denizo254/flood_drought_risk_kenya

This project develops a spatial Flood–Drought Risk Proxy and Displacement Risk Index for Kenya, integrating climate hazards with population exposure to identify potential displacement hotspots. The framework is scalable to IGAD countries and designed to support anticipatory action, early warning systems, and humanitarian planning. # Flood–Drought Risk Kenya A production-ready spatial data science pipeline for computing a **Composite Flood–Drought Displacement Risk Index** at Kenya county level, scalable to the IGAD region. --- ## Architecture ``` flood_drought_risk_kenya/ ├── src/ │ ├── ingestion/ # Data download & SHA-256 checksum verification │ │ ├── checksum.py │ │ ├── download_rainfall.py │ │ └── download_boundaries.py │ ├── preprocessing/ # Raster clipping & alignment │ │ ├── clip_raster.py │ │ └── align_layers.py │ ├── indicators/ # Hazard & exposure computations │ │ ├── rainfall_anomaly.py │ │ ├── ndvi_deviation.py │ │ ├── population_density.py │ │ ├── slope.py │ │ └── risk_index.py ← Composite Risk Index │ └── visualization/ │ └── risk_map.py ├── data/ │ ├── raw/ # Source data (git-ignored, populated by pipeline) │ ├── processed/ # Clipped rasters (git-ignored) │ └── checksums/ # SHA-256 manifests (versioned) ├── outputs/ │ ├── maps/ # PNG risk maps │ └── reports/ # CSV county risk reports ├── config.yaml # All paths, column names, weights — no hardcoding ├── pipeline.py # Single-entry orchestrator └── requirements.txt # Pinned dependencies ``` --- ## Quick Start ```bash # 1. Clone and create virtual environment git clone cd flood_drought_risk_kenya python -m venv venv # Windows: venv\Scripts\activate # macOS/Linux: source venv/bin/activate # 2. Install pinned dependencies pip install -r requirements.txt # 3. Place data files in data/raw/ # Required: # kenya_counties.geojson (auto-downloaded if absent) # rainfall_dec_2025.tif.gz (auto-downloaded from CHIRPS if absent) # # Optional — add for full index (pipeline degrades gracefully without them): # ndvi_current.tif (MODIS MOD13A3 current month) # ndvi_baseline.tif (MODIS long-term mean) # p …