Prospects for Managed Underground Storage of Recoverable Water (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

6 Project Development, Monitoring and Management The general components of a managed underground storage (MUS) system and some of the broad decisions driving the selection of the type of system to be developed are addressed in Chapter 2. Hydrogeological and groundwater engi- neering issues; water quality and geochemical issues; and legal, institutional, and economic issues related to the development, operation, and management of MUS systems are addressed in Chapters 3, 4, and 5, respectively. This chapter provides an opportunity to address some of the issues that face project propo- nents (and opponents) and managers of MUS systems that have not been dis- cussed earlier. It should be noted up front that the entire project development, monitoring, and management process is likely to be more successful if a broad approach to water resources planning is taken. As described in Chapter 7, an integrated strategy in which all measures for managing water scarcity are considered care- fully in a systematic way is highly recommended. Such a strategy would ideally include metrics to allow comparisons of MUS to other water supply and storage options. It would also take into account watershed-wide water quality manage- ment, including control of stormwater, combined sewer overflows, septic tank leaks, agricultural runoff, and coastal water quality issues. The formation of regional water authorities can be a useful step in understanding and incorporat- ing planning in the context of the regional and state water supply and water management options. MUS is not always the solution—or even part of the solution—to water sup- ply and storage challenges. The methodologies described in this chapter assume that all of the options for a given area have been evaluated thoroughly. The list of the geological, hydrologic, geochemical, geotechnical, environmental, public health, water availability, economic, regulatory, and other issues that need to be considered to evaluate the potential suitability of MUS is long; many of these issues have been described in Chapters 3, 4, and 5. No comprehensive MUS site comparison exists; the closest—but an extensive review of aquifer storage and recovery (ASR) planning methodologies—was performed by Brown (2005). The chapter is divided into two main parts. The first part is a summary of the major steps—beginning to end—in the selection, development, management, and oversight of MUS systems. The second part provides an expanded discus- sion of four key issues that a manager, operator, or regulator may need to con- sider. These issues are clogging, monitoring (including the use of surrogates or indicators), public perception, and financing. 223 

224 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER FROM FEASIBILITY TO CLOSURE: STAGES OF AN MUS PROJECT This is a summary of key steps or stages of project development, rather than a “how-to” manual. Many fine references exist on the practical issues of ASR, surface infiltration, and artificial recharge in general. Pyne (2005) discusses in detail the building blocks of an ASR program, including feasibility studies, pilot testing, well design and equipment, plugging issues, and water quality chal- lenges. Brown (2005) does an excellent job of summarizing existing frame- works for both brackish and freshwater ASR projects. He presents a much more detailed framework than the one used in this chapter, including flow charts for the various steps. The Environmental and Water Resources Institute (EWRI, 2001) offers an- other useful and practical guidebook. The Water Environment Federation and the AWWA (1998) provide an overview of planning indirect potable reuse, health and regulatory considerations, treatment technologies, system reliability, and public information outreach programs. Segalen et al. (2005) give a fine summary on that topic. Several recent scientific meeting proceedings are an excellent source for case studies; notable among these are those of the forth and fifth International Symposia on Management of Aquifer Recharge (Dillon, 2002; UNESCO, 2006) and U.S. Geological Survey (USGS) Artificial Recharge Workshop Proceedings (Aiken and Kuniansky, 2002). Finally, Ruetten (2001) deals in depth with pub- lic perception issues. As noted above, there are many different ways to organize an MUS project from beginning to end. The following list is modified only slightly from EWRI/ASCE (2001) Standard Guidelines for Artificial Recharge of Ground Water: • Phase I: Preliminary activities (also called feasibility evaluation), in- cluding data collection; assessment of regulatory, legal, political, and economic feasibility; and conceptual planning; this phase may also in- volve environmental assessment and public involvement • Phase II: Field investigations and testing of pilot sites • Phase III: Design (revision of the conceptual design to reflect results of investigations) • Phase IV: Construction and start-up (systems may require cycle testing to develop recharge systems) • Phase V: Operation and maintenance • Phase VI: Project review and adaptive management • Phase VII: Closure 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 225 Phase I: Feasibility Evaluation The components typically addressed in a feasibility evaluation for an MUS system are summarized as follows: • Site assessment—land availability, ownership, proximity to water sources, suitability of aquifers, preexisting groundwater quality, prox- imity to water use; • Legal and regulatory issues—water rights for source water and owner- ship of water in storage, antidegradation requirements, monitoring re- quirements; • Financial considerations—cost of land, treatment and conveyance fa- cilities, access to capital, availability of grants, loans or other subsidies, costs of operation, and ability to obtain revenue from water users; • Purpose of MUS (duration expected) —seasonal or long-term storage, drought protection, meet summer peak demands, or evening out sur- face water treatment plant flows; • Source water availability and quality—Raw surface water, stormwater, recycled water or treated drinking water, high suspended solids load or clear water, quality comparison to existing groundwater; total dissolved solids (TDS), pH, redox potential, trace elements, microbial quality, trace organic contaminants; • Treatment needs and existing capacity—desilting to prevent clogging by suspended or settleable solids, pH adjustments or other methods to reduce adverse changes in metal concentrations, reverse osmosis for removal of salts and trace organics, and advanced oxidation for destruc- tion of trace organics; unused capacity of existing treatment facilities; • Capture and conveyance facilities needs—stormwater capture im- poundments, temporary storage prior to recharge, pipelines or channels for conveyance, pumping needs; and • Public perception—outreach program needs, existing perceptions of groundwater quality, different outreach needs depending on source wa- ter, especially reclaimed water, different outreach depending on percep- tions of water resource needs and impacts of the MUS project, (e.g., drought protection vs. growth inducement). As noted later in this chap- ter, this step is sometimes left for later in the planning process, to the detriment of all concerned parties. The feasibility of recharging water into the aquifer is perhaps the key issue to explore in a feasibility evaluation. If water cannot be recharged in sufficient quantities, the project will not be possible. For projects that are considering MUS in a geologic area with no prior managed recharge, the feasibility evalua- tion should analyze available data that can be used to evaluate the feasibility of recharge. It should also propose a course of investigation to assess recharge 

226 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER techniques. A phased course of investigation that spans literature review to field-scale testing is recommended. Field investigations and pilot testing for MUS in areas without prior MUS projects will typically have a greater scope and magnitude than in areas where MUS projects previously have been imple- mented. Site selection may also be a major issue. Land ownership or nearby land uses may severely constrain the potential locations of recharge facilities. If, however, there is some degree of freedom in site selection, a location suitability assessment may be useful. An example of one kind of suitability assessment is given for ASR in support of the Florida Everglades restoration in Brown (2005). There, an index was used in which potential sites were ranked by weighting eight factors, including such disparate issues as ecological suitability, existing uses of the aquifer, groundwater quality, road density, access to power lines, and aquifer transmissivity. Given the large number of site-specific variables that must be considered at the beginning of the process, decision trees may be a useful aid. An example of such a tool, in this case for selection of the most appropriate groundwater re- charge technology (which overlaps Phases I and II), is presented in Figure 6-1. As shown in the figure, the first critical question is what aquifer is being considered for use in the MUS system. If a confined aquifer is being consid- ered, then direct recharge using wells is the only feasible alternative. Direct recharge may include either single-use recharge wells or the dual-purpose wells used in ASR systems. If the goal of a groundwater recharge project is to provide short-term storage and the water must be recovered quickly, ASR systems might be the only feasible alternative. If an existing distribution and well system may be utilized as part of an ASR system, then dual-purpose wells might be the best choice. If an unconfined aquifer is being considered, there are no constraints on the choice of recharge method. For unconfined aquifers, one of the constraining variables on the choice of technologies is the depth to groundwater. As depth to groundwater increases, the cost of recharge wells increases. Depths ranging from 100 to 200 m have often been found to be a cutoff point at which recharge wells become more costly than surface recharge systems; however, site-specific factors such as land and drilling costs may make a different depth more appropriate (Bouwer, 2002). Therefore, the effect of depth should be evaluated for each situation, and land availability might make surface recharge basins more economical even with shallow groundwater depths. In situations where the depth to water is greater than 100 m, particular attention should be given to evaluating whether water recharged at the surface will flow down to the aquifer where it is planned to be withdrawn. If the evaluation indicates that a significant portion of the water recharged will not reach the aquifer in a reasonable amount of time, surface re- charge may not be effective. This factor may be particularly important in arid areas where the depth to groundwater is commonly great. The feasibility evaluation should also assess potential impacts on adjacent landowners, including consideration of potential changes in groundwater levels. 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 227 Is the aquifer confined or unconfined? If Confined If Unconfined Direct injection must be used No constraint on recharge method What is the depth to groundwater? If less than 100-200 m, direct injection may be cost competitive with surface If greater than 100-200 m, surface recharge recharge should be considered Is cost-effective land available at an appropriate location? If No, If Yes, Vadose zone injection wells Surface recharge basins may may be appropriate be appropriate FIGURE 6-1 Sample decision tree for selection of groundwater recharge method. Another constraining question that must be considered is the availability of appropriate land. Land price and availability are key considerations. In addition, the location of the land and the cost of the distribution system to deliver water to the land are also important. Such impacts could occur whether groundwater levels rise or fall, affecting ex- isting water users or environmental resources (see Chapter 3). Projects should be designed to minimize the degree of impact to the maximum extent practical. For certain types of water sources such as stormwater, land might be required not only for groundwater recharge, but also as an aid to catch and store the wa- ter. Governance is another important issue. Regional water authorities can be useful in implementing projects but can also create complicated interagency issues that can get in the way of solving water supply issues. In particular, the relationship between a proposed regional authority and participating local agen- cies needs to be carefully considered. For example, if a regional water authority is proposed, consideration should be given to the implications of a publicly elected local board delegating its responsibility for water supply development to an interagency regional water authority whose governing board may be ap- pointed rather than elected. In such a case, local agencies may have concerns about a non-elected board setting policy. 

228 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER The regional authority may also have different interests than the local agen- cies. The manner in which control and ownership of a project is assigned, rights to storage space and stored water are allotted, and related ownership and control issues are decided are important considerations. Thus, regional water authorities may have a useful role in implementing new projects, but the relationships among these authorities and local agencies should be assessed carefully. As listed above, many other factors are involved in feasibility analysis, and this is simply an illustration of an approach that can be taken to guide one through what in the end is a very long and complex decision-making process. Phase II: Components The following text summarizes components typically addressed in an MUS pilot program. Note that these components vary slightly between surface spreading and well recharge systems. Monitoring issues are mentioned only briefly here; they are discussed in more depth later in the chapter. • Pilot program goals—evaluate hydrogeology, including permeability, water quality, pressure or water table gradient, travel time, projected hydrogeological effects. • Soil borings—evaluate the lithology, depth to groundwater, confining zones, aquifer materials, and aquifer properties; in addition to soil bor- ings, core testing, split sampling, and side wall testing should be con- sidered as some of the viable options to better determine aquifer char- acteristics. A lithologic log should be prepared and representative cut- tings should be preserved when conducting soil borings or well drilling. Sieve analyses and mineral classification should be completed for the sediments recovered during drilling. • Tracer studies—employ intrinsic or added tracers, tracer injection, monitoring locations. • Monitoring—assess influent water quality, cycle testing for ASR, downgradient testing, native or background characterization (or upgra- dient). • Valve testing (for ASR) —this should often be a preliminary review, with more detailed assessment after the range of flow rates is better de- fined. • Surface infiltration rates (for surface spreading) —the infiltration rate as a function of time is a key parameter to determine feasibility. • Injection well recharge rates (for ASR) —where the aquifer or associ- ated aquitards are fractured or highly compressible, high injection pres- sures can cause hydrofracturing, and alternation of recharge and dis- charge can lead to irreversible subsidence. 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 229 • Pretreatment evaluation—determine the need for silt removal, pH ad- justment, disinfection, et cetera. • Modeling evaluation—predict impacts of supplemental recharge on gradients and travel time. • Risk Assessment—see Monitoring section of this chapter for an exam- ple. In the field testing program, it is important that the design of the investiga- tion consider carefully the time and spatial scales used in the field testing com- pared to the full-scale MUS project. In most cases, the time and spatial scales used in the field investigations and pilot testing will be smaller than those of the full-scale project. Because of this, it is important to consider how field investi- gation results can be extrapolated for the full-scale MUS project. The manner in which the field data will be extrapolated should be an important consideration in the design of the field investigations and pilot testing. Key issues to consider in the design of the field testing include the following: • How can the recharge water be conveyed to the field test site? • How can the testing program be designed to collect enough data to characterize the spatial heterogeneity of the aquifer materials? • What kinds of water quality transformations, positive or negative, dur- ing storage are likely? Adsorption, filtration, biodegradation, precipita- tion, dissolution, oxidation and reduction, and formation of disinfection by-products must all be considered. Changes in redox conditions as a result of recharging water are important to assess. When a new source of water is recharged, changes in redox conditions may occur that af- fect water quality. Particular attention should be given to redox poten- tial changes and potential changes in concentrations of metals. • What depths of the aquifer(s) should be monitored? Different aquifer depths may have different hydraulic properties and different water quality, thereby requiring monitoring at multiple depths. • Is the well going to be subjected to corrosive environments? What type of well design should be used? What type of well screen and casing materials should be used? What size screen, how many screen inter- vals, and what formation intervals should be targeted? • What are the rates of geochemical reactions that could occur between the recharge water, the aquifer materials, and the native groundwater? • Over what length of time should the pilot testing take place to account for these geochemical reaction rates? • In the case of brackish or saline aquifers, what are the salinity and the density of the groundwater, and what is the dispersivity of the aquifer at various distances from the well (see Brown, 2005)? • How much field data have to be collected to build an accurate model of the full-scale MUS project? 

