This project explores historical wildfire patterns in California from 1984 to 2023, using weather and fire occurrence data. Through exploratory data analysis and hypothesis testing, we aim to understand:
- 📈 How wildfire frequency has changed over time
- 🌡️ How temperature, wind speed, and precipitation influence wildfires
- 📊 Which months are most fire-prone
This dataset contains daily records of wildfire and weather conditions, including:
| Feature | Description |
|---|---|
DATE |
Date of observation |
FIRE_START_DAY |
Binary indicator (1 = Fire started, 0 = No fire) |
MAX_TEMP |
Maximum daily temperature (°F) |
AVG_WIND_SPEED |
Average wind speed (mph) |
PRECIPITATION |
Daily precipitation (inches) |
LAGGED_PRECIPITATION |
Previous 7-day precipitation |
LAGGED_AVG_WIND_SPEED |
Previous 7-day wind speed |
MONTH, YEAR, DAY_OF_YEAR |
Temporal indicators for trend analysis |
Source:Yavas, C. E., Kadlec, C., kim, J., & Chen, L. (2025). California Weather and Fire Prediction Dataset (1984–2025) with Engineered Features [Data set]. Zenodo. https://doi.org/10.5281/zenodo.14712845
🔥 Have wildfires become more frequent over the years?
This analysis examines annual and monthly wildfire trends to identify long-term changes in fire frequency.
🔍 What weather conditions drive wildfires?
This section analyzes temperature, wind speed, and precipitation to determine which factors contribute most to wildfire occurrences.
🧪 Testing if high temperatures significantly increase fire risk
We use statistical hypothesis testing to check if hotter days are more likely to have wildfires.
Ensure you have Python 3.8+ and install the required libraries:
pip install pandas numpy matplotlib seaborn scipy statsmodels jupyter- Clone the repository:
git clone https://github.com/leahdsouza/Wildfire-Analysis-in-California.git- Open Jupyter Notebook:
jupyter notebook- Run the
main.ipynbnotebook.
📂 Wildfire-Analysis-In-California
┣ 📜 main.ipynb # Jupyter Notebook with full analysis
┣ 📜 CA_Weather_Fire_Dataset.csv # Dataset (not included, download separately)
┣ 📜 README.md # Project documentation
┣ 📜 requirements.txt # Required Python packages
✅ Machine Learning: Use Random Forest/XGBoost for wildfire risk prediction
✅ Interactive Dashboard: Create Power BI/Tableau visualizations
✅ Geospatial Analysis: Map wildfire-prone zones using GIS tools
✅ Real-Time Fire Monitoring: Integrate live weather data APIs