Climate is one of the most significant considerations for Australian Agri-businesses profitability. If managed well it can deliver profitability or if managed poorly it can result in economic stress and hardship.

Climate information to support agricultural decision making is required over a range of time-scales e.g. daily, seasonal and decadal, and it must be delivered in a form that is salient, contextualised and relevant to producer needs.

We present a statistical approach for the development of a regionalised seasonal climate forecasting system for southern New South Wales.  We outline the process for selecting appropriate model covariates responsible for variability in this region and the approach used to engage with producers in the region to “road-test” this selection.  We test the performance of the forecast system over the period 1995 to 2011 and compare the predictive performance with other available operational forecasting systems.  Over this period, the seasonal climate forecast predictive performance was superior (i.e. as measured by percent persistence) to other forecasting systems, with results 5 to 17% higher across all months except for the period December to February were results were approximately 5% lower.

This presentation is supported by two other papers submitted to this session that provide a stakeholder perspective on climate information needs and the integration of this information into a systems modelling approach to assess resultant on farm value and utility.