Ground water最新文献

Supercritical CO₂ (sCO₂) removes water from brine held in pumice stone at levels well above the solubility of water in sCO₂. The higher water removal results from a combination of passive emulsification of water in sCO₂ and viscous fingering of sCO₂ through the saturated pumice. This leads to higher levels of salt deposition than that expected from solubility considerations alone. These deposits could impact the injectivity of sCO₂ as well as its movement in the subsurface. The finding that the water concentration in sCO₂ is not necessarily capped at the solubility limit should influence the parametrization of injection models.

In this paper, we review the derivation of the Gauss-Levenberg-Marquardt (GLM) algorithm and its extension to ensemble parameter estimation. We explore the use of graphical methods to provide insights into how the algorithm works in practice and discuss the implications of both algorithm tuning parameters and objective function construction in performance. Some insights include understanding the control of both parameter trajectory and step size for GLM as a function of tuning parameters. Furthermore, for the iterative Ensemble Smoother (iES), we discuss the importance of noise on observations and show how iES can cope with non-unique outcomes based on objective function construction. These insights are valuable for modelers using PEST, PEST++, or similar parameter estimation tools.

Water-table maps are fundamental to hydrogeological studies and a manual, hand-drawn method is still commonly used to produce them. Despite this, the accuracy and variability of such maps have received little attention in international literature. In a unique experiment, 63 groundwater professionals drew water-table equipotential contours based on the same dataset of point measurements and were asked to infer flow directions and predict groundwater elevations at predefined locations. The root mean squared error (RMSE) for the average map calibration data was 10.5 m, which is accuracy comparable to numerical groundwater models. This study confirmed that to produce hand-drawn water-table maps, practitioners seek to not only fit the spatial data, but also to conform to their own cognitive model of hydrogeological concepts and processes. The calibration accuracy increased with experience; from a RMSE of 13.3 m for practitioners with 0-3 years of experience to a RMSE of 9.2 m for those with four or more years. Despite considerable variability in the style of the hand-drawn water-table maps, the maps were consistent in their representation of the dominant regional groundwater flow directions. There was less consensus, however, in predicting the direction of surface water-groundwater interaction for a stream reach. Hand-drawn water-table mapping remains useful and valid, especially as a starting point for hydrogeological conceptualization, yet further work is required to resolve issues around transparency, repeatability, and reproducibility.

This study advances a methodology to estimate effective apertures of fractures in glacial tills based on dye tracer infiltration tests and numerical simulations. The approach uses the visible penetration depth of the dye tracer along fracture flow paths as primary information to calculate effective fracture apertures. Further data used in the calculation are the dye tracer input concentration and retardation, the duration of the tracer injection, and the hydraulic gradient applied to control the infiltrating water fluxes. The method does not require measurement of hydraulic conductivity for the fractured till and enables direct observation of flow and transport patterns within the fractures (e.g., uniform flow and dye tracer distribution, channeling due to aperture variability, and presence of biogenic macropores in fractures). The approach was successfully verified by using the estimated effective fracture aperture values in Large Undisturbed Columns (LUCs) to consistently simulate both the observed LUC effluent breakthrough of a conservative bromide tracer and the water fluxes with the hydraulic gradient applied in the experiments. Sensitivity analyses revealed that estimation of small effective fracture apertures (<10 μm) required accurate determination of the dye tracer retardation factor. By contrast, in the case of larger effective apertures (>20 μm), the sensitivity of the estimated effective fracture aperture to variations in the porous material and solute transport parameters was low compared to the dominant sensitivity to the water flow through the fractures (cubic relation between flow and aperture). The proposed approach may be extended beyond laboratory applications and assist in characterizing field-scale fracture networks.

Faults can fundamentally change a groundwater flow regime and represent a major source of uncertainty in groundwater studies. Much research has been devoted to uncertainty around their location and their barrier-conduit behavior. However, fault timing is one aspect of fault uncertainty that appears to be somewhat overlooked. Many faulted models feature consistent layer offsets, thereby presuming that block faulting has occurred recently and almost instantaneously. Additionally, barrier and/or conduit behavior is often shown to extend vertically through all layers when a fault may in fact terminate well below-ground surface. In this study, we create three plausible geological interpretations for a transect in the Perth Basin. Adjacent boreholes show stratigraphic offsets and thickening which indicate faulting; however, fault timing is unknown. Flow modeling demonstrates that the model with the most recent faulting shows profoundly different flow patterns due to aquifer juxtaposition. Additionally, multiple realizations with stochastically generated parameter sets for layer, fault core, and fault damage zone conductivity show that fault timing influences flow more than layer or fault zone conductivity. Finally, fault conduit behavior that penetrates aquitards has significant implications for transport, while fault barrier behavior has surprisingly little. This research advocates for adequate data collection where faults may cause breaches in aquitards due to layer offsets or conduit behavior in the damage zone. It also promotes the use of multiple geological models to address structural uncertainty, and highlights some of the hurdles in doing so such as computational expense and the availability of seamless geological-flow modeling workflows.