The performance of managed aquifer recharge (MAR) systems is highly dependent upon local hydrogeology, which controls the movement and mixing of stored water and fluid–rock interactions, which can impact recharged water quality. The leading edge of MAR technology is the integration of data obtained using conventional and advanced aquifer characterization technologies into groundwater models that have improved predictive capabilities. Borehole and surface geophysical technologies and geostatistical and stochastic modeling methods, in particular, offer opportunities for improved aquifer characterization and modeling. The objective is to develop more accurate groundwater models that can be used as site-screening tools to identify locations and aquifers that have the greatest potential for successful implementation of MAR and to evaluate various design and operational options to find optimal local solutions.

INTRODUCTION

Fundamental water management challenges, in many areas, lie in the lack of sufficient storage to balance temporal variations in supply and demand and the need for cost-effective water treatment technologies that are correctly sized for local communities. Managed aquifer recharge (MAR) is defined as the intentional banking and treatment of water in aquifers (Dillon 2005). The term MAR was introduced as an alternative to ‘artificial recharge’, which has the connotation that the use of the water was, in some way, unnatural (Dillon 2005). MAR can be divided into technologies concerned primarily with increasing the volume of water stored in an aquifer and those in which solute transport, mixing and fluid–rock interaction, and contaminant attenuation are primary concerns. Examples of the latter are soil-aquifer treatment, riverbank filtration (RBF), salinity barrier, aquifer storage transfer and recovery, and aquifer storage and recovery (ASR) systems that store freshwater in aquifers that contain brackish, saline, or otherwise poor-quality water.

MAR consists of a diversity of technologies that have a long history. MAR systems vary greatly in their scale and technical sophistication. For example, an RBF system has been documented to have been constructed in Glasgow, Scotland in 1810 (Huisman & Olsthoorn 1983; Ray et al. 2002). The first successful test of an ASR system using a brackish aquifer appears to have been performed at Camp Peary, near Williamsburg, VA, USA, in 1946 (Cederstrom 1957). Small-scale, technically unsophisticated systems have long histories and can still have great socioeconomic benefits to local populations. For example, rainwater harvesting and MAR have been continuously practiced in India since at least the 3rd millennium bc (Agarwal & Narain 1997). The most common MAR methods used in India are check dams and storage reservoirs constructed on ephemeral streams. ASR is also performed in India by allowing stormwater to simply flow into existing tube wells in the rainy season after perhaps some basic filtration (Taneja & Khepar 1996; Malik et al. 2002; Goyal et al. 2008). The increase in aquifer recharge and rainwater harvesting in India over the past three to four decades is of such an extent that it has been called a ‘groundwater recharge movement’ (Sakthivadivel 2007).

The implementation of MAR is accelerating because MAR technologies are often less expensive than conventional water storage and treatment options and are often more environmentally friendly in terms of lesser impacts on surface environments and carbon footprints. MAR is thus a leading-edge technology (Dillon et al. 2010) whose implementation may be improved through the application of leading-edge technologies. Much is already understood about MAR. Basic system construction and operational procedures, such as the causes and management of clogging, are well established (Huisman & Olsthoorn 1983; Pyne 2005; Maliva & Missimer 2010, 2012).

A basic challenge for the successful implementation of MAR is that it relies upon heterogeneous natural systems (i.e., aquifers) for which there is inherently a significant element of uncertainty. Aquifer heterogeneity has a much greater effect on flow velocity and travel time than on aquifer heads or water levels (Poeter & Gaylord 1990; de Marsily et al. 2005). MAR projects in which water quality is a concern require a much greater understanding of aquifer heterogeneity than projects concerned only with the volume of water in storage. Flow systems dominated by secondary porosity features may have more rapid, geographically extensive, and unpredictable flows of recharged water, which could adversely impact salinity barrier systems and MAR systems that utilize aquifer filtration and retention time to improve water quality. Similarly, failure to adequately investigate fluid–rock interaction potential resulted in the surprise of significant arsenic and metals leaching into water stored in ASR systems in Florida and elsewhere (Arthur et al. 2002, 2005; Mirecki 2004).

The leading-edge technologies associated with MAR thus involve methods to better characterize utilized aquifers and simulate the flow, mixing, and reactions of recharged water. Objectives for the application of more advanced technologies include the following:

  • determination of the optimal system type and location to meet local water storage and treatment needs;

  • more accurate prediction of the performance of systems prior to incurring the costs of construction of a full-scale system;

  • accurate evaluation (prediction) of the potential and management options for biogeochemical processes that can either improve or degrade the quality of stored water;

  • optimization of system design and operation.

The purpose of this paper is both to provide examples where leading-edge technologies have been applied to MAR projects and to present opportunities where such technologies can be employed in the future. It is recognized that opportunities to apply advanced technologies depend upon project scale and budget. Hence, the paper focuses on large-scale systems.

