precipitation changes in relation to 1961-1990
amount under A2 scenario.
The main feature of the seasonal variations in precipitation is the contrast between
the North - South and winter- summer:
28
A summer drop in the total rainfall over the northern region is expected, only partially
balanced by an increase in winter.
The opposite pattern is evident in SEMR (including the southern part of Turkey):
small increase in autumn rainfall, slightly larger decrease in spring.
All other parts of the Mediterranean (south European and North African countries)
are expected to see a decrease in the summer precipitation and a small decrease or
no change in the other seasons (mainly under B2 scenario).
Figure 22. As Fig. 10, but for percentage precipitation changes in relation to 1961-1990 amount under B2
scenario.
5.4.3 Number of dry and wet days
Figure 23 describes the differences in the yearly number of dry days between the
two examined periods. The dry days are defined when the daily precipitation amount
(RR) is less than 0.5mm. On average, the Mediterranean is expected to feature more dry
days. The increase is expected to be about 2 to 3 weeks in the Northern MR and in the
Maghreb (Algeria/Morocco).
29
The increase is likely to be lower along the coast (~2 weeks), but higher inland (3
weeks in the south of France, the Balkans, Turkey, and Italy, and almost a month in the
Iberian peninsula and in Bulgaria).
This increase is balanced by the situation in the greater SEMR (Libya-Egypt-Israel-
Lebanon-Cyprus) where no significant change or even a slight decrease will be
experienced (between -4 and +5 dry days).
Figure 23: Difference between the average yearly number of dry days in the future and in the control years.
RR <0.5 mm defines dry days. Future scenario is A2.
Note that under the B2 scenario, fewer dry days are expected everywhere, which is
not a significant difference between the two scenarios. The results are also very similar if
RR <1mm is used to define dry days (not shown here).
To summarise, we have an increase in general in the number of dry days, ranging
from ~1 month more dry days within the Iberian peninsula to just few extra dry days in
the SEMR or even a slight decrease in Cyprus.
The number of wet days was also examined. One may ask if a dry day increase is
associated with a decrease in the very or extremely wet days or simply in the wet days.
The number of very wet days (RR > 10 mm) does not change much (+/- <3 days on
average, not shown). No change is also seen in the number of extremely wet days
(RR > 20 mm, not shown).
Finally, variation of rainy days that fall in the 1-10 mm range is shown in Fig. 24. This
corresponds to the middle range of precipitation. It is expected to drop by 2 weeks in the
North Med, and by a less than a week in the southern Mediterranean (except in the
30
Maghreb where the drop will be closer to 2 weeks). This decrease in the middle range is
associated with the increase in dry days, as clearly seen by comparing Figs. 24 and 23,
and taking into account that no change in the number of very wet days is expected.
Figure 24: Variation in the number of days with precipitation between 1 and 10 mm.
5.4.4 Precipitation intensity
Figure 25 shows the annual maximum amount of total rainfall over 3 days. It is worth
noting that some areas in the North Mediterranean are likely to see this parameter
increasing while their total annual precipitation actually decreases and the number of wet
days remains unchanged. This can imply that when it rains it will rain more intensely and
strongly. This is particularly true in Italy, Western Greece, South of France, and the
northwestern part of the Iberian Peninsula. On the contrary, rainfall is likely to become
less intense over the Southern Mediterranean.
5.4.5 Spells
Changes in the length of wet spells and dry spells (also referred to as droughts) are
examined. Our results show little change in the length of wet (RR> 0.5 or 1mm) or
extremely wet (RR> 10 or 20mm) spells (not shown). This is in accordance with our
results under Section 4.4.3 which show little change in the number of days associated
with these ranges of precipitation.
Greater variations are seen in dry spells. Longer dry spells are likely to be common.
The biggest changes for RR<1mm, shown in Figure 26, are likely to be 2 to 4 weeks
31
increase in the south of Italy and the Peloponese region in Greece, from the south of
Iberian Peninsula to Morocco, and in Libya.
Figure 25. Difference in the annual maximum 3-day cumulative rainfall.
Figure 26. Difference in the length of the longest dry spell (RR<1mm defines dry spells, scenario A2).
Other areas in the SEMR and the Cartagena-Algiers axis do not feature longer dry
spells. On the other hand, the northern part of Algeria, which is expected to have more
dry days, is not expected have longer dry spells. The extra dry days are scattered in
time.
The start and the end of these longest dry spells are also of interest since these are
expected to affect agriculture. Fig. 27 (Fig. 28) shows the shift in the end (start) of the
drought.
32
Figure 27. Difference in end of longest drought, for both scenarios (top=A2, bottom=B2).
We note the following noticeable shifts:
In the South of France and Central Spain, drought starts 3 weeks earlier and ends 2
weeks earlier. This is a shift of the dry season towards spring.
North Adriatic (Italy –except far south- and western Balkans), Maghreb: a general
shift of 2-4 weeks later.
Sicily: no major change.
Sardinia sees its driest season shifted towards fall by a month.
Greece and the area east of Greece, see its dry season getting longer towards fall:
no big change for start, but drought end shifts to a later date by 2 weeks or so.
33
It is noteworthy that the droughts of the Maghreb/South of Spain region shift towards
fall under the B2 scenario. The opposite holds true in Tunisia, but with less
amplitude. Continental Greece and Peloponese also show the opposite pattern.
Figure 28. As Fig. 27 but for the longest drought start (top=A2, bottom=B2).
35
5.5 Summary table
High Temperatures Low Temperatures Precipitation
Summer
Days Hot
days Tropical
nights
Days >
90th-
quantile
Nights >
90th-
quantile
Frost
nights Ice
days
(1) Days
< 10th-
quantile
Relative
Precip.
Var.
Dry
days Rainfall
between
1-10mm
Max
3-days
Rainfall
Longes
t dry
spell
NW Iberian
Peninsula 1 1 1 1 -1 -2 2-2 3
South of France
(inland) 3 1 1 2 2 -1 -2 -1 3-2 3
Coast of South of
France 1 2 2 2 -1 -2 -1 2-2 3 1
Corsica 1 1 2 2 2 -1 -2 -1 2-1 2 1
Sardegna 1 3 2 2 -3 2-1 1 2
Sicily/ South Italy 3 3 3 2 3 -3 3-1 3 3
North Adriatic 3 3 2 2 -2 -1 -2 -1 3-2 1 2 1
Central Balkans 3 3 2 2 -2 -1 -2 3-3
Central Greece 2 1 2 2 2 -1 -2 -1 2-2 1 1
Peloponnese 3 3 2 2 -3 -1 2-1 2 2
Crete 3 3 3 3 -3 -1 2-1 1
Coastal Turkey 1 2 1 1 2 2 2 -1 -2 -1 2-1 -1 2 1
Turkey Inland 3 3 2 2 3 -2 -1 -2 3-2 2
Cyprus 1 3 1 1 -3 -1 1 1 -1
Lebanon-Israel-
Nile Delta 1 1 3 3 3 -1 -3 (2) 1-1
E. Egypt – E.
Libya3 1 3 2 3 -3 (2) -1 -1
W. Libya3 1 3 2 3 -3 -3 -1 3
E. Maghreb 2 3 3 2 2 -2 -2 -3 2-2
W. Maghreb 3 3 3 2 2 -2 -2 2-2 -1 3
South Iberian
Peninsula 2 2 2 2 2 -1 -2 -1 2-2 3
Central Spain 3 3 1 2 2 -2 -2 -1 3-2 -1 1
Synoptic Table. Red = Large change=3, Yellow = Moderate change=2, Grey = Small change=1, White = No change, (-)=decrease. Notes: (1) very similar results
for nights; (2) depend on scenario. Typical temporal values: Big is about 1 month or more; Moderate is about 2-3 weeks; Small is about 1 week (few days).
36
6. Impact analysis
6.1 Energy demand
Energy demand is linked to climatic conditions (Giannakopoulos and Psiloglou,
2005) and the relationship of energy demand and temperature is non-linear. The
variability of ambient air temperature is closely linked to energy consumption, whose
maximum values correlate with the extreme values of air temperature (maximum or
minimum). In the Mediterranean region, during January, the maximum values of
energy consumption are related to the appearance of the lowest temperatures.
During the transient season of March-April, energy consumption levels are nearly
constant until about May, while air temperatures are constantly rising. From about
mid-May onwards, and throughout the summer period, any increase in air
temperature translates to an increase in energy consumption mainly due to the
extensive use of air conditioning. The exception is August since most people in the
Mediterranean region tend to take their summer holidays. Another transient period
exists in the months of September and October where energy demand and
consumption are at constant levels. This transient period is followed by a period of
continually increasing energy demand with a peak before the Christmas festive
period. Therefore, it is expected that with warmer weather decreased demand should
be typical in winter and increased demand should be typical in the summer
(Giannakopoulos and Psiloglou, 2005, Valor et al., 2001). Moreover, the effect of
higher temperatures chiefly in the summer is likely to be considerably greater on
peak energy demand than on net demand, suggesting that there will be a need to
install additional generating capacity over and above that needed to cater for
underlying economic growth unless adaptation or mitigation strategies are put into
place.
Since the energy-temperature relationship is non-linear and has distinct winter
and summer branches, it would be more convenient to separate these two branches.
The easiest way to achieve this is to use the idea of Degree-Day, which is defined as
the difference of mean daily temperature from a base temperature.
Base temperature should be the temperature where energy consumption is at its
minimum. If this temperature is chosen, then the degree-day index is positive in the
summer branch and negative in the winter branch. Instead of having both positive
and negative values for this index, the definition of two indices is used: heating
(HDD) and cooling degree days (CDD).
For the calculation of the HDD and CDD indices, the following equations were
used:
37
HDD = max (T* - T, 0) (Eq. 1)
CDD = max (T - T**, 0) (Eq. 2)
where T* and T** are the base temperatures for HDD and CDD respectively, which
can be either the same or different and T is the mean daily temperature as this is
calculated from the daily data of HadCM3 for both the reference and the future
periods.
HDD (CDD) is a measure of the severity of winter (summer) conditions in terms of
the outdoor bry-bulb air temperature, an indication of the sensible heating (cooling)
requirements for the particular location. Kadioğlu et al. (2001) used different base
levels of 15oC and 24oC for the calculations of HDD and CDD in Turkey, respectively.
In our study we use 15oC for the calculation of HDDs and 25oC for the calculation of
CDDs. We identify the changes in energy demand levels by showing differences in
the cumulative numbers of CDDs and HDDs between the reference (1961-1990) and
the future period (2031-2060).
Figure 29. Change in average yearly cumulative CDD between the future and control period.
6.1.1 Cooling energy requirements
In general, more cooling will be required.
Over the year (Fig.29), the increase in CDD will be large in South Mediterranean
(from Gibraltar to Lebanon), i.e., in the Middle East and the North African part of
38
Mediterranean Region. In northern side, the main increase will be in the South
Iberian Peninsula, North-Italy-Balkans-Greece, and South Turkey.
