- Title:
-
Power-law Behaviour of Wildland Fires Reflects
- Date:
- March 2005
- Organisations
- European Geosciences Union
- Authors:
Matthias M. Boer, Pauline F. Grierson and A. Malcolm Gill
- Location:
- UK, United Kingdom
Overview
Power-law frequency-area statistics over several orders of magnitude have been reported for wildland fires in the Mediterranean, the USA, Australia and China. The scale-invariance of real fires has been interpreted as a signature for these landscape systems being in a self-organised critical (SOC) state because the empirical powerlaws resembled those of fires simulated by simple cellular automata models with well-documented SOC characteristics. However, fire behaviour in forest landscapes in Australia suggests that fires in modelled landscapes are driven by a different set of processes than fires in real landscapes. This implies that the power-law behaviour of modelled and observed fires might coincide for reasons other than SOC. An alternative explanation for power-law frequency-area distributions of real fires is sought in the temporal scale-free behaviour of fire-driving weather. Preliminary analyses of time-series of the McArthur’s Forest Fire Danger Index (FFDI), calculated from halfhourly observations at automatic weather stations in Victoria and south-west Western Australia, suggests that scale invariance of fire weather may be a rather common phenomenon and a plausible explanation for the power-law behaviour of real fires. The relationship between time-integrated FFDI and fire size is currently being studied in greater depth using remotely-sensed data and fire behaviour models to provide further support for this hypothesis.
