Evaluation of the Australian Climate Ocean Model (AusCOM) in the tropical oceans

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Thursday, 27 January 2011: 4:00 PM
Evaluation of the Australian Climate Ocean Model (AusCOM) in the tropical oceans
609 (Washington State Convention Center)
Xiaobing Zhou, Bureau of Meteorology, Melbourne, VIC, Australia; and D. Bi, S. Marsland, Z. Sun, A. C. Hirst, O. Alves, and G. Wang

AusCOM (comprising the MOM4p1 ocean model and the CICE4.0 sea ice model) is a component of the Australian Community Climate and Earth-System Simulator (ACCESS) whose results will contribute to the international Coupled Model Intercomparison Project phase 5 (CMIP5). It has been forced with the version 2 Coordinated Ocean-ice Reference Experiments (COREv2) interannual forcing dataset over the period from 1948 to 2007 (referred to as Exp-1), and also run for a 100-year period in the fully coupled atmosphere-ocean-land ACCESS system (referred to as Exp-2). The model mean states in temperature, salinity and ocean currents, along with the climate variability for both experiments have been investigated. Intercomparisons between these model results and observational data show that Exp-1 with atmospheric forcings has good performance in the tropical oceans. Exp-1 results are closer to observations than Exp-2 results as expected, although both experiments have reasonable behaviour in tropical regions. In Exp-2, annual mean sea surface temperature errors off the equatorial Pacific are less than 2oC. The amplitude of El Nino/Southern Oscillation (ENSO) is just a little stronger than observations, and the ENSO period is about 4 years. However, in the ACCESS coupled run there are still some problems such as: the cold tongue extends too far west; the standard deviation of sea surface temperature anomalies along the equator is too strong, particularly in the western Pacific; and the equatorial undercurrent in the Pacific Ocean is too weak. Ongoing work is being undertaken to address these model deficiencies.