covers a wide spectrum of events at spatio-temporal scales
ranging from intense local precipitation lasting a few hours
to exceptional hot years for the whole planet. In addition,
because they are by definition rare, extreme events offer few
case studies, and the detection of possible man-induced
trends requires analysis over long periods of time and/or
large spatial domains. The study of extreme events is thus
often restricted by the availability of datasets of sufficient
spatial and temporal resolution.
Temperature Extremes
Extreme temperature events typically affect large areas
(usually several thousand kilometers), and are often
accompanied by extremes in other climate variables (e.g.
drought during a hot summer, snowstorms during a cold
winter). Their impacts on ecosystems and human activities
are thus particularly important.
One of the most striking recent examples is the European
heat wave in summer 2003 which, with a temperature 2.5 °C
higher than the seasonal mean, exceeded by three standard
deviations the distribution of summer temperatures in Europe. It had dramatic socio-economic and environmental
impacts: high mortality, loss of energy production, acute
urban pollution, fires, accelerated melting of glaciers etc.
The return period of this particular event was estimated at
250 years at the time it occurred (2003); it would have taken
4 times longer (1000 years) without human intervention
(Stott et al. 2004).
This illustrates an expected feature of climate change: hot
extremes become more frequent in a warmer world, and
conversely, cold extremes become less common, but remain
still possible. For instance, in the U.S., the ratio of the record
of high maximum daily temperatures to the record of low
minimum temperatures—which should be 1 to 1 in a stationary climate—is currently about 2 to 1 and is likely to
reach about 50 to 1 by the end of the twenty-first century
(Meehl et al. 2009).
At the global scale the geographical distribution of
changes in the frequency and amplitude of temperature
extremes is consistent with the distribution of average
warming (IPCC 2013). This finding is nonetheless nuanced
by the fact that changes in temperature distributions are often
more complex than a uniform shift towards warmer values:
spreading, tightening, and/or asymmetry of values may also
show up in the statistical distribution. For instance, in Europe and Central U.S., the variability of summer temperatures
is expected to increase in a warmer climate, due to the
increase in evapotranspiration which leads to drier soils
(Douville et al. 2016). This would result in a widening of the
temperature distribution, and further amplify the increase in
the frequency of hot extremes in these regions.
Precipitation Extremes, Tropical Cyclones
and Extra-Tropical Storms
Changes in the frequency and intensity of droughts and
floods, in response to global warming, are also of major
concern, as our industrialized societies become increasingly
vulnerable to rainfall extremes. Although the study of
droughts is similar to that of heat waves, since they both
have impacts over large areas, the analysis of extremes of
intense precipitation is more difficult because it requires data
with finer spatio-temporal resolution.
As illustrated by the record floods in summer 2002 in
Europe, intense precipitation events have been on the
increase since 1950 in the mid-latitudes of the northern
hemisphere. Even in the Mediterranean Basin, where rainfall
is decreasing on average, the episodes of heavy rainfall are
more intense. A recent example is given by the intense precipitation that occurred in South of France in autumn 2018.
This upward trend is noticeable on a global scale, although
the increases are more moderate than those observed for
temperature. Changes in precipitation extremes are consistent
with the Clausius-Clapeyron relationship: warmer air can
hold more humidity, so there is more water to be mobilized
by condensation when rainfall events occur.
Paradoxically, the extent of regions affected by drought is
also increasing, illustrating the fact that climate change not
only affects the mean of statistical distributions, but also its
variability. Africa, southern Eurasia and North America are,
according to the Fifth IPCC Assessment Report, the regions
most affected by these droughts in recent times, as illustrated
by the dry conditions persisting from 2014 to 2018 over the
west of the United States (California). The increase in the
frequency and/or intensity of both intense rainfall episodes
and droughts is consistent with the overall intensification of
the hydrological cycle in a warmer world (wet gets wetter,
dry gets drier).
Lastly, climate change may also affect meteorological
systems like tropical cyclones or extra-tropical storms. In a
warmer world, the former are expected to occur less frequently in general, due to less frequent atmospheric conditions favorable to the cyclogenesis (reduced temperature
difference between the surface and the high-troposphere).
However, once triggered, future cyclones should get more
energy from a warmer ocean: the intensity of the strongest
tropical cyclones is therefore expected to increase. Unfortunately, so far, trends are difficult to detect from the past due to
the lack of homogeneous data. For extratropical storms,
forecasting is even more difficult: projections performed with
climate models do not unanimously agree on their frequency
and/or intensity. The only robust future signal seems to be a
poleward shift of the storm tracks, but again, available
observations are insufficient to capture any past trend.