Fig. 30 shows the seasonal changes in CDDs. As expected the main contribution
is from summer, with no increase in winter, and a very small increase in fall and
spring. The only regions to escape any significant increase in cooling requirements
are: south Italy (including Sicily and Sardinia), south of France, Cyprus, northern part
of Turkey (because of Black Sea), and the North Western tip of Spain.
Figure 30. As Fig.29 but for each season.
Fig. 31 presents another view on the increase of energy demand by showing the
number of days when this requirement will be needed to cool more than 5oC. In the
south side of the Mediterranean Sea, from the southern part of Iberian Peninsula and
the North African coast to Syria, an additional month of heavy cooling will be
required. The 2-3 week increase in the north Aegean area is also worth mentioning.
6.1.2. Heating energy requirements
In general, less heating will be required. Fig. 32 shows the spatial distribution of
the general decrease in HDDs. It can be emphasised that:
The largest decrease occurs in the northern side of the region, from Turkey to
North Italy.
Spain and France will see a smaller but still noteworthy decrease,
39
The SEMR exhibits the lowest decrease, mainly because it is already a
warmer region.
Note also the cooling effect of the sea along the coasts.
Figure 31. Changes in the number of days with large cooling demand (CDD >5) between the
future and control period.
Figure 32. Changes in average yearly cumulative HDDs between the future and control period.
Unlike the CDD rise, the HDD decrease is spread over the year (Fig. 33),
although this probably depends to some extent on the choice of base temperature.
40
Of course, winter will be the season that will require much less heating. The largest
changes will happen along and above the axis North-Italy-Balkans-Greece-Turkey.
Figure 33. As Figure 32, but for each season.
Figure 34. Changes in the number of days with large heating demand (HDD >5) between the future
and control period.
41
As shown in Fig.34, the decrease in the number of days that require warming more
than 5oC (HDD>5) varies from about 2 weeks along the coast to a month inland.
6.1.3 Summary of impacts on energy demand and supply
Summarising about the CDDs and the HDDs, three areas can be identified:
South Side of the Mediterranean (North Africa to Syria): very large rise in CDD,
small drop in HDD
North Side of the Mediterranean (Italy-to-Turkey): small rise in CDD, very large
drop in HDD
Atlantic Side (Spain / France): small rise in CDD, large drop in HDD
As expected, the northern Mediterranean region is likely to reduce energy use in
the winter due to reduced heating needs. However, during summer, substantial
increases in energy demand are expected everywhere and especially in the south.
The peak in energy demand hence falls in the dry season, which is expected to
become even drier in the future (Section 5.4). A low water supply reduces energy
production from hydroelectric plants, as well as from conventional power plants,
which require water for cooling and for driving the turbines. As a result, energy
demands may not be able to be met in the warm period of the year. Additional
capacity may need to be installed unless adaptation or mitigation strategies are to put
into place. On the other hand, conditions for renewable energy production, such as
solar power, may improve under climate change.
Data in Spain show that the response of mean daily demand for electricity to an
increase of 1°C has steadily increased over the past 30 years (Rodriguez et al.,
2005). The energy demand for per degree of cooling is likely to continue to rise as a
society becomes richer and increased incomes allow the population to afford more
comfort. More air conditioning facilities could be installed. In turn, the heat generated
by air conditioning units could raise temperatures further and further increase the
demand for cooling.
6.2 Forest fire risk
6.2.1 About the Fire Weather Index (FWI)
One of the many possible detrimental impacts of anthropogenic climate change is
increased wildfire occurrence. Mediterranean Europe, in particular, has been
identified as likely to suffer hotter, drier summers towards the end of the century
(IPCC, 2001a), and hence potentially increased fire risk (e.g. Pinol et al., 1998,
42
Moriondo et al., 2005). The contribution of meteorological factors to fire risk is
simulated by various non-dimensional indices of fire risk. Viegas et al. (1999)
validated a number of such indices in the Mediterranean against observed fire
occurrence, with the Canadian Fire Weather Index (FWI, van Wagner, 1987)
amongst the best performers. Viegas et al. (2001) demonstrated that in summer, the
slow response of live fine fuel moisture content to meteorological conditions is well
described by the Drought Code sub-component of the FWI system. FWI is also one
of the most widely used indices of fire risk. Hence, it is natural to use output from
climate model simulations (here of HadCM3) of the coming decades (here 2031-
2060) as input to the FWI model to suggest how Mediterranean fire risk may change.
The Canadian Fire Weather Index system is described in detail in van Wagner
(1987). Briefly, it consists of six components that account for the effects of fuel
moisture and wind on fire behaviour. These include numeric ratings of the moisture
content of litter and other fine fuels, the average moisture content of loosely
compacted organic layers of moderate depth, and the average moisture content of
deep, compact organic layers. The remaining components are fire behaviour indices,
which represent the rate of fire spread, the fuel available for combustion, and the
frontal fire intensity; their values rise as the fire danger increases.
Fire risk is low for FWI<15, and increases more rapidly with FWI>15 (Good et al.,
2005). A threshold of FWI>30 was selected as a measure of increased fire risk.
6.2.2 FWI results
Fig. 35 shows the monthly changes of average FWI from May till October
between the future and the control period. We note that:
The increase is higher during the summer, with maximum increase in August in
the North Mediterranean inland.
Balkans, Maghreb, North Adriatic, Central Spain, and Turkey are the most
affected regions.
South of France is as strongly affected as Spain, but only in August and
September.
The SEMR (from Lebanon to Libya) sees no increase or decrease.
The same seems to hold for the islands of Crete, Sardinia, Sicily (southernmost
Italy too), Peloponnese, and Cyprus. Cyprus may even see a small decrease
every month.
The results are very similar under scenario B2 (not shown).
43
Figure 35. Monthly changes of average FWI from May till October between future and control period.
Fig. 36 shows the increase in the number of days with fire risk (top) and extreme fire
risk (bottom). According to this figure, the increase in the mean FWI is translated into:
44
2 to 6 additional weeks of fire risk everywhere, except south Italy and Cyprus and
the SEMR.
The maximum increase is again inland (Spain, Maghreb, Balkans, North Italy,
and Central Turkey), where at least an additional month with risk of fire has to be
expected.
A significant proportion of this increase in fire risk is actually extreme fire risk
(FWI>30).
South of France, Crete, and the coastal area of the rest of Mediterranean Region:
significant increase in the number of days with fire risk (1-4 weeks), but not in the
number of extreme fire risk.
Figure 36. Changes in the number of days with fire risk (top) and extreme fire risk (bottom)between
the future and the control period.
45
To conclude, Fig. 37 shows the number of weeks with fire risk in the future. In the
south part of the Mediterranean, practically the whole year is expected to be a period
of fire risk.
Figure 37. No of weeks with fire risk (FWI>15) in the future period (2031-2060).
6.3 Impacts on tourism
As some measure of the economic importance of summer tourism to the
Mediterranean, 147 million international tourists visited the Mediterranean in 2003
(this is 22% of the international tourism market) and generated US$113 billion for the
region. 70% of these tourists visited just two countries, Italy and Spain. It is very
difficult to model the potential response of tourists to climate change. However, by
discussions with experts at the MICE regional workshop entitled “Impacts of climate
extreme events on Mediterranean tourism and beach holidays”, (which took place in
June 2004 in Crete), it was possible to identify some of the impacts climate change
may have on the tourism industry (Table 1).
Rising temperatures over the Mediterranean region in 2031-2060 will certainly
affect the thermal comfort of tourists and their ability to acclimatise to a region prone
to high temperatures and heatwaves. Rainfall is also projected to decrease, leading
in turn to shortages in the public water supply and more widespread desertification,
which may affect the aesthetics of the region. Water shortages due to extended
46
droughts will also affect tourism flows especially in the SE Mediterranean since the
water use has a strong seasonal cycle. The maximum demand coincides with the
minimum availability. Perry (2001) reports that a tourist in Spain uses 4 times as
much water as a Spanish city dweller, so tourists are not water conscious. He also
states that in Crete water shortages could be experienced in 5 years out of 6 by
2010. As noted earlier in this report, the decreased rainfall and increased
temperatures in combination will lead to a greater frequency of forest fires. Intense
rainfall events in winter may increase especially in the Northern part of the region,
leading to greater erosion rates and a higher risk of flooding. Greater heat is also
likely to repel at the same time the important old age population and the residents of
Mediterranean regions, in France, Italy or Spain. Thus mountainous regions could
become appreciated for their relative coolness and the shade of their forests (Ceron
and Dubois, 2000).
Table 1: Summary of climate changes and their probable impact on major travel flows in the
Mediterranean.
MAJOR TOURISM
FLOW ORIGIN
MARKET
CLIMATE
CHANGE
DESTINATION
REGION CLIMATE
CHANGE
IMPLICATIONS FOR
DESTINATION REGION POSSIBLE MARKET
REACTIONS
NORTHERN
EUROPE TO
MEDITERRANEAN
-Much warmer,
wetter winters
-Warmer, drier
summers
-More reliable
summers
-Warmer winters
-Much warmer, drier
summers
- Increased heat
index
-More heatwaves
-More arid landscape
-Small tidal range
means greater sea
level rise impact
-Greater drought and
fire risk
-Increased water
shortages
-Greater personal heat
stress
-Beach degradation
and habitat loss due to
sea level rises
-Vulnerability to more
tropical diseases (e.g.
malaria)
-More flash floods
-Poor urban air quality
in cities
Overwhelmingly a leisure
travel market
-Improvement of
Northern European
summers triggers more
domestic holidays
-Decreased incentive for
Mediterranean summer
holidays
-Increased incentive for
shoulder month
Mediterranean holidays
-Increased incentive for
southerners to go north
It is not, however, just the change in climate over the Mediterranean that will
impact on the region’s tourism. Improvements in the climate of the source regions of
the tourists visiting the Mediterranean will also affect the popularity of the area.
Warmer, drier and more reliable summers in Northern Europe will encourage tourists
to take domestic holidays and will even encourage those in the Mediterranean region
to holiday further north, away from the high temperatures and water deficits likely in
the south during summer. It is also likely that the Mediterranean holiday season will
split into two seasons, in the spring and the autumn, when climate conditions will be
more comfortable.
47
In conclusion:
Warmer northern European summers encourage an increase in domestic
holidays.
In a warmer future, there is an increased likelihood of people from the
Mediterranean holidaying in the north.
More frequent and more intense heat waves and drought are likely to
discourage Mediterranean summer holidays.
There is likely to be a shift in the Mediterranean holiday season to spring and
autumn.
6.4 Impacts on water resources
One of the greatest potential impacts of climate change on human society is
through its effect on water resources. The Mediterranean is already a region,
experiencing moderate to high water stresses and climate change has the potential
to exacerbate further these stresses.
