POAMA Great Barrier Reef SSTA Forecasts

The Predictive Ocean Atmosphere Model for Australia

The Predictive Ocean Atmosphere Model for Australia (POAMA) is a state-of-the-art seasonal to inter-annual seasonal forecast system based on a coupled ocean/atmosphere model and ocean/atmosphere/land observation assimilation systems. POAMA has been running operationally since 2002 and the current version since July 2007.

The coupled model in POAMA (version 1.5; P1.5) consists of the Australian Bureau of Meteorology atmospheric model (BAM 3.0) and the CMAR Australian Community Ocean Model V.2 (ACOM2), coupled using the Ocean Atmosphere Sea Ice Soil (OASIS) coupling software. No corrections are applied to the exchanged fluxes between the models. The atmospheric model component has a horizontal spectral resolution of T47 and 17 vertical levels. The ocean model grid spacing is 2 degrees in the zonal direction and 0.5-1.5 degrees in the meridional direction, with 25 vertical levels, of which 12 are in the upper 185 m.

Atmosphere initial conditions are provided by an Atmosphere Land Initialisation (ALI) scheme, where a BAM3.0 atmosphere-only integration forced with observed SST, is nudged towards the ERA-40 reanalysis. The ocean model is initialised using an ocean data assimilation scheme based on the optimum interpolation (OI) technique described by Smith et al. (1991).

A hind-cast is a retrospective forecast which can be used to measure model performance by assessing the skill of the model in predicting past events. To generate the POAMA hind-cast set, the model was started on the 1st of each month of 1980-2006, initialised only with information available before the start date, and run forward in forecast mode for 9 months. A hind-cast ensemble was then created by running the model multiple times for each start month with slightly varying initial atmospheric conditions each time. An ensemble gives an indication of the potential spread in forecast skill by allowing the stochastic component of the model to evolve differently in each member and reduce the effects of uncertainties in model initial conditions.

Lead-time is defined as the time elapsed between the model start date and the forecast date, i.e. if the model start month is January, the forecasts at lead-times 0, 1, 2 and 3 months are for January, February, March and April respectively.

Further reading

• Spillman C. and Alves O., 2009: Dynamical seasonal prediction of summer sea surface temperatures in the Great Barrier Reef. Coral Reefs, 28, 197-206.
• Spillman C.M., Alves O., and Hudson D.A., 2010: Real-time seasonal SST predictions for the Great Barrier Reef during the summer of 2009/2010. CAWCR Res. Lett., 4, 11-19.
• Spillman C.M., Alves O., Hudson D.A. and Charles A.N., 2009: POAMA SST predictions for the Great Barrier Reef: Summer 2008/2009. CAWCR Res. Lett., 2, 30-34.

For further information visit POAMA group.

Please be aware that all POAMA forecasts are subject to the Bureau of Meteorology's copyright and disclaimer.