AQUIFER CHARACTERIZATION TOOLS

Borehole geophysical logging

Borehole geophysical logging is a standard tool for aquifer characterization, but there has been a great deal of variation in the technical sophistication of its applications. Conventional borehole geophysical logs, such as natural gamma-ray, spontaneous potential, resistivity (dual induction), sonic, temperature, and impeller (spinner) flowmeter logs, can provide qualitative and, in some applications, quantitative information on lithology, porosity, and the location of flow zones. Flowmeter logs can be used to locate flow and relatively confining (low flow) zones and obtain quantitative data on coarse-scale aquifer heterogeneity (e.g., Molz et al. 1989, 1994; Paillet & Crowder 1996). Such data can be used in groundwater modeling to divide an aquifer into several zones based on transmissivity, which results in improved prediction of the rate of movement and horizontal extent of recharged water (e.g., Pavelic et al. 2006). High-resolution flowmeter logging, for example using a heat pulse tool, can also be used to evaluate fine-scale variations in hydraulic conductivity with depth.

Advanced borehole geophysical logs, such as nuclear magnetic resonance (NMR) and microresistivity imaging, are leading-edge technologies for providing high-resolution data on porosity type and distribution and hydraulic conductivity, especially when used in conjunction with conventional aquifer testing data such as from aquifer pumping and packer tests (Maliva et al. 2009a). NMR logs provide information on pore-size distribution, which can be processed to obtain estimates of hydraulic conductivity and permeability. Microresistivity imaging logs provide data on the type and location of secondary porosity features, which may dominate groundwater flow. The combination of NMR logs and microresistivity imaging logs (Fullbore Formation MicroImager; FMI™) were run on a number of MAR and brackish-groundwater supply exploratory wells in Florida to evaluate aquifer heterogeneity (Figure 1). The advanced geophysical logging was more cost-effective than continuous-core and core-plug analyses.

Figure 1

NMR and FMI run on an ASR exploratory well in South Florida (from Maliva et al. 2009b). The top of the planned storage zone (horizontal black line) corresponds to a down-hole transition from microporous limestones to more permeable limestone with inter-granular porosity.

Figure 1

NMR and FMI run on an ASR exploratory well in South Florida (from Maliva et al. 2009b). The top of the planned storage zone (horizontal black line) corresponds to a down-hole transition from microporous limestones to more permeable limestone with inter-granular porosity.

Electrical and electromagnetic surface geophysical methods

Surface geophysical methods have much lower vertical resolution than borehole-based methods, but have the great advantage of being able to cost-effectively provide a much greater spatial coverage. Surface geophysics thus complements, but does not replace, borehole-based field measurements. Field-based (e.g., well) data are needed to calibrate and ground-truth the surface geophysical data. Electrical and electromagnetic methods, such as direct current (DC) resistivity and time-domain electromagnetic (TDEM) surveys, have been used as screening tools to map locations most favorable for MAR. For example, data from a previously performed TDEM survey was used by the author as a screening tool to identify areas in central Florida (USA) in which the target aquifer likely has suitable, mildly brackish water for an ASR system. Surface geophysical methods can also be effective in locating and mapping aquifer and clay-rich confining strata and bedrock. Airborne electromagnetic surveys have been used in Australia as an element of an investigation to develop a conceptual hydrogeological model of the study area and for mapping and assessment of potential MAR targets (Brodie et al. 2013).

Time series of continuous vertical electrical sounding (CVES) and generated electrical resistivity tomograms can be used to map the movement of saline-water interfaces provided that there is a significant resistivity contrast. For example, a combination of CVES, seismic refraction surveys, and TDEM were used to map the position of the saline-water interface in Wadi Al Hawasinah in northern Oman (Abdalla et al. 2010). A seaward movement of the interface between 2002 and 2007 was detected, which was attributed to increased recharge induced by a wadi dam and regulation of groundwater pumping.

Electrical and electromagnetic methods are sensitive to anthropogenic cultural interference (e.g., roads, power lines, fences, buildings, pipelines, and railroads) and may not be appropriate for developed areas. Inverse modeling is recommended to determine whether or not the survey goals are achievable. For example, Minsley et al. (2011) performed an assessment of the applications of hydrogeophysical methods for ASR based on a proposed system in Kuwait. A synthetic aquifer model was used to simulate the advection and dispersion of injected freshwater. Bulk resistivity values were calculated from the modeled salinity distribution, which was then used to estimate DC resistivity responses. The results demonstrated that resistivity-based methods (DC resistivity and TDEM) are sensitive enough to detect the freshwater plume.

