International Journal of Wildland Fire International Journal of Wildland Fire Society
Journal of the International Association of Wildland Fire

Monitoring the impacts of fire regimes on vegetation in northern Australia: an example from Kakadu National Park

Andrew Edwards, Rod Kennett, Owen Price, Jeremy Russell-Smith, Greg Spiers and John Woinarski

International Journal of Wildland Fire 12(4) 427 - 440
Published: 28 November 2003


We describe the rationale, methodology and preliminary results from a major monitoring program in Kakadu National Park, northern Australia. The program aims to assess fire regimes, their impacts upon biodiversity, and the consequences and efficacy of fire management. The program comprises two complementary elements—mapping of fire histories based upon interpretation of satellite imagery, and assessment of vegetation at a large series of permanent monitoring plots. The program commenced formally in 1995, at which time establishment and baseline sampling of vegetation in 134 plots was conducted, with re-sampling proposed at 5-year intervals up to 2010. The monitoring program has an estimated annual cost of about $A140 000 (around 1% of the total annual budget of the Park).

Over the period 1995–2000, the mean annual extent of burning was 40.3%, a marginal reduction in extent from the previous 15 years, particularly for late dry season fires in lowland habitats. From the baseline (1995) and subsequent re-sampling (2000) of the vegetation plots, 963 plant taxa have been recorded. The power of the program to detect change in the frequency or abundance of individual species was poor, especially for ground-layer species, largely because of typically substantial variability in abundance across plots and sampling events, and because of the high proportion of species recorded from few samples. Notwithstanding this constraint, five tree species (of 47 recorded from sufficient samples to test), nine shrub species (from 121) and 27 ground-layer species (from 111) showed significant change in abundance between the baseline and subsequent sampling. However when species were grouped into strata and life-form categories, major changes were evident over this 5 year period, particularly with increases in cover of trees and shrubs. Such changes were related to a range of environmental and fire regime parameters of the plots, with increase in woody cover but reduction in cover and species richness of herbs in those plots experiencing lower frequency of fires.

Keywords: plants; status; statistical power; management.

© IAWF 2003

Export Citation Cited By (35)

View Altmetrics