Abstract: Spain has deep roots in the underground. The territory of myth and legend, of cave and tunnel, has existed in Spain since that ancient time overwhelmed with shreds of fog, where all was myth around heros like Túbal Hércules, Gárgoris and Abidis. The underground evokes strong links with life and death, light and darkness, and has served as a source of inspiration for art through the centuries. The history of tunnels in Spain reflects the mosaic of cultures that have inhabited Iberia from prehistoric times till today. This contribution on the subterranean History of Spain traces the country’s heritage in the form of natural caves, troglodyte dwellings, mining, crypts, galleries in fortresses and castles, aqueducts, qanats, cellars, and other landmarks.

Abstract: This study refines the hydrogeological conceptual model of an area with three interconnected aquifers, namely the Crystalline Aquifer System (CAS – igneous and metamorphic rocks), which is in contact with the Tubarão Aquifer System (TAS – sedimentary rocks) and the Diabase Aquifer System (DAS – diabase rocks). The detailed investigation involved geophysical logging and hydraulic and hydrodynamic characterization with straddle packers in a local tubular well, in which groundwater presents high uranium concentrations. Hydrogeochemical and isotope (δ2H, δ18O, 3H, δ13C, 14C) analysis in this well and in other three neighboring wells, with lower U concentrations, showed that ancient and modern waters (3H from <0.8 to 1.12 TU, 14C from 69.43 to 78.72 pMC) mix within the aquifer. During groundwater pumping, vertical fractures in the diabase aquifer possibly induce water mixing and recharge of the deeper levels of the aquifers from shallow layers. The high [U] are related to ancient waters from a confined aquifer hosted in CAS that reaches the wells through hydraulically active fractures located deeper than 159 m depth. Groundwater apparent ages do not increase systematically with depth, revealing a complex circulation model for CAS. The results obtained from the other wells, which are all located on drainage lineaments, reveal that one extracts modern water from DAS and TAS, another one extracts modern and ancient water from DAS and CAS, and the third extracts only ancient water from CAS, confirming the complexity of the local hydrogeology. Regarding regional groundwater management, the study revealed the need to characterize the sources of groundwater in each well, in order to protect modern waters from anthropogenic contamination and to protect ancient groundwater from overexploitation, as CAS hosts groundwaters recharged thousands of years ago or more.

Abstract: Predicting groundwater levels is pivotal in curbing overexploitation and ensuring effective water resource governance. However, groundwater level prediction is intricate, driven by dynamic nonlinear factors. To comprehend the dynamic interaction among these drivers, leveraging machine learning models can provide valuable insights. The drastic increase in computational capabilities has catalysed a substantial surge in the utilisation of machine learning-based solutions for effective groundwater management. The performance of these models highly depends on the selection of hyperparameters. The optimisation of hyperparameters is a complex process that often requires application-specific expertise for a skillful prediction. To mitigate the challenge posed by hyperparameter tuning’s problem-specific nature, we present an innovative approach by introducing the automated machine learning (AutoML-GWL) framework. This framework is specifically designed for precise groundwater level mapping. It seamlessly integrates the selection of best machine learning model and adeptly fine-tunes its hyperparameters by using Bayesian optimisation. We used long time series (1997-2018) data of precipitation, temperature, evaporation, soil type, relative humidity, and lag of groundwater level as input features to train the AutoML-GWL model while considering the influence of Land Use Land Cover (LULC) as a contextual factor. Among these input features, the lag of groundwater level emerged as the most relevant input feature. Once the model is trained, it performs well over the unseen data with a strong correlation of coefficient (R = 0.90), low root mean square error (RMSE = 1.22), and minimal bias = 0.23. Further, we compared the performance of the proposed AutoML-GWL with sixteen benchmark algorithms comprising baseline and novel algorithms. The AutoML-GWL outperforms all the benchmark algorithms. Furthermore, the proposed algorithm ranked first in Friedman’s statistical test, confirming its reliability. Moreover, we conducted a spatial distribution and uncertainty analysis for the proposed algorithm. The outcomes of this analysis affirmed that the AutoML-GWL can effectively manage data with spatial variations and demonstrates remarkable stability when faced with small uncertainties in the input parameters. This study holds significant promise in revolutionising groundwater management practices by establishing an automated framework for simulating groundwater levels for sustainable water resource management.