Aquifer Storage and Recovery Using Saline Aquifers: Hydrogeological Controls and Opportunities

Aquifer storage and recovery (ASR) is a valuable tool for managing variations in the supply and demand of freshwater, but system performance is highly dependent upon system-specific hydrogeological conditions including the salinity of the storage-zone native groundwater. ASR systems using storage zones containing saline (>10,000 mg/L of total dissolved solids) groundwater tend to have relatively low recovery efficiencies (REs). However, the drawbacks of low REs may be offset by lesser treatment requirements and may be of secondary importance where the stored water (e.g., excess reclaimed, surface, and storm waters) would otherwise go to waste and pose disposal costs. Density-dependent, solute-transport modeling results demonstrate that the RE of ASR systems using a saline storage zone is most strongly controlled by parameters controlling free convection (e.g., horizontal hydraulic conductivity) and mixing of recharged and native groundwater (e.g., dispersivity and aquifer heterogeneity). Preferred storage zone conditions are moderate hydraulic conductivities (5 to 20 m/d), low degrees of aquifer heterogeneity, and primary porosity-dominated siliclastic and limestones lithologies with effective porosities greater than 5%. Where hydrogeological conditions are less favorable, operational options are available to improve RE, such as preferential recovery from the top of the storage zone. Injection of large volumes of excess water currently not needed into saline aquifers could create valuable water resources that could be tapped in the future during times of greater need.     

Relative Recovery of Thermal Energy and Fresh Water in Aquifer Storage and Recovery Systems

This paper explores the relationship between thermal energy and fresh water recoveries from an aquifer storage recovery (ASR) well in a brackish confined aquifer. It reveals the spatial and temporal distributions of temperature and conservative solutes between injected and recovered water. The evaluation is based on a review of processes affecting heat and solute transport in a homogeneous aquifer. In this simplified analysis, it is assumed that the aquifer is sufficiently anisotropic to inhibit density‐affected flow, flow is axisymmetric, and the analysis is limited to a single ASR cycle. Results show that the radial extent of fresh water at the end of injection is greater than that of the temperature change due to the heating or cooling of the geological matrix as well as the interstitial water. While solutes progress only marginally into low permeability aquitards by diffusion, conduction of heat into aquitards above and below is more substantial. Consequently, the heat recovery is less than the solute recovery when the volume of the recovered water is lower than the injection volume. When the full volume of injected water is recovered the temperature mixing ratio divided by the solute mixing ratio for recovered water ranges from 0.95 to 0.6 for ratios of maximum plume radius to aquifer thickness of 0.6 to 4.6. This work is intended to assist conceptual design for dual use of ASR for conjunctive storage of water and thermal energy to maximize the potential benefits.     

Small‐Scale ASR Between Flow Barriers in a Saline Aquifer

Regular aquifer storage recovery, ASR, is often not feasible for small‐scale storage in brackish or saline aquifers because fresh water floats to the top of the aquifer where it is unrecoverable. Flow barriers that partially penetrate a brackish or saline aquifer prevent a stored volume of fresh water from expanding sideways, thus increasing the recovery efficiency. In this paper, the groundwater flow and mixing is studied during injection, storage, and recovery of fresh water in a brackish or saline aquifer in a flow‐tank experiment and by numerical modeling to investigate the effect of density difference, hydraulic conductivity, pumping rate, cyclic operation, and flow barrier settings. Two injection and recovery methods are investigated: constant flux and constant head. Fresh water recovery rates on the order of 65% in the first cycle climbing to as much as 90% in the following cycles were achievable for the studied configurations with constant flux whereas the recovery efficiency was somewhat lower for constant head. The spatial variation in flow velocity over the width of the storage zone influences the recovery efficiency, because it induces leakage of fresh water underneath the barriers during injection and upconing of salt water during recovery.     

