Abstract: The crucial role which groundwater resource plays in our environment and the overall well-being of all living things can not be underestimated. Nonetheless, mismanagement of resources, over-exploitation, inadequate supply of surface water and pollution have led to severe drought and an overall drop in groundwater resources’ levels over the past decades. To address this, a groundwater flow model and several mathematical data-driven models have been developed for forecasting groundwater levels. However, there is a problem of unavailability and scarcity of the on-site input data needed by the data-driven models to forecast the groundwater level. Furthermore, as a result of the dynamics and stochastic characteristics of groundwater, there is a need for an appropriate, accurate and reliable forecasting model to solve these challenges. Over the years, the broad application of Machine Learning (ML) and Artificial Intelligence (AI) models are gaining attraction as an alternative solution for forecasting groundwater levels. Against this background, this article provides an overview of forecasting methods for predicting groundwater levels. Also, this article uses ML models such as Regressions Models, Deep Auto-Regressive models, and Nonlinear Autoregressive Neural Networks with External Input (NARX) to forecast groundwater levels using the groundwater region 10 at Karst belt in South Africa as a case study. This was done using Python Mx. Version 1.9.1., and MATLAB R2022a machine learning toolboxes. Moreover, the Coefficient of Determination (R2);, Root Mean Square Error (RMSE), Mutual Information gain, Mean Absolute Percentage Error (MAPE), Mean Squared Error (MSE), Mean Absolute Error (MAE), and the Mean Absolute Scaled Error (MASE)) models were the forecasting statistical performance metrics used to assess the predictive performance of these models. The results obtained showed that NARX and Support Vector Machine (SVM) have higher performance metrics and accuracy compared to other models used in this study.

Abstract: Argon-37 (37Ar) and Argon-39 (39Ar) are used for groundwater dating on timescales from weeks to centuries. For both isotopes, the quantification of underground sources is essential to accurately infer water residence times from sampled dissolved activities. Subsurface production resulting from interactions with neutrons from the natural radioactivity in rocks and with primary cosmogenic neutrons has been known for a long time. More recently, the capture of slow negative muons and reactions with muon-induced neutrons were documented for 39Ar subsurface production in the context of underground particle detectors (e.g. for Dark Matter research). However, the contribution from these particles was never considered for groundwater dating applications. Here, we reevaluate the importance of all potential depth-related production channels at depth ranges relevant for 39Ar groundwater dating [0 − 200 meters below the surface (m.b.s)]. The production of radioargon by muon-induced processes is considered in this depth range for the first time. The uncertainty on the total depth-dependent production rate is estimated with Monte Carlo simulations assuming a uniform distribution of the parameter uncertainties. This work aims to provide a comprehensive framework for interpreting 39Ar activities in terms of groundwater residence times and for exposure age dating of rocks. The production of 37Ar is also addressed since this isotope is relevant as a proxy for 39Ar production, for the timing of river-groundwater exchanges, and in the context of on-site inspections (OSI) within the verification framework of the Comprehensive Nuclear-Test-Ban Treaty (CTBT). In this perspective, we provide an interactive web-based application for the calculation of 37Ar and 39Ar production rates in rocks.

Abstract: Uranium (U) roll-front deposits constitute a valuable source for an economical extraction by in situ recovery (ISR) mining. Such technology may induce changes in the subsurface microbiota, raising questions about the way their activities could build a functional ecosystem in such extreme environments (i.e.: oligotrophy and high SO4 concentration and salinity). Additionally, more information is needed to dissipate the doubts about the microbial role in the genesis of such U orebodies. A U roll-front deposit hosted in an aquifer driven system (in Zoovch Ovoo, Mongolia), intended for mining by acid ISR, was previously explored and showed to be governed by a complex bacterial diversity, linked to the redox zonation and the geochemical conditions. Here for the first time, transcriptional activities of microorganisms living in such U ore deposits are determined and their metabolic capabilities allocated in the three redox-inherited compartments, naturally defined by the roll-front system. Several genes encoding for crucial metabolic pathways demonstrated a strong biological role controlling the subsurface cycling of many elements including nitrate, sulfate, metals and radionuclides (e.g.: uranium), through oxidation-reduction reactions. Interestingly, the discovered transcriptional behaviour gives important insights into the good microbial adaptation to the geochemical conditions and their active contribution to the stabilization of the U ore deposits. Overall, evidences on the importance of these microbial metabolic activities in the aquifer system are discussed that may clarify the doubts on the microbial role in the genesis of low-temperature U roll-front deposits, along the Zoovch Ovoo mine.