230 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER • Overall, how can the testing program be designed so that it provides sufficient information to select the method of recharge, if more than one method of recharge is under consideration? Overall, the availability of sufficient data for model input is important if the model is to be calibrated with an acceptable level of accuracy. Data should be collected early in the MUS model development and a sensitivity analysis per- formed to help determine any gaps that may be present in the data. In many cases, multiple, deep monitoring wells should be constructed to obtain the nec- essary data required for model calibration. However, the cost involved in doing so is quite high. Instead of executing several pilot studies in different areas, an alternative approach is to select one central area and perform detailed studies based on data from a well-equipped monitoring well. The drawback to this method is that the results may not be applicable across a large area if the system is heterogeneous. There is no easy answer to this issue. It must be kept in mind that recharged water may mobilize some metals, in- organic compounds, and organic compounds bound in the aquifer material. The degree of mobilization will be impacted by a variety of factors, including pH, alkalinity, total dissolved solids, temperature, and the concentration of anions and cations already in solution. During operation of the pilot plant, core sam- ples of representative subsurface strata taken in the vicinity of the boreholes could be obtained for use in soil or rock column tests to determine the likelihood of leaching or mobilization of chemicals. Native groundwater also could be collected during operation of the pilot plant for testing to determine any effects of mixing recharged water with groundwater. A decision tree example for Phase II, in this case for treatment requirements prior to groundwater recharge to maintain hydraulic capacity, is shown in Figure 6-2. The type of treatment depends primarily on the choice of groundwater re- charge method. For recharge basins, the removal of inorganic suspended solids is the major concern for maintaining infiltration rates. When vadose zone or recharge wells are used, infiltration rates may be reduced by suspended solids (especially critical for the former since there is no mechanism to backwash sol- ids from the wells), biological fouling, or gas entrapment. Gas entrapment is not listed in the decision tree since the solution is often operational and is not a treatment issue. Clogging is a dominant operational issue in MUS and is discussed in more detail both in Phase V (operation and maintenance) and in a separate section later in this chapter. Phase III: Design The following text summarizes components typically addressed in the de- sign phase of an MUS system. 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 231 What type of recharge method is used? Direct Injection Wells or Va- Recharge Basins dose Zone Injection Wells Will Suspended Solids Will Inorganic Suspended Solids Reduce Infiltration Rates? Reduce Infiltration Rates? If yes, filtration will be required Will organic carbon result in If yes, then stilling basins, filtration or biofouling? some other form of suspended solids removal should be done If yes, then removal of organic carbon or addition of a disinfectant residual will be necessary FIGURE 6-2 Decision tree for treatment requirements before recharge with respect to hydraulics. Pre-design Results from the field investigations and pilot testing should be evaluated thoroughly and utilized to prepare the project description. Assessment of the depth to groundwater is important to determine whether a vacuum situation could occur. The potential occurrence of a vacuum condition and air entrain- ment are important when considering the design of recharge wells. The method of recharge and the amount of water to be recharged over de- fined time scales (e.g., per month, per year), should be defined in the project description. Average and maximum recharge amounts should be estimated. The aquifer zones where the water will be stored and the duration of stor- age project description should also be defined. Potential losses of water should be estimated. The anticipated quality of the water after recovery should be specified. The type of water treatment before recharge or after recovery should be de- termined if treatment is required. Any unresolved issues that need to be resolved prior to beginning design should be identified. 

232 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER Additional field investigations should be conducted before commencing de- sign if the unresolved issues are fundamental—for example, if there is signifi- cant uncertainty regarding how much water can be recharged or whether adverse water quality changes will occur after recharge. Potential interference among two or more recharge (or production) wells should be carefully investigated in design of an MUS system. This is an ideal area to use models, which depend on the quality of data on hydraulic conductivity, porosity, recharge or leakage, and so forth. Such models may provide fairly good guidance for well placement and discharge limits. For example, Finch and Livingston (1997) discuss the use of a groundwater model for managed underground storage using the La Luz well field in Alamogordo, New Mexico. The pre-design phase typically concludes with preparation of a pre-design report that includes the project description and a description of the primary fea- tures of the project, including the facilities to be constructed and the estimated project cost. Phased Design Design should be implemented in phases to allow adding progressive levels of detail; as progressive levels of detail are added, the design should be reviewed rigorously and compared to the project description to verify that the design will allow implementation of the project. The design and operation of the MUS project should be integrated into the overall water management strategy for the area or region. Rigorous review of the design at its initial stages is as important as, and in some respects more important than, review of the design at the final stage. Re- view in the early stages is critical to minimize subsequent redesign or inclusion of unnecessary or ineffective project elements. If wells are to be constructed, careful consideration should be given to their design and construction. Installation of additional well sounding tubes and gravel feed tubes should be considered for gravel-packed wells. The drilling method should be matched to aquifer conditions. A proper, engineered drilling fluid program is important for drilling methods using drilling fluids. The manner in which well development is conducted after the well has been installed is im- portant to its productivity. Well components such as casing, screen materials, submersible pumps, shafts, and bearings should be considered carefully with respect to corrosion issues. Epoxy-lined, stainless steel, and other materials should be evaluated as needed to minimize corrosion damage. Engineering design of the system that needs to be constructed and specifica- tion of the operations and maintenance activities and their cost should be in- cluded. An operations and maintenance program should be developed. This should include a written operations and maintenance manual and should de- scribe the water quality monitoring needed to assess the quality of the recovered water and any changes in water quality during recharge or storage that are im- 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 233 portant to understand. The operations and maintenance program should also account for the upkeep of recharge facilities and the cost of equipment replace- ment. A key issue, if the project includes a water quality treatment system, is to include sufficient flexibility in the system so that it can be modified if conditions change and additional treatment is needed. If a system is designed to treat the water prior to recharge or after recovery, it needs to have flexibility to be modi- fied if the influent water quality varies beyond the range anticipated during de- sign or if new conditions occur. Flexibility should include the ability to add additional treatment units without having to remove existing units. Availability of space and adaptable piping and instrumentation are important factors to con- sider. If the project is to be constructed in phases, it may be beneficial to construct the backbone water distribution, electrical, and instrumentation systems for the ultimate size of the project during one of the initial project phases. This can minimize the cost of adding future phases to the project. Phase IV: Construction and Start-up The following text summarizes components typically addressed in the con- struction and start-up phase of an MUS system. • Construction—MUS facilities are sometimes constructed in environ- mentally sensitive areas; it is important that construction be imple- mented in accordance with relevant environmental laws and regula- tions. • Commissioning—Generally speaking, this is the process of testing in- dividual components that have been constructed to verify that they function properly; individual components are also tested to verify that they operate in conjunction with related components. • Startup—After commissioning is completed, start-up is the process of operating the entire system to verify that performance criteria are met and the system operates reliably. If there is a permit with water quality limits or criteria, startup should include water quality monitoring to en- sure that the permit limits are satisfied. One key issue in the construction and start-up phase is to operate the system for sufficient time during start-up to ensure that it is reliably producing water according to permit conditions. If a construction project is near completion and over schedule, there may be pressure to shorten the commissioning and start-up phases to recover time in the schedule. In the event that the commissioning and start-up phases are not properly completed, there is an increased risk of produc- ing water that does not meet permit conditions, which can result in loss of public confidence in the project. 

234 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER Another important point is that the entire range of technical staff involved in planning and designing the project should remain involved with it. Since MUS projects often involve a range of technical disciplines, from water quality to engineering to geology, it is important to maintain the involvement of each discipline as the project moves forward. If changed conditions are encountered during construction, commissioning, or start-up, these changes should be com- municated to each discipline. Phase V: Operation and Maintenance Some specific operational challenges identified in MUS systems are dis- cussed below, with separate discussions for surface and well recharge systems. Surface Spreading Operational Challenges Surface spreading using recharge basins is the most common method for re- charging untreated surface water or reclaimed water into MUS systems. Re- moval of clogging material that retards percolation is the main maintenance ac- tivity with recharge basins. Without removal of the clogging material, recharge basins can rapidly foul and become much less effective for groundwater re- charge, as shown in Figure 6-3. The benthic clogging layer (BCL) can be a combination of inorganic and biological material. For systems recharging stormwater, the clogging material is typically composed of fine silts and clays that can form a layer plugging the surface of the recharge basin. Biological ma- terial, such as algae, bacteria, and organic detritus, can contribute to the clogging layer. A relatively thin layer (less than 2 cm) of fine (organic or inorganic) ma- terial is capable of significantly reducing percolation rates. Control of weeds and vectors is also an important function. Weeds and sur- face debris interfere with maintenance of recharge basins and become a visual blight. Insect vectors and nuisances such as midges (chironomids) create prob- lems in neighborhoods near surface spreading facilities. Control of both weeds and insects is an important part of operating spreading facilities. However, re- charge agencies need to be cautious to avoid the use of persistent chemicals that might affect recharge water quality and therefore should focus more on me- chanical methods and biological controls for weed and pest control. Water agencies use several different methods to clean recharge basins. The simplest systems involve disking or ripping of clogging material to restore per- colation capacity. This approach works best with shallow recharge basins that are routinely rotated between wetting and drying cycles and where silt loads are minimal. In these types of systems, biological clogging may predominate and the drying process is sometimes sufficient to restore percolation capacity by allowing the surface to crack and open up to the infiltration of water. 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 235 Just cleaned Clogged basin bottom FIGURE 6-3 Clogging layer in recharge basin operated in Orange County, California, adja- cent to cleaned portion of recharge basin where the clogging layer was removed. Photo courtesy of Adam Hutchinson, Orange County Water District. Where silts and clays are a more significant factor in clogging, ripping and disking can have the adverse effect of driving fine sediments deeper where they may contribute to a long-term decline in percolation capacity that is more diffi- cult to restore. For recharge basins where heavily silt laden stormwater is re- charged, percolation capacity can be reduced very rapidly (Figure 6-4). Draining and scraping these basins is necessary to remove the silts and clays and restore percolation rates. After basins are drained or pumped dry, bulldozers and scrap- ers may be used to push the clogging material up the slope to the shoreline where it can be hauled away. Since the scraping process is imprecise, much of the removed material is sand. The sand can be recovered and returned to the basin bottom after washing and separation from the finer silts and clays. This sand washing and recovery process adds to space requirements and the mainte- nance and operations costs of spreading facilities. With shallow basins, where depths are typically less than 10 feet, the drain- ing and scraping process can be done relatively quickly and the basins can be returned to service without significant downtime. With deep basins, sometimes more than 60 feet in depth, the draining and scraping process is more lengthy 

