Cd. Hewitt et al., The impact of dynamic sea-ice on the climatology and climate sensitivity of a GCM: a study of past, present, and future climates, CLIM DYNAM, 17(9), 2001, pp. 655-668
We assess two parametrisations of sea-ice in a coupled atmosphere-mixed lay
er ocean-sea-ice model. One parametrisation represents the thermodynamic pr
operties of sea-ice formation alone (THERM), while the other also includes
advection of the ice (DYN). The inclusion of some sea-ice dynamics improves
the model's simulation of the present day sea-ice cover when compared to o
bservations. Two climate change scenarios are used to investigate the effec
t of these different parametrisations on the model's climate sensitivity. T
he scenarios are the equilibrium response to a doubling of atmospheric CO2
and the response to imposed glacial boundary conditions. DYN produces a sma
ller temperature response to a doubling of CO2 than THERM. The temperature
response of THERM is more similar to DYN in the glacial case than in the 2
x CO2 case which implies that the climate sensitivity of THERM and DYN vari
es with the nature of the forcing. The different responses can largely be e
xplained by the different distribution of Southern Hemisphere sea-ice cover
in the control simulations, with the inclusion of ice dynamics playing an
important part in producing the differences. This emphasises the importance
of realistically simulating the reference climatic state when attempting t
o simulate a climate change to a prescribed forcing. The simulated glacial
sea-ice cover is consistent with the limited palaeodata in both THERM and D
YN, but DYN simulates a more realistic present day sea-ice cover. We conclu
de that the inclusion of simple ice dynamics in our model increases our con
fidence in the simulation of the anomaly climate.