Microgravity

Time series of relative microgravity measurements can be used to detect changes in the mass of the underlying strata such as those caused by changes in the mass of water resulting from the filling and draining of porosity. Microgravity can thus be used to map increases in water volume resulting from the recharge of unconfined aquifers (Figure 2). Microgravity is not appropriate for confined aquifers in which changes in water levels are not associated with a corresponding change in the mass of water. Time series of microgravity measurements were successfully used to map changes in the volume of stored water in several MAR systems in the western USA (Howle et al. 2002; Davis et al. 2008; Chapman et al. 2008). The primary advantage of using microgravity to monitor water level changes is that it is substantially less expensive to monitor numerous microgravity stations than to install observation wells, particularly if the water table is deep and thus well construction costs are great. The major limitations of the technique are that microgravity-determined water levels are less accurate than standard well measurements and continuous monitoring is not practical. Microgravity monitoring is most effective when used in conjunction with monitoring wells. Microgravity and water level data can also be used to determine the specific yield of aquifers.

Figure 2

Gravity change (microgals) from channel (Rillito River, Arizona) recharge measured between December 1992 and March 1993 (from Pool & Schmidt 1997).

Figure 2

Gravity change (microgals) from channel (Rillito River, Arizona) recharge measured between December 1992 and March 1993 (from Pool & Schmidt 1997).

RECHARGED WATER QUALITY

MAR as a wastewater treatment tool

There is now an extensive body of research that demonstrates that MAR can result in a marked improvement in the quality of water through the removal of pathogens, a variety of organic compounds, and other constituents of concern, which was reviewed by Maliva & Missimer (2010, 2012). The attenuation rates of various microorganisms and chemicals are known to depend upon numerous variables. As a generalization, MAR systems are most effective at improving water quality when the water to be treated passes through different physical and chemical environments, for example, the vadose and phreatic zones, and oxic and anoxic zones. A given constituent may tend to be removed in a specific environment. By having recharged water pass through multiple environments, there is an increased likelihood that most of the constituents of concern will pass through an environment favorable for their removal. However, some trace organic compounds, such as the antiepileptic drugs carbamazepine and primidone, have been shown to persist for long periods in both oxic and anoxic conditions (Drewes et al. 2002, 2003; Drewes 2009; Benotti & Snyder 2009).

Aquifer treatment can be used for pretreatment, to buffer variations in source water quality prior to engineered treatments, and for post-treatment as an additional barrier and to increase public acceptance of wastewater reuse (Dillon et al. 2008). Despite the clear advantages of using MAR as a final treatment (polishing step) for wastewater treatment, it has not been implemented globally to near the extent warranted by its benefits (Missimer et al. 2011). One hindrance to its implementation has been regulatory concerns over causing groundwater contamination (Missimer et al. 2011). MAR of treated wastewater ties into the issue of indirect potable reuse, both planned (intentional) and unplanned. In some jurisdictions, MAR projects involving treated wastewater or other impaired waters are either not permitted or face excessive regulatory requirements over concerns of indirect potable reuse, irrespective of the actual potential for indirect potable reuse to actually occur.

The state-of-the-art and opportunities lie in methods to cost-effectively predict, monitor, and manage the fate and transport of recharged wastewater, so as to ensure that public health is protected. Accurate prediction of the movement and mixing of recharged wastewater requires more detailed and accurate aquifer characterization and modeling. Potential health impacts of MAR of wastewater can be evaluated by quantitative microbial risk assessment (QMRA) procedures, whose application to wastewater and stormwater reuse is presented by Mara et al. (2007). QMRA, and similar risk assessment procedures used for chemical contaminants, are based on evaluations of the likelihood of exposure (frequency and extent) to a contaminant and the dose response for the contaminant. The dose-response function may consider only a single-exposure event (static assessment) or may incorporate complexity associated with indirect exposure, multiple routes of exposure, varying susceptibility to infection or illness, and other factors impacting responses to exposure (dynamic assessment; Asano et al. 2007). Toze et al. (2010) and Page et al. (2010) have documented performance of static QMRAs for wastewater and stormwater MAR projects in Australia.

Although methods are available to quantitatively assess the risks associated with MAR of wastewater, they are data intensive. The risks associated with given MAR systems can be assessed using system-specific aquifer residence times and pathogen and chemical decay or attenuation rates. However, determination of site-specific decay and attenuation rates, likely exposure frequencies and volumes, and dose-response functions are beyond the scope of most MAR project teams and need to be determined in a government or university research environment (Maliva & Missimer 2012). A more practical solution for the lack of site-specific data is the use of existing conservative ‘worst-case’ rates for similar types of systems and hydrogeological and geochemical settings to provide a safety factor. A key question is what is the appropriate level of risk assessment required for a project, which should depend upon project site and resources, the realistic potential for human exposure, and the overall benefits of the system and the costs and practicalities of other water treatment and supply options.

GEOCHEMICAL EVALUATION

Arsenic and metals leaching evaluation and management

MAR commonly involves the introduction into aquifers of water that is in chemical disequilibrium with native groundwater and aquifer minerals. Fluid–rock interactions may thus occur, which can adversely impact the quality of stored water and the hydraulic properties of the aquifer by clogging pores through mineral precipitation and alteration processes. Arsenic leaching into stored water is a particularly important process in some ASR systems in Florida and elsewhere as it has resulted in exceedances of applicable groundwater and drinking-water quality standards. The potential for the operation of an ASR system to cause what is considered by regulatory agencies to be groundwater contamination has greatly dampened enthusiasm for ASR despite its compelling water-resources management benefits.

