Firewatch

A wildfire prediction and awareness platform inspired by the beautiful game Firewatch by Campo Santo.

Comment

UN Sustainable Development Goals This project directly supports:

Goal 13: Climate Action Goal 15: Life on Land Goal 11: Sustainable Cities and Communities Features Location-based Search - Search for wildfire predictions by location Interactive Maps - Visualize predicted fire zones with Leaflet maps ML Predictions - Machine learning model trained on real wildfire data Risk Assessment - Severity and confidence scoring for each prediction AI Chatbot - Gemini AI-powered chatbot for environmental education and wildfire information Beautiful UI - Inspired by Firewatch's aesthetic Tech Stack Frontend React Vite Leaflet CSS3

Backend Python Flask Scikit-learn Pandas NumPy

AI & Machine Learning Google Gemini

Quick Start Prerequisites Node.js (v16 or higher) Python (v3.8 or higher) Git Installation Clone the repository

git clone https://github.com/yourusername/Firewatch.git cd Firewatch Set up the Backend

Install Python dependencies

pip install -r requirements.txt

Start the Flask server

cd backend python run.py The backend will run on http://localhost:5001

Set up the Frontend

In a new terminal, navigate to frontend

cd frontend

Install dependencies

npm install

Start the development server

npm run dev The frontend will run on http://localhost:5173

Open your browser Navigate to http://localhost:5173 and start exploring!

How to Use Search for a location - Try searching for "Vancouver", "Kamloops", or "Prince George" View predictions - The map will display predicted fire zones with severity levels Explore details - Click on markers to see confidence scores and severity Learn about risks - Each prediction includes severity assessment and confidence metrics Chat with Henry - Use the Gemini AI chatbot to ask questions about wildfires, environmental safety, and climate action AI Chatbot The platform features a Gemini AI-powered chatbot that provides:

Environmental Education - Learn about climate change, wildfires, and environmental safety Wildfire Information - Get answers about fire prevention, safety tips, and emergency procedures Interactive Learning - Ask questions about sustainable development goals and environmental awareness Real-time Assistance - Get instant responses to your environmental and safety questions Machine Learning The project uses a Random Forest Classifier trained on real wildfire data from VIIRS (Visible Infrared Imaging Radiometer Suite) satellites. The model predicts wildfire probability based on:

Geographic coordinates Historical fire patterns Environmental factors Seasonal data Design Inspiration The visual design and atmosphere are inspired by Firewatch, the critically acclaimed indie game by Campo Santo. The game's beautiful art style and environmental storytelling served as the primary inspiration for this project's aesthetic.

Background imagery credit: Firewatch by Campo Santo

License This project is licensed under the MIT License - see the LICENSE file for details.

Credit Campo Santo for the beautiful Firewatch game that inspired this project NASA for the VIIRS wildfire data UN Sustainable Development Goals for the environmental mission Leaflet for map tiles

Share this project:

Updates