The implications of climate change for water resources stress in the
Mediterranean were assessed by the UK Meteorological Office (Arnell, 1999). First,
river runoff was simulated with a macro-scale hydrological model. Then changes in
national water resource availability were computed (taking into account imports from
upstream), and the estimated volume of water available for use was compared with
the amount withdrawn by water users. For the period in question in this report (2031-
2060), it was projected that runoff decreases substantially in the Mediterranean
Europe, North Africa and the Middle East. The North Mediterranean will see a
50mm/year reduction in runoff while the South (already dry) will experience a
25mm/year decrease or less. However, these changes can be very large in
percentage terms.
The rise in temperature is expected to also affect the timing of streamflow through
the year with particularly large changes in the North Mediterranean, where the higher
winter temperatures mean that a much smaller proportion of winter precipitation falls
as snow to be stored on the land surface until the spring melt. In these areas winter
flows might increase but it is highly likely that spring flows will decrease.
One measure of national water resource stress is the ratio of water used to water
available (although this hides within-country variations and the risk of stress during
drought conditions), and countries using more than 20% of their total annual water
supply are generally held to be exposed to water stress. Using this measure, all
countries around the Mediterranean are expected see an increase in water stress.
48
The sole exception can be Egypt where river runoff from the Nile may actually
increase due to floods in the Central African Nile springs.
Some countries have conducted further studies to illuminate the impact of such
changes on their countries. In the northern Mediterranean, the Spanish Government
estimates that a 1°C increase in the mean annual temperature is likely to lead to a
reduction of 5-14% in water yields in the country. In the extreme case of a 4°C
increase, water yields could reduce by as much as 22% in some regions (Rodriguez
et al., 2005). In the southern Mediterranean, the Algerian Government estimates that
a 1°C rise in mean annual temperature would lead to decreases in precipitation by
15% and in influx of surface waters by 30%. Subsequently, water demand would
exceed available water resources by 800 million m3 (Government of Algeria, 2001).
In the southeastern Mediterranean, the Lebanon Government estimates that by
2050, climate change would be responsible for nearly doubling the water shortage to
350 million m3 of water (Khawli, 1999).
6.5 Impacts on sea level rise
Results from the HadCM3 give a projection of a global rise in sea level of 21cm
by the 2050s due to rise in greenhouse gases from human activities (IPCCa, 2001).
This estimate includes direct prediction of thermal expansion combined with
estimates of land-based ice-melt.
However, global mean sea level rise does not manifest itself uniformly around the
world. Regional variations in atmospheric circulation, ocean circulation and warming
rates and the interactions between them have led to significant deviations of
regionally sea level change from the globally averaged trend.
Model projections of regional sea level patterns show very little agreement. For
the Mediterranean, the values range from 1 to 2cm of regional sea level rise per 1 cm
of global sea level rise (IPCCa, 2001). This is due to the low tidal range in the
Mediterranean combined with the limited potential for wetland migration. The most
vulnerable region seems to be the Southern Mediterranean from Turkey to Algeria
where flooding impacts can occur particularly in deltaic countries (such as Egypt).
6.6 Impacts on biodiversity
Climate change over the past 30 years has produced numerous shifts in the
distributions and abundances of species. Recent studies have tried to quantify future
changes under different warming scenarios. Thuiller et al. (2005) shows that a 3.6°C
global warming could lead to a loss of over 50% of plant species in the northern
49
Mediterranean and the Mediterranean mountain region, while species loss is likely to
exceed 80% in northcentral Spain and the Cevennes and Massif Central in France.
These results are in the direction of earlier studies (e.g., Thomas et al., 2004)
although estimates of the magnitudes extinction risks are lower than earlier
predictions. Climate change may also have indirect effects on the ecosystem.
Grigulis et al. (2005) shows that increased fires due to climate change could increase
the spread of invasive grass species which in turn, could lead to more frequent and
more intense fires.
References
Abaurrea J, Asín J, Erdozain O, Fernández E, 2001: Climate variability analysis of
temperature series in the Medium Ebro River Basin, In Brunet, M. and López, (Eds.):
Detecting and modelling regional climate change, Springer Verlag, Berlin,
Heidelberg, New York, pp 109-118.
Alpert P, Ben-Gai T, Baharad A, Benjamini Y, Yekutieli D, Colacino M, Diodato L,
Ramis C, Homar V, Romero R, Michaelides S, Manes A, 2002: The paradoxical
increase of Mediterranean extreme daily rainfall in spite of decrease in total values.
Geophys. Res. Lett., 29, art. no.1536.
Bolle, H.-J. (Ed), 2003: Mediterranean Climate – Variability and Trends. Springer
Verlag, Berlin, Heidelberg, New York.
Brunet M, Aguilar E, Saladie O, Sigró J, López D, 1999: Variaciones y tendencias
contemporáneas de la temperatura máxima, mínima y amplitud térmica diaria en el
NE de España. In Raso Nadal, J. M. and Martin-Vide, J. (Eds.): La Climatologia
española en los albores del siglo XXI, Publicaciones de la A.E.C., Serie A, 1,
Barcelona, pp. 103-112.
Brunet M, Aguilar E, Saladie O, Sigró J, López D, 2001a: The variations and trends
of the surface air temperature in the Northeastern of Spain from middle nineteenth
century onwards, In Brunet, M. and López, (Eds.): Detecting and modelling regional
climate change, Springer Verlag, Berlin, Heidelberg, New York, pp 81-93.
Brunet M, Aguilar E, Saladie O, Sigró J, López D, 2001b: A differential response of
Northeastern Spain to asymmetric trends in diurnal warming detected on a global
scale, In Brunet, M. and López, (Eds.): Detecting and modelling regional climate
change, Springer Verlag, Berlin, Heidelberg, New York, pp 95-107.
50
Brunet M, Aguilar E, Saladie O, Sigró J, López D, 2001c: The Spanish Diurnal
Warming: A different pattern to the observed on a global scale, Geophysical
Research Abstracts, 3, 5332.
Brunet M, Aguilar E, Saladie O, Sigró J, López D, 2002: Warming phases in long-
term Spanish temperature change, In 13th Symposium on Global change and
Climate Variations, Orlando 13-17 January 2002, American Meteorological Society,
Boston, pp. 30-32.
Brunetti M, Colacino M, Maugeri M, Nanni T, 2001a: Trends in the daily intensity of
precipitation in Italy from 1951-1996. Int. J. Climatol., 21, 299-316.
Brunetti M, Maugeri M, Nanni T, 2001b: Changes in total precipitation, rainy days and
extreme events in north-eastern Italy. Int. J. Climatol., 21, 861-871.
Ceron, J-P. and G. Dubois, 2000: Tourisme et changement climatique, In Impacts
potentiels du changement climatique en France au XXIème siècle. Premier ministre.
Ministère de l’aménagement du territoire et de l’environnement 1998. pp 104-111.
(deuxième édition 2000).
De Luís M, Raventós J, González-Hidalgo JC, Sánchez JR, Cortina J, 2000: Spatial
analysis of rainfall trends: a case study in Valencia Region (E Spain). Int. J. Climatol.,
20, 1451-1469.
Folland CK, et al., 2001: Observed climate variability and change, in Chapter 2 of
climate change 2001; the scientific basis, Contribution of Working Group I to the
Third Assessment Report of the Intergovernmental Panel on Climate Change, edited
by edited by Houghton JT et al, 881pp., Cambridge University Press, Cambridge, UK
and New York, NY, USA, 2001.
Frei C, Schär C, 2001: Detection probability of trends in rare events: theory and
application to heavy precipitation in the Alpine Region. J. Climate, 14, 1568-1584.
Galán E, Cañada R, Fernández F, Cervera B, 2001: Annual temperature evolution in
the Southern Plateau of Spain from the construction of regional climatic time series,
In Brunet, M. and López, (Eds.): Detecting and modelling regional climate change,
Springer Verlag, Berlin, Heidelberg, New York, pp 119-131
Giannakopoulos C. and J.P. Palutikof, 2004: Modeling climate extreme events: The
EU project MICE for the case study area of Greece The EGGS (11), 30-35.
Giannakopoulos, C. and B.E. Psiloglou, 2005: Trends in energy load demand for
Athens, Greece: Weather and non-weather related factors, submitted to Climate
Research.
Giorgi F, 2002: Variability and trends of sub-continental scale surface climate in the
twentieth century. Part I: Observations, Clim. Dyn., DOI 10.1007/s00382-001-0204-x.
51
Good, P., M.Moriondo, C.Giannakopoulos, M.Bindi, 2005: The meteorological
conditions associated with extreme fire risk in Italy and Greece: relevance to climate
model studies, submitted to Climate Research.
Goodess CM, Jones PD, 2002: Links between circulation and changes in the
characteristics of Iberian rainfall. Int. J. Climatol., 22, 1593-1615.
Government of Algeria, 2001: Elaboration de la stratégie et du plan d’action national
des changements climatiques. 2001. République Algérienne Démocratique et
Populaire, Ministère de l’Aménagement du Territoire et de l’Environnement, Direction
Générale de l’Environnement.Gordon, C., C. Cooper, C.A. Senior, H. Banks, J.M.
Gregory, T.C. Johns, J.F.B. Mitchell and R.A. Wood, 2000: The simulation of SST,
sea ice extents and ocean heat transports in a version of the Hadley Centre coupled
model without flux adjustments. Climate Dynamics, 16, 147-168.
Grigulis, K., Lavorel, S., Davies, I.D., Dossantos, A., Lloret, F. and Vilà M. ,
Landscape-scale positive feedbacks between fire and expansion of the large tussock
grass, Ampelodesmos mauritanica in Catalan shrublands, Global Change Biology
(2005) 11, 1–12.
Hansen J, Ruedy R, Sato M, Imhoff M, Lawrence W, Easterling D, Peterson T, Karl
T, 2001: A closer look at United States and global surface temperature change. J.
Geophys. Res., 106, 23947-23963.
Hewitt C. and D. Griggs, 2004: Ensembles-based predictions of climate changes and
their impacts. Eos, 85, 566.Horcas R, Rasilla D, Fernández-García F, 2001:
Temperature variations and trends in the Segura River Basin. An exploratory
analysis, In Brunet, M. and López, (Eds.): Detecting and modelling regional climate
change, Springer Verlag, Berlin, Heidelberg, New York, pp 133-142.
IPCC, 2001a: Climate Change 2001: The Scientific Basis. Contribution of WG1 to the
Third Assessment Report of the Ingergovernmental Panel on Climate Change.
Cambridge University Press.
IPCC, 2001b: Climate Change 2001: Impacts, Adaptation and Vulnerability.
Contribution of WG2 to the Third Assessment Report of the Ingergovernmental Panel
on Climate Change. Cambridge University Press.
