synchronized with, the original core. This shows how
important it is to synchronize the different cores. This can be
complex. Given the application of the comparison of results
from different cores, it is important to understand how this
synchronization is done.
At this stage of interpretation, climate models (Chap. 25)
can also be useful to better understand the possible links
between changes in climate recorded at different locations.
We now continue our analysis of the impact of a Heinrich
event on climate and vegetation in Western Europe. In a first
series of numerical experiments we hypothesized that for the
climate of Western Europe, the main forcing linked to a
Heinrich event is a cooling by 4 °C of the North Atlantic at
the mid-latitudes. We used the LMDZ atmospheric general
circulation model (developed at the Laboratoire de
Météorologie Dynamique—Dynamic Meteorology Laboratory, Paris) to estimate the impact of such an anomaly in the
context of glaciation, in this case, the Last Glacial Maximum. We therefore carried out two experiments by forcing
the atmospheric model with glacial boundary conditions (see
http://pmip1.lsce.ipsl.fr): ice sheets as reconstructed for the
LGM, concentrations of greenhouse gases as measured for
this period from ice cores, orbital parametres as they were
21,000 years ago. In the first experiment, we used the surface temperatures of the oceans as reconstructed by the
CLIMAP project (Chap. 21). The only difference in the
second experiment is the surface temperatures of the oceans
in the North Atlantic. In this experiment, we decrease these
temperatures by 4 °C between 40 and 50 °N. The second
experiment is a sensitivity experiment to the ocean surface
temperatures of the North Atlantic at the mid-latitudes.
Kageyama et al. (2005) show the results of this experiment for climate in France and the Iberian Peninsula. Figure 29.5 summarizes their results. The model does not
simulate a propagation of the cooling imposed in the North
Atlantic very far inland over the European continent. The
place on the European continent where this cooling is most
important in terms of temperature of the coldest month is in
the northwest of the Iberian Peninsula. It brings about an
increase of only 1 °C in this region, which is low compared
to the 4 °C imposed in the North Atlantic. On the other
hand, the precipitation anomaly simulated by the model in
response to the cooling imposed in the North Atlantic is
much greater: it reaches −200 mm/year (a shortfall of
200 mm/year, a drop of about 30%) over the Iberian
Peninsula. Examining the results of the model for the North
Atlantic and Europe, we see that the band of strong westerly
winds is shifted southward, contributing to the decrease of
precipitation over Europe. The slight precipitation increase
over northwestern Africa can also be partly attributed to this
atmospheric circulation change.
Kageyama et al. (2005) also show the impact of this
change in climate on vegetation, as simulated by the
dynamic ORCHIDEE vegetation model. The climate changes simulated by the climate model for a cold event in the
North Atlantic, as weak as the impacts may seem on the land
masses, result in a significant decrease in vegetation cover,
of both trees and herbaceous plants. This result suggests that
during glaciations vegetation in Europe and the Mediterranean is, as indicated by pollen records, extremely sensitive
to changes in climate.
It may be noted that there is an area on the Mediterranean
side of the Iberian peninsula where precipitation increases in
the sensitivity experiment on colder ocean surface temperatures at the mid-latitudes of the North Atlantic. This
anomaly, in opposition to the existing reconstructions for
this zone (Combourieu-Nebout et al. 2002; Kageyama et al.
2005), shows the limitations of this experiment. The
hypothesis that the factor responsible for the changes in
climate in western Europe was the differences in surface
temperatures of the North Atlantic Ocean at mid-latitudes is
not sufficient to explain all the climate anomalies reconstructed around the Mediterranean region. A first explanation could be that the differences in surface temperature of
the Mediterranean Sea were not taken into account. A second explanation could simply be that the model is not able to
simulate the climate correctly for this region. A third
explanation could be an error in the interpretation of the
records. In a case like this, it is very instructive for both
modelers and palynologists to compare their data. It is also
for this reason that the results of the models should be
analyzed not only from the point of view of surface climate
variables but also from the point of view of the circulation
and physics of the atmosphere, was this makes it possible to
identify the mechanisms responsible for the simulated
changes in climate and to improve the model and future
experiments.
These experiments allow the influence of the various
mechanisms suggested above to be quantified: propagation
of the ocean temperature anomaly to the adjacent continents
by the mean circulation, the shift of the prevailing winds
following the migration of the zones with a strong meridian
temperature gradient, the impact of atmospheric instabilities
and average temperature on precipitation. In this way, a
model can help interpret reconstructions and, in particular,
can confirm or rebut scenarios established using several
records located far away from each other.
It is important to be aware of the limitations of experiments
of this type: regions where the anomaly is imposed, lack of
vegetation or ocean feedback, assumptions in the model
design and experimentation. These aspects can be evaluated
by additional sensitivity experiments (see Chap. 25).
