64
H.-J. Bolle
(Houghton et aI., 1992), the IPCC compared the results of three climate models. The
results show, that in the Mediterranean basin during the summer months a reduction
of precipitation between zero and one mmlday is likely to occur and for the winter
months a variation from an increase of up to one mmlday down to a decrease of one
to two mmlday is possible. Wigley (1992) analysed the results of four independent
climate simulations for the Mediterranean. According to his interpretation "the only
common feature in which we can have any faith" is a large scale warming (of 1.23.5°C) in all seasons. The likely outcome of the precipitation analysis is a an
increase, mainly in autumn, with a chance 1:4 for a decrease. In summer the western
basin shows a tendency to become dryer, while the eastern part may become wetter.
Similar results for the Mediterranean basin are obtained by the coupled oceanatmosphere model ECHAM for the 100 years run, arriving at a threefold CO2increase.
Cubasch et al. (1996) compared the simulation results of five different transient
coupled atmosphere-ocean models oflow resolution and concluded that "the climate
change predicted (for Southern Europe) by the 5 models is inconsistent and gives no
clear result". The authors could, however, show that the quality of the simulation
improves with increased model resolution. This is understandable in view of the
importance of the complex topography of the area. In a model with higher resolution
then also improved land cover information could be implemented. These should be
effective land cover types representing the annual variability of albedo and latent
heat flux.
The ACACIA project (Parry 2000) evaluated the scenarios underlying the
estimate of the climate change impact on vegetation of five climate models of which
finally one was selected for most of the future yield estimates. To give an idea of the
obtained results some figures are extracted for Greece and Spain targeted at the year
2050. For Greece three of four versions of one model and two other independent
models predict a decrease of winter precipitation with increasing temperature. For
a 2 K temperature increase the winter precipitation should decrease by in the
average 10%. For the summer precipitation all models predict a decrease of
precipitation of in the average 18%. For Spain three out of five models suggest an
increase of winter precipitation of 10% and all models show a decrease of summer
precipitation of the order of 15% at 2 K temperature increase. There is a linear
dependence between precipitation and temperature change with the effect that the
precipitation change doubles if the temperature increase would be 4 K instead of 2
K.
The climate change signal for the persistence of droughts shows an upward trend
of about ten percent for all seasons that is independent of resolution though its
amplitude is resolution dependent. For precipitation a downward trend resulted for
the spring, summer and autumn months with increasing amplitude as the resolution
is improved. During the winter months, however, the trend is upwards. As an
example, for the T42 ECHAM3 model of the DKRZ (German Climate Computer
Center) the expected changes for southern Europe are of the order of + 0.2 mmld in
January/February, - 0.2 mm/d in May/June and -O.lmmld throughout summer and
autumn. In comparison the simulated present precipitation is of the order of 1.0 ±
H.-J. Bolle
(Houghton et aI., 1992), the IPCC compared the results of three climate models. The
results show, that in the Mediterranean basin during the summer months a reduction
of precipitation between zero and one mmlday is likely to occur and for the winter
months a variation from an increase of up to one mmlday down to a decrease of one
to two mmlday is possible. Wigley (1992) analysed the results of four independent
climate simulations for the Mediterranean. According to his interpretation "the only
common feature in which we can have any faith" is a large scale warming (of 1.23.5°C) in all seasons. The likely outcome of the precipitation analysis is a an
increase, mainly in autumn, with a chance 1:4 for a decrease. In summer the western
basin shows a tendency to become dryer, while the eastern part may become wetter.
Similar results for the Mediterranean basin are obtained by the coupled oceanatmosphere model ECHAM for the 100 years run, arriving at a threefold CO2increase.
Cubasch et al. (1996) compared the simulation results of five different transient
coupled atmosphere-ocean models oflow resolution and concluded that "the climate
change predicted (for Southern Europe) by the 5 models is inconsistent and gives no
clear result". The authors could, however, show that the quality of the simulation
improves with increased model resolution. This is understandable in view of the
importance of the complex topography of the area. In a model with higher resolution
then also improved land cover information could be implemented. These should be
effective land cover types representing the annual variability of albedo and latent
heat flux.
The ACACIA project (Parry 2000) evaluated the scenarios underlying the
estimate of the climate change impact on vegetation of five climate models of which
finally one was selected for most of the future yield estimates. To give an idea of the
obtained results some figures are extracted for Greece and Spain targeted at the year
2050. For Greece three of four versions of one model and two other independent
models predict a decrease of winter precipitation with increasing temperature. For
a 2 K temperature increase the winter precipitation should decrease by in the
average 10%. For the summer precipitation all models predict a decrease of
precipitation of in the average 18%. For Spain three out of five models suggest an
increase of winter precipitation of 10% and all models show a decrease of summer
precipitation of the order of 15% at 2 K temperature increase. There is a linear
dependence between precipitation and temperature change with the effect that the
precipitation change doubles if the temperature increase would be 4 K instead of 2
K.
The climate change signal for the persistence of droughts shows an upward trend
of about ten percent for all seasons that is independent of resolution though its
amplitude is resolution dependent. For precipitation a downward trend resulted for
the spring, summer and autumn months with increasing amplitude as the resolution
is improved. During the winter months, however, the trend is upwards. As an
example, for the T42 ECHAM3 model of the DKRZ (German Climate Computer
Center) the expected changes for southern Europe are of the order of + 0.2 mmld in
January/February, - 0.2 mm/d in May/June and -O.lmmld throughout summer and
autumn. In comparison the simulated present precipitation is of the order of 1.0 ±