Geohydrology of the four largest spring systems in the Ozarks of Missouri and Arkansas, USA

ABSTRACT The four largest spring systems in the mid-continent receive recharge through large interconnected voids in fractured and solution-weathered dolostones of the Ordovician and Cambrian systems. Cumulative thickness of the carbonate bedrock aquifer ranges up to 700 m in the Ozark region. Recharge from the surface occurs through weathered overburden, sinkholes, and losing streams and has been traced up to 60 km (straight-line horizontal distance) using fluorescent dyes. Mean discharge of the combined flow of these four spring systems is ~1400 cubic feet/second (ft3/s) or 40 m3/second (m3/s). All four spring systems will be visited while discussing the karst terrane that recharges them. Environmental and engineering challenges in the region will be discussed, such as wastewater treatment systems, solid waste disposal, and failed reservoirs. Hodgson Mill Spring represents a branch of the Rainbow/North Fork/Hodgson Mill System. While it receives base flow from the main system, it also receives local recharge that Rainbow and North Fork springs do not. A portion of the Mammoth Spring recharge system will be viewed at Grand Gulf State Park in Missouri, where a cave collapse has created cliffs and a natural bridge and exposed a small losing tributary that flows into a cave that has been traced to the spring. Mammoth Spring State Park in Arkansas offers a historical perspective of the development and use of large springs. Greer Spring in Missouri was used as a power source for grist, flour, and lumber mills, but has now largely returned to its predevelopment state and is managed by the U.S. Forest Service. Big Spring, featured in a former state park in Missouri, is now part of the Ozark National Scenic Riverways.

Occurrence, Geochemistry and Speciation of Elevated Arsenic Concentrations in a Fractured Bedrock Aquifer System

The presence of elevated arsenic concentrations (≥ 10 µg L−1) in groundwaters has been widely reported in areas of South-East Asia with recent studies showing its detection in fractured bedrock aquifers is occurring mainly in regions of north-eastern USA. However, data within Europe remain limited; therefore, the objective of this work was to understand the geochemical mobilisation mechanism of arsenic in this geologic setting using a study site in Ireland as a case study. Physicochemical (pH, Eh, d-O2), trace metals, major ion and arsenic speciation samples were collected and analysed using a variety of field and laboratory-based techniques and evaluated using statistical analysis. Groundwaters containing elevated dissolved arsenic concentrations (up to 73.95 µg L−1) were characterised as oxic-alkali groundwaters with the co-occurrence of other oxyanions (including Mo, Se, Sb and U), low dissolved concentrations of Fe and Mn, and low Na/Ca ratios indicated that arsenic was mobilised through alkali desorption of Fe oxyhydroxides. Arsenic speciation using a solid-phase extraction methodology (n = 20) showed that the dominant species of arsenic was arsenate, with pH being a major controlling factor. The expected source of arsenic is sulphide minerals within fractures of the bedrock aquifer with transportation of arsenic and other oxyanion forming elements facilitated by secondary Fe mineral phases. However, the presence of methylarsenical compounds detected in groundwaters illustrates that microbially mediated mobilisation processes may also be (co)-occurring. This study gives insight into the geochemistry of arsenic mobilisation that can be used to further guide research needs in this area for the protection of groundwater resources.

Occurrence, Geochemistry and Speciation of Elevated Arsenic Concentrations in a Fractured Bedrock Aquifer System

The presence of elevated arsenic concentrations (≥10 µg L-1) in groundwaters has been widely reported in areas of south east Asia with recent studies showing its detection in fractured bedrock aquifers mainly in regions of north-eastern United States. Data within Europe remains limited; therefore, the objective of this work was to understand the geochemical mobilisation mechanism of arsenic in this geologic setting. Physiochemical (pH, Eh, d-O2), trace metals, major ion and arsenic speciation samples were collected and analysed using a variety of field and laboratory-based techniques and evaluated using statistical analysis including multivariate analysis. Elevated dissolved arsenic concentrations (up to 73.95 µg L-1) were observed in oxic-alkali groundwaters with the co-occurrence of other oxyanions (e.g. Mo, Se, Sb and U), low dissolved concentrations of Fe and Mn and low Na/Ca ratios indicating that arsenic was mobilised through alkali desorption of Fe oxyhydroxides. Arsenic speciation using a solid-phase extraction methodology (n=20) showed that the dominant species of arsenic present in groundwater was arsenate, with pH being a major controlling factor. The expected source of arsenic is sulfide minerals within fractures of the bedrock aquifer with transportation of arsenic and other oxyanion-forming elements facilitated by secondary Fe mineral phases. However, the presence of methylarsenical compounds detected in the groundwaters illustrates that microbially mediated mobilisation processes were also (co)-occurring. This study demonstrates how field speciation of arsenic can be utilised to overcome analytical limitations of conventional laboratory speciation and to facilitate in the interpretation of the environmental mobility of arsenic within groundwaters.