236 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER 10 9 8 7 6 Infiltration Rate (feet 5 per day) 4 3 2 1 0 0 10 20 30 40 50 60 70 80 Day FIGURE 6-4 Recharge rate data from Kraemer Basin (Orange County, California). The decrease in recharge rate is due to formation of a clogging layer on the basin bottom, pri- marily as a result of fine-grained sediment transported with the recharge water. SOURCE: Reprinted, with permission, from Greg Woodside, Orange County Water District. Copyright 2007 by Orange County Water District. and results in longer downtime, with the potential for greater water loss while the basin is out of service. Because percolation rates have slowed markedly by the time that cleaning is initiated, especially in the bottom area of a clogged ba- sin, the basin may not drain naturally and pumping to transfer the water to other recharge basins may be required. Wind agitation and varying water levels can provide a natural cleaning process for the sidewalls of some deeper recharge basins that are not amenable to more conventional cleaning processes. Abandoned gravel pits have been used in some areas for groundwater recharge, and very steep slopes preclude the use of surface scraping to remove clogging material. Drying of steep sidewalls can help restore percolation rates. Resting the sidewalls by lowering water levels works in the same way as the rotation and resting of shallow basins to restore recharge capacity. In deep basins, sidewall percolation may be much more sig- nificant than percolation through the basin bottom. While fine sediments tend to sink and clog the bottom most quickly, sidewall percolation can be sustained for much longer periods. 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 237 Because of the potential for greater sustainability of sidewall percolation, some recharge agencies have constructed recharge ponds that utilize a system of ridges with steep enough slopes to allow fine sediments to drop into the troughs while wind and wave action and water level changes naturally clean the slopes of the ridges. This approach helps to maintain percolation capacity for longer periods without draining and scraping to remove clogging material. The system also provides greater surface area for water to infiltrate. Other innovative approaches are being tested by groundwater management agencies, including the use of submersible devices that disturb the clogging layer and pump out the fine sediments, leaving behind the coarser sediments that allow percolation. The basin cleaning vehicle (BCV) developed by Orange County Water District is shown in Figure 6-5. This type of system allows clog- ging material to be removed without interrupting the percolation process and avoids the water loss and expense associated with the downtime for draining and cleaning deep basins. Since heavy silt loads from storm flows can overwhelm the system, draining the basin and rehabilitating the basin bottom are still re- quired on a periodic basis. FIGURE 6-5 Basin cleaning vehicle (BCV) developed by Orange County Water District, California. This type of system allows clogging material to be removed while avoiding the water loss and expense associated with the downtime for draining and cleaning basins. Photo courtesy of Greg Woodside, Orange County Water District. 

238 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER Since scraping the clogging material from the bottom of dried basins also results in the removal of clean sands, alternative approaches for removal of clogging material after drying are being evaluated. As clogged basins are dried, the layer of fine sediments on the surface cracks and curls up into chip-like structures. Certain types of beach cleaning equipment may offer the potential to remove only curled chips of clays and silts and leave the clean sands behind. The greatest operating expense for recharge basins is the cost of removing the clogging material that retards percolation. For recharge of stormwaters, ba- sin cleaning may be the only significant operating expense. Recharge basins are usually cleaned when percolation rates have declined to a point that groundwater cannot be effectively recharged. The overall capacity of the recharge facilities and the amount of water available for recharge will determine how quickly basin cleaning must be accomplished. In areas where seasonal storm flows can be anticipated, basin cleaning is generally done just prior to the storm season to facilitate capture of the maximum amount of stormwater. Well Recharge Operational Challenges Recharging with wells gives rise to specific challenges. One of the primary challenges is the rate at which wells can recharge and the decline in recharge rate through time at wells. Another challenge is controlling the rate of water flow into a well to prevent adverse flow conditions that will exacerbate the de- cline in the recharge rate. These issues are addressed in Pyne (2005), Segalen et al. (2005), and Brown (2005) and are summarized here: The drilling method appears to be important to well productivity in many cases. For example, Segalen et al. (2005) found that production wells drilled using cable tool tended to outperform reverse circulation rotary drilled wells in the same formation. They also found that wells drilled using biodegradable mud gave rise to less clogging than when bentonite-based mud was used in the same formation and that residual mud near the borehole wall greatly limited recharge capacity. Finally, they concluded that wire-wrapped screens and natural gravel pack wells significantly enhanced well production relative to slotted casing and emplaced gravel pack in the same formation. Clogging during recharge is as important for wells as it is for recharge ba- sins and can be more difficult to overcome. In wells, clogging may be caused by physical factors such as suspended sediment or air entrainment, chemical factors that involve precipitation on the well screen or in the formation next to the well, or biofilms. Clogging is discussed in detail later in the chapter. Cascading control is one of the more important components in recharge well design and operation. Cascading occurs when the water level in the re- charge piping does not rise to ground surface during recharge. Allowing water to cascade down the well can lead to significant plugging problems due to air entrainment in the storage zone and induced geochemical or bacterial activity (Pyne, 2005). Cascading can also cause structural problems due to cavitation 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 239 damage to pipes, valves, and fittings. To avoid these problems, water can be introduced into a well through the pump column, the annulus between the pump column and the casing, one or more injection tubes inside the casing, or some combination of these approaches. The following factors dictate selection among these alternatives (Pyne, 2005): casing diameter; static water level; type, size, and capacity of the pump; specific capacity and specific injectivity of the well, expected production rate; and range of recharge rates. A careful management and balancing of recharge and recovery rates is im- portant to the long term viability and integrity of the gravel pack. Compromising the gravel pack can result in unacceptable levels of fine sand or debris produc- tion, thereby increasing wear and tear of the components and limiting the life of the well. Corrosion of the wells can be a major concern under certain conditions. These include low pH or high dissolved oxygen; hydrogen sulfide, chloride, or other salts; carbon dioxide; or temperature (EWRI/ASCE, 2001). Finally, recovery efficiency—that is, the volume of water recovered as a percentage of volume recharged—is of critical interest, especially in areas where the aquifer salinity is fairly high or where the aquifer structure or geometry is complex. This topic is covered in detail in Chapter 3. Phase VI: Project Review and Adaptive Management Adaptive management is a key principle for the development, regulation, and operation of MUS systems. Assumptions made in the feasibility evaluation may have to be adjusted based on experiences during the pilot testing stage. New information acquired during operation and maintenance of MUS systems may result in continuing refinement and development of new approaches and technologies to minimize risk and increase the efficiency of recharge operations, improve water quality, or enhance the sustainability of the recharge and recov- ery of water stored underground. Such basic operational parameters as percola- tion rate (e.g., because of plugging of the aquifer), pretreatment (e.g., pH control to prevent aquifer pore space clogging from manganese release, and posttreat- ment (e.g., to prevent damage to water distribution systems or for aesthetics) may need adjustment. However, adaptive management has to function at more than just the opera- tional level. Even the regulation of MUS systems needs to evolve as more in- formation is developed about the effects of introduced water on underground systems. Both regulators and MUS project managers need to reevaluate on a regular basis the effectiveness of existing procedures and regulations designed to protect water resources in light of the performance of MUS systems. For exam- ple, adaptive management may lead to changes in permitting requirements. This could mean reducing requirements because no impacts have been seen or a con- taminant of concern has not been detected. It could also lead to new require- ments if our understanding of an MUS system’s dynamics has changed or be- 

240 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER cause new contaminants have been identified based on trends and results of monitoring. In some situations, establishment of an independent advisory panel can be useful to offer guidance and counsel regarding design, operation, maintenance, and monitoring strategies and parameters to ensure water quality integrity. It is best to have an independent third party administrate such a panel to ensure unbi- ased input and avoid conflict of interest implications. In addition to its role in optimizing operations, an independent panel can increase public acceptance of and confidence in the system (see “Public Perception and Involvement” later in this chapter). As an example, under an agreement with California’s Orange County Water District (OCWD), the National Water Research Institute appointed a panel of experts with various areas of expertise to assist OCWD in developing the Groundwater Replenishment (GWR) System program. The panel has met—and continues to meet—on a routine basis as the project has evolved. A report de- tailing the panel’s findings and recommendations is prepared after each meeting and sent to OCWD and the California Department of Health Services. In this case, the panel was a requirement in the draft California Department of Health Services groundwater recharge regulations because of the high percentage of reclaimed water proposed to be recharged and it became a requirement of the GWR System permit issued by the California Regional Water Quality Control Board, Santa Ana Region. A complementary exercise to the above would be a comprehensive status report approximately three to five years after the commencement of operations of an MUS project. Again, there should be considerable involvement of external assessors to contribute to the overall confidence of the community in the project. Phase VII: Closure Proper destruction of unused recharge wells or ASR wells is critical since abandoned wells in drinking water source aquifers can easily provide conduits for surface water or shallow groundwater contamination to reach deeper, nor- mally more protected groundwater. Most jurisdictions have standards for proper well destruction to prevent vulnerability to migration of contamination from poorer-quality zones to higher-quality groundwater zones. In ASR systems where freshwater has been injected into saline or brackish water aquifers to cre- ate a freshwater zone, abandoned ASR wells also offer the risk of carrying higher-salinity water into freshwater zones if not properly sealed and destroyed. Recharge basins or recharge ponds can readily be converted to other land uses if land values become too great to justify retaining large-scale spreading facilities. Ease of conversion will depend on the depth of the excavated ponds and the cost of fill to reestablish historical grades or elevations suitable for de- velopment. Subsurface recharge alternatives such as exfiltration galleries (analogous to leach lines for recharge water) or vadose zone wells may be em- 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 241 ployed to retain some recharge capacity on land that will be used for parking or landscaping. The following sections explore in more detail four key operational issues associated with MUS. These progress from a very practical issue (clogging) to a scientific and regulatory issue (monitoring and indicators), a societal issue (pub- lic perception), and finally to financial considerations. PREDICTION, REDUCTION, AND PREVENTION OF CLOGGING Prediction The broad spectrum of interrelated factors and the relatively frequent occur- rence of clogging render it an important consideration during MUS planning, design, testing, and operation. Clogging may be predicted by bench-scale test- ing, multiscale field testing, indices calculations, and modeling. In a recharge basin for example, one of the most common clogging potential estimates in- volves use of infiltrometers, which are installed in the field to measure local infiltration rates. Results of laboratory column studies are particularly useful as well. Types of empirical clogging potential methods, such as the parallel filter index (PFI) column study, are summarized in Table 6-1. More sophisticated column studies can be designed to assess interrelationships among all clogging factors (Rinck-Pfeiffer et al., 2000), such as the dynamics of precipitation and microbial activity. Limitations of the membrane filtration index (MFI), assimilable organic carbon content (AOC) and PFI methods have been described by Bouwer (2002): Experience has shown that MFI, AOC, and PFI are useful parameters for comparing relative clogging potentials of various waters, but that they cannot be used to predict clogging and declines in injection rates for actual recharge wells, which also depend on well construction and aquifer characteristics. Thus full- scale studies on recharge test wells are still necessary to determine feasibility and design and management criteria for operational recharge wells. Practical as- pects such as a varying flow in the water-supply pipes to the recharge project and associated possibility of fluctuating suspended-solids contents in the water also play a major role in well clogging. The suspended-solids fluctuations can be caused by formation of biofilms in the pipelines during periods of low flow, and by erosion of the biofilms during high flow. Treatment of the water at the recharge site to remove suspended solids before well injection might then be necessary. Biuk and Willemson (2002) also note lack of quantifiable reproducibility of results at the field scale; however, their preliminary study exhibits potential cali- bration of the MFI. By accounting for the mathematical relation between the MFI and aquifer media characteristics, a satisfactory correlation between ex- pected versus observed clogging rates is observed for a limited data set. 