Non-Darcian flow to a well in an aquifer–aquitard system

In this study, we use a linearization procedure and a finite difference method to solve non-Darcian flow to a well in an aquifer–aquitard system. The leakage effect is considered. Flow in the aquifer is assumed to be non-Darcian and horizontal, whereas flow in the aquitard is assumed to be Darcian and vertical. The Izbash equation [Izbash SV. O filtracii V Kropnozernstom Materiale. USSR: Leningrad; 1931 [in Russian]] is employed to describe the non-Darcian flow. The wellbore storage is also considered in this study. An approximate semi-analytical solution has been obtained by the linearization procedure, and a numerical solution has been obtained by using a finite difference method. The previous solutions for Darcian flow case and non-Darcian flow case without leakage can be described as special cases of the new solutions. The error caused by the linearization procedure has also been analyzed. The relative error caused by the linearization procedure is nearly 100% at early times, and decreases to zero at late times. We have also compared the results in this study with Wen et al. [Wen Z, Huang G, Zhan H. A numerical solution for non-Darcian flow to a well in a confined aquifer using the power law function. J Hydrol, 2008d [in revision]] in which the leakage effect is not considered, and Hantush and Jacob [Hantush MS, Jacob CE. Non-steady radial flow in an infinite leaky aquifer. Trans Am Geophys Union 1955;36(1):95–100] who investigated a similar problem in Darcian flow case. The comparison of this study and Wen et al. (2008d) indicates the dimensionless drawdown in the aquifer with leakage is less than that without leakage, and the leakage has little effect at early times. The comparison between the results of this study and that of Hantush and Jacob (1955) indicates that the dimensionless drawdown in the aquifer for non-Darcian flow is larger at early times and smaller at late times, than their counterparts for Darcian flow. A larger dimensionless non-Darcian conductivity k_D results in a smaller dimensionless drawdown in the aquifer at late times, and leads to a larger dimensionless drawdown in the aquifer at early times. A smaller dimensionless leakage parameter B_D results in a smaller drawdown at late times, and the leakage does not affect the early-time drawdown. The analysis of the dimensionless drawdown inside the well has also been included in this study when the wellbore storage is considered.     

Hydrogeophysical Methods for Analyzing Aquifer Storage and Recovery Systems

Hydrogeophysical methods are presented that support the siting and monitoring of aquifer storage and recovery (ASR) systems. These methods are presented as numerical simulations in the context of a proposed ASR experiment in Kuwait, although the techniques are applicable to numerous ASR projects. Bulk geophysical properties are calculated directly from ASR flow and solute transport simulations using standard petrophysical relationships and are used to simulate the dynamic geophysical response to ASR. This strategy provides a quantitative framework for determining site‐specific geophysical methods and data acquisition geometries that can provide the most useful information about the ASR implementation. An axisymmetric, coupled fluid flow and solute transport model simulates injection, storage, and withdrawal of fresh water (salinity ～500 ppm) into the Dammam aquifer, a tertiary carbonate formation with native salinity approximately 6000 ppm. Sensitivity of the flow simulations to the correlation length of aquifer heterogeneity, aquifer dispersivity, and hydraulic permeability of the confining layer are investigated. The geophysical response using electrical resistivity, time‐domain electromagnetic (TEM), and seismic methods is computed at regular intervals during the ASR simulation to investigate the sensitivity of these different techniques to changes in subsurface properties. For the electrical and electromagnetic methods, fluid electric conductivity is derived from the modeled salinity and is combined with an assumed porosity model to compute a bulk electrical resistivity structure. The seismic response is computed from the porosity model and changes in effective stress due to fluid pressure variations during injection/recovery, while changes in fluid properties are introduced through Gassmann fluid substitution.     

An assessment of aquifer storage recovery using ground water flow models

Owing to increased demands on ground water accompanied by increased drawdowns, technologies that use recharge options, such as aquifer storage recovery (ASR), are being used to optimize available water resources and reduce adverse effects of pumping. In this paper, three representative ground water flow models were created to assess the impact of hydrogeologic and operational parameters/factors on recovery efficiency of ASR systems. Flow/particle tracking and solute transport models were used to track the movement of water during injection, storage, and recovery. Results from particle tracking models consistently produced higher recovery efficiency than the solute transport models for the parameters/properties examined because the particle tracking models neglected mixing of the injected and ambient water. Mixing between injected and ambient water affected recovery efficiency. Results from this study demonstrate the interactions between hydrogeologic and operational parameters on predictions of recovery efficiency. These interactions are best simulated using coupled numerical ground water flow and transport models that include the effects of mixing of injected water and ambient ground water.     

Recovery of Injected Freshwater from a Brackish Aquifer with a Multiwell System

Herein we propose a multiple injection and recovery well system strategically operated for freshwater storage in a brackish aquifer. With the system we call aquifer storage transfer and recovery (ASTR) by using four injection and two production wells, we are capable of achieving both high recovery efficiency of injected freshwater and attenuation of contaminants through adequately long residence times and travel distances within the aquifer. The usual aquifer storage and recovery (ASR) scheme, in which a single well is used for injection and recovery, does not warrant consistent treatment of injected water due to the shorter minimum residence times and travel distances. We tested the design and operation of the system over 3 years in a layered heterogeneous limestone aquifer in Salisbury, South Australia. We demonstrate how a combination of detailed aquifer characterization and solute transport modeling can be used to maintain acceptable salinity of recovered water for its intended use along with natural treatment of recharge water. ASTR can be used to reduce treatment costs and take advantage of aquifers with impaired water quality that might locally not be otherwise beneficially used.     