Jacobeit J, Wanner H, Luterbacher J, Beck C, Philipp. A, Sturm K, 2003:
Atmospheric circulation variability in the North-Atlantic-European area since the mid-
seventeenth century. Clim. Dyn. 20, 341-352.
Kadioğlu M, Şen Z, Gültekin L, 2001: Variations and trends in turkish seasonal
heating and cooling degree-days. Climatic Change 49: 209-223.
52
Khawli, 1999: Water Resources: Assessment of Lebanon’s Vulnerability to Climate
Change. Lebanon’s National Communication on Climate Change. Ministry of
Environment, UNDP, GEF. Beirut.
Mariotti A, Struglia MV, 2002: The hydrological cycle in the Mediterranean region and
implications for the water budget of the Mediterranean Sea, J. Climate, 15, 1674-
1690.
Met Office, 1998: Climate change and its impacts, Report: A first look at results from
the Hadley Centre’s new climate model.
Moriondo M., P. Good, R. Durao, M. Bindi, C. Giannakopoulos and J. Corte Real,
2005: Potential impacts of climate change on forest fire risk in the Mediterranean
area, submitted to Climate Research.
Nakicenovic N, Alcamo J, Davis G, de Vries B, Fenhann J, Gaffin S, Gregory K,
Grübler A, Jung TY, Kram T, La Rovere EL, Michaelis L, Mori S, Morita T, Pepper W,
Pitcher H, Price L, Raihi K, Roehrl A, Rogner HH, Sankovski A, Schlesinger M,
Shukla P, Smith S, Swart R, van Rooijen S, Victor N and Dadid Z. 2000. Emissions
scenarios. A Special Report of Working Group III of the Intergovernmental Panel on
Climate Change. Cambridge University Press.
New M, 2005: Arctic climate change with a 2°C global warming. In Rosentrater L
(Ed.): Evidence and Implications of Dangerous Climate Change in the Arctic, WWF
International Arctic Programme..New MG, Hulme M, Jones PD, 2000: Representing
twentieth-century space time climate fields. Part II: development of a 1901-1996
mean monthly terrestrial climatology, J. Climate, 13, 2217-2238.
New M, Todd M, Hulme M, Jones PD, 2001: Precipitation measurements and trends
in the twentieth century, Int. J. Climatol., 21, 1899-1922.
Palutikof J.P, 2004: Understanding the impacts of climate extremes in Europe:
Progress in the MICE project, Geophysical Research Abstracts, Vol. 6, 06200.
Perry, A., 2001: More heat and drought - Can Mediterranean tourism survive and
prosper?, In Proceedings of the First International Workshop on Climate, Tourism
and Recreation, A. Matzarakis and C.R. De Freitas, eds., 1-6, International Society of
Biometeorology.
Pinol, J., Terradas, J. and Lloret, F., 1998: Climate warming, wildfire hazard, and
wildfire occurrence in coastal eastern Spain Climatic Change, 38, 345–357.
Rodriguez and 48 others, 2005: A preliminary assessment of the impacts in Spain
due to the effects of climate change, Ministry of the Environment, Government of
Spain.
53
Pope, V. D., M. L. Gallani, P. R. Rowntree and R. A. Stratton, 2000: The impact of
new physical parametrizations in the Hadley Centre climate model -- HadAM3.
Climate Dynamics, 16: 123-146.
Ramos MC, 2001: Rainfall distribution patterns and their change over time in a
Mediterranean area. Theor. Appl. Climatol., 69, 163-170.
Romero R, Guijarro JA, Alonso S, 1998: A 30-year (1964-1993) daily rainfall data
base for the Spanish Mediterranean regions: First exploratory study. Int. J. Climatol.,
18, 541-560.
Romero R, Ramis C, Guijarro JA, 1999: Daily rainfall patterns in the Spanish
Mediterranean area: an objective classification. Int. J. Climatol., 19, 95-112.
Thomas C.D. and 18 others. 2004. Extinction risks from climate change. Nature, 427,
145–148.
Thuiller W., Lavorel S., Araújo M.B., Sykes M.T. and Prentice I.C. 2005. Climate
change threats to plant diversity in Europe. PNAS 102 (23), 8245–8250.
Türkes M, 1996: Spatial and temporal analysis of annual rainfall variations in Turkey,
Int. J. Climatol., 16, 1057-1076.
Türkes M, 1998: Influence of geopotential heights, cyclone frequency and southern
oscillation on rainfall variations in Turkey, Int. J. Climatol., 18, 649-680.
Valor E., Meneu V. and Caselles V., 2001: Daily air temperature and electricity load
in Spain. J. Appl. Meteor., 40(8), 1413-1421.
van Wagner, C.E., 1987: Development and structure of the canadian forest fire
weather index system. Canadian Forestry Service, Forestry Technical Report35.
Viegas, D.X., Bovio G. Ferreira A. Nosenzo A. and B. Sol, 1999: Comparative Study
of Various Methods of Fire Danger Evaluation in Southern Europe. Int. J. Wildland
Fire, 9, 235–246.
Viegas, D.X., Pinol J. Viegas M.T. and R. Ogaya, 2001: Estimating live fine fuels
moisture content using meteorologically-based indices. Int. J. Wildland Fire, 10, 223–
240.
Xoplaki E, 2002: Climate Variability over the Mediterranean, Ph.D. Thesis, University
of Bern, 193pp. Available at: http://sinus.unibe.ch/klimet/docs/phd_xoplaki.pdf.
Xoplaki E, González-Rouco JF, Luterbacher J, Wanner H, 2003: Mediterranean
summer air temperature variability and its connection to the large-scale atmospheric
circulation and SSTs, Clim. Dyn., 20: 723-739, DOI 10.1007/s00382-003-0304-x.
Impact of a 2° C global temperature rise on the Mediterranean region:
Agriculture analysis assessment.
M. Bindi & M. Moriondo
Department of Agronomy and Land Management, Florence, Italy
d changes in
ime-slice
3 and C4
severe in the
ing
is likely to
locations in
ange and its associated increase in
CO2 may have little or small positive impacts on yields, provided that additional water
demands can be met. The adoption of specific crop management options (e.g. changes in
sowing dates or cultivars) may help in reducing the negative responses of agricultural
crops to climate change. However, such options could require up to 40% more water for
irrigation, which may or may not be available in the future.
Summary
The climate change impact analysis on agriculture showed that the expecte
temperature and precipitation for SRES-IPCC scenarios A2 and B2 in the t
2031-2060 determined a general reduction in yield of agricultural crops (e.g. C
summer crops, legumes, cereals, tuber crops). Reductions in yields are more
southern Mediterranean than in the northern Mediterranean, even when the fertiliz
effect of increased CO2 is taken into account. The southern Mediterranean
experience an overall reduction of crop yields due to climate change. In some
the northern Mediterranean, the effects of climate ch
54
Materia
1 2
345
6
7
8
9
11
12
13
10
ls
imulation
Hadley Centre.
rise on the
ered to
narios were
(SRES)
greenhouse gas emission; ii)
different areas of
p simulation
model (i.e. CropSyst) to reproduce crop yields, the data reported in the FAOSTAT
tries.
ep crop
e soil water budget,
el allows
etic
leaf area index,
n management,
to simulate crop
ent between
d,
alysis
hirteen grid cells (called ‘hot
spots’) of the HadCM3 has been selected (Fig. 1). More specifically these grid cells (2.5°
latitude by 3.75° longitude) were selected to provide an homogenous cover of the study
area and to study the areas where changes in precipitation and temperature patterns are
expected to be substantial, accordin odel simulations.
Figure 1 – Grid cells of Had CM3 selected for impact assessment on agriculture
Climate data. The daily climate data used for this study was obtained from the s
results of the HadCM3 model developed by the UK Meteorological Office
In particular, in order to reproduce the impact of a 2° C global temperature
Mediterranean region, two time-slices 1961-1990 and 2031-2060 were consid
represent present and future climate, respectively. Further, two emission sce
selected among those proposed by the Special Report on Emissions Scenarios
(IPCC 2000): i) scenario A2, characterised by medium-high
scenario B2 characterised by medium-low greenhouse gas emission.
Statistical yield data. In order to select the most important crops for the
the Mediterranean basin and to evaluate the capacity of the selected cro
database (http://faostat.fao.org/) has been collected for Mediterranean coun
Crop simulation model. CROPSYST is a multi-year, multi-crop, daily time st
growth simulation model (Stockle et al, 2003). The model simulates th
soil-plant nitrogen budget, crop canopy and root growth, phenology, dry matter
production, yield, residue production and decomposition, and erosion. The mod
the user to specify management parameters such as sowing date, cultivar gen
coefficients (photoperiodic sensitivity, duration of grain filling, maximum
etc. ), soil profile properties (soil texture, thickness), fertilizer and irrigatio
tillage, atmospheric CO2 concentration etc. The capability of the model
yields has been evaluated in numerous field studies conducted in the Mediterranean,
United States and Australia (Stockle et al., 2003). In general, the agreem
simulated and measured yields was good. When properly calibrated and applie
CropSyst has proved to be a suitable tool for simulating cropping systems.
Methods: Selection of the hot spots. Following the results of the climate an
assessment performed in Section 5, by C. Giannakopoulos, t
g to HadCM3 m
55
Selection of the crops. The data reported on the FAOSTAT database was us
the agricultural crops for each ‘hot spot’. In particular, the selection of the sp
based on the different characteristic of agricultural crops: i) photosynthesis
and C4 crops)1, ii) growing period (winter an summer seasons), iii) food
(protein, e.g. legumes and carbohydrates, e.g. tuber crops); and on the basis o
extension of the cultivated area in the countries that are included the grid cells
No account ha
ed to select
ecies was
pattern (C3
composition
f the
(Tab. 1).
s been taken of the introduction of new crops which may becomes suitable
ble 1 – Type of cr ulated i of the selected grid
cell
numbe
Country C4 summer
crop
C3 summer
crop
Legumes Tuber Cereals
in these regions.
