Advancing AI-based pan-European groundwater monitoring

Abstract

The main challenge of pan-European groundwater (GW) monitoring is the sparsity of collated water table depth ( wtd ) observations. The wtd anomaly ( wtd _a ) is a measure of the increased wtd due to droughts. Combining long short-term memory (LSTM) networks and transfer learning (TL), we propose an AI-based methodology LSTM-TL to produce reliable wtd _a estimates at the European scale in the absence of consistent wtd observational data sets. The core idea of LSTM-TL is to transfer the modeled relationship between wtd _a and input hydrometeorological forcings to the observation-based estimation, in order to provide reliable wtd _a estimates for regions with no or sparse wtd observations. With substantially reduced computational cost compared to physically-based numerical models, LSTM-TL obtained wtd _a estimates in good agreement with in-situ wtd _a measurements from 2569 European GW monitoring wells, showing r ⩾ 0.5, root-mean-square error ⩽1.0 and Kling-Gupta efficiency ⩾0.3 at about or more than half of the pixels. Based on the reconstructed long-term European monthly wtd _a data from the early 1980s to the near present, we provide the first estimate of seasonal wtd _a trends in different European regions, that is, significant drying trends in central and eastern Europe, which facilitates the understanding of historical GW dynamics in Europe. The success of LSTM-TL in estimating wtd _a also highlights the advantage of combining AI techniques with knowledge contained in physically-based numerical models in hydrological studies.

Keywords

• groundwater
• water table depth anomaly
• artificial intelligence (AI)
• long short-term memory (LSTM) network
• transfer learning
• LSTM-TL