1 Promote the growth of the model development community :

1
(No Transcript)
2
(No Transcript)
3
(No Transcript)
4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7
What do we need and what should we do better ? (I)
(1) Promote the growth of the model development
community -gt reaffirm the importance of
improving basic atmospheric and oceanic
components of models -gt ensure long-term
appointments and funding for development
activities -gt promote ways of having a critical
mass of model developers within modeling
centers -gt organize regular workshops (cf WGNE,
US-CLIVAR) and trainings (cf ECMWF)
8
What do we need and what should we do better ? (I)
(1) Promote the growth of the model development
community -gt reaffirm the importance of
improving basic atmospheric and oceanic
components of models -gt ensure long-term
appointments and funding for development
activities -gt promote ways of having a critical
mass of model developers within modeling
centers -gt organize regular workshops (cf WGNE,
US-CLIVAR) and trainings (cf ECMWF) (2) Organize
systematic and coordinated investigations
(physical / statistical) of the link
between model errors and prediction errors -
to prioritise areas for model development and
improvement - to reduce empiricism of the
development/evaluation process (what biases
are most important/acceptable, how good a model
needs to be?) -gt develop activities devoted
to the understanding of model differences,
analyses revealing the primary cause of
model differences (e.g. regimes) -gt promote
systematic investigations of the impact of
resolution, strato/tropo coupling, eddies,
etc. -gt encourage idealized frameworks and
maintain a hierarchy of models within each
modeling group (e.g. 1D, aqua-planet, slab)
9
What do we need and what should we do better ?
(II)
(3) Reduce the gap between large-scale
modeling/processes/observations communities -gt
encourage process-oriented evaluations/diagnostics
of models (cf CFMIP, CCMVal) -gt associate
modelers with field expts/process studies right
from the beginning (cf ARM, VOCALS),
including for the terrestrial biosphere -gt
golden opportunities already there (e.g. AMMA,
YOTC, VOCALS) -gt organize and coordinate
integrating efforts on the long term (e.g. DIME)
10
What do we need and what should we do better ?
(II)
(3) Reduce the gap between large-scale
modeling/processes/observations communities -gt
encourage process-oriented evaluations/diagnostics
of models (cf CFMIP, CCMVal) -gt associate
modelers with field expts/process studies right
from the beginning (cf ARM, VOCALS),
including for the terrestrial biosphere -gt
golden opportunities already there (e.g. AMMA,
YOTC, VOCALS) -gt organize and coordinate
integrating efforts on the long term (e.g.
DIME) (4) Reduce the gap between
climate/NWP/assimilation communities -gt
encourage NWP modes of evaluation of climate
models (cf CAPT, Transpose AMIP) -gt share
expertise on initialisation methodologies
(workshops) -gt be aware of observational
constraints derived from data assimilation
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)