HR: 1330h
AN: A42B-0753 [PDF]
TI: Performance of a Southern Ocean sea ice forecast model
AU: Heil, P
EM: Petra.Heil@utas.edu.au
AF: University of Tasmania, Sandy Bay, Hobart, Tas 2001
Australia
AU: * Roberts, A
EM: andrew.roberts@utas.edu.au
AF: University of Tasmania, Sandy Bay, Hobart, Tas 2001
Australia
AU: * Roberts, A
EM: andrew.roberts@utas.edu.au
AF: Antarctic Climate and Ecosystems Cooperative Research Centre, Private Bag 80, Hobart, Tas 7001
Australia
AU: Budd, W
EM: w.f.budd@utas.edu.au
AF: University of Tasmania, Sandy Bay, Hobart, Tas 2001
Australia
AB:
The presentation examines the forecast peformance of an oriented fracture sea ice model applied to the Southern Ocean to
predict sea ice state up to five days in advance. The model includes a modified Coulombic elastic-viscous-plastic rheology,
enthalpy conserving thermodynamics and a new method of parameterising thickness distribution mechanics. 15 ice thickness
classes are employed within each grid cell with a horizontal resolution of 50km. The model provides considerable insight
into the thickness evolution and climatology of Antarctic sea ice. To date, thickness evolution of the Southern Ocean sea
ice zone has mostly been assessed using course two-category models in climate simulations and results presented in this talk
provide much greater detail over some existing model output. Simulations are presented from the model driven with NCEP-2
atmospheric analyses, NOAA sea surface temperatures, and mean climatogological currents generated using an eddy resolving
ocean model. Analyses are generated by nudging ice concentrations with daily satellite derived open water fractions, and
simulations using this method are compared to those without. There are important considerations in assimilating passive
microwave ice concentration data into thickness distribution models, and particular attention is given to the treatment of
lead ice and the impact this has on estimated total Southern Ocean sea ice volume. It is shown that nudging the model with
satellite derived concentrations has an impact on ice mechanics as judged from simulated buoy tracks. A comparison with sonar
soundings of sea ice draft is also favourable but shows variation with location. Whilst 5 day forecasts are reasonably
skilled, predictive performance changes with season. Application of this research to operational ocean data assimilation
systems is discussed in the final stages of the talk.
DE: 4263 Ocean prediction
DE: 4540 Ice mechanics and air/sea/ice exchange processes
SC: Atmospheric Sciences [A]
MN: 2003 Fall Meeting