Emma Burgess, Murray Haseler and Martine Maron
Introduction. A study investigating the response of bird assemblages to mosaic burning is being conducted on 60,000 hectares private nature reserve in the Brigalow Belt bioregion of Queensland (Fig 1). The Brigalow Belt has recently experienced high rates of native vegetation clearing, motivating Bush Heritage Australia (BHA) to purchase and protect the property in 2001. The subsequent removal of cattle and horses from Carnarvon Station Reserve has increased grass and herb biomass. The seasonal surge in productivity the property now experiences however, increases the potential for more intense, frequent and extensive fires in hot, dry conditions. The risk of such wildfires needs to be managed, and a common approach to such management is prescribed burning. But how to ensure nature conservation objectives are still met?
In fire ecology, there is a common assumption that if we introduce a range of burn conditions to produce a mosaic of patches with different fire histories (pyrodiversity) – then the resulting diversity in fire histories and the greater representation of successional stages of vegetation is expected to accommodate more species in a given area (Fig. 2). Reducing the spatial scale at which fire history turns over- the “breaking up” of country- is also known as the patch mosaic burning approach.
Whilst we assume that pyrodiversity will give us increased habitat diversity, and therefore greater animal diversity, there is uncertainty as to the scale (alpha, beta or gamma diversity) at which pyrodiversity might influence biodiversity (Fig. 3). Alpha diversity is the total number of different species within a site or habitat; beta diversity is the difference in species composition between sites or habitats; and gamma diversity is the number of different species across all sites or habitats in the area of interest. At what spatial scale do we see the benefit for birds of mosaic burning (Fig. 3)?

Fig 4. Fire-sensitive semi-evergreen vine-thicket extending into Mountain Coolibah (Eucalyptus orgadophila) woodland, Carnarvon Station Reserve
Methods: We examine the relative influence of the diversity of fire histories, spatial configuration of these fire histories, spatial extent of particular fire histories and other measures of environmental heterogeneity on:
- Aggregated measures of bird species richness at both the landscape- (100 ha) and local-scale (1 ha); and
- Response of different bird foraging guilds to mosaic burning, at both the landscape- and local-scale.
So what did we find? The diversity of fire regimes in the 100-ha landscape did not correlate with average site (alpha) or landscape- (gamma) diversity of birds. Rather, the total area of longer-unburnt vegetation was important for increasing bird richness at the landscape-scale, and sites in longer-unburnt vegetation had more species.
Although areas burnt in prescribed burns supported lower bird diversity compared to long-unburnt areas, prescribed burns are still necessary to reduce the risk of extensive wildfire. Such burns should focus on breaking up areas of high fuel at the beginning of the dry season (Fig. 4). The extent of long-unburnt vegetation that can be maintained with careful fire management is yet to be determined, but its importance as bird habitat is clear.
Acknowledgements: This work could not have been completed without funding and logistical support provided by AndyInc Foundation, Bush Heritage Australia and UQRS. Thanks to Peta Mather and Donna Oliver who assisted with field work. This study was carried out with approval from the Animal Ethics Committee at the University of Queensland (approval no. SGPEM/325/11/UQ).

Fig 4. Fire-sensitive semi-evergreen vine-thicket extending into Mountain Coolibah (Eucalyptus orgadophila) woodland, Carnarvon Station Reserve
Contact: Dr Emma Burgess University of Queensland, Email: e.burgess4@uq.edu.au
[This project summary is a precis of a talk presented to the Nature Conservation Council of NSW’s 10th Biennial Bushfire Conference, ‘Fire and Restoration: Working with Fire for Healthy Lands’ 26-27 May 2015. For full paper see: http://www.nature.org.au/healthy-ecosystems/bushfire-program/conferences/%5D