A New Operational Paradigm for Small‐Scale ASR in Saline Aquifers

A new operational paradigm is presented for small‐scale aquifer storage and recovery systems (ASR) in saline aquifers. Regular ASR is often not feasible for small‐scale storage in saline aquifers because fresh water floats to the top of the aquifer where it is unrecoverable. In the new paradigm, fresh water storage is combined with salt water extraction from below the fresh water cone. The salt water extraction counteracts the buoyancy due to the density difference between fresh water and salt water, thus preventing the fresh water from floating up. The proposed approach is applied to assess the feasibility of ASR for the seasonal storage of fresh water produced by desalination plants in tourist resorts along the Egyptian Red Sea coast. In these situations, the continuous extraction of salt water can be used for desalination purposes. An analytical Dupuit solution is presented for the steady flow of salt water toward a well with a volume of fresh water floating on top of the cone of depression. The required salt water discharge for the storage of a given volume of fresh water can be computed with the analytical solution. Numerical modeling is applied to determine how the stored fresh water can be recovered. Three recovery approaches are examined. Fresh water recovery rates on the order of 70% are achievable when salt water is extracted in high volumes, subsurface impermeable barriers are constructed at a distance from the well, or several fresh water recovery drains are used. The effect of ambient flow and interruptions of salt water pumping on the recovery efficiency are reported.     

Denitrification in a Deep Basalt Aquifer: Implications for Aquifer Storage and Recovery

Aquifer storage and recovery (ASR) can provide a means of storing water for irrigation in agricultural areas where water availability is limited. A concern, however, is that the injected water may lead to a degradation of groundwater quality. In many agricultural areas, nitrate is a limiting factor. In the Umatilla Basin in north central Oregon, shallow alluvial groundwater with elevated nitrate‐nitrogen of <3 mg/L to >9 mg/L is injected into the Columbia River Basalt Group (CRBG), a transmissive confined aquifer(s) with low natural recharge rates. Once recovery of the injected water begins, however, NO₃‐N in the recovered water decreases quickly to <3 mg/L (Eaton et al. 2009), suggesting that NO₃‐N may not persist within the CRBG during ASR storage. In contrast to NO₃‐N, other constituents in the recovered water show little variation, inconsistent with migration or simple mixing as an explanation of the NO₃‐N decrease. Nitrogen isotopic ratios (δ¹⁵N) increase markedly, ranging from +3.5 to > +50, and correlate inversely with NO₃‐N concentrations. This variation occurs in <3 weeks and recovery of <10% of the originally injected volume. TOC is low in the basalt aquifer, averaging <1.5 mg/L, but high in the injected source water, averaging >3.0 mg/L. Similar to nitrate concentrations, TOC drops in the recovered water, consistent with this component contributing to the denitrification of nitrate during storage.     

Arsenic Control During Aquifer Storage Recovery Cycle Tests in the Floridan Aquifer

Implementation of aquifer storage recovery (ASR) for water resource management in Florida is impeded by arsenic mobilization. Arsenic, released by pyrite oxidation during the recharge phase, sometimes results in groundwater concentrations that exceed the 10 µg/L criterion defined in the Safe Drinking Water Act. ASR was proposed as a major storage component for the Comprehensive Everglades Restoration Plan (CERP), in which excess surface water is stored during the wet season, and then distributed during the dry season for ecosystem restoration. To evaluate ASR system performance for CERP goals, three cycle tests were conducted, with extensive water‐quality monitoring in the Upper Floridan Aquifer (UFA) at the Kissimmee River ASR (KRASR) pilot system. During each cycle test, redox evolution from sub‐oxic to sulfate‐reducing conditions occurs in the UFA storage zone, as indicated by decreasing Fe²⁺/H₂S mass ratios. Arsenic, released by pyrite oxidation during recharge, is sequestered during storage and recovery by co‐precipitation with iron sulfide. Mineral saturation indices indicate that amorphous iron oxide (a sorption surface for arsenic) is stable only during oxic and sub‐oxic conditions of the recharge phase, but iron sulfide (which co‐precipitates arsenic) is stable during the sulfate‐reducing conditions of the storage and recovery phases. Resultant arsenic concentrations in recovered water are below the 10 µg/L regulatory criterion during cycle tests 2 and 3. The arsenic sequestration process is appropriate for other ASR systems that recharge treated surface water into a sulfate‐reducing aquifer.     

