HR: 1340h
AN: A23B-0792    [Abstracts]
TI: Spin-Up Procedures for a Coupled Climate System Model
AU: * Phipps, S J
EM: sjphipps@utas.edu.au
AF: Antarctic Climate and Ecosystems CRC, University of Tasmania, Private Bag 80, Hobart, TAS 7001 Australia
AU: * Phipps, S J
EM: sjphipps@utas.edu.au
AF: Institute of Antarctic and Southern Ocean Studies, University of Tasmania, Private Bag 77, Hobart, TAS 7001 Australia
AU: Roberts, J L
AF: Antarctic Climate and Ecosystems CRC, University of Tasmania, Private Bag 80, Hobart, TAS 7001 Australia
AU: Power, S B
AF: Bureau of Meteorology Research Centre, GPO Box 1289K, Melbourne, VIC 3001 Australia
AU: Budd, W F
AF: Institute of Antarctic and Southern Ocean Studies, University of Tasmania, Private Bag 77, Hobart, TAS 7001 Australia
AU: Bindoff, N L
AF: Antarctic Climate and Ecosystems CRC, University of Tasmania, Private Bag 80, Hobart, TAS 7001 Australia
AU: Bindoff, N L
AF: Institute of Antarctic and Southern Ocean Studies, University of Tasmania, Private Bag 77, Hobart, TAS 7001 Australia
AU: Bindoff, N L
AF: CSIRO Marine Research, GPO Box 1538, Hobart, TAS 7001 Australia
AB: Natural climate variability arises both from the variability within the components of the climate system, and from the feedbacks between them. Coupled climate system models aim to represent both of these sources of variability. In order to use a coupled model to study natural variability on decadal and longer timescales, the simulated climate should be both realistic and stable over time. In the case of the coarse-resolution models used for millennial-scale climate studies, this can generally only be achieved through the use of flux adjustments. Both the control climate of such a model, and the magnitude of the flux adjustments required to eliminate drift, depend on the manner in which the model is "spun up", or initialised. The traditional approach used to initialise a coupled model, whereby the atmospheric and oceanic components are spun up independently, has a number of problems. Extensive use of flux adjustments is required in order to prevent the atmosphere and ocean models from exhibiting unrealistic drift upon coupling. The use of a relaxation boundary condition to spin up the ocean model also causes it to exhibit both a phase lag and a modulation of the seasonal cycle relative to observations. This paper assesses a variety of modifications to the traditional procedure for initialising a coarse-resolution atmosphere-sea ice-ocean general circulation model. As well as spinning up the atmosphere and ocean models independently, we assess the relative merits of spinning up the coupled model as a whole. We also attempt to modify the surface temperatures and salinities used to spin up the ocean model, in order to bring the model response into line with observations. In general, we find that the spin-up procedure involves a compromise between the degree of realism of the simulated climate and the magnitude of the flux adjustments required to eliminate drift in the coupled model.
DE: 3337 Numerical modeling and data assimilation
DE: 3339 Ocean/atmosphere interactions (0312, 4504)
DE: 4215 Climate and interannual variability (3309)
DE: 3309 Climatology (1620)
DE: 1620 Climate dynamics (3309)
SC: Atmospheric Sciences [A]
MN: 2004 AGU Fall Meeting