Sensitivities and uncertainties in the simulation of Arctic sea ice with a coupled regional atmosphere-ocean-ice model
A series of sensitivity experiments has been performed with thecoupled regional atmosphere-ocean-ice model HIRHAM-NAOSIM in order toassess the importance of uncertainties in the physical processdescriptions for the simulation of Arctic sea ice with a fully coupledmodel. The significance of the follo...
| Main Authors: | , , , |
|---|---|
| Format: | Conference Object |
| Language: | unknown |
| Published: |
2007
|
| Subjects: | |
| Online Access: | https://epic.awi.de/id/eprint/16770/ https://hdl.handle.net/10013/epic.26600 |
| Summary: | A series of sensitivity experiments has been performed with thecoupled regional atmosphere-ocean-ice model HIRHAM-NAOSIM in order toassess the importance of uncertainties in the physical processdescriptions for the simulation of Arctic sea ice with a fully coupledmodel. The significance of the following physical parameterizationshas been addressed: snow and ice albedo, snow cover, cloud cover,ocean heat transfer, lateral freezing of sea ice as well as the heatflux treatment in the ice growth scheme at large. It has been foundthat the simulation of Arctic summer sea ice responds very sensitivelyto the parameterization of snow and ice albedo but also to thetreatment of ice growth. Large sensitivities also arise with respectto the parameterizations of cloud cover, snow cover, and ocean heattransfer, but the response is not as strong as compared to the albedoand ice growth schemes. Arctic winter sea ice is only little affectedby all these parameterizations, but the parameterization of lateralfreezing controls the rate of increase in ice thickness, whichsignificantly influences summer sea ice conditions in multi-yearsimulations. Uncertainties in the simulation of the atmospheric heatfluxes and the atmospheric circulation hamper the accurate simulationof Arctic sea ice and may decrease the overall model performance dueto feedback effects. |
|---|