Field, Laboratory and Modeling Evidence for Strong Attenuation of a Cr(VI) Plume in a Mudstone Aquifer Due to Matrix Diffusion and Reaction Processes

This study uses a combination of conventional and high resolution field and laboratory methods to investigate processes causing attenuation of a hexavalent chromium (Cr(VI)) plume in sedimentary bedrock at a former industrial facility. Groundwater plume Cr(VI) concentrations decline by more than three orders of magnitude over a 900 m distance down gradient from the site. Internal plume concentrations generally exhibit stable to declining trends due to diffusive and reactive transport in the low permeability matrix as fluxes from the contamination source dissipate due to natural depletion processes and active remediation efforts. The strong attenuation is attributed to diffusion from mobile groundwater in fractures to immobile porewater in the rock matrix, and reactions causing transformation of aqueous Cr(VI) to low-solubility Cr(III) precipitates, confirmed by high spatial resolution rock matrix contaminant concentrations and comparisons with groundwater concentrations from multi-level sampling within the plume. Field characterization data for the fracture network and matrix properties were used to inform 2-D discrete-fracture matrix (DFM) numerical model simulations that quantify attenuation due to diffusion and reaction processes, which show consistency with field datasets, and provide insights regarding future plume conditions. The combination of field, laboratory and modeling evidence demonstrates effects of matrix diffusion and reaction processes causing strong attenuation of a Cr(VI) plume in a sedimentary bedrock aquifer. This approach has important implications for characterization of sites with Cr(VI) contamination for improved site conceptual models and remediation decision-making.

Contribution of Sentinel Radar Images 1Ato the Extraction of Lineaments from the Lobo Watershed (Center-West of the Ivory Coast)

The improvement of the conditions of access to drinking water for the populations of the Lobo watershed requires the exploitation of groundwater because of the surface water which remains exposed to climatic hazards and the impacts of anthropic activities. And yet, these underground waters finds himself bedrock aquifer which are complex aquifers. Thus, the objective of this study is to characterize those fissure aquifers that who govern underground run-off in the aquifer system of the Lobo catchment area. The methodology adopted consisted in using 1A sentinel radar images to map fractures and their spatial distribution. The validation of the lineaments first consisted in comparing and highlighting the lineaments from the radar images and the fractures revealed from photo-geological images. Next, we superimposed the map of lineaments on the map of boreholes with a flow rate greater than or equal to 5 m3/h, which were considered as productive boreholes. To determine the traffic corridors, this fracturing map was superimposed on the piezometric map. This work made it possible to extract 9,753 lineaments over a surface area of 7,000 km². The various validation techniques enabled us to confirm 121 major fractures with an average length of 9 km. In addition, the most productive boreholes are located on average less than 300 m from the fractures. The analysis of the distribution of the orientations of these fractures revealed a heterogeneity of direction and a predominance of the N-S; NW-SE and NE-SW families. The Fracturing density maps and density of the number of fracture crossing points highlight the spatial heterogeneity of the fracture network which is controlled by geomorphology, geological formations and lithological contacts. The river Lobo and its main tributary the Dé, flow preferentially in fractures. This river and its tributary drains the aquifer system. The results obtained from different thematic maps are useful for the realisation of future high-yield hydraulic wells (Q ≥ 5 m3/h).

A Modeling Comparison of Groundwater Potential Mapping in a Mountain Bedrock Aquifer: QUEST, GARP, and RF Models

In some arid regions, groundwater is the only source of water for human needs, so understanding groundwater potential is essential to ensure its sustainable use. In this study, three machine learning models (Genetic Algorithm for Rule-Set Production (GARP), Quick Unbiased Efficient Statistical Tree (QUEST), and Random Forest (RF)) were applied and verified for spatial prediction of groundwater in a mountain bedrock aquifer in Piranshahr Watershed, Iran. A spring location dataset consisting of 141 springs was prepared by field surveys, and from this three different sample datasets (S1–S3) were randomly generated (70% for training and 30% for validation). A total of 10 groundwater conditioning factors were prepared for modeling, namely slope percent, relative slope position (RSP), plan curvature, altitude, drainage density, slope aspect, topographic wetness index (TWI), terrain ruggedness index (TRI), land use, and lithology. The area under the receiver operating characteristic curve (AUC) and true skill statistic (TSS) were used to evaluate the accuracy of models. The results indicated that all models had excellent goodness-of-fit and predictive performance, but that RF (AUCmean = 0.995, TSSmean = 0.89) and GARP (AUCmean = 0.957, TSSmean = 0.82) outperformed QUEST (AUCmean = 0.949, TSSmean = 0.74). In robustness analysis, RF was slightly more sensitive than GARP and QUEST, making it necessary to consider several random partitioning options for preparing training and validation groups. The outcomes of this study can be useful in sustainable management of groundwater resources in the study region.