Built for WeatherWise Hack 2026 — Smarter Weather. Safer World.

ClimFire is a real-time forest fire risk intelligence system for Victoria, Australia — one of the most fire-prone regions on Earth.

The 2019–2020 Black Summer fires burned over 5.8 million hectares in Victoria alone, killed billions of animals, and displaced thousands of people. Yet many communities still lack access to reliable, actionable fire risk information.

ClimFire addresses this by combining real-time climate data with a Machine Learning model trained on 10 years of historical weather patterns to predict daily fire risk levels — and estimate the environmental impact on endangered wildlife before a fire event occurs.

Open-Meteo API NASA-grade climate data ↓ Data Processing pandas · feature engineering ↓ ML Model Random Forest Classifier (scikit-learn) ↓ Risk Prediction LOW / MODERATE / HIGH ↓ Flask Backend REST API serving predictions ↓ Web Dashboard Real-time map · Wildlife alerts · Impact calculator

• 7-Day Fire Risk Forecast — daily predictions powered by ML
• Interactive Risk Map — Victoria's protected areas color-coded by risk level
• Wildlife Impact Dashboard — 6 endangered species monitored with real-time alerts
• Environmental Impact Calculator — estimates hectares, trees, CO₂, and animal populations at risk
• Live Data — climate data refreshed on every page load via Open-Meteo API

• Algorithm: Random Forest Classifier
• Training data: Open-Meteo historical archive 2015–2024 (3,653 days)
• Features used:
  • Maximum daily temperature (°C)
  • Total precipitation (mm)
  • Maximum wind speed (km/h)
  • Maximum relative humidity (%)
  • Consecutive dry days (engineered feature)
• Output classes: LOW · MODERATE · HIGH fire risk
• Test accuracy: 99% on 731 unseen days
• Labeling: Based on Bureau of Meteorology (BOM) Fire Weather Index thresholds

git clone https://github.com/EmiCrack132/ClimFire.git
cd ClimFire

python -m venv venv

# Windows
venv\Scripts\activate

# Mac/Linux
source venv/bin/activate

pip install -r requirements.txt

python fase2_datos_historicos.py
python fase2_etiquetar.py
python fase2_modelo.py

python servidor.py

http://127.0.0.1:5000

ClimFire/ ├── static/ │ └── index.html # Frontend dashboard ├── fase1_api.py # Phase 1 — API exploration ├── fase2_datos_historicos.py # Phase 2 — Historical data download ├── fase2_etiquetar.py # Phase 2 — Data labeling ├── fase2_modelo.py # Phase 2 — ML model training ├── servidor.py # Flask backend ├── modelo_incendio.pkl # Trained ML model ├── forecast_victoria.csv # Sample forecast output ├── requirements.txt # Python dependencies ├── Procfile # Render deploy config ├── LICENSE # MIT License └── README.md # This file

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

EmiCrack132 Built with 🔥 for WeatherWise Hack 2026