As reviewed by Maliva & Missimer (2010) and Mirecki et al. (2103), the primary cause of the arsenic leaching in ASR systems is the oxidation of trace arsenic-bearing iron sulfide minerals (e.g., arsenopyrite) caused by the introduction of dissolved oxygen (DO) in the recharged water. The released arsenic may then be sequestered by sorption onto or co-precipitation with iron oxyhydroxide or iron sulfide minerals. The fate of leached arsenic depends upon the overall chemistry of the recharged water. The reestablishment of reducing conditions can result in the reductive dissolution of iron oxyhydroxides and release of sorbed arsenic, as reported in the Bolivar, Australia ASR system (Vanderzalm et al. 2011).

Recharged water with relatively high dissolved organic carbon and dissolved iron concentrations and low nitrate and manganese concentrations (e.g., some reclaimed and surface waters) is conducive for the reestablishment of sulfate-reducing conditions and the sequestration of arsenic by co-precipitation with iron sulfide (Mirecki et al. 2013). In aquifers in which aerobic conditions are maintained, experimental results suggest that iron oxyhydroxide may subsequently be transformed into more stable iron oxides (hematite), which could lead to irreversible sorption of arsenic (Neil et al. 2014).

Although the general processes involved in arsenic and metals leaching in MAR systems are understood, there is a need for a means to accurately evaluate the potential for significant arsenic and metals leaching to occur in advance of full-scale system construction and to develop a means to prevent or manage leaching. The interaction of the recharged water with native groundwater and aquifer minerals can be evaluated using aqueous geochemical modeling software (e.g., PHREEQC) provided that high-quality and appropriately comprehensive water chemistry and aquifer mineralogy data are collected. However, arsenic and metals leaching appears to be caused by reactions involving trace redox reactive minerals, such as pyrite, which may be missed in bulk rock analyses. Reactive solute-transport modeling using codes such as PHAST and PHT3D can be used to simulate fluid–rock interaction during groundwater flow. Reactive solute-transport modeling has been used to provide insights into the geochemical and biogeochemical processes active in MAR systems (e.g., Saaltink et al. 2003; Prommer & Stuyfzand 2005, 2006; Greskowiak et al. 2005). Reactive solute-transport modeling has had more limited value to date in predicting future water quality mainly due to uncertainty in the values of the numerous flow, transport, and reaction parameters (i.e., non-uniqueness of solutions).

The Florida Geological Survey developed a bench-scale leaching and sequential extraction procedure to approximate conditions in the field and determine the likely source of the leachable arsenic and metals (Arthur et al. 2005, 2007). The developed leaching procedure was employed in the exploratory phase of two ASR projects in central Florida, USA, using core samples. Arsenic and metals leaching at concentrations exceeding the groundwater quality standard of 10 μg/L occurred during the bench-scale tests and subsequently occurred during operational testing of the systems. The bench-scale leaching procedure could thus determine whether or not significant leaching will likely occur, but is not accurate as to actual concentrations. The inaccuracy is presumably due to the laboratory procedures not exactly matching actual field conditions and perhaps, more importantly, because the testing is performed on a very limited number of grab samples (six per system) rather than the entire storage zone. Bench-scale testing may also be used to evaluate the efficacy of various pretreatment options to prevent arsenic and metals leaching such as DO removal techniques.

Another promising option to evaluate fluid–rock interaction early in the implementation of MAR systems is single-well tracer tests (also referred to as single-well pulse and push–pull tests). The tests consist of the injection of a known volume of water having a known tracer concentration into a well and then pumping the well to recover the tracer. The volume and concentrations of the tracer, and other constituents of concern in the recovered water, are recorded as a function of time and recovered water volume. Single-well tracer tests have been used to estimate distribution coefficients (Pickens et al. 1981) and the occurrence and rate of various physical and microbial reactions (Istok et al. 1997; Haggerty et al. 1998). Fluid–rock interaction can be evaluated by injecting water with a composition similar to that of the planned recharged water with or without a non-reactive tracer added. Single-well tracer tests have the great advantage of being performed under in situ conditions and could be performed early in a MAR investigation using an exploratory well or even an existing monitoring well. The tests can also provide information on aquifer transport properties. Single-well tracer tests can be used to evaluate the longitudinal dispersivity of an aquifer, which is a key parameter in solute transport (Gelhar & Collins 1971; Pickens & Grisak 1981). However, dispersivity values are scale dependent and values obtained from the injection and recovery of a small volume of water may not be accurately scaled up (Domenico & Schwartz 1998).

