Search

The nonlinearity of equations governing flow in unconfined aquifers poses challenges for numerical models, particularly in field-scale applications. Existing methods are often unstable, do not converge, or require extremely fine grids and small time steps. Standard modeling procedures such as automated model calibration and Monte Carlo uncertainty analysis typically require thousands of model runs. Stable and efficient model performance is essential to these analyses. We propose a new method that offers improvements in stability and efficiency and is relatively tolerant of coarse grids. It applies a strategy similar to that in the MODFLOW code to the solution of Richard’s equation with a grid-dependent pressure/saturation relationship. The method imposes a contrast between horizontal and vertical permeability in gridblocks containing the water table, does not require “dry” cells to convert to inactive cells, and allows recharge to flow through relatively dry cells to the water table. We establish the accuracy of the method by comparison to an analytical solution for radial flow to a well in an unconfined aquifer with delayed yield. Using a suite of test problems, we demonstrate the efficiencies gained in speed and accuracy over two-phase simulations, and improved stability when compared to MODFLOW. The advantages for applications to transient unconfined aquifer analysis are clearly demonstrated by our examples. We also demonstrate applicability to mixed vadose zone/saturated zone applications, including transport, and find that the method shows great promise for these types of problem as well. [ABSTRACT FROM AUTHOR]

Abstract: A generalized dual porosity method (GDPM) has been developed to incorporate sub-grid scale heterogeneity into large-scale flow and transport simulations. The method is spatially variable in the sense that the method can be applied with different levels of resolution for different spatial nodes in the simulation. The method utilizes the nodal connectivity structure and linear equation solvers of unstructured grids like those used in the finite element method, and can be applied to any problem without externally modifying the numerical grid. The algorithm scales linearly in CPU time and storage with the number of GDPM nodes. We demonstrate the utility and computational efficiency of the technique with two verification problems and an example problem of a field site. [Copyright &y& Elsevier]

Modern ground water characterization and remediation projects routinely require calibration and inverse analysis of large three-dimensional numerical models of complex hydrogeological systems. Hydrogeologic complexity can be prompted by various aquifer characteristics including complicated spatial hydrostratigraphy and aquifer recharge from infiltration through an unsaturated zone. To keep the numerical models computationally efficient, compromises are frequently made in the model development, particularly, about resolution of the computational grid and numerical representation of the governing flow equation. The compromise is required so that the model can be used in calibration, parameter estimation, performance assessment, and analysis of sensitivity and uncertainty in model predictions. However, grid properties and resolution as well as applied computational schemes can have large effects on forward-model predictions and on inverse parameter estimates. We investigate these effects for a series of one- and two-dimensional synthetic cases representing saturated and variably saturated flow problems. We show that “conformable” grids, despite neglecting terms in the numerical formulation, can lead to accurate solutions of problems with complex hydrostratigraphy. Our analysis also demonstrates that, despite slower computer run times and higher memory requirements for a given problem size, the control volume finite-element method showed an advantage over finite-difference techniques in accuracy of parameter estimation for a given grid resolution for most of the test problems. [ABSTRACT FROM AUTHOR]

This paper presents several different conceptual models of the Large Hydraulic Gradient (LHG) region north of Yucca Mountain and describes the impact of those models on groundwater flow near the potential high-level repository site. The results are based on a numerical model of site-scale saturated zone beneath Yucca Mountain. This model is used for performance assessment predictions of radionuclide transport and to guide future data collection and modeling activities. The numerical model is calibrated by matching available water level measurements using parameter estimation techniques, along with more informal comparisons of the model to hydrologic and geochemical information. The model software (hydrologic simulation code FEHM and parameter estimation software PEST) and model setup allows for efficient calibration of multiple conceptual models. Until now, the Large Hydraulic Gradient has been simulated using a low-permeability, east–west oriented feature, even though direct evidence for this feature is lacking. In addition to this model, we investigate and calibrate three additional conceptual models of the Large Hydraulic Gradient, all of which are based on a presumed zone of hydrothermal chemical alteration north of Yucca Mountain. After examining the heads and permeabilities obtained from the calibrated models, we present particle pathways from the potential repository that record differences in the predicted groundwater flow regime. The results show that Large Hydraulic Gradient can be represented with the alternate conceptual models that include the hydrothermally altered zone. The predicted pathways are mildly sensitive to the choice of the conceptual model and more sensitive to the quality of calibration in the vicinity on the repository. These differences are most likely due to different degrees of fit of model to data, and do not represent important differences in hydrologic conditions for the different conceptual models. [Copyright &y& Elsevier]