414
M. Kageyama et al.
important it is to synchronize the different cores. This can be
complex. Given the application of the comparison of results
from different cores, it is important to understand how this
synchronization is done.
At this stage of interpretation, climate models (Chap. 25)
can also be useful to better understand the possible links
between changes in climate recorded at different locations.
We now continue our analysis of the impact of a Heinrich
event on climate and vegetation in Western Europe. In a first
series of numerical experiments we hypothesized that for the
climate of Western Europe, the main forcing linked to a
Heinrich event is a cooling by 4 °C of the North Atlantic at
the mid-latitudes. We used the LMDZ atmospheric general
circulation model (developed at the Laboratoire de
Météorologie Dynamique—Dynamic Meteorology Laboratory, Paris) to estimate the impact of such an anomaly in the
context of glaciation, in this case, the Last Glacial Maximum. We therefore carried out two experiments by forcing
the atmospheric model with glacial boundary conditions (see
http://pmip1.lsce.ipsl.fr): ice sheets as reconstructed for the
LGM, concentrations of greenhouse gases as measured for
this period from ice cores, orbital parametres as they were
21,000 years ago. In the first experiment, we used the surface temperatures of the oceans as reconstructed by the
CLIMAP project (Chap. 21). The only difference in the
second experiment is the surface temperatures of the oceans
in the North Atlantic. In this experiment, we decrease these
temperatures by 4 °C between 40 and 50 °N. The second
experiment is a sensitivity experiment to the ocean surface
temperatures of the North Atlantic at the mid-latitudes.
Kageyama et al. (2005) show the results of this experiment for climate in France and the Iberian Peninsula. Figure 29.5 summarizes their results. The model does not
simulate a propagation of the cooling imposed in the North
Atlantic very far inland over the European continent. The
place on the European continent where this cooling is most
important in terms of temperature of the coldest month is in
the northwest of the Iberian Peninsula. It brings about an
increase of only 1 °C in this region, which is low compared
to the 4 °C imposed in the North Atlantic. On the other
hand, the precipitation anomaly simulated by the model in
response to the cooling imposed in the North Atlantic is
much greater: it reaches −200 mm/year (a shortfall of
200 mm/year, a drop of about 30%) over the Iberian
Peninsula. Examining the results of the model for the North
Atlantic and Europe, we see that the band of strong westerly
winds is shifted southward, contributing to the decrease of
precipitation over Europe. The slight precipitation increase
over northwestern Africa can also be partly attributed to this
atmospheric circulation change.
Kageyama et al. (2005) also show the impact of this
change in climate on vegetation, as simulated by the
dynamic ORCHIDEE vegetation model. The climate changes simulated by the climate model for a cold event in the
North Atlantic, as weak as the impacts may seem on the land
masses, result in a significant decrease in vegetation cover,
of both trees and herbaceous plants. This result suggests that
during glaciations vegetation in Europe and the Mediterranean is, as indicated by pollen records, extremely sensitive
to changes in climate.
It may be noted that there is an area on the Mediterranean
side of the Iberian peninsula where precipitation increases in
the sensitivity experiment on colder ocean surface temperatures at the mid-latitudes of the North Atlantic. This
anomaly, in opposition to the existing reconstructions for
this zone (Combourieu-Nebout et al. 2002; Kageyama et al.
2005), shows the limitations of this experiment. The
hypothesis that the factor responsible for the changes in
climate in western Europe was the differences in surface
temperatures of the North Atlantic Ocean at mid-latitudes is
not sufficient to explain all the climate anomalies reconstructed around the Mediterranean region. A first explanation could be that the differences in surface temperature of
the Mediterranean Sea were not taken into account. A second explanation could simply be that the model is not able to
simulate the climate correctly for this region. A third
explanation could be an error in the interpretation of the
records. In a case like this, it is very instructive for both
modelers and palynologists to compare their data. It is also
for this reason that the results of the models should be
analyzed not only from the point of view of surface climate
variables but also from the point of view of the circulation
and physics of the atmosphere, was this makes it possible to
identify the mechanisms responsible for the simulated
changes in climate and to improve the model and future
experiments.
These experiments allow the influence of the various
mechanisms suggested above to be quantified: propagation
of the ocean temperature anomaly to the adjacent continents
by the mean circulation, the shift of the prevailing winds
following the migration of the zones with a strong meridian
temperature gradient, the impact of atmospheric instabilities
and average temperature on precipitation. In this way, a
model can help interpret reconstructions and, in particular,
can confirm or rebut scenarios established using several
records located far away from each other.
It is important to be aware of the limitations of experiments
of this type: regions where the anomaly is imposed, lack of
vegetation or ocean feedback, assumptions in the model
design and experimentation. These aspects can be evaluated
by additional sensitivity experiments (see Chap. 25).
414
M. Kageyama et al.