Abstract: The Southern High Plains (SHP) is a new and emerging U.S. uranium province. Here, uranyl vanadates form deposits in Pliocene to Pleistocene sandstone, dolomite, and limestone. Fifteen calcrete uranium occurrences are identified; two of these, the Buzzard Draw and Sulfur Springs Draw deposits, have combined in-place resources estimated at about 4 million pounds of U3O8. Ore minerals carnotite and finchite are hosted in dolomite at the Sulfur Springs Draw deposit, with accessory fluorite, celestine, smectite/illite, autunite, and strontium carbonate. Host carbonate at the Sulfur Springs Draw deposit is ∼190 ka and mineralization mobilized as recently as 3.8 ka. Ash collected near the deposit is 631 ka and erupted from the Yellowstone caldera complex. The Triassic Dockum Group that contains sandstone-hosted uranium deposits throughout the region and underlies the SHP is a potential source for uranium and vanadium. Regional uplift and dissection reintroduced oxygenated groundwater into the Dockum Group, mobilizing uranium. Additional uranium may have been contributed to groundwater by weathering of volcanic ash in Pliocene and Pleistocene host rocks. The locations of the uranium occurrences are mostly in modern drainage systems in the southeast portion of the SHP. Modelling of modern groundwater in the SHP carried out in a parallel study shows that a single fluid could form carnotite through evaporation, and that fluids of the requisite composition are more prevalent in the southern portion of the SHP. The southeastern portion of the SHP hosts more uranium occurrences due to a variety of factors including (1) upward transport of groundwater and connectivity between source and host rock, (2) higher uranium and vanadium content of groundwater, (3) higher rates of groundwater recharge in this region to drive the mineralizing system, and (4) shallower groundwater facilitating surface evaporation. Ongoing erosion of host rocks challenges preservation of deposits and may limit their size.

Abstract: This paper reviews critical features of basin-related uranium mineral systems in Australia. These mineral systems include Proterozoic unconformity-related uranium systems formed predominantly from diagenetic fluids expelled from sandstones overlying the unconformity, sandstone-hosted uranium systems formed from the influx of oxidised groundwaters through sandstone aquifers, and calcrete uranium systems formed from oxidised groundwaters flowing through palaeochannel aquifers (sand and calcrete). The review uses the so-called ‘source-pathway-trap’ paradigm to summarise critical features of fertile mineral systems. However, the scheme is expanded to include information on the geological setting, age and relative timing of mineralisation, and preservation of mineral systems. The critical features are also summarised in three separate tables. These features can provide the basis to conduct mineral potential and prospectivity analysis in an area. Such analysis requires identification of mappable signatures of above-mentioned critical features in geological, geophysical and geochemical datasets. The review of fertile basin-related systems shows that these systems require the presence of at least four ingredients: a source of leachable uranium (and vanadium and potassium for calcrete-uranium deposits); suitable hydrological architecture enabling connection between the source and the sink (site of accumulation); physical and chemical sinks or traps; and a post-mineralisation setting favourable for preservation. The review also discusses factors that may control the efficiency of mineral systems, assuming that world-class deposits result from more efficient mineral systems. The review presents a brief discussion of factors which may have controlled the formation of large deposits in the Lake Frome region in South Australia, the Chu-Sarysu and Syrdarya Basins in Kazakhstan and calcrete uranium deposits in the Yilgarn region, Western Australia.

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.