Fuel reduction burning is often patchy as a result of fuel and climatic conditions and inherent landscape-related features such as topography and soil moisture, with a strong sampling design required to capture this variation. With bushfires becoming more intense the larger they become, affecting soils and vegetation more. It is unknown if the same situation arises with prescribed burning.

The relationships between burn size and soil, water, vegetation and fuel outcomes has yet to be quantified. The ability to predict the effects of prescribed burns of different size across landscapes is currently negligible.

To design an a priori sampling scheme of prescribed burns with appropriate statistical power, it is important to define what a ‘small’ fire is compared to a ‘big’ fire. Logically, larger fires will need to be sampled at a different scale and frequency than smaller fires.

To determine historical fire size, data relating to fire size, location and timing for the last 10 years will be used from New South Wales, Victoria, South Australia, Western Australia and Tasmania. Patterns in fire size and timing that will provide valuable information for the project’s sampling design are emerging.

Sampling has been undertaken in sites in mixed-species eucalypt forest in southern Victoria, the ACT and western and southern NSW. This sampling will determine the effect of prescribed fire on changes in fuel load, carbon pools and tree water use. The sampling scheme investigates ‘burn units’ – pairs of sites that have been measured and compared. The pair of sites can be burnt and unburnt areas near each other, sampled at the same time, or are a single site sampled at different times before and after prescribed burning. Nearly 50 burn units have been sampled across south eastern Australia. The data collected has been used to test if environmental variability is adequately captured for measurements made at different spatial scales and if fire size affects the optimal number of samples required for characterising burnt and unburnt areas.

The predictive model developed by this project will quantify the optimisation of environmental service outcomes for water and carbon management against the effectiveness of the fuel reductions outputs. This will assist fire and land management agencies by giving them greater confidence in forecasting results for their actions.

Ultimately, this project will move research and management capabilities to its next logical focus – building a predictive model and framework for planning of prescribed burns.

This will help predict the effects of fuel reduction burning on fuel loads, broad vegetation types and carbon and water potential (for example, capacity for carbon sequestration, water yield) of forests at a manageable spatial scale.