31 From the Climates of the Past to the Climates …
449
ranging from intense local precipitation lasting a few hours
to exceptional hot years for the whole planet. In addition,
because they are by definition rare, extreme events offer few
case studies, and the detection of possible man-induced
trends requires analysis over long periods of time and/or
large spatial domains. The study of extreme events is thus
often restricted by the availability of datasets of sufficient
spatial and temporal resolution.
Temperature Extremes
Extreme temperature events typically affect large areas
(usually several thousand kilometers), and are often
accompanied by extremes in other climate variables (e.g.
drought during a hot summer, snowstorms during a cold
winter). Their impacts on ecosystems and human activities
are thus particularly important.
One of the most striking recent examples is the European
heat wave in summer 2003 which, with a temperature 2.5 °C
higher than the seasonal mean, exceeded by three standard
deviations the distribution of summer temperatures in Europe. It had dramatic socio-economic and environmental
impacts: high mortality, loss of energy production, acute
urban pollution, fires, accelerated melting of glaciers etc.
The return period of this particular event was estimated at
250 years at the time it occurred (2003); it would have taken
4 times longer (1000 years) without human intervention
(Stott et al. 2004).
This illustrates an expected feature of climate change: hot
extremes become more frequent in a warmer world, and
conversely, cold extremes become less common, but remain
still possible. For instance, in the U.S., the ratio of the record
of high maximum daily temperatures to the record of low
minimum temperatures—which should be 1 to 1 in a stationary climate—is currently about 2 to 1 and is likely to
reach about 50 to 1 by the end of the twenty-first century
(Meehl et al. 2009).
At the global scale the geographical distribution of
changes in the frequency and amplitude of temperature
extremes is consistent with the distribution of average
warming (IPCC 2013). This finding is nonetheless nuanced
by the fact that changes in temperature distributions are often
more complex than a uniform shift towards warmer values:
spreading, tightening, and/or asymmetry of values may also
show up in the statistical distribution. For instance, in Europe and Central U.S., the variability of summer temperatures
is expected to increase in a warmer climate, due to the
increase in evapotranspiration which leads to drier soils
(Douville et al. 2016). This would result in a widening of the
temperature distribution, and further amplify the increase in
the frequency of hot extremes in these regions.
Precipitation Extremes, Tropical Cyclones
and Extra-Tropical Storms
Changes in the frequency and intensity of droughts and
floods, in response to global warming, are also of major
concern, as our industrialized societies become increasingly
vulnerable to rainfall extremes. Although the study of
droughts is similar to that of heat waves, since they both
have impacts over large areas, the analysis of extremes of
intense precipitation is more difficult because it requires data
with finer spatio-temporal resolution.
As illustrated by the record floods in summer 2002 in
Europe, intense precipitation events have been on the
increase since 1950 in the mid-latitudes of the northern
hemisphere. Even in the Mediterranean Basin, where rainfall
is decreasing on average, the episodes of heavy rainfall are
more intense. A recent example is given by the intense precipitation that occurred in South of France in autumn 2018.
This upward trend is noticeable on a global scale, although
the increases are more moderate than those observed for
temperature. Changes in precipitation extremes are consistent
with the Clausius-Clapeyron relationship: warmer air can
hold more humidity, so there is more water to be mobilized
by condensation when rainfall events occur.
Paradoxically, the extent of regions affected by drought is
also increasing, illustrating the fact that climate change not
only affects the mean of statistical distributions, but also its
variability. Africa, southern Eurasia and North America are,
according to the Fifth IPCC Assessment Report, the regions
most affected by these droughts in recent times, as illustrated
by the dry conditions persisting from 2014 to 2018 over the
west of the United States (California). The increase in the
frequency and/or intensity of both intense rainfall episodes
and droughts is consistent with the overall intensification of
the hydrological cycle in a warmer world (wet gets wetter,
dry gets drier).
Lastly, climate change may also affect meteorological
systems like tropical cyclones or extra-tropical storms. In a
warmer world, the former are expected to occur less frequently in general, due to less frequent atmospheric conditions favorable to the cyclogenesis (reduced temperature
difference between the surface and the high-troposphere).
However, once triggered, future cyclones should get more
energy from a warmer ocean: the intensity of the strongest
tropical cyclones is therefore expected to increase. Unfortunately, so far, trends are difficult to detect from the past due to
the lack of homogeneous data. For extratropical storms,
forecasting is even more difficult: projections performed with
climate models do not unanimously agree on their frequency
and/or intensity. The only robust future signal seems to be a
poleward shift of the storm tracks, but again, available
observations are insufficient to capture any past trend.
31 From the Climates of the Past to the Climates …
449