This project aims to leverage geospatial (GIS) and climatic data to develop a tool based on machine learning modeling to assess the static depth to water (DTW) in crystalline basement aquifers of Africa. Accurate water table depth predictions are crucial for sustainable water resource management, especially in regions dependent on rainfed agriculture.
The model is trained using one-time measurements of water table depth in the following countries: Benin, Burkina Faso, Guinea, Mali, Niger, Togo, TChad
The training data includes:
- Climatic variables (daily precipitation and NDVI).
- Topographic and hydrological features derived from Digital Elevation Models (DEMs) and GIS analysis.
This project was developed by Les solutions géostack, Inc. as part of a research initiative for The World Bank Group.
For inquiries, contact: info@geostack.ca
Repository: https://github.com/geo-stack/hydrodepthml License: MIT