akvo/eswatini-droughtmap-hub-cdi

# Eswatini Drought Map Hub - CDI Automation This repository contains the automation scripts for generating and updating drought-related data for the **Eswatini Drought Map Hub**. The script is designed to run periodically (ideally on a monthly basis) to ensure the data remains up-to-date. ## Table of Contents 1. Overview 2. Prerequisites 3. Setup Instructions 4. Running the Script 5. Adjusting CDI Weights 6. Environment Variables 7. Automation 8. Contributing 9. License --- ## Overview The Eswatini CDI Automation script (`job.sh`) downloads satellite/climate data, runs the Combined Drought Indicator (CDI) algorithm over Eswatini, and uploads the resulting GeoTiff rasters to a GeoNode instance. It is designed to run monthly via cron on a Linux server. The CDI integrates four drought indices into a single percentile-ranked map. All inputs come from the NDMC Regional Percentiles endpoint as pre-ranked GeoTIFFs — **no authentication required**: | Index | Source | Replaces | |-------|--------|----------| | ESI — Evaporative Stress Index | `era5_esi_1mn` | MODIS LST | | EVI2 — Enhanced Vegetation Index | `evi2_1mn` | MODIS NDVI | | SPI 3-month | `chirps_spi_3mn` | Locally computed SPI | | NOAH Soil Moisture | `noah_soilm_1mn` | FLDAS SM | For more information about the CDI methodology, visit the NDMC website. --- ## Prerequisites Before running the script, ensure the following prerequisites are met: 1. **Operating System**: Linux-based system. 2. **System tools**: `wget`, `pup`, `curl` ```bash sudo apt-get install wget curl # pup (HTML parser): github.com ``` 3. **Python virtual environment** with packages from `src/data-processing/cdi-scripts/requirements.txt`. The default location is `~/.myenv`, but you can point at any path via the `PYTHON_VENV` env var (see Environment Variables). If no venv is found, the system `python3` is used as a fallback. ```bash sudo apt-get install -y python3-venv python3-full # Debian/Ubuntu: required for …