The state-of-the-art with respect to the geochemistry of MAR systems ultimately lies in developing means to effectively manage geochemical processes. Arsenic and metals leaching in ASR systems can be prevented by reducing the DO concentration of recharged water so that pyrite dissolution and associated arsenic releases do not occur. Potential DO removal methods for ASR systems have been reviewed by ASR Systems (2006), CH2M Hill (2007), Bell et al. (2009), and Maliva & Missimer (2010). Operational testing at the Bradenton, FL, USA, ASR system demonstrated that oxygen removal, using a membrane degasification system, and dechlorimination using sodium bisulfite, could successfully reduce leached arsenic concentrations to below the applicable 10 μg/L standard (Norton et al. 2012). Arsenic leaching in ASR systems naturally decreases over time due to there being a finite amount of leachable arsenic. A system-specific technical and economic question is whether to prevent arsenic leaching by pretreatment or to allow concentrations to naturally decline over time.

MODELING OF MAR SYSTEMS

Simulation of aquifer heterogeneity

Groundwater flow, solute transport, and geochemical modeling serve two main purposes in MAR projects. The development of calibrated models can provide valuable insights into the hydrogeological and geochemical processes that will control system performance. Inverse modeling in itself is a valuable aquifer characterization method. However, the ultimate goal is development of models that can accurately predict the movement, mixing, and reaction of recharged water. The models can be used for the optimization of system design and operation and for economic analyses.

The sophistication of the modeling required for MAR projects varies depending upon project scale (and thus budget) and whether or not solute transport is of primary concern. Depending upon the local hydrogeology and project information requirements, a single- or equivalent-continuum modeling approach employing widely used codes, such as MODFLOW, MT3DMS, and SEAWAT, may be sufficient. Where flow is dominated by secondary porosity, a dual-continuum or perhaps a discrete-fracture-network approach may be needed. The state-of-the-art is advanced codes, such as Eclipse™, which are capable of simulating complex density-dependent dual-porosity systems. Eclipse can simulate the conduction of fluids in both the matrix and fractures. Guo et al. (2014) demonstrated the use of the Eclipse code to simulate density-dependent movement of water in fractures in a hypothetical dual-porosity ASR system.

Developing models that adequately represent spatial heterogeneity in aquifer properties remains a challenge in groundwater investigations, in general, principally because of a paucity of data in the horizontal direction. In systems with a relatively low degree of heterogeneity relative to the area affected by the MAR system, basic geostatistical techniques, such as ordinary kriging, may be satisfactorily used to estimate parameter values between measurement points. Parameter values may then be adjusted during model calibration either manually or using automated calibration software such as PEST.

In areas with a high degree of heterogeneity, accurate modeling of groundwater flow and solute transport necessitates fuller incorporation of the heterogeneity into the model. Indicator statistics are particularly well-suited to characterizing and modeling sedimentary architecture where the strata can be subdivided into sedimentary facies (hydrofacies) that can be coded using integers (Proce et al. 2004). For example, binary indicator kriging may be used to map the distribution of high- and low-permeability strata. Either a single model can be developed in which a cutoff value is used to determine which facies is assigned to a grid cell (Johnson & Dreiss 1989) or a stochastic approach can be used to generate numerous realizations. Several types of geostatistical methods are available for simulating facies and hydrofacies distribution, including:

  • sequential indicator simulation, which is based on indicator variograms for the different facies;

  • transition probability geostatistical simulation, which utilizes the Markov chain method;

  • multiple point simulation, which is a pixel-based simulation technique that utilizes training images of the heterogeneity to be reproduced rather than two-point statistics (variograms).

Once the facies distribution is modeled, hydraulic parameters are assigned to the geological model cells, and the data can then be up-scaled to the groundwater model grid. Parameter values may subsequently be adjusted during model calibration.

Probabilistic evaluation of performance and economics of ASR systems

One hindrance to the implementation of MAR that has been expressed by decision-makers is the absence of a rigorous economic case for the investment. Evaluation of the economic feasibility of MAR projects should be based on cost–benefit analysis (CBA), as is the case for other investments in water and wastewater infrastructure. It is recognized that some of the benefits of MAR projects may be difficult to quantify in monetary terms (e.g., ecological system maintenance and restoration, and human health benefits). Any objective evaluation of the historic performance of MAR systems would indicate that the results have been mixed. While there has been many successful projects, there have also been projects that have failed to meet expectations, such as with a less than anticipated recovery efficiency for an ASR system. Underperformance of MAR systems has often been due to site-specific hydrogeological conditions turning out to be less favorable than anticipated.

State-of-the-art evaluation of prospective MAR systems, therefore, needs to include an uncertainty analysis, which evaluates the probability of different system outcomes. Typically, economic evaluation of MAR systems, where they are indeed actually performed, have been based on the assumption of a specific, usually successful outcome. The possibility of a less than successful outcome is not explicitly considered, which biases the CBA in favor of the project. Methods are available for evaluating parameter uncertainty, such as Monte Carlo techniques. The problem of conceptual model uncertainty is more intractable, yet is often more important than parameter uncertainty. If the conceptual model is wrong, then the predictions based on the model are likely also wrong (Bredehoeft 2005).

