Editors’ Highlights are summaries of recent papers by AGU’s journal editors.
Source: Water Resources Research
Shallow groundwater dynamics are reflected in spatiotemporal variations in groundwater table depth. Monitoring these variations over large areas is challenging. As a result, available data are often sparse and fragmented, posing difficulties for water managers.
Cunha Teixeira et al. [2025] address this challenge by using seismic wave information from passing trains, recorded with a geophone sensor array, along with measurements from two piezometers, to estimate groundwater table depth at a daily time scale using a multilayer perceptron. Their method enables spatially continuous monitoring while requiring a minimal amount of invasive borehole data.
Building on their approach, a degree of generalization is achieved, allowing extrapolation beyond the training area. This innovative approach potentially paves the way for cost-effective, high-resolution groundwater monitoring, providing water managers with the data they need to make informed decisions on a regional scale.
Citation: Cunha Teixeira, J., Bodet, L., Rivière, A., Hallier, A., Gesret, A., Dangeard, M., et al. (2025). Physics-guided deep learning model for daily groundwater table maps estimation using passive surface-wave dispersion. Water Resources Research, 61, e2024WR037706. https://doi.org/10.1029/2024WR037706
—Stefan Kollet, Editor, Water Resources Research