Analytical solution of transient flow in a sloping soil layer with recharge

Abstract

An analytical solution of planar flow in a sloping soil layer described by the linearized extended Boussinesq equation is presented. The solution consists of the sum of steady-state and transient-series solutions, the latter in a separation-of-variables form, and can satisfy an arbitrary initial condition via collocation; this feature reduces the number of series terms, making the solution efficient. Key parameter is the dimensionless linearization depth η _o (R), R being the dimensionless recharge. The variable η _o (R), not the slope, characterizes the flow as kinematic or diffusive, and R ≈ 0.2 demarcates the two regimes. The transient series converges rapidly for large η _o (large R, near-diffusive flow) and slowly as η _o → 0 (kinematic flow). The quasi-steady (QS) state method of Verhoest & Troch is also analysed and it is shown that the QS depth profiles approximate the transient ones well, only if Δt exceeds a system-dependent transition time between flow states (possibly >>1 day). In an application example for a 30-day recharge series, the QS solution differs from the transient one by as much as 20% (RMSE = 15%), does not track recharge changes as well and fails to conserve mass.     

Boundary depletion rate and drawdown in leaky wedge‐shaped aquifers

This study presents analytical solutions of the three‐dimensional groundwater flow to a well in leaky confined and leaky water table wedge‐shaped aquifers. Leaky wedge‐shaped aquifers with and without storage in the aquitard are considered, and both transient and steady‐state drawdown solutions are derived. Unlike the previous solutions of the wedge‐shaped aquifers, the leakages from aquitard are considered in these solutions and unlike similar previous work for leaky aquifers, leakage from aquitards and from the water table are treated as the lower and upper boundary conditions. A special form of finite Fourier transforms is used to transform the z‐coordinate in deriving the solutions. The leakage induced by a partially penetrating pumping well in a wedge‐shaped aquifer depends on aquitard hydraulic parameters, the wedge‐shaped aquifer parameters, as well as the pumping well parameters. We calculate lateral boundary dimensionless flux at a representative line and investigate its sensitivity to the aquitard hydraulic parameters. We also investigate the effects of wedge angle, partial penetration, screen location and piezometer location on the steady‐state dimensionless drawdown for different leakage parameters. Results of our study are presented in the form of dimensionless flux‐dimensionless time and dimensionless drawdown‐leakage parameter type curves. The results are useful for evaluating the relative role of lateral wedge boundaries and leakage source on flow in wedge‐shaped aquifers. This is very useful for water management problems and for assessing groundwater pollution. The presented analytical solutions can also be used in parameter identification and in calculating stream depletion rate and volume. Copyright © 2011 John Wiley & Sons, Ltd.     

Implications of rate-limited mass transfer for aquifer storage and recovery

Pressure to decrease reliance on surface water storage has led to increased interest in aquifer storage and recovery (ASR) systems. Recovery efficiency, which is the ratio of the volume of recovered water that meets a predefined standard to total volume of injected fluid, is a common criterion of ASR viability. Recovery efficiency can be degraded by a number of physical and geochemical processes, including rate-limited mass transfer (RLMT), which describes the exchange of solutes between mobile and immobile pore fluids. RLMT may control transport behavior that cannot be explained by advection and dispersion. We present data from a pilot-scale ASR study in Charleston, South Carolina, and develop a three-dimensional finite-difference model to evaluate the impact of RLMT processes on ASR efficiency. The modeling shows that RLMT can explain a rebound in salinity during fresh water storage in a brackish aquifer. Multicycle model results show low efficiencies over one to three ASR cycles due to RLMT degrading water quality during storage; efficiencies can evolve and improve markedly, however, over multiple cycles, even exceeding efficiencies generated by advection-dispersion only models. For an idealized ASR model where RLMT is active, our simulations show a discrete range of diffusive length scales over which the viability of ASR schemes in brackish aquifers would be hindered.     

Optimizing Multiwell Aquifer Storage and Recovery Systems for Energy Use and Recovery Efficiency

As more aquifer storage and recovery (ASR) systems are employed for management of water resources, the skillful operation of multiwell ASR systems has become very important to improve their performance. In this study, we developed MODFLOW and MT3DMS models to simulate a multiwell ASR system in a synthetic aquifer to assess effects of hydrogeological and operational factors on the performance of the multiwell ASR system. We evaluated a simplified (dual well) ASR system in comparison with complex system (three-, four-, five-, and seven-well systems). Recovery and energy efficiencies were calculated using the model simulations. Factors such as higher hydraulic conductivity and longitudinal dispersivity significantly reduced the recovery and energy efficiencies of the system. In contrast, increasing the volume of recharged water increased the recovery efficiency; however, the energy efficiency was reduced. Recovery and energy efficiencies also plummet when there is an increase in the underlying regional gradient and the designed storage duration. Operating the system multiple times can yield higher volume of potable water, but the energy efficiency may not vary significantly after the second operating cycle. Single-well systems and multiwell systems exhibit similar responses to changes in physical factors, although operational factors have a more pronounced effect on the multiwell systems. One of the major findings was that fewer wells in a multiwell ASR system can yield higher volume of potable water and better output with respect to the electrical power being consumed. The results provide design engineers with guidelines for optimizing performance of the multiwell ASR systems.     

