Scott Goodrick

Bio [2007]
I live in Athens, Georgia where I work as a research meteorologist with the United States Forest Service studying forest fire-atmosphere interactions, smoke dispersion, and how climate influences the occurence of forest fires. Much of my work centers on numerical modeling of these processes. I received my doctorate and master's in atmospheric science from the University of Alabama, Huntsville, and my bachelor's degree in Marine Biology from Auburn Unviersity. I became interested in Stephen Wolfram's New Kind of Science when I was given the book as a gift. As I read the book, I continually recognized patterns that were strikingly similar to patterns I have observed in my work.

Project Title
Process-Based Cellular Automaton Model of Forest Fire Growth

Project
Forest fires are very complex phenomena controlled by processes spanning a wide range of spatial and temporal scales. The majority of models currently being used in the United States to predict the spread of forest fires fall into two classes. The first class relies upon simple empirical relationships that are computationally simple but unable to reproduce the complexity of behaviors observed in nature. The second class of models treat fire as a chemically reactive fluid flow using the full three dimensional Navier-Stokes equations. While these models are capable of modelling complex fire-atmosphere interactions, they are also so computationally expensive that they are useless as predictive tools for real world events.

This project seeks to model forest fires based on the physical processes of radiative and convective heat transfer. The model is based on a coupled set of cellular automata (CA), one representing the distribution of temperature in the model domain, and the other representing the changes in the wind field forced by the temperature changes. A simple representation of combustion releases heat which the temperature CA must redistribute through radiation or convection. Radiative heat transport is accomplished via diffusion, while a separate CA develops thermally induced winds to handle the convective heat transport. While cellualr automatons have been used previously to simulate forest fire spread, they have either used empirical fire spread relationships in determining their rules or have tended to rely upon a probabalistic rule set to achieve complexity. This is a first attempt at modelling fire spread with a cellular automaton as a coupled system that links heat released by the fire to an atmospheric response.

Favorite Outer Totalistic 3-Color Rule
Rule chosen: 3292834

The 3-color outer totalistic CA that I settled on is rule 3292834. My reason for selecting this one is that in the searches I conducted with a series of random initial conditions, this rule did the best job of producing a time series that correlated reasonably well with a time series of sea surface temperatures that serves as an index for the El Nino Southern Oscillation (ENSO).