242 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER TABLE 6-1 Selected Plugging Potential Empirical Methods Clogging Predictor Abbreviation Method Membrane filtration MFI Describes suspended solids captured via mi- 2 index crofilter in units of time per volume Assimilable organic AOC Based on microbial growth in terms of carbon carbon content concentrations Parallel filter index PFI Passing recharge water through columns filled with aquifer media, measured in flow rate per unit area Bypass filter test BFT Passing recharge water through spun polyes- ter cartridges while monitoring flow rates; al- lows for calculation of suspended solids SOURCE: Olsthoorn (1982); Bouwer (2002); Pyne (2005). Results of column studies, index calculations, and infiltrometer data can be scaled up to address field or operational conditions for recharge basins. How- ever, the transfer of these data to full scale may require an intermediate step to validate the data in terms of heterogeneity and temporal variations. Bouwer (2002) suggests that test basins on the order of 30 m × 30 m should be employed to address this concern. Numerical models can also be used to forecast clogging potential. A par- ticularly robust method, Easy Leacher® 4.6 (Stuyfzand, 2002), predicts the ac- cumulation rate and chemical composition of clogging sludge layers in recharge basins. While this application does not model the hydrologic clogging process, it does predict the rate of sludge accumulation and its composition. Moreover, the model allows for sensitivity analyses to assess optimal conditions to reduce sludge development. Because complexities of geochemical and microbiological processes are not fully understood and uncertainties exist when scaling up from laboratory to field scale, other numerical models exist to facilitate sustainable design and operation of MUS systems (e.g., CLOG; Perez-Paricio and Carera, 1998). Reduction and Prevention During MUS system design, clogging potential may be reduced by the addi- tion of pre-treatment systems to remove suspended solids, nutrients, or mi- crobes. In the case of a recharge basin, design of the basin floor, including sediment grain size and morphology (e.g., ridge and furrow; or flat surface), can also reduce compounding long-term effects of clogging. 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 243 Monitoring the effects of clogging is an important practice during MUS sys- tem operation. Of primary significance are observed reductions in infiltration or recharge rates. Other monitored parameters include water quality (e.g., dis- solved oxygen, pH, TDS) and rates of change in hydraulic head within recharge zone monitor wells. The bypass filter test (BFT) and PFI methods can provide early warning regarding the onset of clogging. In addition, video analysis of the borehole and well screen (i.e., biofilm or precipitate buildup) is also a useful method. MUS system maintenance can be categorized as either physical-mechanical or chemical. In recharge wells, physical maintenance practices often include backflushing (e.g., backwashing or redevelopment) or some other form of physical agitation to loosen and remove plugging materials such as (1) com- pressed air jetting, (2) controlled sonic blasting, and (3) pressurized CO2 injec- tion (liquid and/or gas). Other physical methods could include brushing or swabbing the screen. Backflushing practices can be optimized by establishing the relation between injection rates, total suspended solids, and backwashing frequency. It may be necessary to monitor the quality of the water produced during maintenance and construct storage systems to contain it. Physical main- tenance in a recharge basin generally includes breakup and/or removal of the low-permeability “cake” layer. Breakup practices include disc harrows or rotary tillers, or a “dry-and-crack” technique. Several methods exist with respect to cake removal, such as scraping; however, care must be taken to minimize com- paction of the basin floor. The goal of chemical additives is generally to dissolve clogging constitu- ents. Heat may be used to augment the process. With regard to biomass buildup, chlorine or related chemicals can be added to recharge wells; however, this is weighed against the potential for formation of disinfection by-products such as trihalomethanes. Adjustment of pH during recharge and borehole acidi- fication is among the techniques used to remove precipitates. In addition, these practices may be complemented by physical well agitation to optimize removal of clogging material. Flocculation and removal of swelling clays is also accom- plished by chemical additives. It is noteworthy that changes to the borehole environment with respect to acid-base or redox reactions may adversely affect the quality of the stored or recovered water. As such, awareness of the potential hydrogeochemical reactions and subsequent monitoring of constituents of con- cern is an important consideration. For more detail on MUS well clogging issues, texts such as Mansuy (1998), Bloetscher et al. (2005), and Pyne (2005) can be consulted. MONITORING ISSUES Monitoring is an integral part of MUS site selection, design, and operation. Successful MUS involves careful and thorough project-specific assessment that includes chemical and microbiological monitoring to document system perform- 

244 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER ance and evaluate the reliability of the process. There are several roles for monitoring, as outlined below: • Establish the feasibility of the site by characterizing the hydrogeology, and identify pertinent water quality issues. Knowledge from this phase of monitoring is used to develop a site conceptual model. • Obtain parameters for design and operation, such as recharge and ex- traction well placement, hydraulic capacity and recovery, and appropri- ate travel time or residence time. • Determine the need for pre- or posttreatment of the water, such as re- moval of particles and biodegradable organic matter from the source water or removal of excessive dissolved constituents in the extracted water. • Comply with regulatory requirements. • Document the performance to build trust with consumers and improve public perception. Monitoring provides an opportunity to become pro- active for emerging contaminants and issues. • Adjust system operation in the future in response to what has been learned from ongoing monitoring (i.e., part of adaptive management). A number of the roles for monitoring identified above are addressed else- where in this report. For example, the need for information on the hydro- geologic and water quality parameters is addressed in Chapters 3 and 4. The nature of the present regulatory framework is mentioned in Chapter 5. Earlier sections in this chapter cover some of the important factors in design and opera- tion and describe the benefits of adaptive management and risk management. The sections below elaborate on three remaining issues pertinent to monitoring. The first is where to monitor? The second is what should be monitored? The final topic is the frequency of monitoring. Where to Monitor? The subsurface has the capacity to remove or attenuate many chemicals and pathogens, thus improving the quality of the source water. A monitoring pro- gram is needed to document the water quality behavior and establish the reliabil- ity of the MUS system. This will involve installation of monitoring points to track the behavior of the water and the constituents in the water as the source water is introduced, stored, and eventually extracted. A number of reports on the topic of monitoring well installation and networks for characterizing subsur- face processes exist for the interested reader (NRC, 1994, 1999, 2000, 2003). For ASR, it is recommended that several wells in addition to the ASR well be monitored to establish the physical extent of the introduced water and help pinpoint the recovery efficiency. The exact number of monitoring wells will 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 245 vary with their purpose and the degree of existing knowledge of the site, but in all cases dialogue with regulators beginning with the inception of a project is recommended. Since federal and state agencies are charged with protecting an entire aquifer for all users, when an MUS project is proposed, they will wish obtain information on the impacts over a broad area of an aquifer, not just the “bubble” around the ASR well. Since it is unlikely that the physical or geo- chemical behavior of the stored water can be accurately assessed or modeled using information from a single monitoring point, the requirements of science and regulatory authorities are not necessarily in conflict. For systems that involve recharge at one point and extraction at a downgra- dient point, it is also recommended that several monitoring wells be installed downgradient from the recharge site prior to initiation of recharge to document changes in the quality of the water as a result of mixing with native groundwa- ter. Placement of such monitoring wells is also advantageous for confirming estimates for travel time from recharge wells to existing and proposed extraction wells. The depth interval(s) in monitoring wells for MUS systems should relate to the depths of injection zones or other important hydrostratigraphic units. In areas where the aquifer properties, groundwater velocity, and groundwa- ter quality are poorly understood, multiple monitoring wells are commonly needed. The number of monitoring wells required also generally increases with the areal size of the project and the likely risk to other users of the aquifer. In karst aquifers, it may be necessary to utilize geophysical methods to characterize hydrogeologic conditions. Unless geophysical methods or other detailed studies are conducted, there may be considerable uncertainty regarding the direction and rate of movement of stored water in karst systems. The point of compliance for many regulatory programs is the location at which the source water is first introduced into the subsurface. This means that the source water must meet all regulatory requirements prior to introduction and storage. Such antidegradation approaches can in some cases be overly restric- tive. They do not provide credit for improvements in water quality that can oc- cur while the source water passes through the subsurface or resides in the stor- age zone. As noted in Chapter 5, there is precedent for balancing the benefits of the MUS project against a strict antidegradation policy. The content of Box 5-3 provides some language from the State of California that offers an opportunity to weigh the benefits of subsurface storage and water quality improvements against an antidegradation policy. Serious consideration needs to be given to allowing flexibility in the regulatory framework so that the point of compliance can be at the extraction well or downgradient monitoring well if the extent of water quality improvements merits such a designation and human health is not compromised. 

246 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER What to Monitor? Chapter 3 discusses several basic ideas for what needs to be monitored in the context of recovery efficiency. The degree of mixing between the intro- duced source water and the native groundwater can be assessed using a tracer test. The hydraulic properties of the subsurface are gleaned from step- drawdown pump tests. Cycle test monitoring is used to determine the recovery efficiency and look for potential water quality changes that occur after mixing. For many types of recharge operations, water quality monitoring require- ments are relatively limited. Most systems recharging drinking water through ASR wells or river water through channel beds or recharge basins into under- ground storage are essentially unregulated. Water quality monitoring require- ments are therefore very limited. In some areas of the country, water deliber- ately introduced into the subsurface must meet water quality objectives that pro- tect beneficial uses of the groundwater. The most restrictive use is typically for drinking water. Monitoring to ensure compliance with drinking water standards in the extracted water is therefore often the most basic requirement, regardless of the source of recharge water. So a frequent answer to the question about what to monitor is the list of contaminants that have drinking water standards. As noted in Chapter 4, a change in redox conditions in an MUS system can impact water quality. It is important to consider monitoring redox indicators and inor- ganics such as manganese, arsenic, and other trace metals that might be mobi- lized during MUS operations. For waters of more impaired origin, such as reclaimed water, urban runoff, or agricultural runoff, there may be additional contaminants of concern in the recharge water. Examples include pharmaceuticals and personal care products (PPCPs), hormones, and other trace organic chemicals. These are usually called emerging contaminants (principally trace organic compounds of anthropogenic origin and may be better labeled trace organics), and most are presently unregu- lated. Chemicals that interfere with endocrine systems of humans and wildlife are termed endocrine disrupting compounds (EDCs). Chemicals that elicit a pharmaceutical response in humans are termed pharmaceutically active com- pounds (PhACs). EDCs and PhACs are not mutually exclusive classifications, because some, but not all, EDCs are also PhACs. Thousands of compounds have been reported to show endocrine disrupting properties, primarily in relation to estrogen effects (Global Water Research Coalition, 2003a), and more than 60 PhACs have been identified that impact the endocrine system of animals or hu- mans in nanogram per liter or lower concentrations in the ecosystem. PPCPs comprise a very broad, diverse collection of thousands of chemicals, including prescription and over-the-counter drugs, fragrances, cosmetics, sunscreen agents, diagnostic agents, and many other compounds. PPCPs and EDCs are found in many watercourses, usually at extremely low concentrations. In one 1999-2000 survey of the occurrence of organic contami- nants in the United States, for example, samples collected from 139 streams in 30 states tested for 95 pharmaceuticals, personal care products, and known or 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 247 potential endocrine disruptors found that 80 percent of the streams sampled con- tained at least one of the chemicals (Kolpin et al., 2002). Although measured concentrations were generally low and rarely exceeded drinking water guide- lines, drinking water advisories, or aquatic life criteria, many of the compounds have not been subjected to toxicological testing to establish drinking water lim- its. Unfortunately the information on occurrence of unregulated chemicals is much greater than the understanding of their significance. Several lists of emerging contaminants have evolved on the basis of new analytical techniques that have made it possible to examine extraordinarily low levels of compounds used in everyday life. We are essentially at the point of testing because we can, without knowing whether there is any significance to the findings. There are compounds with ecologic significance that may also have human health signifi- cance, such as some of the EDCs. There are many other compounds such as commonly prescribed or over-the-counter pharmaceuticals whose detection at parts-per-trillion levels may have no significance to either wildlife or humans. Nevertheless testing for a growing array of these compounds continues. Their detection by one researcher often prompts testing by others. As lower and lower detection levels are pursued, this problem of detecting compounds such as PPCPs and EDCs without understanding the significance of the findings be- comes more serious. One of the most difficult questions facing both regulators and groundwater management agencies is the appropriate testing requirements for such compounds including detection levels. Additional information is needed on the levels of PPCPs and EDCs that are of potential health concern. Once those levels of can be determined, appropriate detection levels and test methods can be developed and potentially included for MUS systems using wa- ters from impaired sources. In the interim, some states such as California have established guidance levels for some of the compounds for which maximum contaminant levels (MCLs) have not been established. N-Nitrosodimethylamine (NDMA) is an ex- ample that is described in Box 6-1. As of 2006, the California Department of Health Services had established notification levels and response levels for roughly 40 compounds, and the list may continue to expand as additional chemicals are evaluated for guidance. The concept behind the testing is to de- velop information on the occurrence of representative compounds that may be used to guide future regulations regarding treatment requirements for indirect potable reuse. The notification levels and response levels are based on potential health effects, but without the comprehensive review necessary for establish- ment of MCLs. Some of these compounds will eventually be regulated with MCLs, but many will have only guidance levels for many years. Where waters used for MUS are derived from more impaired sources subject to a wider range of contaminants than more protected surface waters, monitoring for compounds with such guidance levels may be appropriate since these compounds are targets of concern in drinking water supplies. 

