Unveiling hidden dynamics: fine-scale mapping of groundwater-dependent ecosystems using multi-source Earth observations

Groundwater-dependent ecosystems (GDEs) provide valuable ecosystem services but are threatened by climate change and increased human groundwater use. Sustainable groundwater management requires knowledge of the spatial distribution of GDEs, but this remains a challenge. Here we developed an effective workflow for mapping GDEs and their dependence on groundwater at 30 m resolution over the Hexi Corridor, northwestern China, by integrating multi-source Earth observations. The method integrates seasonal Landsat NDVI and NDWI indices with an unsupervised ISODATA clustering algorithm and groundwater depth datasets to classify GDEs. The accuracy of the mapping results was validated by multiple measurements. Results show that 34.23% of the study area is GDE, mainly in shallow water table and arid regions, making them vulnerable to exploitation. We also found that the rate of decline of the GRACE terrestrial water storage anomaly (TWSA), a satellite proxy for groundwater dynamics, was faster in areas with high GDE occurrence, suggesting that urgent groundwater management actions are needed to avoid the potential loss of valuable services provided by these ecosystems to the human society. While demonstrated in northwestern China, the workflow is adaptable to other global regions, enabling consistent and dynamic monitoring of GDEs. The GDE map has been uploaded to the data repository for free public access: (http://datadryad.org/stash/share/mODbi5SlXI_oeV1E_rWs5SD7eZEz-HfDKgB8JxET-ZA).