precipitation. This sudden change in state is therefore associated with a strong positive feedback from the vegetation.
Liu (2007) have another interpretation based on coupled
ocean-atmosphere simulations in which the rapid disappearance of vegetation is not accompanied by sudden aridification implying no strong positive feedback induced by
vegetation. The mechanism in this case is linked to the
internal variability of precipitations. This variability can be
considered to be a stochastic forcing that generates slow
stochastic variability of soil moisture. This variability
interacts with the nonlinearities of the plant system to generate a rapid decrease in vegetation. It is therefore the nonlinear response of the vegetation to the strong internal
variability that is responsible for the sudden change of state
of the vegetation when a bioclimate threshold is reached. All
these studies provide insights into the interactions between
long-term insolation variations, climate and vegetation.
Nevertheless, we do not yet know the exact reasons and the
dominant mechanisms in these African regions. Progress is
essential to better represent the semi-arid zones and to obtain
hydrological and ecological data at high temporal resolution.
The Recent Climate (Recent Centuries)
Simulating the climate variations of the last millennium is a
major scientific challenge for the beginning of the
twenty-first century. The first problem is a technological
one: these simulations require several months of intensive
calculations on the most powerful computers available. This
explains why the number of available simulations have
gradually increased from one IPCC report to the other (IPCC
2007, 2013). The second problem is how to define an initial
state for the ocean and the atmosphere. This problem is
avoided by using the output from an ocean model output
using pre-industrial conditions and for which no drift is
observed. The idea behind this approximation is that the
climate will adjust sufficiently quickly to forcing conditions
so that the initial conditions of the beginning of the last
millennium are not critical. The third problem is purely
physical and is the definition of the climate forcings over
time. On this time scale, the main forcings are:
• Changes in solar activity. The role of solar activity is not
yet clear. Fluctuations in the energy received in the troposphere are theoretically very low. Fluctuations in solar
activity also affect the chemistry of the stratosphere. The
modulation of reactions with ozone, in particular, can
cause heating of the stratosphere and its expansion, thus
favoring atmospheric circulation patterns in the troposphere (Shindell 2001). Variations in solar activity in
numerical climate simulations can have a purely radiative
effect. More sophisticated models include chemical
reactions with stratospheric ozone (Shindell 2001) which
can produce a warming. This option is more complicated
to implement because most of the models used for this
exercise do not include stratospheric chemistry, which is
very demanding on computing time.
• Volcanic eruptions, particularly the most powerful, which
emit sulfurous SO 2 and dust into the stratosphere. The
solid particles are quickly pulverized but the sulfurous
gas turns into sulfuric acid, which is only washed away
after several months. Large volcanic eruptions (e.g.
Pinatubo 1991) have the effect of cooling the planet for
two to three years (Robock 2000). Volcanism is generally
accounted for rather crudely in climate models (Jansen
2007) and is often done by decreasing the solar constant
by a few W/m
2 for two years. This representation,
although it saves calculation time, cannot account for the
dispersion of the volcanic particles across latitude in the
stratosphere and therefore for its impact on atmospheric
circulation.
• Greenhouse gases added to those already present in the
atmosphere. This additional greenhouse effect is noticeable from the twentieth century onwards. These are
almost exclusively emissions related to human activity.
Sulfate aerosols are also linked to human activity and
have a cooling effect on the atmosphere locally.
Estimates of the radiative effects of these forcings during
the last millennium are shown in Fig. 30.2.
The details of the atmospheric circulation responses vary
widely from one model to another, particularly for GCMs.
Simpler models such as EMICs, on the other hand, show
similar behaviors because the atmospheric responses have
few degrees of freedom, which constrains these models to
behave in the same way.
Characteristics of Climate Variability
Extratropical Circulation
The North Atlantic Oscillation (NAO) is the climate regime
of the atmospheric circulation dominating the North Atlantic
region. This regime was first detected in temperature
anomaly structures around the North Atlantic basin. A temperature anomaly, the term used by meteorologists, is the
difference between the observed temperature and the average
temperature. Sir Gilbert Walker, the British meteorologist,
noticed opposing temperature anomalies between Labrador
and North Africa on the one hand and the eastern United
States and Western Europe on the other. Thus, when it is
warmer than usual in the American East (a positive temperature anomaly), it is generally colder than usual in the
south of France (a negative anomaly).