This study is an extension of the stochastic analysis of transient two-phase flow in randomly heterogeneous porous media (Chen et al. in Water Resour Res 42:W03425, ), by incorporating direct measurements of the random soil properties. The log-transformed intrinsic permeability, soil pore size distribution parameter, and van Genuchten fitting parameter are treated as stochastic variables that are normally distributed with a separable exponential covariance model. These three random variables conditioned on given measurements are decomposed via Karhunen–Loève decomposition. Combined with the conditional eigenvalues and eigenfunctions of random variables, we conduct a series of numerical simulations using stochastic transient water–oil flow model (Chen et al. in Water Resour Res 42:W03425, ) based on the KLME approach to investigate how the number and location of measurement points, different random soil properties, as well as the correlation length of the random soil properties, affect the stochastic behavior of water and oil flow in heterogeneous porous media. [ABSTRACT FROM AUTHOR]

Abstract: The Hijiori hot dry rock (HDR) system, Japan, consists of a shallow and a deep reservoir, both in fractured granitic rocks. During a long-term circulation test (LTCT) lasting approximately 18 months and which tested different fluid production scenarios, large changes were observed in output fluid temperatures, pressures, and flow rates. A multi-reservoir, numerical model of the Hijiori HDR system was developed using the finite element heat and mass (FEHM) transfer code and applied to simulations of the LTCT. The model reproduced the pressure, temperature, and flow data observed during the test. Based on the modeling study, it was concluded that most of the produced fluid came from the shallow reservoir, that the permeability of the deep reservoir changed during the initial part of the LTCT, and that the redistribution of injected water between the two reservoirs had little impact on the relative amounts of deep and shallow fluid production. After validating the model on the LTCT, it was used to optimize injection rates in both reservoirs. The model was also used in simulations of reservoir performance where an additional heat transfer surface area has been created in the subsurface through hydraulic fracturing. [Copyright &y& Elsevier]

Three-dimensional grids representing a heterogeneous, ground water system are generated at 10 different resolutions in support of a site-scale flow and transport modeling effort. These grids represent hydrostratigraphy near Yucca Mountain, Nevada, consisting of 18 stratigraphic units with contrasting fluid flow and transport properties. The grid generation method allows the stratigraphy to be modeled by numerical grids of different resolution so that comparison studies can be performed to test for grid quality and determine the resolution required to resolve geologic structure and physical processes such as fluid flow and solute transport. The process of generating numerical grids with appropriate property distributions from geologic conceptual models is automated, thus making the entire process easy to implement with fewer user-induced errors. The series of grids of various resolutions are used to assess the level at which increasing resolution no longer influences the flow and solute transport results. Grid resolution is found to be a critical issue for ground water flow and solute transport. The resolution required in a particular instance is a function of the feature size of the model, the intrinsic properties of materials, the specific physics of the problem, and boundary conditions. The asymptotic nature of results related to flow and transport indicate that for a hydrologic model of the heterogeneous hydrostratigraphy under Yucca Mountain, a horizontal grid spacing of 600 m and vertical grid spacing of 40 m resolve the hydrostratigraphic model with sufficient precision to accurately model the hypothetical flow and solute transport to within 5% of the value that would be obtained with much higher resolution. [ABSTRACT FROM AUTHOR]

The use of low-temperature geothermal resources is important from the viewpoint of global warming. In order to evaluate various underground projects that use low-temperature geothermal resources, we have estimated the parameters of a typical underground system using the two-well model. By changing the parameters of the system, six different heat extraction scenarios have been studied. One of these six scenarios is recommended because of its small energy loss. [Copyright &y& Elsevier]

