Southern Oscillation Reproduction Project

Reproduction of the landmark 1981 study by John D. Horel and John M. Wallace on "Planetary-Scale Atmospheric Phenomena Associated with the Southern Oscillation" using modern ERA-5 reanalysis data.

📖 About

This project provides a complete, reproducible workflow for analyzing the Southern Oscillation (now known as ENSO - El Niño Southern Oscillation) and its global teleconnections using state-of-the-art climate reanalysis data.

Original Paper: Horel, J. D., and J. M. Wallace, 1981: Planetary-Scale Atmospheric Phenomena Associated with the Southern Oscillation. Monthly Weather Review, 109, 813-829.

✨ Features

• 🌍 Comprehensive Analysis: Reproduces all major figures from the original 1981 paper
• 📊 Modern Data: Uses ERA-5 reanalysis (1950-present) from Copernicus Climate Data Store
• 🔧 MCP Server: Custom server for efficient ERA-5 data access and caching
• 📓 Interactive Notebook: Jupyter notebook with detailed explanations and visualizations
• 🎨 Publication-Quality Figures: All figures saved in high-resolution format
• 🧩 Extensible: Easy to modify time periods, regions, and analysis methods

🚀 Quick Start

Prerequisites

1. Python 3.9 or higher
2. Copernicus Climate Data Store Account
   • Register at https://cds.climate.copernicus.eu
   • Get your API key from https://cds.climate.copernicus.eu/user

Installation

Option 1: Using Conda (Recommended)

# Clone the repository
git clone https://github.com/yourusername/southern-oscillation-demo.git
cd southern-oscillation-demo

# Create conda environment
conda env create -f environment.yml
conda activate southern-oscillation

# Configure CDS API key
echo "url: https://cds.climate.copernicus.eu/api/v2" > ~/.cdsapirc
echo "key: YOUR_UID:YOUR_API_KEY" >> ~/.cdsapirc

Option 2: Using pip

# Clone the repository
git clone https://github.com/yourusername/southern-oscillation-demo.git
cd southern-oscillation-demo

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Configure CDS API key (same as above)

Run the Analysis

# Start Jupyter notebook
jupyter notebook notebooks/southern_oscillation_reproduction.ipynb

Or run from command line:

python -m notebooks.southern_oscillation_reproduction

📁 Repository Structure

southern-oscillation-demo/
├── CLAUDE.MD                          # Detailed project documentation
├── README.md                          # This file
├── LICENSE                            # MIT License
├── environment.yml                    # Conda environment
├── requirements.txt                   # Python dependencies
├── .gitignore                         # Git ignore rules
│
├── mcp_server/                        # MCP server for ERA-5 data
│   ├── __init__.py
│   ├── era5_server.py                # Main server implementation
│   ├── config.json                   # Server configuration
│   └── README.md                     # Server documentation
│
├── notebooks/                         # Jupyter notebooks
│   └── southern_oscillation_reproduction.ipynb
│
├── skills/                            # Custom analysis skills
│   ├── climate_analysis/
│   │   └── SKILL.md                  # Climate analysis methods
│   └── teleconnection_viz/
│       └── SKILL.md                  # Visualization methods
│
├── data/                              # Data directory (gitignored)
│   ├── era5/                         # Downloaded ERA-5 data
│   ├── station/                      # Station data
│   └── processed/                    # Processed datasets
│
├── figures/                           # Generated figures
│   └── reproductions/                # Reproduced figures
│
└── tests/                             # Unit tests
    └── test_analysis.py

🎯 What's Included

Data Analysis

1. EOF Analysis: Empirical Orthogonal Function analysis of Pacific SST
2. Seasonal Averaging: DJF (December-January-February) means computation
3. Normalization: Standardized anomalies for comparison
4. Correlation Analysis: Spatial correlation maps
5. Composite Analysis: Warm vs cold episode composites
6. Teleconnection Indices: PNA, WP, SOI computation

Figures Reproduced

• Figure 1: EOF of Pacific SST and time series
• Figure 2: Normalized SST Index
• Figure 3: Rainfall at equatorial Pacific stations
• Figure 4: Southern Oscillation Index
• Figure 5: Station location map
• Figure 6: 200 mb height time series at stations
• Figure 7: Tropical 200 mb height index
• Figure 8: Composite 200 mb height maps
• Figure 9: Correlation maps (700 mb height)
• Figure 10: Teleconnection pattern indices

🔬 Scientific Background

The Southern Oscillation is a large-scale atmospheric pressure pattern in the tropical Pacific that is coupled with El Niño/La Niña ocean temperature variations. Key features:

• Warm Episodes (El Niño):

  • Above-normal SST in equatorial Pacific
  • Weakening of trade winds
  • Enhanced rainfall in central Pacific
  • Teleconnections to mid-latitudes

• Cold Episodes (La Niña):

  • Below-normal SST in equatorial Pacific
  • Strengthening of trade winds
  • Suppressed rainfall in central Pacific
  • Opposite teleconnection patterns

Historical Significance

The original 1981 paper was groundbreaking because it:

• Documented global-scale atmospheric response to tropical SST
• Established teleconnection patterns for seasonal forecasting
• Provided physical basis for ENSO prediction
• Connected tropical and extratropical variability

📊 Example Results

Key Findings (1951-1978)

• SST EOF1: Explains ~23% of variance in Pacific SST
• SOI Correlation: r ≈ 0.80 with SST Index
• Warm Episodes: 1951-53, 1957-59, 1963-64, 1965-66, 1969-70, 1972-73, 1977-78
• 200 mb Height: Above-normal throughout tropics during warm episodes
• Teleconnections: Strong PNA and WP patterns during El Niño

🛠️ Customization

Extending the Analysis

1. Different Time Period

# In notebook configuration cell
START_YEAR = 1979
END_YEAR = 2023

2. Additional Variables

# Download other ERA-5 variables
await era5_server.download_monthly_means(
    variable='u_component_of_wind',
    pressure_level=200,
    years=YEARS
)

3. Different Region

# Focus on Indian Ocean
DOMAIN = [30, 40, -30, 120]  # [north, west, south, east]

4. Seasonal Selection

# Analyze JJA (June-July-August) instead
SEASON_MONTHS = [6, 7, 8]

🧪 Running Tests

pytest tests/

📚 Documentation

Detailed documentation is available in:

• CLAUDE.MD - Complete project overview and methodology
• mcp_server/README.md - MCP server usage
• skills/climate_analysis/SKILL.md - Analysis methods
• skills/teleconnection_viz/SKILL.md - Visualization techniques

🤝 Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

Areas for contribution:

• Additional validation against original results
• Optimization of data processing
• Enhanced visualizations
• Documentation improvements
• Bug fixes

📖 Citation

If you use this code in your research, please cite:

@article{horel1981southern,
  title={Planetary-Scale Atmospheric Phenomena Associated with the Southern Oscillation},
  author={Horel, John D and Wallace, John M},
  journal={Monthly Weather Review},
  volume={109},
  number={4},
  pages={813--829},
  year={1981}
}

📄 License

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

🙏 Acknowledgments

• Original Authors: John D. Horel and John M. Wallace for their pioneering work
• Copernicus Climate Data Store: For providing ERA-5 reanalysis data
• xarray Community: For excellent climate data tools
• Cartopy Team: For geospatial visualization capabilities

📧 Contact

For questions or issues:

• Open an issue on GitHub
• Email: your.email@example.com

🔗 Related Resources

• Copernicus Climate Data Store
• ERA-5 Documentation
• NOAA ENSO Information
• Climate Prediction Center

---

Note: First data download may take significant time. Subsequent runs will use cached data for much faster execution.