Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 108
Desalination: A National Perspective 5 Environmental Issues Desalination has been used around the world on a municipal scale for many decades, yet it is still considered by many to be a “new” option for addressing water supply needs. Part of the hesitancy to accept this technology comes from concerns over potential environmental impacts of desalination, which have not been fully quantified. The environmental issues surrounding desalination fall into four general categories, which are reviewed in this chapter: (1) impacts from the acquisition of source water, (2) impacts from the management of waste products and concentrate from the desalination process, (3) issues with desalinated product waters, and (4) the impacts of greenhouse gas emissions from these energy-intensive processes. Technologies and approaches to mitigate these impacts are also discussed. Environmental impact assessments for any project also include concerns that are not addressed in this chapter, such as environmental effects of plant construction, material use, potential releases to the air, disposal of used membranes, and socioeconomic considerations. These issues are discussed in a recent World Health Organization report “Desalination for Safe Water Supply” (WHO, 2007). Human use of any water supply will have some environmental impacts; ultimately, consideration of the potential impacts of desalination will need to be weighed against the impacts from other water supply alternatives. SOURCE WATER ACQUISITION Desalination technologies can provide high-quality water tailored to the user’s needs, and many otherwise unusable sources of water (e.g., oceans, estuaries, brackish aquifers, wastewater) can be treated by desalination technologies. For each type of source water, there are distinct environmental considerations when that water is withdrawn. In coastal surface waters, issues of impingement and entrainment of marine organ-
OCR for page 109
Desalination: A National Perspective isms are paramount. For inland aquifer systems, the renewability of the resource and land subsidence over time are significant issues. Marine Water Intake Issues: Impingement and Entrainment Pumps bringing large volumes of ocean or estuary water into desalination plants can cause impingement and entrainment. Impingement, defined as the pinning and trapping of fish or other larger organisms against the screens of the intake structures, can cause severe injury and death to organisms. Entrainment occurs when intake pipes take in small aquatic organisms, including plankton, fish eggs, and larvae, with the intake water. Organisms that are pulled into the system will die if they are subjected to high temperatures or are crushed by high-pressure membranes. Intakes for desalination plants co-located with power plants are regulated under Section 316B of the Clean Water Act, although states may choose to apply these regulations to stand-alone plants as well (see Box 5-1). Power plants have been well studied with regard to impingement and entrainment of organisms. Most desalination plants will take in far less water, roughly an order of magnitude lower than medium-sized power plants. However, very large stand-alone desalination plants might require comparable quantities of intake water if substantial volumes of water are needed for concentrate dilution. Intakes from a single large power plant are estimated to kill billions of juvenile-stage fishes each year and may affect recruitment of juvenile fish and invertebrates into the adult populations (Brining et al., 1981). It has been estimated that the magnitude of loss from one large power plant is equivalent to the loss of biological productivity of thousands of acres of habitat (York and Foster, 2005). The decomposition of the dead organisms can reduce the oxygen in the water, causing an additional stress in the area. However, the population-level impacts of mortality due to entrainment of marine organisms may or may not be substantial because the normal mortality of larval organisms in the marine environment is generally very high. The impacts and the acceptability of this loss will likely vary from place to place. There are technologies and practices that can be applied to reduce the amount of impingement and entrainment associated with coastal desalination. To reduce the amount of entrainment, it is possible to reduce intake during the times when eggs and larvae are abundant in the water, and windows of operation can be set to minimize the entrainment of eggs and larvae of the species of concern. If intake pipes are located in deeper parts of a bay, there will also be fewer organisms that could be impinged or entrained (San Francisco Bay Conservation and Development Commission, 2005). Entrainment can also be reduced substantially by
OCR for page 110
Desalination: A National Perspective BOX 5-1 Environmental Regulatory Framework for Desalination Several national regulations serve as the legal framework to minimize environmental impacts from desalination processes. The most pertinent regulations are associated with the Clean Water Act, although the Safe Drinking Water Act and its Underground Injection Control program are also described here. Additional environmental regulations that need to be considered in the permit process are described in Chapter 7. Clean Water Act Under Section 316(b) of the Clean Water Act (CWA; P.L. 92-500), the Environmental Protection Agency (EPA) has developed regulations that require that the location, design, construction, and capacity of cooling water intake structures reflect the best technology available for minimizing adverse environmental impacts. Phase 1, promulgated in 2001, addresses the intake structures of new power plants. Phase 2 addresses the intake structures of large existing electric generating plants and requires these plants to meet impingement and entrainment standards for reducing the number of organisms affected. As of July 2007, the Phase 2 regulations were suspended while the EPA addresses several issues remanded by the 2nd Circuit Court of