MelissaMatindi/ndvi-anomaly-analysis

🌍 A decade-long deep dive into Ethiopia’s vegetation health using NDVI anomalies in Google Earth Engine. Tracks drought patterns, visualizes climate trends, and leverages Python, ML & GIS to support SDGs 2, 13 & 15. 📊🌾 # 🌍 Drought Monitoring Using NDVI Anomalies in Google Earth Engine This project leverages **Google Earth Engine (GEE)** and **Python** to analyze NDVI anomalies across Ethiopia (2015–2024) to detect drought-prone regions and support Sustainable Development Goals (**SDGs 2, 13, 15**). It was developed as part of the **UNDP-sponsored FTL Cybersecurity Training**, blending **GIS**, **Machine Learning**, and **Cybersecurity** principles. --- ## 🎯 Objectives - Monitor drought severity using NDVI anomaly maps and time-series graphs - Apply ML (K-Means) for drought classification - Ensure data security with OAuth and optional encryption --- ## 🔧 Tools & Technologies - **Python**, **Google Earth Engine**, **geemap**, **matplotlib**, **GDAL**, **GeoPandas** - **Landsat 8 Surface Reflectance** data - **scikit-learn**, **cryptography**, **Jupyter Notebooks** --- ## 📈 Outputs - NDVI anomaly maps (2021–2024) - Annual NDVI trend chart - ML-classified drought severity map - Exported **GeoTIFFs** & analytical report --- ## 🔐 Security - OAuth-based GEE access - Optional encryption using `cryptography.Fernet` for data protection --- ## 🤖 Optional ML Implementation - **K-Means clustering** for anomaly classification (severe, moderate, none) --- ## 🧠 Skills Demonstrated Geospatial Data Analysis | Remote Sensing | ML in GIS | Visualization | Cybersecurity | Python scripting | SDG-aligned decision support