248 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER BOX 6-1 NDMA Testing in Waters for MUS N-Nitrosodimethylamine (NDMA), a highly toxic compound, has been found in various types of water, including reclaimed water recharged into aquifers used for potable reuse. NDMA is one of the few unregulated trace organic chemicals with toxicological data indicat- ing risks at part-per-trillion levels. Under some conditions, NDMA has been found to be persistent in groundwater and therefore a potential threat for MUS systems. NDMA is a contaminant in some industrial wastewaters and is a disinfection by-product, particularly with chloramine disinfection. Because of these factors, NDMA is a particular concern for recharge of reclaimed wa- ter and should be included in monitoring programs to verify suitable water quality for MUS systems intended for drinking water supply. Although the federal government has yet to develop regulatory limits for NDMA in water, the State of California has developed a notifi- cation level and a public health goal for the compound. In the context of other exposures to NDMA, through foods, beverages, and rubber and plastic products, NDMA exposures at part-per-trillion levels in drinking water may not be significant, but in the context of regula- tory limits for other compounds in drinking water, testing and control of NDMA in waters for MUS appears appropriate. There is an inherent conflict—both in time and money—between the desire for complete and comprehensive information and the need to keep costs reason- able and commensurate with risks. This raises the question, If one cannot moni- tor everything everywhere, continuously, and forever, how can one feel confi- dent that the risks are manageable at a reasonable cost? One approach to answer this question is to develop a biomonitoring system to signal the presence of tox- icity. A second approach is to employ surrogates and indicators for the many compounds and microorganisms of interest. Inorganic chemical analyses are not unduly expensive, and the demand for inorganic chemical indicators has there- fore not been overwhelming. There are cases where conductivity or chloride is used as an indicator of salinity. Most of the need, however, has been to under- stand the risks from organic compounds and pathogens. A third, and often com- plementary, approach is to implement a risk management system. The following sections focus on online biomonitoring and surrogates and indicators for trace organics and microorganisms, and minimizing risk within MUS systems from source to supply. Online Biomonitoring Toxicological data have yet to be developed for many of the EDCs and pharmaceuticals found in water. A National Research Council (NRC, 1998) re- port on potable reuse, recommended development of fish biomonitoring meth- ods to address unidentified chemicals and lack of toxicological data for identi- fied chemicals. At least one study has been conducted to evaluate online bio- monitoring methods (Schlenk et al., 2007) using Japanese medaka exposed to 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 249 river water after recharge under ambient conditions in flow-through systems, which are subsequently examined for tumors and other anomalies. This online biomonitoring procedure, while promising, is not yet developed to the point where it can be implemented to evaluate the safety of potable water. Surrogates and Indicators for Trace Organics Traditional measures of organic matter, such as biochemical oxygen de- mand (BOD), chemical oxygen demand (COD), and total organic carbon (TOC), have been used as measures of treatment efficiency and indicate the presence of wastewater in a water supply. This in turn can signal the likelihood that specific trace compounds of health concern are present. TOC is composed mainly of natural organic matter, organic chemicals of anthropogenic origin, and soluble microbial products generated during biological wastewater treatment from the decomposition of organic matter (Drewes and Fox, 2000). The contributions can vary depending on location and season. Different approaches have been proposed to distinguish between naturally occurring and wastewater-derived organic constituents using differences in functional groups, structural properties, molecular size distribution, aromaticity, reactivity, or acid-base solubility (Drewes et al., 1999; Leenheer et al., 2001; Leenheer, 2003; Müller and Frimmel, 2002; Her et al., 2003). These approaches are promising and provide more insight into the origin of organic matter, but they often are semiquantita- tive and require a high degree of expertise for proper assessment. Further, TOC (and other bulk parameter) measurements are not a useful predictive tool for tracking the behavior of very low levels of some health-significant chemicals, and identification of one or more constituents in water that can be used as surro- gates for unregulated chemicals is needed. Analytical techniques such as ELISA (enzyme-linked immunosorbent as- say), gas chromatography-mass spectrometry (GC-MS), and liquid chromatog- raphy-mass spectrometry (LC-MS) are available to identify and quantify indi- vidual trace organic chemicals, and quantitative structural activity relationship (QSAR) and quantitative structure activity-property relationship (QSPR) models have been used to predict the behavior of EDCs, PPCPs, and other chemicals. Such advanced analytical techniques require highly trained chemists to operate equipment that is expensive to purchase and use and, thus, are used mainly by university researchers and a limited number of agency and commercial laborato- ries. The growing number of trace organic constituents, particularly EDCs and PPCPs, make it impractical to routinely test water for the entire suite of known or suspected constituents of concern. The lack of adequate indicators of surrogates is particularly important to MUS where the extracted groundwater is to be used as a potable supply. Some chemicals that exhibit unique characteristics or are poorly removed during engi- neered treatment or soil aquifer treatment (SAT), such as carbamazepine, may not be readily identified or quantified through the use of indicators or surrogates 

250 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER and specific testing for these compounds may be needed. Because of to their chemistry or behavior, other constituents are amenable to evaluation via indica- tors or surrogates. Clearly, there is no single constituent, indicator, or surrogate that is representative of the vast array of trace organic constituents present in recharge waters, and monitoring will have to include a suite of parameters. The selection of which specific chemicals, indicators, or surrogates to monitor is dependent on several factors, including the following: • Type of recharge water (e.g., stormwater, river water, reclaimed water) • Treatment prior to recharge, if any • Type of recharge (i.e., direct recharge or surface spreading) • Regulatory requirements (e.g., drinking water standards, antidegrada- tion requirements) • Specific trace organics known or suspected to be present in the re- charge water • Known toxic chemicals not amenable to detection or quantification by indicators or surrogates • Time lapse between sample collection and completion of analyses • Validity of analytical techniques used and confidence that the suite of parameters measured is indicative of water quality In recognition of the need to identify appropriate indicators or surrogates for organic constituents, several research efforts have been undertaken in recent years. One promising approach has been advocated by Drewes and Dickinson (2007) and others to target the presence and concentration of many individual or types of organic compounds having known or suspected health significance us- ing indicators and surrogates. In this case, an indicator is defined as an individ- ual chemical occurring at quantifiable levels that represents certain physico- chemical and biological characteristics of a family of trace constituents and pro- vides a conservative assessment of removal (e.g., ibuprofen, NDMA), while a surrogate is defined as a quantifiable change of a bulk parameter that can serve as a measurement of the performance of individual unit processes or operations regarding their removal of trace compounds (e.g., change in TOC or conductiv- ity through a treatment process). The proposed methodology entails identifying several “treatment bins” (biodegradation, chemical oxidation, physical separation, etc.) into which chemicals are listed as to their removal. Removal of surrogates such as biode- gradable organic carbon (BDOC) can then be compared to the removal of the various constituents of concern, and where the removal of surrogates corre- sponds to the removal of constituents or classes of constituents, the surrogates can be used for monitoring purposes. Another approach is to establish a priority list of chemicals for monitoring. In recognition that monitoring the entire spectrum of potential EDCs in water and wastewater would be cost-prohibitive, the Global Water Research Coalition 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 251 (2003b) developed a targeted list of EDCs that would provide a basis for credi- ble analytical determination of EDCs in water. It is understood that the priority list of EDCs is dynamic and additions or deletions to the list may be made as additional information becomes available. The advantage of the above approaches is that easily measured bulk pa- rameters can be used to simplify the trace organic analytical monitoring effort and provide a conservative assessment of removal. The disadvantages are that the indicator occurrence pattern may change, indicator selection requires regular review, and operational conditions determining removal can change over time. While the methodology used by Drewes and Dickinson and the Global Water Research Coalition (and similar methodologies being developed by others) may eventually prove to be appropriate for MUS systems using either wells or sur- face spreading, further evaluation and refinement is needed to validate the con- cept in practice. Using targeted indicators and surrogates to evaluate water qual- ity and safety in lieu of intense monitoring for the plethora of unregulated or- ganic constituents potentially present in water is a reasonable and realistic goal that is achievable with our current state of knowledge. Microbial Indicators The quality and safety of drinking water and groundwater have always been measured via fecal indicator organisms and in some cases the presence of vi- ruses and other surface water associated pathogens such as Cryptosporidium and Giardia Environmental Protection Agency (EPA) proposed groundwater rule). As already mentioned, the nature (perceived as protected) of groundwater and its use (as a potable supply) dictate the absence of “indicator” microorganisms. Indicator organisms most commonly used include total coliform bacteria and Escherichia coli. These are a part of the regulatory targets for drinking water; however they are now known to have disadvantages and cannot be used as broadly as once intended. Fecal indicator bacteria are generally harmless them- selves, but are found in high numbers in the gut of humans and other warm- blooded animals, including birds. These are excreted daily in the feces of people and mammals. It should be noted that total coliforms are found in soils and are generally used as a disinfection process control target and have yet to be associ- ated directly with pathogens or human health risks. • Fecal indicator bacteria including E. coli, enterococci, and virus indica- tors such as coliphage are also found in many environments, such as sewage (even treated sewage), septic tank effluent (liquid from a septic tank), septage (solids from a septic tank), manure and animal waste la- goons, and bird and other animal droppings. Heavy rainfall can wash the fecal wastes and associated indicator bacteria into nearby water bodies. 