The expected net benefit analysis method (Boardman et al. 1996) is suitable for more robust CBAs of MAR projects (Maliva 2014). The expected net benefit involves consideration of the probabilities, costs, and benefits of all possible outcomes or contingencies. For a hypothetical ASR project, possible scenarios or outcomes, based on probabilistic modeling results, might be that there is a 70% percent probability that the system would perform as desired and provide the target water volume at an 80% or greater recovery efficiency, a 20% probability that the recovery efficiency would be between 50% and 80%, and a 10% probability that recovery efficiency would be less than 50%. The expected net benefit would be the probability-weighted sum of the net benefits of the three considered outcomes. The CBA might alternatively, or in addition, consider the probability of and costs associated with the need for different pretreatment options.

CONCLUSIONS

New technologies and methods are continually being developed in the groundwater field to characterize and model aquifers. The leading edge for technologies utilized in MAR does not necessarily involve development of new discipline-specific tools, but rather finding the optimal implementation of existing technologies. MAR projects have budgetary constraints, and advanced technologies typically can only be employed where it can be demonstrated that their benefits exceed their costs. Professionals involved in MAR projects, therefore, need to be familiar with the range of technologies available, the information they provide, and their limitations, benefits, and costs. Target employment of technology offers the opportunity to improve the implementation of MAR and thus obtain greater water resource management and socioeconomic benefits.

