Fire Management of Woody Vegetation in the Gulf Region

Summary

Researchers and extension officers from CSIRO and Queensland Department of Primary Industries and Fisheries worked directly with pastoralists from the northern Gulf Savannas in an area stretching from Mt Surprise through to Karumba. The approach was to establish a small number of "core" sites and a larger number of "satellite" sites and to impose prescribed fires at these sites. Changes in tree and shrub populations on burned areas were compared with those on nearby unburned plots. 
The research was originally planned for the period 2003-2006. However, low fuel loads, as a result of drier than average conditions during the 2003-2004 and 2004-2005 wet seasons, meant that fires in the subsequent dry seasons were of low intensity. The effect of these dry years on the prospects for burning and on the intensity of fires that were conducted, were exacerbated at several sites by the fact that the sites were stocked, in accordance with the needs of the various co-operating land-holders. For this reason, further resources were obtained in order to take advantage of the relatively wet season of 2005-2006. Where fuel loads permitted sites were burned each year between 2003 and 2005. 
Decisions about whether to burn were made following discussions between researchers and land-holders. Effects were quantified at five core sites and photographic records made at twelve satellite sites. All five core sites and seven out of twelve satellite sites were burned during the initial experimental period (2003-2006). At unburned sites, fuel loads were inadequate to carry effective fires during this period. Most sites were burned only once during 2003-2006, two sites were burned twice, and three were burned three times. One site had been burned immediately prior to the beginning of the study in 2002. In 2006, four core sites were burned. The study found that: - Fuel loads during the study period were generally below 2000 kg/ha. - Single fires resulted in increased mortality of a range of tree and shrub species, including two species considered increasers of greatest concern, gutta percha (Excoecaria parvifolia) and breadfruit (Gardenia vilhelmii). 
Mortality rates due to burning were up to three times those recorded in unburned plots though they were highly variable between sites and species. - Shrubs at the few sites that were subject to more than one fire experienced higher mortality than sites subject to a single fire. - A large proportion of trees and shrubs that were subject to fire sprouted from the base after the fire. - Fires reduced the average size of living trees and shrubs. - There was a widespread acceptance of the principle of using fire to manage tree and shrub populations but a significant challenge in applying this principle is to incorporate into property management a capacity to manage fuel grass loads.