Impact of model complexity and multi-scale data integration on the estimation of hydrogeological parameters in a dual-porosity aquifer

This study investigates the impact of model complexity and multi-scale prior hydrogeological data on the interpretation of pumping test data in a dual-porosity aquifer (the Chalk aquifer in England, UK). In order to characterize the hydrogeological properties, different approaches ranging from a traditional analytical solution (Theis approach) to more sophisticated numerical models with automatically calibrated input parameters are applied. Comparisons of results from the different approaches show that neither traditional analytical solutions nor a numerical model assuming a homogenous and isotropic aquifer can adequately explain the observed drawdowns. A better reproduction of the observed drawdowns in all seven monitoring locations is instead achieved when medium and local-scale prior information about the vertical hydraulic conductivity (K) distribution is used to constrain the model calibration process. In particular, the integration of medium-scale vertical K variations based on flowmeter measurements lead to an improvement in the goodness-of-fit of the simulated drawdowns of about 30%. Further improvements (up to 70%) were observed when a simple upscaling approach was used to integrate small-scale K data to constrain the automatic calibration process of the numerical model. Although the analysis focuses on a specific case study, these results provide insights about the representativeness of the estimates of hydrogeological properties based on different interpretations of pumping test data, and promote the integration of multi-scale data for the characterization of heterogeneous aquifers in complex hydrogeological settings.     

Spatial resolution of transport parameters in a subtropical karst conduit system during dry and wet seasons

Karst aquifers are characterized by a high degree of hydrologic variability and spatial heterogeneity of transport parameters. Tracer tests allow the quantification of these parameters, but conventional point-to-point experiments fail to capture spatiotemporal variations of flow and transport. The goal of this study was to elucidate the spatial distribution of transport parameters in a karst conduit system at different flow conditions. Therefore, six tracer tests were conducted in an active and accessible cave system in Vietnam during dry and wet seasons. Injections and monitoring were done at five sites along the flow system: a swallow hole, two sites inside the cave, and two springs draining the system. Breakthrough curves (BTCs) were modeled with CXTFIT software using the one-dimensional advection-dispersion model and the two-region nonequilibrium model. In order to obtain transport parameters in the individual sections of the system, a multi-pulse injection approach was used, which was realized by using the BTCs from one section as input functions for the next section. Major findings include: (1) In the entire system, mean flow velocities increase from 183 to 1,043 m/h with increasing discharge, while (2) the proportion of immobile fluid regions decrease; (3) the lowest dispersivity was found at intermediate discharge; (4) in the individual cave sections, flow velocities decrease along the flow direction, related to decreasing gradients, while (5) dispersivity is highest in the middle section of the cave. The obtained results provide a valuable basis for the development of an adapted water management strategy for a projected water-supply system.     

Hybrid Gaussian-cubic radial basis functions for scattered data interpolation

Scattered data interpolation schemes using kriging and radial basis functions (RBFs) have the advantage of being meshless and dimensional independent; however, for the datasets having insufficient observations, RBFs have the advantage over geostatistical methods as the latter requires variogram study and statistical expertise. Moreover, RBFs can be used for scattered data interpolation with very good convergence, which makes them desirable for shape function interpolation in meshless methods for numerical solution of partial differential equations. For interpolation of large datasets, however, RBFs in their usual form, lead to solving an ill-conditioned system of equations, for which, a small error in the data can cause a significantly large error in the interpolated solution. In order to reduce this limitation, we propose a hybrid kernel by using the conventional Gaussian and a shape parameter independent cubic kernel. Global particle swarm optimization method has been used to analyze the optimal values of the shape parameter as well as the weight coefficients controlling the Gaussian and the cubic part in the hybridization. Through a series of numerical tests, we demonstrate that such hybridization stabilizes the interpolation scheme by yielding a far superior implementation compared to those obtained by using only the Gaussian or cubic kernels. The proposed kernel maintains the accuracy and stability at small shape parameter as well as relatively large degrees of freedom, which exhibit its potential for scattered data interpolation and intrigues its application in global as well as local meshless methods for numerical solution of PDEs.     

Convergence analysis of fixed stress split iterative scheme for anisotropic poroelasticity with tensor Biot parameter

We perform a convergence analysis of the fixed stress split iterative scheme for the Biot system modeling coupled flow and deformation in anisotropic poroelastic media with tensor Biot parameter. The fixed stress split iterative scheme solves the flow subproblem with all components of the stress tensor frozen using a multipoint flux mixed finite element method, followed by the poromechanics subproblem using a conforming Galerkin method in every coupling iteration at each time step. The coupling iterations are repeated until convergence and Backward Euler is employed for time marching. The convergence analysis is based on studying the equations satisfied by the difference of iterates to show that the fixed stress split iterative scheme for anisotropic poroelasticity with Biot tensor is contractive. We also demonstrate that the scheme is numerically convergent using the classical Mandel’s problem solution for transverse isotropy.     

