 |
| Titel |
Sea ice-Ocean modeling at CNRM-GAME for global climate studies: recent improvements of the sea-ice component |
| VerfasserIn |
Matthieu Chevallier, David Salas y Mélia, Vincent Vionnet, Gilles Garric, Sophie Belamari, Eric Brun |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250047038
|
|
|
|
|
|
| Zusammenfassung |
| GELATO model is the sea-ice component of the CNRM-CM Global Coupled model
(GCM). The current version of CNRM-CM (CNRM-CM5) is used at CNRM for
Coupled Model Intercomparison Project phase 5 (CMIP5) experiments. It includes the
Arpege-Climat atmospheric component and the NEMO-OPA9 ocean dynamic model,
including the dynamic-thermodynamic GELATO multicategory sea-ice model. We
developed a forced configuration of the coupled ice-ocean model NEMO-GELATO
in order to produce an hindcast of the Arctic and Antarctic sea-ice covers. The
ice/ocean model is forced by atmospheric fields from the ERAinterim reanalysis
during the period 1990-2009. Performance of the model is addressed, including the
sensitivity to the forcing data. Additionally, some physical processes, not much
represented in current GCM, were implemented in GELATO to improve sea-ice mean
state and variability. To better capture the transformation of the ice surface during
the melt season, we implemented a semi-empirical parametrization of melt ponds,
which are known to play a crucial role in the high-latitude feedbacks. Performance
of this parametrization is investigated, in both forced and coupled modes. In the
forced configuration, the evolution of the surface albedo during the spring-summer
seasons compares well with data sampled during the 1997-1998 SHEBA campaign.
We also introduced a simple parametrization of snow densification through wind
effect. This process induces a thickening of the ice, mainly in the Antarctic where
the snow cover is the thickest. The sea-ice/ocean hindcast is meant to be used for
initialization of seasonal forecasts using the fully coupled model CNRM-CM5. |
|
|
|
|
|
|