REFERENCES

REFERENCES
Abdalla
O. A. E.
Ali
M.
Al-Higgi
K.
Al-Zidi
H.
El-Hussain
I.
Al-Hinai
S.
2010
Rate of seawater intrusion estimated by geophysical methods in an arid region: Al Khabourah, Oman
.
Hydrogeol. J.
18
,
1437
1445
.
Agarwal
A.
Narain
S.
1997
Dying Wisdom. Rise, Fall and Potential of India
’s Traditional Water Harvesting Systems.
India Centre for Science and Environment
,
New Delhi, India
.
Arthur
J. D.
Dabous
A. A.
Cowart
J. B.
2002
Mobilization of arsenic and other trace elements during aquifer storage and recovery, southwest Florida
. In:
U.S. Geological Survey Artificial Recharge Workshop Proceedings, April 2–4, 2002, Sacramento, California
(
Aiken
G. R.
Kuniansky
E. K.
, eds),
US Geological Survey Open-File Report 02–89
, pp.
47
50
.
Arthur
J. D.
Dabous
A. A.
Cowart
J. B.
2005
Water-rock geochemical considerations for aquifer storage and recovery: Florida case studies
. In:
Underground Injection Science and Technology
(
Tsang
C.-F.
Apps
J. A.
, eds),
Elsevier, Amsterdam
, pp.
65
77
.
Arthur
J. D.
Dabous
A. A.
Fischler
C.
2007
Aquifer storage and recovery in Florida: geochemical assessment of potential storage zones
. In:
Management of Aquifer Recharge for Sustainability. Proceedings of the 6th International Symposium on Managed Aquifer Recharge of Groundwater
(
Fox
P.
, ed.),
Acacia Publishing, Phoenix
,
USA
, pp.
185
197
.
Asano
T.
Burton
F. L.
Leverenz
H. L.
Tsuchihashi
R.
Tchobanoglous
G.
2007
Wastewater Reuse. Issues, Technologies and Applications
.
McGraw-Hill
,
New York
,
USA
.
ASR Systems, LLC
2006
Evaluation of Arsenic Mobilization Processes Occurring during Aquifer Storage Recovery Activities. Task 2 – Technical Memorandum, Literature Review, Arsenic Mobilization Processes during ASR Operations.
ASR Systems, LLC, Report to Southwest Florida Water Management District
,
Brooksville, Florida
,
USA
.
Bell
K. Y.
Wiseman
L.
Turner
L. A.
2009
Designing pretreatment to control arsenic leaching in ASR facilities
.
J. AWWA
101
(
6
),
74
84
.
Boardman
A.
Greenberg
D.
Vining
A.
Weimer
D.
1996
Cost-Benefit Analysis: Concepts and Practice
.
Prentice Hall
,
Upper Saddle River, NJ
,
USA
.
Brodie
R. S.
Lawrie
K.
Dillon
P.
Tan
K.
Gibson
D.
Clarke
J.
Magee
J.
Christensen
N. B.
Halas
L.
Somerville
P.
Gow
L.
Apps
H.
Vanderzalm
J.
Page
D.
Brodie
R. C.
Hostetler
S.
Abraham
J.
Miotlinski
K.
2013
An integrated approach to developing hydrogeological conceptual models to underpin assessment of managed aquifer recharge options, Darling floodplain, N.S.W., Australia
. In:
Proc. 8th International Symposium on Managed Aquifer Recharge
,
Beijing
,
China
.
Cederstrom
D. J.
,
1957
Geology and Ground-water Resources of the York-James Peninsula
,
US Geol. Surv. Water Supply Pap.
1361
.
CH2M Hill
2007
Arsenic Mobilization in Two Suwanee Limestone Aquifer Storage Recovery Systems
.
Final Technical Report to the Southwest Florida Water Management District
,
Brooksville, Florida
,
USA
.
de Marsily
G.
Delay
F.
Goncalvès
J.
Renard
Ph.
Teles
V.
Violette
S.
2005
Dealing with spatial heterogeneity
.
Hydrogeol. J.
13
,
161
183
.
Dillon
P.
2005
Future management of aquifer recharge
.
Hydrogeol. J., 13
(
1
),
313
316
.
Dillon
P.
Page
D.
Vanderzalm
J.
Pavelic
P.
Toze
S.
Bekele
E.
Sudhu
J.
Prommer
H.
Higginson
S.
Regel
R.
Rinck-Pfeiffer
S.
Purdie
M.
Pitman
C.
Wintgens
T.
2008
A critical evaluation of combined engineered and aquifer treatment systems in water recycling
.
Water Sci. Technol.
57
(
5
),
753
762
.
Dillon
P.
Toze
S.
Page
D.
Vanderzalm
J.
Bekele
E.
Sidhu
J.
Rinck-Pfeiffer
S.
2010
Managed aquifer recharge: rediscovering nature as a leading edge technology
.
Water Sci. Technol.
62
,
2338
2345
.
Domenico
P. A.
Schwartz
F. W.
1998
Physical and Chemical Hydrogeology
, 2nd edn.
John Wiley & Sons
,
New York
,
USA
.
Drewes
J. E.
Heberer
T.
Reddersen
K.
2002
Fate of pharmaceuticals during indirect potable reuse
.
Water Sci. Technol.
46
(
3
),
73
80
.
Drewes
J. E.
Heberer
T.
Rauch
T.
Reddersen
K.
2003
Fate of pharmaceuticals during groundwater recharge
.
Ground Water Monit. Remediation
23
(
3
),
64
72
.
Gelhar
L. W.
Collins
M. A.
1971
General analysis of longitudinal dispersion in nonuniform flow
.
Water Resour. Res.
7
,
1511
1521
.
Goyal
V.
Jhorar
B. S.
Malik
R. S.
Streck
T.
2008
Performance evaluation of aquifer storage recovery for conjunctive water management as influenced by buffer storage volume and storage time
.
Curr. Sci.
94
,
465
472
.
Greskowiak
J.
Prommer
H.
Vanderzalm
J.
Pavelic
P.
Dillon
P.
2005
Modelling of carbon cycles and biogeochemical changes during injection and recovery of reclaimed water at Bolivar, South Australia
.
Water Resour. Res.
31
,
W10418
.
Guo
W.
Coulibaly
K.
Maliva
R. G.
2014
Simulated effects of aquifer heterogeneity on ASR system performance
.
Environ. Earth Sci.
10.1007/s12665-014-3822-4
.
Howle
J. F.
Phillips
S. P.
Ikehara
M. E.
2002
Estimating water-table change using microgravity surveys during an ASR program in Lancaster, California
. In:
Management of Aquifer Recharge for Sustainability
(
Dillon
P.
, ed.),
Swets & Zeitlinger, Lisse
,
The Netherlands
, pp.
269
272
.
Huisman
L.
Olsthoorn
T. N.
