Forecast demonstration projects (FDPs) have provided valuable opportunities to expose recent research outcomes in a near-operational setting. Operational forecasters and researchers work together each day over a period of weeks or months to trial new techniques and technologies in real-time forecasting situations. Researchers and new systems become better attuned to operational realities, and forecasters have an early look at, and chance to shape, tools that they will be using in the future.

The Sydney 2014 FDP was conducted by the Bureau of Meteorology as a test bed for improvements to the underlying science for extreme weather forecasting. A centrepiece of the new techniques was a Rapid Update Cycle (RUC) numerical weather prediction (NWP) model, which issued a new forecast every hour, incorporating the latest weather observations much more frequently than traditional NWP models that update every 6 or 12 hours. It was run at 1.5-kilometre horizontal resolution, significantly finer than current operational predictions.

NWP models are crucial to forecasting natural hazards such as dangerous fire weather conditions. They provide input to the Graphical Forecast Editor, which is used to provide gridded and text forecast services including fire weather products.  We will describe how RUC guidance was incorporated into the Graphical Forecast Editor, how forecasters managed the increased volume of information and how they applied the extra spatial resolution of the RUC guidance to spot fire forecasts.

The past decade of research into meteorology surrounding bushfires has many learnings which are yet to be formally incorporated into operational fire weather forecasts and services.  Studies have shown the importance of the structure of the atmosphere in three dimensions and the dynamical interactions between the fire, atmosphere and terrain. Through forecast demonstration projects, researchers and operational practitioners can take up the challenge of transitioning such research outcomes into the operational sphere.