252 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER • Fecal indicators can be found in most waters and the indicator levels generally reflect the amount of fecal pollution. However, even in pris- tine waters there is a background level of fecal indicators. They are presumed to be absent from groundwaters. • These fecal indicators are used to indicate the potential presence of pathogens, microorganisms that come from the gut and cause diseases such as diarrhea. • Disadvantages and limitations in the use of these “indicators” includes the fact that that sources of fecal contamination cannot be determined with routine methods; regrowth of the fecal bacterial indicators occurs, and there is a poor relationship of the indicators to the presence of vi- ruses, parasites, Legionella, and cyanobacteria. The quality and safety of drinking water and groundwater have always been measured via these fecal indicator organisms, with MCLs set only for total coli- forms and E. coli. Viruses and other surface water-associated pathogens such as Cryptosporidium and Giardia have been addressed through treatment technol- ogy rules based on removal and inactivation associated with filtration and disin- fection (EPA long-term enhanced surface water treatment rule and ground water rule). All surface waters will have some level of algae, bacteria, and parasites in them, and with increasing sewage inputs there will also be enteric viruses and other microbes of fecal origin. Thus, unless this water is pretreated to drinking water standards or infiltration systems are used to effectively remove some per- centage of the microorganisms, the source or stored water will contain these microbes. The native groundwater could also contain some bacteria (Le- gionella) and protozoa (Naegleria) that pose a risk to human health (outbreaks and associated deaths have occurred for both of these microorganisms due to the use of groundwater; see Appendix A). The targeted microbial contamination level associated with acceptable risks would depend on the use of the recovered water. For potable purposes a maximum contaminant level goal (MCLG) of zero is the target for those pathogenic microorganisms. Finally, regrowth of bacteria and the free-living protozoa can occur depending on the conditions, but more importantly, attenuation (usually due to inactivation of bacteria, parasites, and viruses) occurs. For enteric viruses and protozoa, long-term survival is of concern and interest. As a part of the attenuation via filtration or dilution (diffu- sion) the concentrations of the microorganisms that may migrate and be trans- ported into other aquifers has also been an area of research. Monitoring for the wide range of microorganisms in source, stored, and re- covered water has not been widely implemented. Thus, there is often a pre- sumption of microbial water quality based on the monitoring of selected “indica- tor” species. As mentioned, the primary research has focused on drinking water MUS systems, thus, those microbes associated with fecal pollution and stan- dards and rules for potable water have been the target of most of the controversy and studies. Bacterial pathogens are rarely monitored, a select group of viruses 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 253 may be monitored on occasion, and protozoa are monitored in surface waters but not groundwaters. None of these groups of microbes are monitored in reclaimed waters on a routine basis (the exception being in the State of Florida, which re- quires monitoring of Cryptosporidium and Giardia in reclaimed wastewaters albeit at a low frequency). A full description of the indicator bacteria (coliforms and alternative indica- tors such as enterococci and coliphages) other pathogenic bacteria (such as Le- gionella, Arcobacter, and Cyanobacteria), parasites (Cryptosporidium and Giardia; free-living amoebae), and enteric viruses (norovirus) is found in Ap- pendix A. Several key scientific data gaps have been identified that hinder the moni- toring of microbial water quality in the assessment of MUS as well as other groundwater projects: 1. The type of microorganism to be used in studies. Most studies have used laboratory strains, and the survival rate is questionable as it relates to either less or more resilient groups of pathogens or naturally occur- ring fecal indicators on which monitoring programs may be focused. In addition, better surrogates of pathogens may be needed. Some have found, for example, that PRD-1 survival may be a good model for that of hepatitis A in groundwater (Blanc and Nasser, 1996), and PRD-1 has been used as an indicator of virus transport and as a resilient tracer in field studies (Harden et al., 2003; Paul et al., 1995; Ryan et al., 1999). 2. The influence of the native microflora in surface and groundwaters on the inactivation rates of fecal indicators, along with redox conditions and nutrients; 3. The impacts on fecal microbial survival of infiltration into aquifer envi- ronments conducive for storage and inactivation associated with pore waters; and 4. In-situ studies in general, because most work has been done in the labo- ratory. A Risk Assessment Approach Each MUS system has associated risks of physical, chemical and biological hazards. In a stormwater recharge basin, for example, hazards include spills, floods, or land-use changes that impact stormwater quality. The degree of risk is related to the likelihood and consequence of the hazard, or combination of hazards. Identifying hazards and assessing risk are important toward develop- ment of a sustainable water resource for the end-user or the environment. An example of this assessment process is the Hazard Analysis and Critical Control Point (HACCP) plan implemented for a stormwater to drinking water project in 

254 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER Salisbury, Australia (Swierc and others, 2005).1 The project employs ASTR as the final treatment step within a larger system that includes stormwater catch- ment/management, surface storage and cleansing wetlands to produce potable water (Figure 6-6). Rather than monitoring at the end-point of the system, hazards and controls are evaluated along the entire system flowpath. The hazard analysis component of HACCP is based on a verified understanding of the processes, and identifica- tion of potential hazards and preventative measures. With this knowledge, each step along the process is assessed as a potential critical control point (CCP), which is defined as a point, step or procedure “… at which control can be ap- plied and is essential to prevent or eliminate a hazard or reduce it to an accept- able level.” For example, a disinfection system or a water quality monitoring station is a CCP. Each CCP has associated hazards that may be unique to that step, and each is examined to identify the following requirements: 1) monitor- ing: allows for tracking of the operation and trend analysis to flag potential loss of control, 2) corrective action or response: in the event of loss of control, and 3) data/record management: evidence of adherence to procedures and events re- flecting loss of control. These requirements are assessed in the context of criti- cal limits (e.g., EPA MCLs) established for each CCP. Once these limits are established, monitoring requirements and corrective action procedures are de- veloped that are specific to each CCP. With regard to the ASTR CCP in this particular system, aquifer characteri- zation as well as physical, chemical, radiological and microbiological processes are identified and considered in relation to variability in quality of input waters from the wetland treatment (reedbed cleansing) step. Contaminant attenuation can be modeled along the ASTR flow path, recognizing that the aquifer has fi- nite sorption capacity. The modeling facilitates system understanding, including travel times along the flow path, which helps in the effective design of a moni- toring plan. Monitoring wells are placed and sampled to allow performance tracking of the system; considerations for water-quality sample parameters and frequency are discussed in other sections within this chapter. How Frequently to Monitor? In most instances, the frequency of monitoring will be dictated by the regu- latory jurisdiction that is overseeing the project. An efficient monitoring pro- gram is one that involves a frequent sampling schedule at the start of operation to develop a historical record of the hydraulic characteristics and water quality trends. As clear trends in performance emerge along with consistent sampling results, the monitoring frequency can be decreased with confidence. A frequent sampling schedule in the early stages of operation will help to build trust of con- sumers and improve public perception of the project. An early, frequent 1 This section modified from Swierc and others (2005). 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT FIGURE 6-6 ASTR project site and scope of the HACCP plan. Available online http://www.clw.csiro.au/publications/techni- cal2005/tr20-05.pdf. Accessed December 19, 2007. Reprinted with permission from CSIRO. Copyright by CSIRO. 255 

256 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER monitoring program can document that the system is performing as intended. To the extent possible, predetermined decision rules should be established prior to commencing monitoring. The predetermined rules should reflect the goals of the monitoring program and provide criteria for the constituents to monitor, ini- tial monitoring frequencies, when monitoring frequencies should be increased or decreased, and how results from the monitoring program should be acted upon. Proactive monitoring combined with the ability to adapt the system is im- portant to maintain a successful MUS project. Water quality is an evolving sci- ence and unknowns exist. For example, the manager of an MUS system needs to pay close attention to the range of trace organic contaminants that are candi- dates for monitoring. An ongoing water quality plan that takes measures to in- crease knowledge risks (e.g., occurrence of emerging contaminants) and to im- prove quality over time helps to gain public trust for an MUS project. PUBLIC PERCEPTION AND INVOLVEMENT As Seerley (2003) has pointed out, public involvement in MUS projects is not automatic—projects tend to be technically complex and regulatory processes not entirely clear. Nevertheless, although underground storage of water has not been a high-profile issue in many communities in the United States, public edu- cation and involvement constitute an important step in any type of water man- agement undertaking. The general public has legitimate interests in, and con- cerns about, the quality and reliability of its water supplies. Any plans to initiate or enlarge a water storage project should consult with and educate the public in the process. As public trust in public agencies and private corporations has declined over the past half-century in the United States (Pew Research Center, 2001), there is a possibility that the public might react negatively to projects and decisions on which there has been insufficient information and consultation. When some event or development regarding a water storage or treatment project does gain public attention, a variety of organized interests can be expected to engage in a struggle for media attention and public support (Seerley, 2003), widening the scope of involvement in decision making. Failure to engage the public at the outset of a project planning process may reinforce public mistrust when plans are publicized later. Under those circumstances, project planning and the deci- sion-making processes can progress into competition for public opinion among organized interests. Loss of public trust in water providers and/or regulatory agencies can influence public opinion on MUS and other projects for a long time (Seerley, 2003). Trust takes time to build, can be lost overnight, and is espe- cially difficult to restore. When public outreach, education, and involvement have been taken seri- ously and pursued conscientiously, there are notable success stories of public engagement with water management activities. From decades of experience and research, some general principles have emerged for public participation in water 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 257 management decision making. The Water Environment Research Foundation (WERF) published a useful review of the state of knowledge on this topic (Hart- ley, 2003). Although the WERF review was undertaken with particular refer- ence to storage, recovery, and reuse of treated wastewater, its conclusions and recommendations are applicable to a much broader range of projects. The “core principles” presented there were • Manage information for all; • Maintain individual motivation and demonstrate organizational com- mitment; • Promote communication and public dialogue; • Ensure fair and sound decision making and decisions; and • Build and maintain trust. The WateReuse Foundation (2004) summarized what it termed “best prac- tices” to ensure that “well planned indirect potable reuse projects receive fair consideration in water supply decisions.” It listed 25 such best practices, many of which overlap with others discussed here. Some of those considered to be the most critical are summarized below. • Stakeholders will likely support a project if they understand that the project will improve their quality of life. Stakeholders must be able to perceive the value of the project. This is done through public education and collaboration. • A water agency must take the leadership to clearly articulate the prob- lem to stakeholders. For example, a continuing dialogue with the community about water supply and drought resistance should be held prior to identifying a solution. A water agency should help communi- ties define the value of the project. Communication between the water agency and the stakeholders should be continuous and should start early in the process. A water agency should develop and maintain good relationships with key audiences such as elected officials, the media, the community, and other official decision makers. • Once a problem has been defined and understood, all alternatives should be reviewed. The value of each solution is assessed in relation to other alternatives. Potential for conflicts are possible when a par- ticular solution or project appears to have been forced on stakeholders. When conflicts do occur, the water agency should endeavor to under- stand the issues underlying conflict and opposition and deal with them constructively. Water agencies should be advocates for solving the problem, not advocates for a particular project. • Water agencies should be cautious of any environmental justice issues that may arise. A project that is first implemented in neighborhoods 

258 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER where community leaders reside may be perceived better than one where they do not. • Establishing that a water agency is a trusted source of water quality is important. While the public generally credits the quality of water to its original source (i.e., a spring, a river, groundwater), the reality is that the safety of water is generally due more to the diligence of a water utility and its investments in testing and treatment. It is a high priority for the water agency to be perceived as the source of that quality. It should be noted, of course, that many of these practices could be also use- fully employed by opponents of potable reuse or related projects. Forester (1999) provides a thorough discussion of participatory decision making processes that is quite consistent with these principles. The NRC (2005) also appointed a committee to study the state of knowledge on public participa- tion in environmental decision making, and the committee’s observations simi- larly stressed the importance of early involvement, open sharing of information, and solicitation of citizens’ opinions in environmental and natural resource deci- sions. Typically more concerns are raised regarding well recharge systems, par- tially due to association of these programs with disposal of wastes, versus man- aged underground storage of water intended for later recovery. Two examples of public involvement and public perception concerns that have arisen in spe- cific MUS projects follow below. Orange County, California The Orange County Water District in Orange County, California, has im- plemented a successful public involvement program as part of its Groundwater Replenishment System, a managed underground storage project. In addition to community research and program evaluation activities, the public outreach effort included community presentations, appearances on local and public access cable television programs, distribution of materials to and through libraries and other public gathering places, a media relations program, and site and project tours (Wildermuth, 2001). District staff made an average of 120 presentations per year for seven years, to a wide array of civic groups, not only environmental, business, or other obviously interested organizations. The district’s program of information and engagement is generally credited with the general public ap- proval of the groundwater storage project, even though reclaimed wastewater is a significant source of the water being stored for later recovery, blending, and further treatment (Boxall, 2006). 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 259 Georgia and Florida Public perception of a technology such as ASR can be influenced positively or negatively depending on the degree of open and well-communicated scientific facts and uncertainties. In January 2001, a bill was before the Georgia General Assembly that would place a moratorium on injection of treated river water into the Floridan Aquifer System (FAS) through ASR. This moratorium gained the support of environmental advocates; one politician wanted to make the moratorium perma- nent and statewide. Another politician recognized that “not enough is known scientifically on whether the technology is safe or potentially harmful” and therefore was uncertain regarding whether “we need to close that door com- pletely right now” (Florida Times Union, January 13, 2001). The bill was not signed into law by the governor (to update before publication). A few months later in Florida, a different message was being communi- cated. Florida’s ASR operations had existed for several years with few known problems. The recharged water was (and is) required to be treated to drinking water standards. However, legislation before Florida state lawmakers was being proposed to relax water quality standards (e.g., fecal coliforms) prior to recharge into the FAS under certain conditions. Concurrent with these legislative actions was the release of a report by the NRC (2001) that identified issues of uncer- tainty regarding the role of ASR in the $7.8 billion Everglades restoration plan. Moreover, the New York Times (April 13, 2001) reported that “the state is in the midst of a drought that is the worst in 50 years,” and forecasts say that by 2020 without new sources Florida “would face a water deficit of as much as 30 per- cent.” As such, the ASR issue was very high profile. As the Florida bill successfully moved forward during the 2001 legislative session, newspaper articles described the bill as allowing “untreated,” “pol- luted,” or “tainted” water to be injected into the FAS. The bill was amended often to address environmental and scientific concerns as they arose; however, public opposition was building. The message from the legislature differed from that of environmental advocates, and scientists could not provide definitive an- swers. On April 24, 2001, an editorial in the St. Petersburg Times by a state environmental protection official addressed misperceptions regarding the bill and acknowledged that “opponents of the ASR plan have done a masterful job of offering sound bites that would ignite most who were hearing of the legisla- tion for the first time. The problem is, many of these sound bites are false.” After considerable debate among lawmakers and strong opposition by environ- mental advocacy groups, the bill was withdrawn. This occurred even though many believed the bill as amended in its final version provided conditions to ensure protection of Florida’s groundwater resources. In Georgia, uncertainty led an effort to remove ASR as a water-resource management option, while in Florida, this established technology experienced a perception “backslide” due to: (1) introduction of a legislative bill that would have benefited from additional input by scientific and technical experts, and (2) 

