A homogeneous, 40-year (1972-2012), hourly four-kilometre gridded climate dataset for Victoria has been generated using a combination of mesoscale modelling, global re-analysis data, surface observations, and historically observed rainfall analyses. The primary purposes of this dataset are to optimise planned burning and land-management strategies, and scenario planning for major fire events.

Outputs include fire weather and fire danger variables. The output data are created using the Weather Research and Forecast (WRF) model. Error correction techniques are applied to minimise any model biases. Outputs provide an almost limitless opportunity for hitherto unavailable analyses – fields of percentiles of forest fire danger index (FDDI) values, analysis of periods exceeding thresholds at any location, inter-annual and regional variations of fire season characteristics, analysis of prescribed burning windows, of atmospheric dispersion climates, and various atmospheric stability measures that might affect fire behaviour, and to assess climatologies of more esoteric mesoscale weather events, such as mountain waves and wind change structures, that may affect fire behaviour.

Finally, the hourly mesoscale wind fields provide a previously unavailable long-period homogeneous data set with which to drive fire-spread models such as Phoenix. This presentation describes the generation of the dataset, shows examples of output, and highlights use and relevance for fire management.