Jacopo Bizzotto, Stefano Bolatto, Stefania Minardi

Computer Science and Simulation for Economics

Project work on

"A Fire in Yellowstone Park"

A study about the frequency and the dimension of the phenomena of nature


The applet requires Java 1.4.1 or higher to run. It will not run on Windows 95 or Mac OS 8 or 9. Mac users must have OS X 10.2.6 or higher and use a browser that supports Java 1.4 applets (Safari works, IE does not). On other operating systems, you may obtain the latest Java plugin from Sun's Java site.


created with NetLogo

view/download model file: yellowstonemodel.nlogo


This project simulates the dynamic of fires in a forest and it's inspired by
"Fire", a model present in the Netlogo Models Liibrary and made by U.Wilensky (1998).
This model doesn’t take into account factors such as the characteristics of the trees, the atmospheric conditions or the existence of barriers, natural or artificial.
Our purpose is to create a simulation that, even without any specification, shows the same kind of relation between dimension and frequency of fires as observed in nature.
In our model the fires spread through the forest, burning the patches where they can find a tree. The fires have a certain chance to extend at the adjacent wooded areas (patches), but they can not skip empty patches.
The final dimension of a fire depends on the density of the trees and on placement of the initial patch.

Click the setup button to set up the trees (green).
Click the go button to start the simulation.
With the slider INITIAL-DENSITY you can choose the density of the forest at the beginning of the simulation.
The slider REAFFORESTATION controls the frequency at which new trees are planted on the unwooded areas at every period.
You can set the risk of having a new fire on a green patch with RISK-OF-FIRE, and the chances of expansion of the fire with PROPAGATION-PROBABILITY.

In our model you can not have two fires at the same time. What would happen if different fires were allowed to take place at the same time?
With the chooser TIME-MANAGEMENT you can decide between two different ways of dealing with the creation of new fires. The option ONLY-ONE doesn’t allow a fire to be created, before the last one has gone out. The option SIMULTANEOUS allows different fires to burn at the same time.
By choosing the second option, you can control the number of periods in which a new fire can start with CYCLES-AT-RISK out of the total number of periods set by the slider NUMBER-OF-CYCLES.

The grey border around the edge of the screen prevents the fire from spreading beyond the borders of the screen.
The simulation has also a graph where the dimension of fires is plotted, so that you can immediately check the result of changing any of the initial setting.

Specifically the sliders PROPAGATION-PROBABILITY and REAFFORESTATION have been set in order to allow you to try the effect of different policies.
Will a fire prevention policy, that lowers the chances of expansion, have as a consequence a lower average dimension of the fires?

It could be interesting to add some of these factors above-mentioned, such as the characteristics of the trees, the atmospheric conditions or the existence of natural or artificial barriers.

Trees are represented by green patches; burned areas and areas where no trees have been planted are black.
The turtles are the matches that let the fire start.
Two breeds of turtles are used, "matches" and "embers". When a match-turtle (red) is created a tree catches fire, the adjacent trees have a certain chance (set by PROPAGATION-PROBABILITY) to create a new match-turtle. The matches turn into embers after a period.
Notice how the program gradually darkens the color of embers to achieve the visual effect of burning out.
The border patches are grey to show that they can not burn for any reason.
The data for the graph (dimension of fires) are collected at the end of every period, if you choose the ONLY-ONE option.
Otherwise, the program will check the burned area at the end of every NUMBER-OF-CYCLES. You should choose a NUMBER-OF-CYCLES larger than CYCLES-AT-RISK, in order to get a sufficient number of periods devoted to the development of fires, and to avoid the risk of having fires still burning when the program measures the burned area.