260 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER communication of a mixture of facts, uncertainty, and inconsistent, misleading, or misunderstood (thus poorly communicated) information. Although the bill rapidly evolved to address concerns, change in public perception outran its pro- gress. Today, the pros and cons of ASR in Florida are widely understood (e.g., NRC, 2001) and have had the benefit of increased scientific study, open com- munication, and the support of a more informed citizenry. FINANCIAL DRIVERS AND RELATED CONSIDERATIONS Chapter 5 discusses many of the economic issues associated with MUS. Fi- nancial considerations are an important component in the development of MUS. Major financial considerations are capital costs and operating costs. Availability of grants, loans, and other subsidies and rate-paying schemes are also other fi- nancial considerations. Capital costs include the cost of the land needed for recharge facilities, the cost of constructing recharge facilities, and the cost of surface water retention and conveyance facilities necessary to capture and move the water to recharge facilities. Operational and maintenance costs include cost of the water to be stored, cost of any additional treatment required, cost of acquiring the necessary easements and permits, and monitoring costs. For some systems using re- claimed water, monitoring costs may be a significant factor affecting the final cost of the stored water and the feasibility of the MUS project. The large initial capital costs of such projects may be beyond what can be covered by tax increases, special assessments, or user fees. In such cases, water agencies tend to finance some of the capital costs through bonds, loans, and grants. Of these, bonds have been one of the most commonly employed methods of public finance (Howitt et al., 1999). In some cases, local agencies can utilize tax-exempt bonds as an effective approach to generate funding. With this ap- proach, revenue from the project is used to support the debt. Debt servicing is usually the largest cost component of these projects. The incremental increase in user fees associated with project costs will depend on the cost of the project, the size of the user base, and other factors. A major challenge to MUS is the ability of water providers to secure the fi- nancing necessary to develop a project where infrastructure is needed to bring surface water into a site that is feasible for recharge. In some cases, the distance between the location of available surface water and the recharge site may be large and there may be no existing infrastructure to convey the water. In other cases, the cost of land for recharge facilities is prohibitive. Small water provid- ers may be limited in gaining access to such infrastructure and resources without the support of larger-scale water providers or without institutional coordination. Individual states sometimes impose limitations on the powers that local governments have to raise funds to secure needed financing. Even where public entities recognize opportunities to possibly pool resources or coordinate funding, 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 261 if the legal mechanisms are absent to facilitate this, then project implementation is less likely. As noted in Chapter 5, it is possible for a project to be financially justified but not economically justified. It is also possible for a project to be economi- cally justified but not financially feasible. Important factors that relate to finan- cial feasibility include whether institutional opportunities are available to make a project feasible and whether the legal authority exists to support it. If institu- tional opportunities are not available or legal authority does not exist to support it, then project proponents may need to evaluate other institutional arrange- ments. Such arrangements may involve the creation of a new agency to sponsor the project, which may entail increased costs for the project. Historically, surface storage projects have been heavily subsidized through agencies such as the Bureau of Reclamation and various state agencies, espe- cially in the western United States. These subsidies can take the form of grants or low-interest loans for capital improvements, and operating subsidies based on the amount of water stored underground. Issues concerning subsidies have been addressed in Chapter 5. Revenues from MUS projects are obtained through the sale of the stored water. Collecting revenues from water users is a critical financial consideration when planning for MUS. Particularly, it is not clear how MUS will affect water rates. Pyne (2005) identifies two issues: timing of when consumers pay for wa- ter stored and not yet recovered; and how costs can be distributed among users with very different demands. Some rate-paying arrangements are evolving as in Pasadena, California, where the city has to provide storage capacity for the Met- ropolitan Water District of Southern California (MWD) for $3.00 per year per acre foot payable upon recovery (Pyne, 2005). One major incentive cited by ASR facility owners or operators for using this technology is as a means of maximizing the use of water treatment facilities. Shrier (2002) found that ASR facilities intended for potable uses typically treat water to primary and secondary drinking water standards prior to recharge; a few (27 percent of responding facilities) perform some additional pre-recharge treatment at the wellhead (e.g., pH adjustments) to improve injection operations and prevent geochemical interactions between the stored and native waters un- derground. Most ASR facilities perform no additional post-recovery treatment before introducing the recovered water into their water supplies. 42percent of the responding facilities perform minimal post-recovery treatment prior to re- charge (e.g., pH adjustments, iron and manganese removal, filtration or turbidity reduction). Thus, ASR enables facility owners and operators to shift the demand on treatment facilities to non-peak periods by treating the stored water to drinking water standards prior to recharge. The capacity of water treatment facilities is typically designed to meet peak treatment demands. Increasing non-peak use and decreasing peak use of water treatment facilities enable water providers to delay the need for capital investments for increased treatment capacity. 

262 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER Another financial incentive for groundwater-dependent utilities is the way in which wholesale water agencies purchase contracts or rates are structured. For example, in Wildwood, New Jersey, retail water agencies pay for treated water from the wholesale water agency whether they use the water now, store it for later use, or do not use it at all. This rate environment creates an incentive to capture the water during periods of low demand and store it underground for use during periods of high demand. (Wildwood has highly seasonal demand as a coastal resort area with a much greater summer than winter population.) Until 2005 the water stored underground in New Jersey had to be extracted within one year of storage or be claimed by the state. New Jersey has recently allowed studies on water banking for longer periods in the northeast part of the state. In an effort to restore seriously overdrafted regional aquifers the state al- lows only 85 percent of the water stored underground to be extracted and recov- ered in designated critical zones of the state. The remaining 15 percent of stored water reverts to the state. To ensure reliable customers for the new 30 million-gallon-per-day water treatment and conveyance facilities, purchase contracts from the regional whole- sale agencies in New Jersey American and South Jersey have been structured based on consistent use over 365 days each year. The contracts are essentially take-or-pay contracts. Since the water must be paid for regardless of whether it is used, the water purchasers have virtually no marginal cost for the water during low-demand periods. In some areas the demand rate is set at 90 percent of the full water rate, so the marginal cost is essentially 10 percent of the full cost for the water. Under these circumstances storing the water underground until it is needed the following summer has become an easy decision for Wildwood. In Southern California, the water stored in the ground for MWD can be called under various constraints, but it is generally expected to be stored for sev- eral years and available under drought conditions. Such storage is being called upon along with surface storage during 2007 to cope with the severe shortfall in water available from imported water sources, the Colorado River, and the State Water Project. In addition to relatively new storage agreements, MWD has long relied on the availability of groundwater from basins that receive discounted replenishment water. The replenishment water program has involved no formal obligation to increase groundwater withdrawals during periods of need, but agencies such as OCWD have historically cooperated in increasing available groundwater during drought conditions affecting imported water supplies. CONCLUSIONS AND RECOMMENDATIONS Conclusion: The development of an MUS system from project conception to a mature, well functioning system is a complex, multistage operation requir- ing interdisciplinary knowledge of many aspects of science, technology, and institutional issues. 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 263 Recommendation: A comprehensive decision framework should be de- veloped to assist in moving through the many stages of project development in an organized, rational way. Professionals from many fields, including chemists, geologists, hydrologists, microbiologists, engineers, economists, planners, and other social scientists should be involved in developing this framework. Conclusion: Growing experience with MUS systems indicates that hydro- geological feasibility analysis including aquifer characterization is one of sev- eral important components in their development and implementation. The bene- fits of doing so include establishing the hydraulic capacity, recharge rates, resi- dence times, and recoverable fraction of the introduced water—all of which help identify the optimum design and viability of the MUS system. Some types of aquifers have matrix, hydrogeologic, and geochemical char- acteristics that are better suited to MUS systems than others. For example, the aquifer characteristics may dictate recharge, storage, and recovery methods. For an unconfined aquifer, source water can be recharged into the aquifer through recharge basins, vadose zone recharge wells, and deep recharge wells. Stored water can be recovered by production wells or ASR wells, or it can enhance baseflow to neighboring streams. For confined aquifers, however, source water can only be injected through deep recharge wells, including ASR wells. The stored water is usually recovered through ASR wells or downgradient produc- tion wells. As another example, water quality benefits are likely to be greater with alluvial systems compared to fractured or dual porosity systems. Recommendation: Multiple factors should be assessed and monitored dur- ing design, pilot tests, and operations, including spatial and hydrogeological characterization of storage zones, temporal variation in quality and quantity of recharged, stored, and recovered water and factors that constrain sustainability of the MUS system, including hydrogeochemical, microbiological, and eco- nomic conditions. Uncertainty reduction is the ultimate goal. Conclusion: An independent advisory panel can provide objective, third- party guidance and counsel regarding design, operation, maintenance, and moni- toring strategies for an MUS project. An independent panel can increase public acceptance of and confidence in the system, if such trust is warranted. It can also be a catalyst for altering a plan if changes appear to be necessary. Recommendation: Water agencies should highly consider the creation of an independent advisory panel or equivalent at an early stage of planning for an MUS system. Conclusion: Relatively little research has been done to characterize the ex- tent of vertical migration of fine-grained particles into the sediments beneath surface spreading facilities. Likewise, the science and technology of cleaning recharge basins is not well developed. Recommendation: Research is recommended to develop new approaches to optimizing surface recharge, including assessing the extent of migration of 