Impact of recharge water temperature on bioclogging during managed aquifer recharge: a laboratory study

To investigate the effect of recharge water temperature on bioclogging processes and mechanisms during seasonal managed aquifer recharge (MAR), two groups of laboratory percolation experiments were conducted: a winter test and a summer test. The temperatures were controlled at ~5±2 and ~15±3 °C, and the tests involved bacterial inoculums acquired from well water during March 2014 and August 2015, for the winter and summer tests, respectively. The results indicated that the sand columns clogged ~10 times faster in the summer test due to a 10-fold larger bacterial growth rate. The maximum concentrations of total extracellular polymeric substances (EPS) in the winter test were approximately twice those in the summer test, primarily caused by a ~200 μg/g sand increase of both loosely bound EPS (LB-EPS) and tightly bound EPS (TB-EPS). In the first half of the experimental period, the accumulation of bacteria cells and EPS production induced rapid bioclogging in both the winter and summer tests. Afterward, increasing bacterial growth dominated the bioclogging in the summer test, while the accumulation of LB-EPS led to further bioclogging in the winter test. The biological analysis determined that the dominant bacteria in experiments for both seasons were different and the bacterial community diversity was ~50% higher in the winter test than that for summer. The seasonal inoculums could lead to differences in the bacterial community structure and diversity, while recharge water temperature was considered to be a major factor influencing the bacterial growth rate and metabolism behavior during the seasonal bioclogging process.     

Hydrogeology of a montane headwater groundwater system downgradient of a coal-mine waste rock dump: Elk Valley,British Columbia,Canada

Steelmaking-coal waste rock placed in mountain catchments in the Elk Valley, British Columbia, Canada, drain constituents of interest (CIs) to surface water downgradient of the waste rock dumps. The role of groundwater in transporting CIs in the headwaters of mountain catchments is not well understood. This study characterizes the physical hydrogeology of a portion of a 10-km² headwater catchment (West Line Creek) downgradient of a 2.7-km² waste rock dump placed over a natural headwater valley-bottom groundwater system. The study site was instrumented with 13 monitoring wells. Drill core samples were collected to determine subsurface lithology and geotechnical properties. The groundwater system was characterized using field testing and water-level monitoring. The valley-bottom sediments were composed of unconsolidated glacial and meltwater successions (<64 m thick) deposited as a series of cut and fill structures overlying shale bedrock. An unconfined basal alluvial aquifer located above fractured bedrock was identified as the primary conduit for groundwater flow toward Line Creek (650 m from the toe of the dump). Discharge through the basal alluvial aquifer was estimated using the geometric mean hydraulic conductivity (±1 standard deviation). These calculations suggest groundwater discharge could account for approximately 15% (ranging from 2 to 60%) of the total water discharged from the watershed. The residence time from the base of the waste rock dump to Line Creek was estimated at <3 years. The groundwater system was defined as a snowmelt (i.e., nival) regime dominated by direct recharge (percolation of precipitation) across the catchment.     

Prediction of water inflow to mechanized tunnels during tunnel-boring-machine advance using numerical simulation

An accurate estimate of the groundwater inflow to a tunnel is one of the most challenging but essential tasks in tunnel design and construction. Most of the numerical or analytical solutions that have been developed ignore tunnel seepage conditions, material properties and hydraulic-head changes along the tunnel route during the excavation process, leading to inaccurate prediction of inflow rates. A method is introduced that uses MODFLOW code of GMS software to predict inflow rate as the tunnel boring machine (TBM) gradually advances. In this method, the tunnel boundary condition is conceptualized and defined using Drain package, which is simulated by dividing the drilling process into a series of successive intervals based on the tunnel excavation rates. In addition, the drain elevations are specified as the respective tunnel elevations, and the conductance parameters are assigned to intervals, depending on the TBM type and the tunnel seepage condition. The Qomroud water conveyance tunnel, located in Lorestan province of Iran, is 36 km in length. Since the Qomroud tunnel involved groundwater inrush during excavating, it is considered as a good case study to evaluate the presented method. The groundwater inflow to this tunnel during the TBM advance is simulated using the proposed method and the predicted rates are compared with observed rates. The results show that the presented method can satisfactorily predict the inflow rates as the TBM advances.     