1983
Artificial Groundwater Recharge
.
Pitman Advanced Publishing
,
Boston
,
USA
.
Istok
J. D.
Humphrey
M. D.
Schroth
M. H.
Hyman
M. R.
O'Reilly
K. T.
1997
Single well, ‘push-pull’ test for in situ determination of microbial activities
.
Ground Water
35
,
619
631
.
Johnson
N. M.
Dreiss
S. J.
1989
Hydrostratigraphic interpretation using indicator geostatistics
.
Water Resour. Res
.
25
(
12
),
2501
2510
.
Malik
R. S.
Jhorar
B. S.
Jhorar
R. K.
Streck
T.
Richter
J.
2002
Long-term successful operation of existing brackish cavity wells for ASR to improve quality for irrigation by Indian farmers
. In:
Management of Aquifer Recharge for Sustainability
, (
Dillon
P.
, ed.),
A. A. Balkema
,
Lisse, The Netherlands
, pp.
465
468
.
Maliva
R. G.
Missimer
T. M.
Clayton
E. A.
Dickson
J. A. D.
2009b
Diagenesis and porosity preservation in Eocene microporous limestones, South Florida, USA
.
Sediment. Geol.
217
,
85
94
.
Maliva
R. G.
Missimer
T. M.
2010
Aquifer Storage and Recovery and Managed Aquifer Recharge Using Wells: Planning, Hydrogeology, Design, and Operation
.
Schlumberger Water Services
,
Houston, Texas
,
USA
.
Maliva
R. G.
Missimer
T. M.
2012
Arid Lands Water Evaluation and Management
.
Springer
,
Berlin, Germany
.
Minsley
B. J.
Ajo-Franklin
J.
Mukhopadhyay
A.
Morgan
F. D.
2011
Hydrogeophysical methods for analyzing aquifer storage and recovery systems
.
Ground Water
49
,
250
269
.
Mirecki
J. E.
2004
Water–Quality Changes during Cycle Tests at Aquifer Storage Recovery (ASR) Systems of South Florida
.
Engineer Research and Development Center Report ERDC/EL TR-04–8 US Army Corps of Engineers, Jacksonville, FL
.
Mirecki
J. E.
Bennett
M. W.
López-Baláez
M. C.
2013
Arsenic control during aquifer storage recovery cycle tests in the Floridan Aquifer
.
Groundwater
51
,
539
549
.
Molz
F. J.
Boman
G. K.
Young
S. C.
Waldrop
W. R.
1994
Borehole flowmeters: field applications and data analysis
.
J. Hydrol.
163
(
3–4
),
347
371
.
Molz
F. J.
Morin
R. H.
Hess
A. E.
Melville
J. G.
Güven
O.
1989
The impeller meter for measuring aquifer permeability variations: evaluation and comparison with other tests
.
Water Resour. Res.
25
,
1677
1683
.
Norton
S.
Ellison
D.
Kohn
S.
2012
Minimizing arsenic mobilization during aquifer storage and recovery by source water degasification
. Paper presented at
2012 NGWA Ground Water Summit. May 6–10, 2012
,
Garden Grove, CA
,
USA
(
abstract)
.
Page
D.
Dillon
P.
Vanderzalm
J.
Toze
S.
Sidhu
J.
Barry
K.
Levett
K.
Kremer
S.
Regel
R.
2010
Risk assessment for aquifer storage transfer and recovery with urban stormwater for producing water for a potable quality.
J. Environ. Qual.
39
,
2029
2039
.
Pickens
J. F.
Grisak
G. E.
1981
Scale-dependent dispersion in a stratified granular aquifer
.
Water Resour.
Res.
17
,
1191
1211
.
Pickens
J. F.
Jackson
R. E.
Inch
K. J.
Merritt
W. F.
1981
Measurement of distribution coefficients using a radial injection dual-tracer test
.
Water Resour. Res.
17
,
529
544
.
Pool
D. R.
Schmidt
W.
1997
Measurement of Ground-Water Storage Change and Specific Yields Using the Temporal-Gravity Method near Rillito Creek, Tucson, Arizona
.
Water Resources Investigations Report
97
4125
,
US Geological Survey
,
Tucson, AZ
.
Proce
C. J.
Ritzi
R. W.
Dominic
D. F.
Dai
X.
2004
Modelling multiscale heterogeneity and aquifer interconnectivity
.
Ground Water
42
,
658
670
.
Prommer
H.
Stuyfzand
P. J.
2005
PHT3D modelling of water quality changes during deep well injection at Someren
. In:
Water Quality Improvements during Aquifer Storage and Recovery. Volume 1: Water Quality Improvement Processes
(
Dillon
P.
Toze
S.
, eds),
AWWA Research Foundation Report 91056F
,
Denver, Colorado
,
USA
, pp.
229
244
.
Prommer
H.
Stuyfzand
P.
2006
On the use of reactive multicomponent transport modelling for assessing water quality changes during managed aquifer recharge
. In:
Recharge Systems for Protecting and Enhancing Groundwater Resources, Proceedings of the 5th International Symposium on Management of Aquifer Recharge, Berlin, Germany
,
11–16
June 2005
,
UNESCO
,
Paris
, pp.
415
420
.
Pyne
R. D. G.
2005
Aquifer Storage Recovery: A Guide to Groundwater Recharge through Wells
.
ASR Systems
,
Gainesville, FL
.
Ray
C.
Schubert
J.
Linsky
R. B.
Melin
G.
2002
Introduction
. In:
Riverbank Filtration. Improving Source-Water Quality
(
Ray
C.
Melin
G.
Linsky
R. B.
, eds),
Kluwer Academic Press
,
Dordrecht
,
The Netherlands
, pp.
1
18
.
Saaltink
M. W.
Ayora
C.
Stuyfzand
P. J.
Timmer
H.
2003
Analysis of a deep well recharge experiment by calibrating a reactive transport model with field data
.
J. Contamin. Hydrol.
65
(
1
),
1
18
.
Sakthivadivel
R.
,
2007
,
The groundwater recharge movement in India
. In:
The Agricultural Groundwater Revolution: Opportunities and Threats to Development
(
Giordano
M.
Villholth
K. G.
, eds),
CAB International
,
Wallingford
,
UK
, pp.
195
210
.
Vanderzalm
J. L.
Dillon
P. J.
Barry
K. E.
Miotlinski
K.
Kirby
J. K.
Le Gal La Salle
C.
2011
Arsenic mobility and impact on recovered water quality during aquifer storage and recovery using reclaimed water in a carbonate aquifer
.
Appl. Geochem.
26
,
1946
1955
.