Ta
Grid
op sim n each cell
r
crops
m sun er an potato
m sun er ntil potato
m sun er ybean potato
m sun er ybean potato
m sun er ybean potato
m sun er an potato
7 Turk m sun er ntil potato
8 Jord m sun er ntil potato
9 Egypt m sun er an potato
1 Portugal aize flow be wheat
2 Spain aize flow le barley
3 France aize flow so wheat
4 Italy aize flow so wheat
5 Serbia aize flow so wheat
6 Greece aize flow be wheat
ey aize flow le wheat
an aize flow le barley
aize flow be wheat
10 Libya maize sunflower bean potato wheat
potato wheat
wheat
11 Tunisia maize sunflower bean potato wheat
12 Algeria maize sunflower bean
13 Morocco maize sunflower bean potato
Impact analyses. First of all the CropSyst model (Stockle et al. 2003) was ca
fit as much as possible the data
librated to
lues of
1990 and
one setting
s for the
future climate scenarios were done without and with the effects of increasing CO2 (470
CO2 was
r CO2
005). In
ffect of
reported on the FAOSTAT database
(http://faostat.fao.org/). Then, for each ‘hot spot’ and crop type the annual va
development stages and yields were calculated for the two time-slices (1961-
2031-2060). In particular, for the present climate the simulation runs were d
the atmospheric concentration of CO2 at 350 ppm; whilst the simulation run
ppm scenario B2 and 520 ppm scenario A2). The crop responses to increasing
introduced in the simulation runs following the results obtained in the Free Ai
Enrichment (FACE2) experiments (Kimball et al. 2002; Ainsworth and Long 2
general, results from FACE experiments show, as under field conditions, the e
1 Almost all plant life on Earth can be broken into two categories based on the way they assim
dioxide into their systems. C3 plants
ilate carbon
include more than 95 percent of the plant species on earth (e.g. trees,
t steps in CO2
plants, on the other hand, initially
form four carbon-atom molecules. It turns out that under present CO2 concentration ( 360 ppm) C4 plants
are more economical in the use of water and have a much higher yield potential than C3 plants. However, it
is expected that C3 plants can respond readily to higher CO2 levels, and C4 plants can make only limited
responses. Thus, increased CO2 likely will mean that some plant species will be stronger, more prolific, and
may overwhelm those less able to benefit.
2 In FACE experiments, jets of CO2 are released over vegetation plots in the open-air through an array of
pipes. Natural wind and diffusion disperse the CO2 across the experimental area.
wheat, sunflower). C4 plants include such crop plants as sugar cane and corn. During the firs
assimilation, C3 plants form a pair of three carbon-atom molecules. C4
56
CO2 on crop yields is significantly lower than that obtained in controlled environmental
studies.
strategies (e.g.
f climate
r crops (i.e.
t the rest
rainfed crops”. Moreover, all the crops were considered
fully fertilised (i.e. no nitrogen stress).
t climate
ported in
mer crop,
the different
ica which is
expected to have lower water availability. Whilst for the irrigated crops (C4 summer
re less
apacity of
ach
error is the
ses and can
ths study.
put with
for individual grid cells
while the statistical data from FAO has been collected at the nation level. The climate
on over a
gical trend,
pest and disease stresses, etc.) that cannot be reproduced by the model.
e 2. M Absolute r (MAE, % een observ nd simulated crop yields for the
Medi nean regi N-W = Por l, Spain, France and Italy, N-E = Serbia, Greece and
ey, S ordan, E nd Libya, sia, Algeria and Morocco
Mean olute Error
Finally, the CropSyst model was rerun introducing adaptation management
changes in sowing dates, cultivar, etc.) that may reduce the negative impact o
change or enhance positive impacts. In all the simulation runs the C4 summe
maize) and tuber crops (i.e. potato) were considered as “irrigated crops”, whils
of the crops were considered as “
Results:
Crop yields under present climate The simulated crop yields for the presen
obtained by calibrating CropSyst model on the basis of the FAO data were re
Fig. 2. The results showed a pattern of crop yields in rainfed crops (C3 sum
legumes, cereals) that is strongly correlated to the precipitation regimes of
grid cells. Hence, lower crop yields were seen in regions of Northern Afr
crops and tuber crops) the differences in crop yields among the grid cells we
evident and these were mainly driven by the temperature regimes.
In order to provide a quantitative estimate of the uncertainties related to the c
CropSyst to simulate crop yields, Mean Absolute Errors (MAEs) were calculated for e
crop type in four main regions in the Mediterranean (Table 2). Mean absolute
average of the difference between modeled and observed value in all test ca
be considered as the average prediction error for CropSyst simulations for
MAEs also include inherent errors that arise from the comparison of model out
observed data. Simulated values from CropSyst are calculated
conditions of the grid cells cannot completely represent the average conditi
nation, and the statistical data also include sources of variability (e.g. technolo
Tabl ean Erro ) betw ed a
main terra ons: tuga
Turk -E = J gypt a S-W = Tuni
Abs (MAE, %)
Region C4 summer Legumes C3 summer Tuber crops Cereals
.8
N-W 9.3 4.6 10.8 8.7 9
N-E 4.4 1.4 2.1 6.7
S-E 12.1 13.8 4.4 11.7
S-W 13.4 0.9 2.7 3.3
4.3
1.3
6.4
Table 2 shows that CropSyst results match quite closely with the statistical data collected
by FAO, with MAEs ranging from less than 1% to under 14%. The MAEs can also be
used as a measure of the uncertainties of crop yield estimates due to the bias between
observed and simulated yields. The total uncertainty is likely to be larger due to the
uncertainties cascaded from the emission scenarios and the global climate model (see
Section 4.4 for a discussion of uncertainty cascades).
57
Changes in crop yields without the effect of CO2
The results of the CropSyst simulation runs (without including the effect of C
present and future climate scenar
O2) for
io were reported in Fig. 3. On the basis of the different
ion of yields
es (e.g.
5% (Morocco)
e) in B2
ith the
as very small. More
2 scenario
s in all the grid
se was very small.
in A2
rops showed a general reduction of yields in all the grid cells with
s ranges from
+15% (Spain)
eneral reduction of yields, even if in a few grid cells the
yields increased (Turkey, Greece and Spain). More specifically these changes ranges
a) to +8%
l the ‘hot
ermined a
lopment) (Tab.
higher in
reductions in
precipitation are expected to be higher during the summer periods. Moreover, in general
the reductions were higher in A2 scenarios, since for this scenario larger changes in
temperature (increase) and precipitation (reduction) regimes were expected. However, the
results from the two scenarios agree closely in terms of maximum decreases in yields.
crop types these are the main results:
- C4 summer crops. C4 summer crops showed an almost systematic reduct
with the exception of a few grid cells located in the EU-Mediterranean countri
Italy, France, Spain and Portugal). More specifically, these ranges from -1
to + 10% (France) in A2 scenario and from -13% (Morocco) to +9% (Franc
scenarios (Fig. 3a).
- Legumes. Legumes showed a general reduction of yields in all the grid cells w
exception of that on Spain, where, however, the increase w
specifically, these changes ranges from -44% (Morocco) to -13% (Turkey) in A
and from -45% (Egypt) to +3% (Spain) in B2 scenarios (Fig. 3b).
- C3 summer crops. C3 summer crops showed a general reduction of yield
cells with the exception of that on Spain, where, however, the increa
More specifically, these changes ranges from -32% (Portugal) to -6% (Jordan)
scenario and from -23% (Morocco) to +1% (Spain) in B2 scenarios (Fig. 3c).
- Tuber crops. Tuber c
the exception of those on Jordan and Spain. More specifically these change
-35% (Morocco) to -1% (Jordan) in A2 scenario and from -31% (Egypt) to
in B2 scenarios (Fig. 3d).
- Cereals. Cereals showed a g
from -23% (Morocco) to -2% (Turkey) in A2 scenario and from -24% (Liby
(Turkey) in B2 scenarios (Fig.3e).
The general picture showed a substantial reduction for all the crop types in al
spots’ grid cells. This was due to the increases in temperature and reduction in
precipitation predicted for both future climate scenarios (A2 and B2), that det
reduction of the length of the growing period (i.e. higher rate of crop deve
3) and the water available for crop growth, respectively. These decreases were
summer and not irrigated crops (e.g. C3 summer crops, legumes), since the
58
0246
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
8
b
0246
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
a
8
c
0246
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
8
d
0246
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
8
e
0246
Portugal
Spain
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
France
Ital y
8
Figure 2- Crop yields (t ha-1) in the selected ‘hot spots’ grid cells: a) C4 summer crop, b) legumes,
c) C3 summer crop, d) tuber crops, e) cereals.
59
-55 -45 -35 -25 -15 -5 5 15
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
B2
A2
b
-55 -45 -35 -25 -15 -5 5 15
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
B2
A2
a
-55 -45 -35 -25 -15 -5 5 15
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
B2
A2
d
-55 -45 -35 -25 -15 -5 5 15
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
B2
A2
c
e
-55 -45 -35 -25 -15 -5 5 15
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
Figure 3- Impact of climate change on crop yields without CO2 effect: a) C4 summer crop, b)
legumes, c) C3 summer crop, d) tuber crops, e) cereals. The changes reported in the figures were
expressed as % and obtained as differences between the mean yields of the two futures and the
present yields.
Portugal
Spain
France
Ital y B2
A2
60
Table 3 – Mean length of the growing season for the different crop types under the present
cenario and me hange ce
Crop type Length (days) A2 (% change) B2 (% change)
climate s an percentage c d under future s narios
C4 summer crop 126.3 -8.0 -7.5
mer crop 131.2 -6.7 -6.4
108.0 -5.9 -5.7
crops 170.4 -5.5 -4.7
-1.7 -1.8
C3 sum
Legu
Tuber
mes
Cereals 214.0
Syst simulation runs (including the effect of CO2) for present and
ops types
lds in the grid
, Tunisia,
increase.
ce) in A2
a general reduction of yields in all the grid cells with the
anges ranges
) to +15%
the grid
ls located on Northern and South-western shores of the Mediterranean basin; whereas
antially
al) to +6%
B2 scenarios (Fig.
r crops showed an inconsistent response among selected grid cells,
nd
cenario and from -27% (Egypt) to +30% (Jordan) in
e changes
m -19% (Libya) to
n tropical
lizing effect
an,
er climate,
but is not able to completely offset the losses. In the cooler northeastern Mediterranean,
CO2 increase and the associated climate change result in little net effect on most crops,
provided that the increase in water demands, especially for irrigated crops, can be
satisfied (Table 4). Similarly in the northwestern Mediterranean, yields of irrigated crop
may increase if water demands can be met. However, rainfed summer crops are likely to
experience a net reduction in yield, even when the fertilizing effect of CO2 is considered.
Changes in crop yields with the effect of CO2
The results of the Crop
future climate scenario were reported in Fig. 4. On the basis of the different cr
these are the main results:
- C4 summer crops. C4 summer crops showed a prevalent reduction in yie
cells located in the African and Asian Mediterranean countries (e.g. Morocco
Jordan, etc.); whereas on the European grid cells yields showed a consistent
More specifically, these changes ranges from -12% (Morocco) to + 16% (Fran
scenario and from -11% (Egypt) to +13% (France) in B2 scenarios (Fig. 4a).
- Legumes. Legumes showed
exception of those on Spain, Turkey and Greece. More specifically, these ch
from -39% (Morocco) to +2% (Turkey) in A2 scenario and from -41% (Egypt
(Spain) in B2 scenarios (Fig. 4b).
- C3 summer crops. C3 summer crops showed a general reduction of yields in
cel
in the grid cells located on the Southern or Eastern shores, yields were subst
unchanged. More specifically, these changes ranges from -21% (Portug
(Turkey) in A2 scenario and from -16% (Morocco) to +11% (Spain) in
4c).
