Fire Farmers | Devpost

Inspiration

With global warming and climate change causing more extreme weather, forest fires are also becoming a globally urgent issue. Additionally, with the recent fires in LA, we were inspired to create a fire obstruction creation model to reduce the impact of future forest fires.

Problem Statement

Clearings are the most common ways to prevent a wildfire from spreading. However, they are often expensive to build and harmful to the environment. We want to demonstrate the best places to clear for the most effective fire prevention with the least amount of land cleared.

What it does

Our model simulates the spread of a fire on an imported rural environment based on wind, topology, and fuel type. Using these results, we generate clearings to best limit the damage caused by the fire.

How we built it

We use a physics based approach to model the effect of wind spread in the fire simulation. We calculate the probability the fire spreads to another piece of land based on the fuel material and the effect of wind. We compile the results of 100 simulations into a burn probability matrix and feed this into our obstacle-creation heuristic to generate our clearings based on the pieces of land most likely to end up burnt.

Challenges we ran into

We ran into issues when trying to increase the complexity of our project. There were several factors like humidity and a more land-efficient edge detection that we tried to implement but were too complicated for the scope of this project.

Accomplishments that we're proud of

We are proud to have a working model with some amazing visualizations. We are also proud to help address a current pressing issue.

What we learned

We learned how to leverage mathematics and physics to model complex processes, and we learned how to create a heuristic to fit our fire-prevention needs

What's next for Fire Farmers

We look to further improve upon this model by adding future complexities to the fire simulation and clearing creation models. Additionally, we would like to add an implementation for different types of environments (urban fires, ground fires, etc.).