J. van den Berg,R. Brouwer
Abstract: We consider a forest-fire model which, somewhat informally, is described as follows: Each site (vertex) of the square lattice is either vacant or occupied by a tree.Vacant sites become occupied at rate 1. Further, each site is hit by lightningat rate lambda. This lightning instantaneously destroys (makes vacant) the occupied cluster of the site. This model is closely related to the Drossel-Schwabl forest-fire model, which has received much attention in the physics literature. The most interesting behaviour seems to occur when the lightning rate goes to zero. In the physics literature it is believed that then the system has so-called self-organized critical behaviour.
What problem does this paper attempt to address?
The main problem that this paper attempts to address is the behavior of the forest-fire model as the lightning rate approaches zero. Specifically, the authors focus on whether the system exhibits so-called self-organized critical behavior when the lightning rate λ is very small. In the physics literature, it is generally believed that as the lightning rate approaches zero, the system will exhibit characteristics similar to those of ordinary statistical mechanics systems near the critical point, particularly that the cluster size distribution will follow a power-law behavior.
To investigate this issue, the authors start from a state where all lattice points are empty and study the behavior near the "critical time" tc. Here, tc is defined by the relation 1 - exp(-tc) = pc, where pc is the critical probability for site percolation. The authors explore whether, for a fixed time t > tc, the probability that a tree at a distance less than m from the origin O will be burned before time t approaches 1 as λ approaches 0. However, through a percolation-like hypothesis (which they cannot prove but believe to be correct), the authors show that this intuitive expectation is actually incorrect.
Furthermore, the authors propose several natural open questions and hope that their results will provide a foundation for further research and clarification of these issues. For example, they pose two main open questions:
1. For all t > tc, does there exist
\[
\limsup_{\lambda \downarrow 0} \limsup_{n \to \infty} P_\lambda(\eta^n \text{ experiences a fire at O before time t}) > 0?
\]
2. For all t > tc and each ε > 0, does there exist an m such that
\[
\limsup_{\lambda \downarrow 0} \limsup_{n \to \infty} P_\lambda(\eta^n \text{ experiences a fire within B(m) before time t}) > 1 - \epsilon?
\]
Solving these questions is crucial for understanding the self-organized critical behavior of the forest-fire model.