- Tuber crops. Tube
with a general reduction of yields in those on the African shores of the basin, a
prevalent increase in the rest. More specifically these changes ranges from -25%
(Morocco) to +16% (Jordan) in A2 s
B2 scenarios (Fig. 4d).
- Cereals. Cereals showed a prevalent increase in yields. More specifically thes
ranges from -15% (Morocco) to +13% (Tunisia) in A2 scenario and fro
+19% (Turkey) in B2 scenarios (Fig. 4e).
The general picture showed that the effect of climate change on agriculture i
areas are likely to be more severe than in temperate areas, even when the ferti
of increased CO2 is considered (Table 4). In the warmer southern Mediterrane
increases in CO2 help to reduce the loss in yield arising from a warmer and dri
61
-40 -30 -20 -10 0 10 20 30
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
B2
A2
b
-40 -30 -20 -10 0 10 20 30
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
B2
A2
a
c
-40 -30 -20 -10 0 10 20 30
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
B2
A2
d
-40 -30 -20 -10 0 10 20 30
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
B2
A2
e
-40 -30 -20 -10 0 10 20 30
Portugal
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
Figure 4- Impact of climate change on crop yields with CO2 effect: a) C4 summer crop, b)
legumes, c) C3 summer crop, d) tuber crops, e) cereals. The changes reported in the figures were
expressed as % and obtained as differences between the mean yields of the two futures and the
present yields.
Spain
France
Ital y B2
A2
62
Table 4 – Percentage changes of crop yields for the main Mediterranean regions: N-W
Spain, France and Italy, N
= Portugal,
-E = Serbia, Greece and Turkey, S-E = Jordan, Egypt and Libya, S-W =
Tunisia, Al ria a
it OWO2
ge nd Morocco
W hout C2 ith C
A2-A -A B2-A
C4 summe 0.1 5 4 8.78
B2 A2-A
r N-W 9 .80 .19
N
S
-4.43
-1
.54
.26
0.60
7.89 1. 4 -9 -
S -1 .87 8.94 9.38
-2 .90 - 3.42 4.38
N -1 .59 8.11 -7.19
2. -2
S -3 .26 - 5.81 3.92
r N -2 .79 - 0.44 2.41
5. -6 -
S -7.44 9 .66
S -1 4 - .81 .33
-1 .37 4.24 4.87
2. -6 -
S
S
-1 2 - .77 4.31
-2 .88 - .10 3.28
-E -2 - 0.21
-E 4 -6.70
-W 2 - - -6.37
Legumes N-W 4 1 -1 -4.86
-E 8 - 0.97
S-E -3 72 -36.43 3.30 -30.15
-W 3 2 -2 -18.48
C3 summe -W 1 1 -1 -2.85
N-E -1 57 .92 5.44 0.96
-E -8.1 3 -0.41
-W 9.9 11 -10 -4.34
Tubers N-W 0 - 7.53
N-E -2 50 .80 9.33 4.39
-E 8.2 15 - -5.66
-W 5 12 -1 -1.55
Cereals N-W -10.97 -3.49 -0.29 4.68
9 3.71 4.39 12.49
S-E -15.08 -17.17 -4.88 -10.15
N-E -6.7
S-W -13.77 -11.29 -3.42 -3.77
Impacts of adaptation options
r present
nt crops
ultivars with a
d positive effects
– 1% with
with longer growing
tandard
rage
+4% with an
.
early sowing of
standard cultivars determined positive effects on crop yields. More specifically, the
average changes of yields passed from – 3% with standard cultivar to +5% and -1% with
an early sowing and cultivar with longer growing cycles, respectively (Fig. 5c).
- Tuber crops. Tuber crops showed that the introduction of cultivars with a shorter
growing cycle or the early sowing of standard cultivars determined positive effects on
crop yields. More specifically, the average changes of yields passed from – 3% with
The results of the CropSyst simulation runs (including adaptation strategies) fo
and future climate scenarios were reported in Fig. 5. On the basis of the differe
types these are the main results:
- C4 summer crops. C4 summer crops showed that the introduction of c
longer growing cycle or the early sowing of standard cultivars determine
on crop yields. More specifically, the average changes of yields passed from
standard cultivar to +1% and +9% with an early sowing and cultivar
cycles, respectively (Fig. 5a).
- Legumes. Legumes showed that the introduction of an early sowing of s
cultivars determined positive effects on crop yields. More specifically, the ave
changes of yields passed from – 15% with standard cultivar to -10% and
early sowing and cultivar with longer growing cycles, respectively (Fig. 5b)
- C3 summer crops. C3 summer crops showed that the introduction of an
63
standard cultivar to +16% and +6% with cultivars with longer growing cycles an early
ing cycle
p yields. More
ltivar to
d +9% with cultivars with longer growing cycles and early sowing, respectively
(Fig. 5e).
gative
n of the sowing date
owing cycle.
of
he increasing
dditional water for irrigation. In
particular, the effective use of long cycle cultivars can demand 25 – 40 % more water
Table 5 - M n o p ga e narios for
different adaptation option
Crop type cenario A2 Scenario B2
sowing, respectively (Fig. 5d).
- Cereals. Cereals showed that the introduction of cultivars with a longer grow
or the early sowing of standard cultivars determined positive effects on cro
specifically, the average changes of yields passed from +1% with standard cu
+8% an
The introduction of adaptation strategies showed the possibility to reduce the ne
effects determined by the changes in climate conditions. Anticipatio
may allow the crops to escape the water stress during the late period of the gr
Cultivars with longer growing period may increase the length of the filling
reproductive organs that under future climate is expected to be shorter for t
temperature. Both options, however, would require a
(Table 5), which may or may not be available in the future.
ean perce tage changes
s
f water sup ly for irri ted crops und r future sce
S
Standard
(%)
ly sowin
(%)
ong cy
(%)
Standa
(%)
Early sowing
(%
Long cycle
(%)
Ear g L cle rd
)
C4 summer
crop
2.9
3.6
25.1
2.6
2.6
26.5
Tuber crops 7.0
9.6
43.7
7.8
8.3
41.0
64
-30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
early sowing
long cycle
standard crop
b
-40 -35 -30 -25 -20 -15 -10 -5 0 5 10 15 20
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
early sowing
long cycle
standard crop
a
-30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
early sowing
long cycle
standard crop
d
-30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30
Portugal
Spain
France
Ital y
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
early sowing
long cycle
standard crop
c
e
-30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30
Serbia
Greece
Turkey
Jordan
Egypt
Libya
Tunisia
Algeria
Morocco
early sowing
Figure 5 - Impact of different crop adaptation options on crop responses under climate change: a)
C4 summer crop, b) legumes, c) C3 summer crop, d) tuber crops, e) cereals. The changes reported
in the figures were expressed as % and obtained as differences between the mean yields of the
two futures and the present yields.
Portugal
Spain
France
Ital y long cycle
standard crop
65
Discussions and Conclusions
ion
reduction of
uction of the
ons in
ler northern
account.
ields due to
editerranean, the effects of climate
change and its associated increase in CO2 may have little or small positive impacts on
ates may be
crop yield determined
re up to 40%
heric
losses under a
of ground
d (Morgan et
d by it,
zone concentrations are expected to increase globally. Thus, if the effects of
to be included in an assessment of crop yields in the Mediterranean under a
n presented in
References
-air CO2
tosynthesis,
-372
2001) Ecosystems and their goods and
services, In: Climate Change 2001: Impacts, Adaptation and Vulnerability, McCarthy, JJ
BA, Kobayashi K, Bindi M (2002) Responses of agricultural crops to free-air
CO enrichment. Advances in Agronomy, 77: 293-368
Long SP (2003) How does elevated ozone impact soybean?
A meta-analysis of photosynthesis, growth and yield. Plant, Cell and Environment 2003
26:8 1317
Stockle CO, Donatelli M, Nelson R (2003) CropSyst, a cropping systems simulation
model. Eur J Agron 18: 289-307
Our study showed that the increases in temperature and reduction in precipitat
predicted for both future climate scenarios (A2 and B2) lead to a substantial
yields for all the crop types in all the ‘hot spots’ grid cells, through the red
length of the growing period and the water available for crop growth. Reducti
yields are more severe in the warmer southern Mediterranean than in the coo
Mediterranean, even when the fertilizing effect of increased CO2 is taken into
The southern Mediterranean is likely to experience an overall reduction of y
climate change. In some locations in the northern M
yields, provided that additional water demands can be met.
Strategies such as early sowing dates or cultivar with slower development r
considered as helpful options to reduce some of the reductions in
by the changes in climate conditions. However, such options could requi
more water for irrigation, which may or may not be available in the future.
Moreover, according to recent studies on the effects on crop yields of troposp
pollutants such ozone, there is reason to believe that our estimates of yield
future scenario may be conservative. Current and increased concentrations
level ozone have been shown to lead to decreases in plant biomass and yiel
al., 2003; Gitay et al., 2001). Independently of climate change, but exacerbate
surface o
ozone are
future climate scenario, the results are likely to greater yield reductions tha
this report.
Ainsworth EA, Long SP (2005) What have we learned from 15 years of free
enrichment (FACE)? A meta-analytic review of responses to rising CO2 in pho
canopy properties and plant production. New Phytol., 165:351
Gitay H, Brown S, Easterling W and Jallow B. (
Canziani, OF, Leary, NA, Dokken DJ and White KS (ed.), p.735-800.
Kimball
2
Morgan PB, Ainsworth EA,
66
67
... Over the past decades, a surge in the number of large and uncontrolled wildfires has occurred on all terrestrial ecosystems, and this can be observed irrespectively of national fire-fighting capacities or management policies (Westerling et al. 2006;Lohman et al. 2007;Forsyth and Van Wilgen 2008;Bowman et al. 2009). The global warming predicted by climate models for the twenty-first century and beyond may amplify this trend and threaten most ecosystems worldwide (Giannakopoulos et al. 2005;Pachauri and Reisinger 2008;Varela et al. 2019). ...
... Corsica harbours a high biodiversity, many endemic species (Médail and Verlaque 1997;Médail 2017), and preserved Mediterranean forest ecosystems (Medail and Quezel 1997;Vogiatzakis et al. 2016). This renders Corsican ecosystems vulnerable to climate and disturbance-regime changes as shown in its past environmental history and the predictions for the future (Giannakopoulos et al. 2005;Giorgi 2006;Vogiatzakis et al. 2016;Lestienne et al. 2020b). Recent studies have suggested that the fire danger could be three times higher when summer temperature anomaly exceeds + 2 °C ) and that the fire danger could be doubled in Corsica (Varela et al. 2019). ...
