This model simulates the spread of wildfires in a regrowing forest.
All trees grow over time, and have a small chance of being struck by lightning. If a tree is struck by lightning, or is adjacent to a fire, it sets alight. After burning for one step, the tree is reduced to an 'ember'.
Embers have a small chance of regrowing into a new tree each step, and that chance increases linearly with the count of its adjacent trees.
In this model, we can play with the effects of changing forest density, regrowth rate, and lighting probability in order to observe the health of our 'regrowing' forest.
Consider what metrics we might evaluate to determine the health of our forest: Average tree height and the number of trees in our forest exhibit periodic fluctuations. We could assess the frequency of these fluctuations or their amplitude.
If we define a "wildfire" as a step in the model during which there are more than a certain critical percentage of trees on fire, we can assess the frequency with which they occur. Is it periodic, or does the time between them increase?
See also the unbounded Forest model.