436
P. Braconnot and P. Yiou
Liu (2007) have another interpretation based on coupled
ocean-atmosphere simulations in which the rapid disappearance of vegetation is not accompanied by sudden aridification implying no strong positive feedback induced by
vegetation. The mechanism in this case is linked to the
internal variability of precipitations. This variability can be
considered to be a stochastic forcing that generates slow
stochastic variability of soil moisture. This variability
interacts with the nonlinearities of the plant system to generate a rapid decrease in vegetation. It is therefore the nonlinear response of the vegetation to the strong internal
variability that is responsible for the sudden change of state
of the vegetation when a bioclimate threshold is reached. All
these studies provide insights into the interactions between
long-term insolation variations, climate and vegetation.
Nevertheless, we do not yet know the exact reasons and the
dominant mechanisms in these African regions. Progress is
essential to better represent the semi-arid zones and to obtain
hydrological and ecological data at high temporal resolution.
The Recent Climate (Recent Centuries)
Simulating the climate variations of the last millennium is a
major scientific challenge for the beginning of the
twenty-first century. The first problem is a technological
one: these simulations require several months of intensive
calculations on the most powerful computers available. This
explains why the number of available simulations have
gradually increased from one IPCC report to the other (IPCC
2007, 2013). The second problem is how to define an initial
state for the ocean and the atmosphere. This problem is
avoided by using the output from an ocean model output
using pre-industrial conditions and for which no drift is
observed. The idea behind this approximation is that the
climate will adjust sufficiently quickly to forcing conditions
so that the initial conditions of the beginning of the last
millennium are not critical. The third problem is purely
physical and is the definition of the climate forcings over
time. On this time scale, the main forcings are:
• Changes in solar activity. The role of solar activity is not
yet clear. Fluctuations in the energy received in the troposphere are theoretically very low. Fluctuations in solar
activity also affect the chemistry of the stratosphere. The
modulation of reactions with ozone, in particular, can
cause heating of the stratosphere and its expansion, thus
favoring atmospheric circulation patterns in the troposphere (Shindell 2001). Variations in solar activity in
numerical climate simulations can have a purely radiative
effect. More sophisticated models include chemical
reactions with stratospheric ozone (Shindell 2001) which
can produce a warming. This option is more complicated
to implement because most of the models used for this
exercise do not include stratospheric chemistry, which is
very demanding on computing time.
• Volcanic eruptions, particularly the most powerful, which
emit sulfurous SO 2 and dust into the stratosphere. The
solid particles are quickly pulverized but the sulfurous
gas turns into sulfuric acid, which is only washed away
after several months. Large volcanic eruptions (e.g.
Pinatubo 1991) have the effect of cooling the planet for
two to three years (Robock 2000). Volcanism is generally
accounted for rather crudely in climate models (Jansen
2007) and is often done by decreasing the solar constant
by a few W/m
2 for two years. This representation,
although it saves calculation time, cannot account for the
dispersion of the volcanic particles across latitude in the
stratosphere and therefore for its impact on atmospheric
circulation.
• Greenhouse gases added to those already present in the
atmosphere. This additional greenhouse effect is noticeable from the twentieth century onwards. These are
almost exclusively emissions related to human activity.
Sulfate aerosols are also linked to human activity and
have a cooling effect on the atmosphere locally.
Estimates of the radiative effects of these forcings during
the last millennium are shown in Fig. 30.2.
The details of the atmospheric circulation responses vary
widely from one model to another, particularly for GCMs.
Simpler models such as EMICs, on the other hand, show
similar behaviors because the atmospheric responses have
few degrees of freedom, which constrains these models to
behave in the same way.
Characteristics of Climate Variability
Extratropical Circulation
The North Atlantic Oscillation (NAO) is the climate regime
of the atmospheric circulation dominating the North Atlantic
region. This regime was first detected in temperature
anomaly structures around the North Atlantic basin. A temperature anomaly, the term used by meteorologists, is the
difference between the observed temperature and the average
temperature. Sir Gilbert Walker, the British meteorologist,
noticed opposing temperature anomalies between Labrador
and North Africa on the one hand and the eastern United
States and Western Europe on the other. Thus, when it is
warmer than usual in the American East (a positive temperature anomaly), it is generally colder than usual in the
south of France (a negative anomaly).
436
P. Braconnot and P. Yiou