... accidental or intentional ignitions) (Curt et al. 2016;Costafreda-Aumedes et al. 2018). Associated to the warmer and dryer climate, these conditions will strongly contribute to increase wildfire frequency and intensity in the Mediterranean region and likely the occurrence of large fires and megafires (Mouillot et al. 2002;Pausas 2004;Giannakopoulos et al. 2005;Tedim et al. 2013;Batllori et al. 2013;Lahaye et al. 2018;Lestienne et al. 2020b). In terms of burned area, the increase would reach 15 to 140% by the end of the twenty-first century, depending upon the RCP scenario (Amatulli et al. 2013; Dupuy et al. 2020). ...
Article
Full-text available
In the Mediterranean basin, Corsica (French island) harbours among the best-preserved Mediterranean forest ecosystems. However, its high biodiversity could be threatened by the climate and disturbance-regime changes due to the global warming. This study aims (i) to estimate the future climate-related fire hazard in Corsica for the current century (2020–2100) based on two RCP scenarios (RCP4.5 and RCP8.5) and (ii) to compare the predicted trends with the entire Holocene period for which fire hazard has previously been assessed. An ensemble of future climate simulations from two IPCC RCP scenarios has been used to compute the Monthly Drought Code (MDC) and the Fire Season Length (FSL) and to assess the level of fire hazard. Here, we show that the MDC and the FSL would both strongly increase over the next decades due to the combined effect of temperature increase and precipitation decrease in the Corsica region. Moreover, the maximum Holocene FLS (7000 to 9000 years ago) will be reached (and even exceeded depending upon the scenario) after 2040. For the first time in the Holocene, we may be confronted to an increase in the number of fire-prone months driven by climate combined with many human-caused ignitions. This combination should increase the burned area from 15 to 140% according to scenarios. For the next 30 years, the game seems to be already played as both RCP scenarios resulted in similar increase in fire hazard in terms of drought and duration. It is thus mandatory to reconsider fire-management and fire-prevention policy to mitigate the future fire risk and its catastrophic consequences for ecosystems, population, and economy.
... The coastal areas of the Mediterranean Basin are also among the world's leading tourist centres and primary sources of income and employment in the region (Brochier & Ramieri, 2001). In 2003, the Mediterranean countries received 147 million tourists and generated US$113 billion (Giannakopoulos et al., 2005). In 2007, about 275 million tourists visited the Mediterranean countries. ...
Article
The Mediterranean Basin covers more than 2 million square kilometres and is surrounded by three continents: Africa, Asia and Europe. The Basin that is rich in biodiversity has tilted towards warmer and drier conditions over the last decades. The emerging climatic conditions particularly the increase in the number of climate extremes are bringing new threats and risks that will exacerbate existing pressures. The present study thoroughly reviewed the recent scientific literature and synthesized existing body of knowledge on the impacts (direct and indirect) of climate change on forest ecosystem services in the Mediterranean Basin. Despite many uncertainties about climate change in the Basin, there appears to be a consensus among a number of studies that climate change is having and will continue to have mostly negative impacts on the Mediterranean forest ecosystem services (wood and non-wood forest products, water resources, carbon storage and recreation and tourism) with possible substantial impacts in the future. Further, evidence is mounting that climate-induced natural disturbances (fires, insect pests, and pathogenic diseases) are becoming frequent and severe. The Mediterranean plants are known for their resilience to natural disturbances. However, the novel climatic conditions may exceed their resilience and alter the ecosystem services. Therefore, there is the need to mitigate the challenges posed by climate change and adapt forest management practices to impending changes to sustain the forest ecosystem services.
... Yağış miktarındaki azalış ve dengesizlikler ile sulu tarımın yaygınlaşması ile bu bölgede önemli bir tarımsal potansiyele sahip Konya havzasında su kaynaklarında azalma eğilimi olmuştur. İleriye dönük yapılan iklim tahmini çalışmalarında Türkiye'nin içinde bulunduğu alanda küresel ısınma ile birlikte kuraklık sorunu ile karşılaşılabileceği ve Orta Anadolu bölgesinin bu sorunu daha fazla yaşayabileceği öngörülmektedir [1,2]. ...
Article
Full-text available
Determination of The Effects of Different Irrigation Levels on Biomass Yield and Vegetation Day and GDD Value Required for Different Growth Periods in Switchgrass Varieties Abstract: This study was conducted to determine biomass yield and the days with growing degree days (GDD) values required for some development periods in different irrigation levels conditions of switchgrass varieties in Konya ecological conditions. The data in the study were obtained in the 4th year (2018), when the potential of the yield and other agricultural characteristics from the area established in 2015 was fully revealed. The study was carried out in randomized blocks – split plots experimental design. In the study the main parcels were irrigated levels (S1: Full irrigation; S2: Irrigation 75% of S1; S3: Irrigation 50% of S1) and the sub plots were six switchgrass varieties (Alamo,Kanlow, Shelter, Trailblazer, Shawnee, Cave in rock). In the study, the highest and lowest biomass yield were obtained from Kanlow variety (4271 kg da-1) on S1 and Shawnee variety (1670 kg da-1) on S3 irrigation, respectively. In the study, as the irrigation level decreased, the number of days and GDD required for flowering and harvest increased, and the highest and lowest number of flowering days and the required GDD value for this period were obtained from Alamo variety at S3 irrigation level (137 days and 1465 ℃) and Shelter at S1 irrigation level, respectively. (91 days and 892 ℃). The highest and lowest number of harvest days and the required GDD value for this period were obtained from Alamo variety on S3 (144 days and 1564 ℃) and Shelter variety on S1 (98 days and 981℃), respectively.
... This is grounded on the spectacular scenery that is built upon the remarkable natural and cultural richness and diversity of the Mediterranean's mainland, which is most important for the tourism sector's coastal, insular, and marine counterparts. Concurrently, however, this area is also marked as a hot spot in sustainability terms in many respects, e.g., overtourism, urbanization, water scarcity, climate change, environmental degradation, floods and droughts [33,[36][37][38][39]. Such threats menace the distinguishable land and underwater cultural heritage of the Mediterranean region (abbreviated as CH and UCH, respectively, in the following of this work), namely the key resources upon which cultural tourism in the destinations of this area is built, but also the destinations per se, e.g., in coastal areas. ...
Article
Full-text available
Tourism trends towards authentic, experience-based products have brought to the fore front lesser-known destinations, e.g., small and medium-sized Mediterranean islands. These can gain competitiveness on the ground of their distinguished land and Underwater Cultural Heritage (UCH), thus opening up new directions for their future sustainable development. Sustainable exploitation of UCH, though, calls for dedicated planning endeavors and respective UCH data acquisition/management. However, while the planning discipline can offer effective approaches and tools for properly handling cultural heritage, a deficit of sufficient, pertinent and well-documented UCH data is noticed, e.g., data on shipwrecks as part of UCH. The latter is the focus of this work, aiming, as a first step, to illuminate aspects of a shipwreck database, relevant to planners’ requests. Towards this end, global-wide distributed knowledge is explored in an effort to identify potentialities and limitations, content commonalities, shipwreck attributes so far recorded, (spatial) planning requirements, to name but a few. Knowledge acquired from this exploration can feed conceptualization of a planning-oriented shipwreck database. This can be replicable to various national/regional contexts; and is capable of providing well-structured shipwrecks’ content that can steer strategic planning efforts towards authentic ensembles of people’s history and culture in remote and less-privileged Mediterranean islands.
... In general, a trend of maize yield reduction is in line with Giannakopoulos et al. (2005), Gallo (2015), and Tubiello et al., (2000). The impact of climate change on future ET c , ET a and ET aw for maize in the Mediterranean countries, under both RCPs, is expected to be on average about 4%, 3% and 13% higher than the baseline, respectively. ...
Article
The impact of climate change on agriculture is projected to be more severe over the coming years due to changing intensity, magnitude and distribution of precipitation, soil water content, atmospheric water vapor, higher temperatures, and thus larger evapotranspiration. This will have significant consequences for irrigation requirements, especially in semi-arid area of Southern Europe, which is recognised as a climate change hotspot. Since the total water use to satisfy agricultural demand is currently about 25% of total water withdrawal in Europe and up to 80% in some Mediterranean countries, improvements in water management are needed to cope with worsening climate conditions. Although several crop models (e.g. EPIC, WOFOST, CERES-Maize) and hydrological models (e.g., DREAM, SWAT, HYDRUS) have been developed, the call for developing models able to couple crop growth, soil water balance, and irrigation practices to assess agricultural water management both at local and regional scale is increasing. In this work, the Simulation of Evapotranspiration of Applied Water (SIMETAW#) model is implemented in R programming language with two new versions able to estimate crop water consumption, irrigation demand and scheduling at local (SIMETAW_R) and regional scale (SIMETAW_GIS platform) using extensive climate and environmental geospatial datasets. SIMETAW_R was validated in ten experimental sites, and SIMETAW_GIS performance in Mediterranean countries was assessed by estimating the impact of climate change on maize, wheat, and wine grape water requirements in the past (1976–2005) and future climate conditions (2036–2065), under RCP4.5 and RCP8.5 scenarios. Results show that in Mediterranean countries, maize, wheat, and grape production will require on average about 13%, 16%, and 10% more water, respectively, under future climate. This represents a considerable challenge for water resources management, especially with demand increasing in other sectors. The tool showed great potential in estimating climate change impact on crop water consumption and irrigation requirements, both at local and regional scale, and offers new analytical skills for water resources management planners for improved decision-making.
... Climate is the typical Mediterranean climate characterized by having mild winters, warm autumns, rainy springs, and dry and hot summers, in which forest fires abound. As for all Mediterranean regions, Valencia is classified as one of the most negatively affected by climate change in Europe [30]. ...
Article
Full-text available
Forest management is an untapped tool, yet to realize its full potential to fight against climate change. The capability of forests to act as carbon sinks makes them a key resource to reduce CO2 concentration in the atmosphere. However, carbon which has been fixed can be suddenly emitted again as a consequence of disturbances such as pests or wildfires. Mediterranean plant phenology, climatic conditions, and the accumulation of fuel biomass due to abandonment of traditional forest uses generate a scenario prone to large wildfires and consequently large greenhouse gases (GHG) emissions. The abandonment could be offset by considering the economic value of forest ecosystem services, principally carbon fixation. Nevertheless, currently existing forest carbon markets consider only anthropogenic fixation based on a business as usual scenario without disturbances that cannot be applied to Mediterranean forest reality. Thus, a methodology to monetize carbon fixed has been developed and applied. A range between 55.5 and 250 million € produced by the monetization of 16.5 million potential carbon credits has been obtained based on anthropogenic avoided emissions produced over a 10 year-period. Thereby, the potential for offsetting emissions of the pilot region was between 1.2% and 5.6% of total diffuse GHG emissions. Consequently, sustainable forest management represents an important opportunity to combat climate change, taking advantage of the margin of improvement that the Mediterranean forests currently have to avoid GHG emissions through forest fire prevention silviculture.