The effect of heterogeneity on the character of density-driven natural convection of CO2 overlying a brine layer

The efficiency of mixing in density-driven natural-convection is largely governed by the aquifer permeability, which is heterogeneous in practice. The character (fingering, stable mixing or channeling) of flow-driven mixing processes depends primarily on the permeability heterogeneity character of the aquifer, i.e., on its degree of permeability variance (Dykstra-Parsons coefficient) and the correlation length. Here we follow the ideas of Waggoner et al. (1992) [13] to identify different flow regimes of a density-driven natural convection flow by numerical simulation. Heterogeneous fields are generated with the spectral method of Shinozuka and Jan (1972) [13], because the method allows the use of power-law variograms. In this paper, we extended the classification of Waggoner et al. (1992) [13] for the natural convection phenomenon, which can be used as a tool in selecting optimal fields with maximum transfer rates of CO₂ into water. We observe from our simulations that the rate of mass transfer of CO₂ into water is higher for heterogeneous media.     

Study of the feasibility of an aquifer storage and recovery system in a deep aquifer in Belgium

Abstract

The feasibility of aquifer storage and recovery (ASR) was tested in a deep aquifer near Koksijde, Belgium. To achieve this, oxic drinking water was injected into a deep aquifer (the Tienen Formation) that contains anoxic brackish water. The hydraulic properties of the aquifer were determined using a step-drawdown test. Chemical processes caused by the injection of the water were studied by two push—pull tests. The step-drawdown test was interpreted by means of an inverse numerical model, resulting in a transmissivity of 3.38 m²/d and a well loss coefficient of 0.00038 d²/m⁵. The push—pull tests identified mixing between the injection and pristine waters, and cation exchange, as the major processes determining the quality of the recovered water. Mobilization of DOC, aerobic respiration, denitrification and mobilization of phosphate were also observed.     

Solutions for Non‐Darcian Flow to an Extended Well in Fractured Rock

We have investigated non‐Darcian flow to a vertical fracture represented as an extended well using a linearization procedure and a finite difference method in this study. Approximate analytical solutions have been obtained with and without the consideration of fracture storage based on the linearization procedure. A numerical solution for such a non‐Darcian flow case has also been obtained with a finite difference method. We have compared the numerical solution with the approximate analytical solutions obtained by the linearization method and the Boltzmann transform. The results indicate that the linearized solution agrees generally well with the numerical solution at late times, and underestimates the dimensionless drawdown at early times, no matter if the fracture storage is considered or not. When the fracture storage is excluded, the Boltzmann transform solution overestimates the dimensionless drawdown during the entire pumping period. The dimensionless drawdowns in the fracture with fracture storage for different values of dimensionless non‐Darcian hydraulic conductivity β approach the same asymptotic value at early times. A larger β value results in a smaller dimensionless drawdown in both the fracture and the aquifer when the fracture storage is included. The dimensionless drawdown is approximately proportional to the square root of the dimensionless time at late times.     

Mixing at the interface between fresh and salt waters in 3D steady flow with application to a pumping well in a coastal aquifer

We consider 3D steady flow of fresh water over a salt water body in a confined aquifer of constant thickness D, with application to a pumping well in a coastal aquifer. With neglect of mixing, a sharp interface separates the two fluid bodies and an existing analytical solution, based on the Dupuit assumption, is adopted. The aim is to solve for the mixing between the fresh and salt waters for α_T/D  1 (α_T transverse dispersivity), as field studies indicate that α_T = O(10⁻³ − 10⁻² m). The mixing zone around the interface is narrow and solutions by existing codes experience numerical difficulties. The problem is solved by the boundary layer (BL) approximation, extending a method, applied previously to two-dimensional flows. The BL equations of variable-density flow are solved by using the Von Karman integral method, to determine the BL thickness and the rate of entrainment of salt water along the interface. Application to the pumping well problem yields the salinity of the pumped water, as function of the parameters of the problem (well discharge, seaward discharge, well distance from the coast and density difference).