We present a numerical model that improves our capability to simulate multiphase, non-isothermal flow in variably saturated porous and fractured media at magmatic temperatures and shallow crustal pressures. Simulations of heat and fluid flow in variably saturated host rock near a magmatic intrusion provide insight into contact metamorphic processes, including dryout, condensation, and resaturation effects and implications for host-rock alteration. The numerical code, an enhanced version of FEHM, uses a finite-element/finite-volume technique incorporating implicit Newton–Raphson iteration to solve non-linear conservation equations for mass and energy, using thermodynamic properties of water and air in the ranges 10°C≤T≤1500°C, 0.00123≤P≤1000 MPa and 10°C≤T≤1500°C, 0.00123≤P≤22 MPa, respectively. The study area is located at Paiute Ridge, eastern Nevada Test Site, Nevada, USA, where hypabyssal mafic intrusions were emplaced at about 8.5–8.6 Ma (Ar/Ar age estimate) and cooled contemporaneously with part of a geomagnetic field reversal, inferred from paleomagnetic data from over 100 sites in intrusions and remagnetized host ash-flow tuffs. We used a radial model of heat flow and multiphase pore fluid flow adjacent to a 1200°C intrusion to characterize the thermal evolution of the contact metamorphic system. For likely initial pore saturations of 0.4–0.6, an expanding dryout zone near the intrusion and a condensation zone of enhanced saturation (S≤0.8) extends 150–400 m from the intrusion. Host-rock temperatures reach 800°C near the contact and cool below 100°C within 2000 yr after emplacement, two to four times faster than predicted by a simple conduction model. The thermal history of the system is very sensitive to initial saturation. The multiphase thermal model allows bounds to be placed on the rate of change of the transitional part of the geomagnetic field during the field reversal recorded at Paiute Ridge. We assume that magnetization acquisition took place during the life of the thermal system that developed in the intrusions and contact rocks and that the paleomagnetic data provide a quasi-continuous record of the transitional part of the reversal. Sites in intrusions and thermally annealed ash-flow tuffs reveal subtle yet systematic variations in paleomagnetic directions. We combine the directional data with robust thermal (temperature/time) models to estimate the rate of change of the geomagnetic field. Modeled times of 140–290 yr and 215–440 yr for the duration of magnetization acquisition at two different sites correspond to estimated rates of change of 0.06–0.13°/yr for the field during the transitional part of the reversal. [Copyright &y& Elsevier]

Abstract: In the last few decades hydrologists have made tremendous progress in using dynamic simulation models for the analysis and understanding of hydrologic systems. However, predictions with these models are often deterministic and as such they focus on the most probable forecast, without an explicit estimate of the associated uncertainty. This uncertainty arises from incomplete process representation, uncertainty in initial conditions, input, output and parameter error. The generalized likelihood uncertainty estimation (GLUE) framework was one of the first attempts to represent prediction uncertainty within the context of Monte Carlo (MC) analysis coupled with Bayesian estimation and propagation of uncertainty. Because of its flexibility, ease of implementation and its suitability for parallel implementation on distributed computer systems, the GLUE method has been used in a wide variety of applications. However, the MC based sampling strategy of the prior parameter space typically utilized in GLUE is not particularly efficient in finding behavioral simulations. This becomes especially problematic for high-dimensional parameter estimation problems, and in the case of complex simulation models that require significant computational time to run and produce the desired output. In this paper we improve the computational efficiency of GLUE by sampling the prior parameter space using an adaptive Markov Chain Monte Carlo scheme (the Shuffled Complex Evolution Metropolis (SCEM-UA) algorithm). Moreover, we propose an alternative strategy to determine the value of the cutoff threshold based on the appropriate coverage of the resulting uncertainty bounds. We demonstrate the superiority of this revised GLUE method with three different conceptual watershed models of increasing complexity, using both synthetic and real-world streamflow data from two catchments with different hydrologic regimes. [Copyright &y& Elsevier]

Abstract: A flexible, robust method for linking parent (regional-scale) and child (local-scale) grids of locally refined models that use different numerical methods is developed based on a new, iterative ghost-node method. Tests are presented for two-dimensional and three-dimensional pumped systems that are homogeneous or that have simple heterogeneity. The parent and child grids are simulated using the block-centered finite-difference MODFLOW and control-volume finite-element FEHM models, respectively. The models are solved iteratively through head-dependent (child model) and specified-flow (parent model) boundary conditions. Boundary conditions for models with nonmatching grids or zones of different hydraulic conductivity are derived and tested against heads and flows from analytical or globally-refined models. Results indicate that for homogeneous two- and three-dimensional models with matched grids (integer number of child cells per parent cell), the new method is nearly as accurate as the coupling of two MODFLOW models using the shared-node method and, surprisingly, errors are slightly lower for nonmatching grids (noninteger number of child cells per parent cell). For heterogeneous three-dimensional systems, this paper compares two methods for each of the two sets of boundary conditions: external heads at head-dependent boundary conditions for the child model are calculated using bilinear interpolation or a Darcy-weighted interpolation; specified-flow boundary conditions for the parent model are calculated using model-grid or hydrogeologic-unit hydraulic conductivities. Results suggest that significantly more accurate heads and flows are produced when both Darcy-weighted interpolation and hydrogeologic-unit hydraulic conductivities are used, while the other methods produce larger errors at the boundary between the regional and local models. The tests suggest that, if posed correctly, the ghost-node method performs well. Additional testing is needed for highly heterogeneous systems. [Copyright &y& Elsevier]