... As is known that climate change is one of the most serious environmental problems that directly affect the economy and human life (Peker et al., 2019). In present reports on climate change and potential effects on water resources, it was envisaged that Mediterranean, Aegean and Central Anatolia regions of Turkey will face to serious drought problems by the year 2050 (Giannakopoulos et al., 2005;Yurekli and Anli, 2008;Yilar et al., 2020). Safe water potential of Konya province with insufficient and irregular precipitations is about 1.8 billion m 3 , but already used quantity is 2.6 billion m 3 , so there is a deficit of 0.8 billion m 3 . ...
Article
Full-text available
İklim, turizm destinasyonunun se iminde, turizm faaliyetlerinin ve turizm periyodunun belirlenmesinde nemli bir fakt rdür. İklim değişikliği ve turizm ilişkisine bu perspektiften yaklaşıldığı zaman, dünyadaki turizm dinamiklerinin, popüler turizm destinasyonlarının, turist eken ve turist üreten merkezlerin yeniden tanımlanma sürecine girecekleri ve bu süre te farklı b lgeler i in farklı tehditler ve fırsatların ortaya ıkacağı g rülmektedir. Bugün dünyanın en popüler turizm destinasyonlarından biri olan ve Türkiye’nin de i inde bulunduğu Akdeniz havzası i in oluşacak fırsat ve tehditleri ng rebilmek ve buna uyumlu turizm planlaması yapmak bu b lgedeki turizm sekt rünün sürekliliği a ısından olduk a kritiktir. Türkiye i in nemli bir ekonomik sekt r olan turizmin iklim değişikliğine karşı dayanıklılığının arttırılması, gelecekteki turizm gelişme, yatırım ve teşvik politikalarının ng rülen tehdit ve fırsatları g zeterek oluşturulması ile mümkündür. Bu alışma, Türkiye i in ng rülen iklim değişikliği senaryolarını dikkate alarak turizm sekt rünün karşı karşıya kalacağı fırsat ve tehditleri belirlemeyi ve gelecek turizm planlaması i in altlık oluşturacak neriler geliştirmeyi ama lamaktadır. Bu ama er evesinde detaylı bir literatür alışması yapılarak, elde edilen bulgular GZFT (SWOT) analizi y ntemi ile değerlendirilmiştir. Bu alışma, iklim değişikliğine karşı Türkiye turizminin güncel ve gelecekteki durumunu ortaya koyması ve bu temelde neri geliştirmesi bakımından nemlidir.
Chapter
Full-text available
The fundamental objective of this report is to analyse the variations and tendencies of the annual average maximum and minimum temperatures during the XX century in the Southern Plateau (Spain). To that end, we have gathered data for a regional study by applying the space aggregation method proposed by JONES et al. (1986), and then, we have undertaken the study of the temporary thermal changes. The results obtained reveal that the maximum temperature shows a greater long-term tendency than the minimum. Nevertheless, both variables have displayed a very similar and significant tendency increase since the 70´s.
Chapter
Full-text available
http://books.google.es/books?hl=es&lr=&id=RBsnNRE6ATIC&oi=fnd&pg=PA109&ots=9MYluD7bR_&sig=_zvBIZPUcUs5Gb1IRoPyEhh4W6c#v=onepage&q&f=false Moberg and Alexandersson’s (1997) methodology has been applied to homogenize series of monthly mean values of maximum and minimum daily temperature recorded in observatories located at the medium Ebro river basin. For each season, regional anomaly series have been constructed with the purpose of reflecting the evolution of temperatures in this area. The underlying trends have been extracted by smoothing, and then studied in order to appreciate the signs of a possible climate change
Article
Full-text available
Earlier reports indicated some specific isolated regions exhibiting a paradoxical increase of extreme rainfall in spite of decrease in the totals. Here, we conduct a coherent study of the full-scale of daily rainfall categories over a relatively large subtropical region- the Mediterranean- in order to assess whether this paradoxical behavior is real and its extent. We show that the torrential rainfall in Italy exceeding 128 mm/d has increased percentage-wise by a factor of 4 during 1951-1995 with strong peaks in El-Nino years. In Spain, extreme categories at both tails of the distribution (light: 0-4 mm/d and heavy/torrential: 64 mm/d and up) increased significantly. No significant trends were found in Israel and Cyprus. The consequent redistribution of the daily rainfall categories -torrential/heavy against the moderate/light intensities - is of utmost interest particularly in the semi-arid sub-tropical regions for purposes of water management, soil erosion and flash floods impacts.
Book
Mohamed Larbi Selassi Deputy Director of the National Meteorology, Morocco Welcome address (translated from French) WMO, WCRP, Medias-France and scientific institutions representatives, ladies and gentlemen, to thank WMO and MEDIAS-France, who have honoured us by I want first organizing the two workshops, climate indices in Africa and data assessment for global change research in the Mediterranean region, in Casablanca and I welcome all of you here in Morocco. It is with great pleasure that I open these two workshops on behalf of myself and on behalf of the Direction of the Meteorologie N ationale of Morocco. Climate change is becoming the focus of the international community because of its global scale and unpredictable effects, the numerous impacts it causes, its global feature and the complexity of the solutions that can mitigate its impacts. Global warming and the greenhouse effect became a subject of study at the international level since the United Nations Conference on the human environment that was held in Stockholm in 1972. The research and coordination efforts that have been made in this area have led to an "International Scientific consensus". High level meetings like those held in Toronto in 1988, in Lahaye in 1989 and in Geneva in 1990, did confirm the greenhouse threat and the emergency to treat it.
Chapter
A new and more extensive homogenised and gridded data set of surface air temperature (NESAT v.2) has been generated for Northeastern Spain. The aim of this compilation has been to extend, temporally, the preliminary estimations of the long-term variations and trends of the Catalonian temperature (Brunet, et al., 1999). The improved meteorological network for this region, composed of 23 records of non-urban stations, has been used to analyse the 1869–1998 period. To obtain a new regional time-series of monthly average of daily mean temperature, the gridded procedures proposed by Jones et al. (1986) have been applied to the adjusted data. Hence, a rectangular 3° × 4° latitude-longitude box, that covers our region from 40°–43° N and 0°–4° E, has been employed to accomplish this purpose. The seasonal and annual variations and trends of these data have been estimated by using usual statistical techniques (linear trends, low-pass filters, change point detection, etc.). The results show a moderate but highly significant regional warming (0.9°C) in the whole period, largely contributed by winter trend, in agreement with results observed at the global scale (i. e. Peterson, et al., 1999; Jones, et al., 1999). However, this warming has not occurred in a steady, monotonic fashion, but different thermal periods have been recorded. First, an initial long steady period of moderate but highly significant warming (0.01° C/year) has been observed between 1869–1949. Second, and intermediate short cool phase lasting until mid1970s, associated with significant (95% confidence level) but moderate cooling (-0.03°C/year). Third, an unprecedented fast, abrupt, strong and highly significant warming (0.07°C/year) has been recorded from the mid-1970s to the present. Furthermore, six out of the ten warmest years of this record have been registered in the last decade (1989–1998).
Chapter
The aim of this paper is to detect, estimate and characterise the existing variations and trends in surface air temperature in the Northeast of Spain, from the mid-nineteenth century onwards. To achieve this purpose, a new homogenised and gridded dataset of monthly mean maximum and minimum temperatures has been generated improving the spatial and temporal coverage of a previous study (Brunet, et al., 1999). The annual and seasonal averages of this data are analysed for the entire period (1869–1998) and partial time subsections in search of time variations and trends, by using usual statistical techniques (linear regression, filtering, change point detection, etc.). Our results show a moderate warming over the Catalonian region during the overall period, either for daily maximum temperature (0.96° C) or daily minimum temperature data (0.82° C). However, the warming has not occurred in a steady and linear fashion, and several periods can be distinguished, the fast, abrupt, strong and highly significant warming observed from the mid-seventies to the present being the most remarkable. However, both maximum and minimum temperatures present quasi similar amounts of long-term warming leading to a slight but non-significant increase in DTR over the total period (0.14°C), in contrast with the reported ongoing reduction of day-night temperature differences found over several continental areas of the globe. Also, no diurnally asymmetrical changes have been observed for this region, in agreement with other regional-scale studies, but in disagreement with hemispherical and global findings.
Article
This paper examines the spatial and temporal rainfall characteristics of the region of Valencia, Western Mediterranean Basin least Spain), during the World Meteorological Organization (WMO) normal period 1961-1990. The study used a dense and homogeneous daily precipitation database comprising 97 rain-gauge stations. Total and monthly rainfall concentrations have been studied in the context of their mean values, interannual variability and spatial diversification. Trends have been analysed using both parametric and non-parametric tests. In order to establish the spatial distribution of rainfall patterns and to detect homogeneous areas with similar rainfall evolution, a statistic based on the Cramer-von Mises test is proposed. The kriging interpolation methods for characterizing the magnitude of observed changes is used. Areas with contrasting rainfall evolution are identified. In more humid areas, a significant decrease in annual rainfall associated with significant increases in interannual rainfall variability is observed. In inland zones, decreases in total annual rainfall and increases in interannual variability are less clear, but there are indications of an increase in monthly rainfall concentration. In these inland zones, where more forest and woodland areas are located and forest fires are frequent, the observed trends could greatly affect desertification through changes in the disturbance regime. In more arid areas, local variability in rainfall evolution is higher and no significant changes can be defined. Copyright (C) 2000 Royal Meteorological Society.
Article
A statistical framework is presented for the assessment of climatological trends in the frequency of rare and extreme weather events. The methodology applies to long-term records of event counts and is based on the stochastic concept of binomial distributed counts. It embraces logistic regression for trend estimation and testing, and includes a quantification of the potential/limitation to discriminate a trend from the stochastic fluctuations in a record. This potential is expressed in terms of a detection probability, which is calculated from Monte Carlo-simulated surrogate records, and determined as a function of the record length, the magnitude of the trend and the average return period (i.e., the rarity) of events.Calculations of the detection probability for daily events reveal a strong sensitivity upon the rarity of events:in a 100-yr record of seasonal counts, a frequency change by a factor of 1.5 can be detected with a probability of 0.6 for events with an average return period of 30 days; however, this value drops to 0.2 for events with a return period of 100 days. For moderately rare events the detection probability decreases rapidly with shorter record length, but it does not significantly increase with longer record length when very rare events are considered. The results demonstrate the difficulty to determine trends of very rare events, underpin the need for long period data for trend analyses, and point toward a careful interpretation of statistically nonsignificant trend results.The statistical method is applied to examine seasonal trends of heavy daily precipitation at 113 rain gauge stations in the Alpine region of Switzerland (1901-94). For intense events (return period: 30 days) a statistically significant frequency increase was found in winter and autumn for a high number of stations. For strong precipitation events (return period larger than 100 days), trends are mostly statistically nonsignificant, which does not necessarily imply the absence of a trend.