Impact of irrigated agriculture on groundwater-recharge salinity: a major sustainability concern in semi-arid regions

Intensive irrigated agriculture substantially modifies the hydrological cycle and often has major environmental impacts. The article focuses upon a specific concern—the tendency for progressive long-term increases in the salinity of groundwater recharge derived from irrigated permeable soils and replenishment of unconfined aquifers in more arid regions. This process has received only scant attention in the water-resource literature and has not been considered by agricultural science. This work makes an original contribution by analysing, from scientific principles, how the salinisation of groundwater recharge arises and identifies the factors affecting its severity. If not proactively managed, the process eventually will impact irrigation waterwell salinity, the productivity of agriculture itself, and can even lead to land abandonment. The types of management measure required for mitigation are discussed through three detailed case histories of areas with high-value groundwater-irrigated agriculture (in Spain, Argentina and Pakistan), which provide a long-term perspective on the evolution of the problem over various decades.     

Adsorption of copper (II) on mesoporous silica: the effect of nano-scale confinement

Nano-scale spatial confinement can alter chemistry at mineral–water interfaces. These nano-scale confinement effects can lead to anomalous fate and transport behavior of aqueous metal species. When a fluid resides in a nanoporous environments (pore size under 100 nm), the observed density, surface tension, and dielectric constant diverge from those measured in the bulk. To evaluate the impact of nano-scale confinement on the adsorption of copper (Cu²⁺), we performed batch adsorption studies using mesoporous silica. Mesoporous silica with the narrow distribution of pore diameters (SBA-15; 8, 6, and 4 nm pore diameters) was chosen since the silanol functional groups are typical to surface environments. Batch adsorption isotherms were fit with adsorption models (Langmuir, Freundlich, and Dubinin–Radushkevich) and adsorption kinetic data were fit to a pseudo-first-order reaction model. We found that with decreasing pore size, the maximum surface area-normalized uptake of Cu²⁺ increased. The pseudo-first-order kinetic model demonstrates that the adsorption is faster as the pore size decreases from 8 to 4 nm. We attribute these effects to the deviations in fundamental water properties as pore diameter decreases. In particular, these effects are most notable in SBA-15 with a 4-nm pore where the changes in water properties may be responsible for the enhanced Cu mobility, and therefore, faster Cu adsorption kinetics.     

Natural speciation of nickel at the micrometer scale in serpentine (ultramafic) topsoils using microfocused X-ray fluorescence,diffraction, and absorption

Serpentine soils and ultramafic laterites develop over ultramafic bedrock and are important geological materials from environmental, geochemical, and industrial standpoints. They have naturally elevated concentrations of trace metals, such as Ni, Cr, and Co, and also high levels of Fe and Mg. Minerals host these trace metals and influence metal mobility. Ni in particular is an important trace metal in these soils, and the objective of this research was to use microscale (µ) techniques to identify naturally occurring minerals that contain Ni and Ni correlations with other trace metals, such as Fe, Mn, and Cr. Synchrotron based µ-XRF, µ-XRD, and µ-XAS were used. Ni was often located in the octahedral layer of serpentine minerals, such as lizardite, and in other layered phyllosilicate minerals with similar octahedral structure, such as chlorite group minerals including clinochlore and chamosite. Ni was also present in goethite, hematite, magnetite, and ferrihydrite. Goethite was present with lizardite and antigorite on the micrometer scale. Lizardite integrated both Ni and Mn simultaneously in its octahedral layer. Enstatite, pargasite, chamosite, phlogopite, and forsterite incorporated various amounts of Ni and Fe over the micrometer spatial scale. Ni content increased six to seven times within the same 500 µm µ-XRD transect on chamosite and phlogopite. Data are shown down to an 8 µm spatial scale. Ni was not associated with chromite or zincochromite particles. Ni often correlated with Fe and Mn, and generally did not correlate with Cr, Zn, Ca, or K in µ-XRF maps. A split shoulder feature in the µ-XAS data at 8400 eV (3.7 Å^?1 in k-space) is highly correlated (94% of averaged LCF results) to Ni located in the octahedral sheet of layered phyllosilicate minerals, such as serpentine and chlorite-group minerals. A comparison of bulk-XAS LCF to averaged µ-XAS LCF results showed good representation of the bulk soil via the µ-XAS technique for two of the three soils. In the locations analyzed by µ-XAS, average Ni speciation was dominated by layered phyllosilicate and serpentine minerals (76%), iron oxides (18%), and manganese oxides (9%). In the locations analyzed by µ-XRD, average Ni speciation was dominated by layered phyllosilicate, serpentine, and ultramafic-related minerals (71%) and iron oxides (17%), illustrating the complementary nature of these two methods.

Open image in new window