Six Years of a Climate Model Consensus Forecast Newsletter for Victorian (Australia) Farmers.

Advances in computer capacity have allowed many world meteorological agencies to construct Coupled Global Circulation Models (CGCM’s) for world climate predictions. Large and complex, these models attempt to mimic the behaviour of the atmospheric and ocean phenomena that drive the world’s climatic patterns.  Previous to their development, these agencies used statistical systems of modelling, based on observed historic ocean and atmospheric patterns.

The Australian Bureau of Meteorology had created a summary of CGCM’s predictions of temperature in the NINO 3.4 section of the Pacific Ocean. We thought that more value could be extracted from these models by looking further at their outputs. The challenge was to simplify complex information without compromising usefulness. Apart from NINO3.4, we decided to include the Indian Ocean surface temperature, rainfall and land temperature predictions. Models were selected according to their ease of web based access.

Predictions from five CGCM’s (ECMWF, POAMA, SINTEX, NCEP, GSM-COAPS recently NASA added), three ensembles of CGCM’s (UKMO, IRI, APCC) and three statistical models (BoM SCF, Qld SOI, ESS WA, recently BoM SCF removed) were collated each month since June 2008. The predictions were from the one to three and four to six month outlooks. The model outputs for the state of Victoria were incorporated into a newsletter called the “Fast Break”, effectively bringing two hours of web searching onto an A4 page. At the end of each year the models performance was rated against the actual rainfall. Consensus of models was found to be more useful, as individual model accuracy was erratic. Best overall model performance was for the 2010 La Nina event.

The newsletter has a distribution list of 2480 people, 62% of whom are farmers with 37% agribusiness and government advisors. The newsletter has also been popular with bankers and fire agencies. Apart from the model outputs, the newsletter explains current climate indicators with graphics and commentary.