264 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER fine-grained sediment into the subsurface, its impact on the long-term sustain- ability of surface recharge, and more efficient methods to clean recharge basins after clogging occurs Conclusion: Successful MUS involves careful and thorough chemical and microbiological monitoring to document system performance and evaluate the reliability of the process. Each MUS project needs real-time monitoring of the quality of the waters being introduced into underground storage and of waters being extracted from storage for use. Recommendation: Water quality monitoring programs should be designed on a case-by-case basis to assess water quality changes for elements, com- pounds, and microbes of concern, optimizing the potential for documenting any improvement in the quality of the source water and to collect samples represent- ing any adverse water quality changes. A proactive monitoring plan is needed to respond to emerging contaminants and increase knowledge about potential risks. Conclusion: New surrogates or indicators of pathogen and trace organic contaminant presence are needed for a variety of water quality parameters to increase the certainty of detecting potential water quality problems through monitoring. The categorization of chemicals and microorganisms into groups with similar fate and transport properties and similar behavior in treatment steps is one approach to streamline the list of potential contaminants to be monitored. It is unclear whether we can continue to rely on total coliform and E. coli indica- tor bacteria to characterize the microbial quality of water as the drinking water industry has done for decades. Such methodologies will improve the ability of MUS systems of a variety of sizes to engage in sound monitoring practices. Recommendation: Research should be conducted to understand whether we can rely on monitoring surrogate or indicator parameters as a substitute for analysis of long lists of chemicals and microorganisms. Conclusion: Surface spreading facilities sometimes require large amounts of land, particularly where large amounts of water are recharged or the geology is not ideal. Recharge well systems require less land, but may have as many different factors to consider in their placement. Optimization of recharge facil- ity placement is important but not always well understood. Recommendation: If there is some degree of freedom in site selection for recharge wells or basins, a location suitability assessment may be useful in site optimization. Factors such as ecological suitability, existing uses of the aquifer, groundwater quality, aquifer transmissivity, road density, land use and owner- ship, and access to power lines can be weighed in such an analysis. 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 265 REFERENCES Aiken, G. R., and E. L. Kuniansky (eds.). 2002. U.S. Geological Survey Artifi- cial Recharge Workshop Proceedings, Sacramento, California. U.S. Geo- logical Survey Open-File Report 02-89. Reston: USGS. Available online at http://water.usgs.gov/ogw/pubs/ofr0289/index.htm. Biuk, N. A., and a. Willemson. 2002. Clogging rate of recharge wells in porous media. Dillon, P. (ed.) Management of Aquifer Recharge for Sustainability. Lisse, The Netherlands: A.A.Balkema, 195-198. Blanc, R., and A. Nasser. 1996. Effect of effluent quality and temperature on the persistence of viruses in soil. Water Science and Technology 33(10- 11):237-242. Bloetscher, F., A. Muniz, and G. M. Witt. 2005. Groundwater Injection— Modeling, Risks and Regulations. New York; McGraw-Hill.Boxall, B. 2006. Doubts still swirl to surface. The Los Angeles Times. May 7. Brown, C. 2005. Planning decision framework for brackish water aquifer, stor- age and recovery (ASR) projects. Ph. D. dissertation, Department of Civil and Coastal Engineering. University of Florida, Gainesville. Bouwer, H.. 2002. Artificial recharge of groundwater: hydrogeology and engi- neering. Hydrogeology Journal 10: 121-142. Dillon, P. J. (ed.). 2002. Management of Aquifer Recharge for Sustainability. Proceedings of the fourth International Symposium on Artificial Recharge (ISAR 4), Adelaide, 22-25 Sept. 2002. Lisse, The Netherlands: A.A. Balkema Publishers. Drewes J. E., M. Sprinzl, A. Soellner, M. D. Williams, P. Fox, and P. Wester- hoff. 1999. Tracking residual dissolved organic carbon using XAD- fractionation and 13C-NMR spectroscopy in indirect potable reuse systems. Vom Wasser 93:95-107. Drewes, J. E., and P. Fox. 2000. Effect of drinking water sources on reclaimed water quality in water reuse systems. Water Environ. Res. 72(3):353-362. Drewes, J. E., and E. Dickinson. 2007. State-of-the-art approaches to monito- runregulated and regulated trace organics in indirect potable reuse. In Pro- ceedings of the 2007 WateReuse Foundation Research Conference, El Paso, Texas, June 4-6.. EWRI/ASCE (Environmental and Water Resources Institute/American Society of Civil Engineers). 2001. Standard Guidelines for Artificial Recharge of Ground Water. EWRI/ASCE 34-01. Reston, VA: ASCE. Finch, S. T., and E. Livingston. 1997. Aquifer storage and recovery study for the La Luz Well Field, City of Alamogordo, New Mexico. In D.R. Kendall (ed.) Proceedings of American Water Resources Symposium on Conjunc- tive Use of Water Resources: Aquifer Storage and Recovery. Herndon, VA: American Water Resources Association. Forester, John. 1999. The Deliberative Practitioner: Encouraging Participatory Planning Processes. Cambridge, MA: MIT Press. 

266 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER Global Water Research Coalition. 2003a. Endocrine Disrupting Compounds: Occurrence of EDC in Water Systems. Report prepared by the Water Re- search Commission (South Africa) in cooperation with Kiwa Water Research (Netherlands) and TWZ-Water Technology Center (Germany) for the Global Water Research Coalition, London, UK. Global Water Research Coalition. 2003b. Endocrine Disrupting Compounds: pri- ority list of EDCs. Report prepared by the Water Research Commission (South Africa) in cooperation with Kiwa Water Research (Netherlands) and TWZ-Water Technology Center (Germany) for the Global Water Research Coalition, London, UK. Harden H. S., J. P. Chanton, J. B. Rose, D. E. John, and M. E. Hooks. 2003. Comparison of sulfur hexafluoride, fluorescein and rhodamine dyes and the bacteriophage PRD-1 in tracing subsurface flow. Journal of Hydrology 277(1):100-115. Hartley, T. W. 2003. Water Reuse: Understanding Public Perception and Par- ticipation. Alexandria, VA: Water Environment Research Foundation. Her, N., G. Amy, D. McKnight, J. Sohn, and Y. Yoon. 2003. Characterization of DOM as a function of MW by fluorescence EEM and HPLC-SEC using UVA, DOC, and fluorescence detection. Water Res. 37: 4295–4303. Howitt, R. E., J. R. Lund, K. W. Kirby, M. W. Jenkins, A. J. Draper, P. M. Grimes, K. B. Ward, M. D. Davis, B. D. Newlin, B. J. Van Lienden, J. L. Cordua, and S. M. Msangi. 1999. Integrated Economic-Engineering Analysis of California's Future Water Supply. Report for the State of Cali- fornia Resources Agency, Sacramento, California. Available online at http://cee.engr.ucdavis.edu/faculty/lund/CALVIN/Report1/ReportAug99.pdf. Accessed on August 17, 2007. Kolpin, D. W., E. T. Furlong, M. T. Meyer, E. M. Thurman, S. D. Zaugg, L. B. Barber, and H. T. Buxton. 2002. Pharmaceuticals, hormones, and other or- ganic wastewater contaminants in U.S. streams, 1999-2000: A national re- connaissance. Environment Science Technology 36(6)1202-1211. Leenheer, J. 2003. Comprehensive Characterization of Dissolved and Colloidal Organic Matter in Waters Associated with Groundwater Recharge at the Orange County Water District. Denver, CO: U.S. Geological Survey. Leenheer, J. A., C. E. Rostad, L. B. Barber, R. A. Schroeder, R. Anders, and M. L. Davisson. 2001. Nature and chlorine reactivity of organic constituents from reclaimed water in groundwater, Los Angeles County, California. En- vironmental Science and Technology 35:3869-3876. Mansuy N. 1998. Water Well Rehabilitation – A Practical Guide to Understand- ing Well Problems and Solutions. Mission Woods, KS: CRC Press. Müller, M. B., and F. H. Frimmel. 2002. A new concept for the fractionation of DOM as a basis for its combined chemical and biological characterization. Water Research 36:2643-2655. NRC (National Research Council). 1994. Alternatives for Ground Water Cleanup. National Academies Press, Washington, D.C. 

PROJECT DEVELOPMENT, MONITORING AND MANAGEMENT 267 NRC. 1998. Issues in Potable Reuse: The Viability of Augmenting Drinking Water Supplies With Reclaimed Water. Washington, DC: National Acad- emies Press. NRC. 1999. Groundwater and Soil Cleanup: Improving Management of Persis- tent Contaminants. Washington, DC: National Academies Press. NRC. 2000. Natural Attenuation for Groundwater Remediation. Washington, DC: National Academies Press. NRC. 2001. Aquifer Storage and Recovery in the Comprehensive Everglades Restoration Plan: A Critique of the Pilot Projects and Related Plans for ASR in the Lake Okeechobee and Western Hillsboro Areas. Washington, DC: National Academies Press. NRC. 2003. Environmental Cleanup at Navy Facilities: Adaptive Site Manage- ment. Washington, DC: National Academies Press. NRC. 2005. Decision Making for the Environment: Social and Behavioral Sci- ence Research Priorities Washington, DC: National Academies Press. Olsthoorn, T. N. 1982. Clogging of recharge wells, main subjects. KIWA Communications 72. Paul, J. H., J. B. Rose, J. Brown, E. A. Shinn, S. Miller, and S. R. Farrah.1995. Viral tracer studies indicate contamination of marine waters by sewage dis- posal practices in Key-Largo, Florida. Appl Environ Microbiol 61(6):2230- 2234. Perez-Paricio, A. and J. Carera. 1998, A conceptual and numerical model to characterize clogging. Pp. 55-60 In Peters, J.H., et al. (eds) Proceedings of the Third International Symposium on Artifical Recharge of Groundwater. Rotterdam, Netherlands: A.A. Balkema. Pew Research Center for the People and the Press. 2001. Available online at http://www.people-press.org/questionnaires/trustque.htm. Accessed August 17, 2007. Pyne, R. D. G. 2005. Aquifer Storage Recovery: A Guide to Groundwater Re- charge Through Wells. Gainesville, FL: ASR Press. Rinck-Pfeiffer, S., S. Ragusa, P. Sztajnbok, and T. Vandevelde. 2000. Interrela- tionships between biological, chemical, and physical processes as an analog to clogging in aquifer storage and recovery (ASR) wells. Water Research 34(7): 2110-2118. Ruetten, J. 2001. Best Practices for Developing Indirect Potable Reuse Pro- jects: Phase 1 Report. Product Number 01-004-01. Alexandria, VA: Wa- teReuse Foundation. Ryan, J. N., M. Elimelech, R. A. Ard, R. W. Harvey, and P. R. Johnson . 1999. Bacteriophage PRD1 and silica colloid transport and recovery in an iron ox- ide-coated sand aquifer. Environ Sci. Technol 33(1):63-73. Schlenk, D., D. E. Hinton, and G. Woodside. 2007. Online Methods for Evalu- ating the Safety of Reclaimed Water. Water Environment Research Foun- dation Report No. 01-HHE-4A. Alexandria, VA: Water Environment Re- search Foundation. 

268 PROSPECTS FOR MANAGED UNDERGROUND STORAGE OF RECOVERABLE WATER Seerley, D. J. W. 2003. An analysis of the evolution of public policy for aquifer storage and recovery: Experiences in three southeastern states. Ph.D. disser- tation.The University of Georgia, Athens, GA. Segalen, A.-S., P. Pavelic, and P. Dillon. 2005. Review of Drilling, Comple- tion, and Remediation Methods for ASR Wells in Unconsolidated Aquifers. Technical Report No. 04/05. Canberra, Australia: CSIRO Land and Water. Shrier, C. 2002. Survey and Analysis of Aquifer Storage and Recovery (ASR) Systems and Associated Regulatory Programs in the United States. Denver, CO: American Water Works Association. Swierc, J., D. Page, J. Van Leeuwen, and P. Dillon. 2005. Preliminary Hazard Analysis and Critical Control Points Plan (HACCP) - Salisbury Stormwater to Drinking Water Aquifer Storage Transfer and Recovery (ASTR)Project: CSIRO Land and water Techical Report No. 20/05. Available online at http://www.clw.csiro.au/publications/technical2005/tr20-05.pdf. Accessed November 20, 2007. UNESCO (United Nations Educational, Scientific and Cultural Organization). 2006. Recharge Systems for Protecting and Enhancing Groundwater Re- sources. Proceedings of the 5th International Symposium on Management of Aquifer Recharge (ISMAR5), Berlin, Germany, 11–16 June 2005. IHP- VI, Series on Groundwater No. 13. Paris: UNESCO. Stuyfzand. P. J. 2002. Modeling the accumulation rate and chemical composi- tion of clogging sludge layers in recharge basins with Easy-Leacher® 4.6. Pp. 221-224 In Dillon, P. (ed.) Management of Aquifer Recharge for Sus- tainability, Rotterdam, Netherlands: A.A.Balkema. Water Environment Federation and American Water Works Association. 1998. Using Reclaimed Water to Augment Potable Water Resources. Special Pub- lication. Alexandria, VA:Water Environment Federation. WateReuse Foundation. 2004. Best Practices for Developing Indirect Potable Reuse Projects: Phase 1 Report. Report of Project WRF-01-004. Alexan- dria, VA: WateReuse Foundation. Wildermuth, R. October 2001. Groundwater replenishment system: Public out- reach and education program. Presentation at the NWRI National Urban Watersheds Conference, Costa Mesa, Calif., October 17-19.