Geospatial mapping of global groundwater salinity using a data science approach

Water salinity and the salinisation of freshwater due to land-use practices and rising sea levels are global challenges. These issues threaten human health, water and food security, and ecosystems, especially in low-lying coastal regions and deltas. However, limited observation data on groundwater salinity hinders effective global assessment. High-resolution mapping that integrates diverse datasets is critical for sustainable water management in the context of climate change and human development.

This research aims to develop a global-scale geospatial mapping framework for groundwater salinity using machine learning (ML) and artificial intelligence (AI). Sparse in-situ salinity data will be combined with global datasets from Earth Observation (EO) satellites and global hydrological models. By addressing observation data gaps and employing proxy datasets, e.g., rainfall, temperature, the framework will provide new insights into global salinisation patterns.

While maintaining a global focus, regional case studies, such as the Asian mega-deltas, will demonstrate the framework’s applicability across diverse contexts. This research will enhance methodologies for salinity mapping, inform climate adaptation strategies, and support the United Nations Sustainable Development Goal 6 (Clean Water and Sanitation). The outcomes will contribute to global efforts in managing freshwater resources and addressing salinity risks in a changing climate.