Towards a Mediterranean GEWEX
363
activities have to be coordinated and more has to be done, specifically long term
measurements are needed to obtain information on the variability of the investigated
processes.
The approach described here concerns the "measured world" but it has strong
links to the modelling domain which needs these data to construct atmospheric
chemistry models.
4 This leads to the question of the advantage of a dedicated network of
Mediterranean modelling laboratories. A hierarchy of models is used to simulate
various aspects of climate change starting from local SV AT models of high temporal
resolution to low resolution global climate models. It is understood that in general
modelling centres cooperate well in various programs. The goal for a specific
Mediterranean simulation is to
• better simulate regional climate idiosyncrasies such as different developments of
the temperature in adjacent areas,
simulate occurrence, intensity, and frequence of extreme events and to
understand their causes,
simulate the impact of land cover changes on regional climate,
simulate and understand the dependency of the Mediterranean climate on large
scale phenomena such as NAO and the Indian Monsoon (which in turn are
relared to tropical systems),
• understand the causes of the MO,
model the Mediterranean water cycle.
For these purposes the major goal of the Mediterranean modelling community
is the development and improvement of coupled sea - atmosphere models. This is
one goal of the before mentioned MFS project.
The wider scientific community would be interested to know how accurate the
models in use simulate present climate state variables at the regional scale and could
answer the questions formulated above. More contacts should be made with those
groups that could contribute data for validation and consequently could help to
detect discrepancies with observed quantities. Furthermore the cooperation of the
water cycle group with modellers must be very close. From the research into
atmospheric chemistry new challenges are expected for the improvement of models.
5 The satellite remote sensing community could contribute the most integrative
data set for the whole Mediterranean basin though this data set consists of a different
kind of information than provided by all other communities. The first task would be
to standardize calibration, correction, and evaluation methods for the specific
Mediterranean conditions. It must be aimed at the longest possible consistent data
series. Secondly the relationships to measurements made at the ground must be
validated and it must be proved by intercomparison of data sets whether the analysis
of satellite data matches the local observations (this is not obvious in the case of soil
moisture and surface energy budget nor even in the "simple" cases of surface
temperature and albedo). Therefor "Anchor Stations" are inevitable for this
validation task. Thirdly the time series of information inferred from the satellite data
Précédent

- 364/372

Suivant