Appeals.1 In the United States, effluent discharges are federally regulated by the CWA. The regulatory program includes the National Pollutant Discharge Elimination System (NPDES), established by Section 402 of the CWA, which sets limits on the amount of various pollutants that a point source (i.e., the desalination plant) can discharge into a surface water body in a specific time period. Effluent limits can be technology based or water quality based, but they are all performance standards—that is, the permittee is free to use any combination of process modification, end-of-pipe treatment, or other strategies to meet them. Water quality standards can also vary depending on the specified use of the particular water body into which the concentrate is disposed. NPDES permits typically quantify areas—termed mixing zones—where surface waters may exceed water quality standards due to point source discharges. If state regulatory programs meet the EPA requirements, the programs can be delegated to be administered by the states; therefore, regulations may vary somewhat from state to state. Effluent limits for desalination plants may specify pH, metaphosphates, chlorides, dissolved oxygen, conductivity, copper, iron, radium, total dissolved solids (TDS), total nitrogen, sulfide, ammonia, turbidity, radionuclides, selenium, and others. EPA (or the delegated state regulatory program) specifies the monitoring requirements, frequency of testing, and reporting, and the monitored species can be regionally variable. Some states require whole effluent tests for desalination concentrate in addition to chemical-specific numerical limits (Mickley et al., 1993). Whole effluent toxicity testing may include acute tests of 96 hours’ duration using larval or juvenile fish and invertebrates, with survival as the end point, and chronic tests of 7 days in duration using early life stages of a fish and an invertebrate, considering metrics 1 For more information, see http://www.epa.gov/waterscience/316b/index.html.
OCR for page 111
Desalination: A National Perspective such as growth. Local species may also be used instead of “standard” bioassay organisms if a bioassay has been developed for them and is approved by EPA. Safe Drinking Water Act The Safe Drinking Water Act (SDWA) regulates the levels of contaminants permitted in drinking water supplies and applies to every public water system in the United States. Based on data describing how often a particular contaminant occurs in the environment, how humans are exposed to it, and potential health effects of exposure, EPA sets a maximum contaminant level goal (MCLG), the level of a contaminant in drinking water below which there is no known or expected health risk, including a margin of safety. These goals are not enforceable because they do not take available technology into consideration and are sometimes set at levels that public water systems cannot meet. EPA proposes an enforceable standard in the form of a maximum contaminant level (MCL), which is the maximum amount of a contaminant allowed in water delivered to a user of a public water system. Every 5 years, EPA establishes a list of contaminants that are known or anticipated to occur in public water systems and may require future regulation under SDWA.2 EPA oversees deep-well injection of desalination concentrate through its underground injection control program within the SDWA. EPA has developed the following classification for injection wells (EPA, 2007b): Class I: wells that inject hazardous waste; Class II: wells associated with the oil and gas industry; Class III: wells that inject fluids for the extraction of minerals; Class IV: wells that inject hazardous or radioactive waste into a formation within one-quarter mile of a drinking water source; and Class V: all other injection wells not covered by Classes I-IV. These classifications each have associated standards and associated regulations. States, rather than EPA, generally enforce the program and issue permits. Subsurface injection of desalination concentrates are covered by the states under regulations for either Class I or V injection wells. reducing water intake volumes. Reducing the size of mesh of screens in intakes can reduce entrainment but will increase impingement. However, rotating screens and other types of technologies can minimize the intake of aquatic organisms. If intakes are placed below the surface through the use of beach wells or other subsurface intakes (see Chapter 4), the problems of entrainment of marine organisms are largely eliminated. Technologies for reducing impingement and entrainment are discussed in detail in Chapter 4. Co-location of a desalination plant with an existing power plant takes advantage of existing intake structures (see also co-location in Chapter 7). Typically, a co-located desalination plant takes its source water from 2 http://www.epa.gov/safewater/sdwa/30th/factsheets/standard.html#4.
OCR for page 112
Desalination: A National Perspective the power plant discharge water; thus, as long as the power plant is operating, the desalination facility does not increase the impacts from impingement and entrainment. However, should the power plant discontinue operating on an interim or permanent basis or if once-through cooling practices are phased out, water withdrawals would have to continue to provide source water to the desalination plant. It is worth noting that many of the nation’s power plants were sited decades ago, before the adverse environmental impacts of their intake structures were understood and before many of the current federal environmental legislation and regulations were in place, and some of the existing power plant intakes are located in areas where they create considerable environmental damage. Thus, the potential source water impacts of co-located desalination facilities still need to be considered. Brackish Groundwater Source Issues Some inland and coastal communities utilize brackish groundwater as a source for desalination, and withdrawal of brackish groundwater creates a quite different set of environmental concerns, including the physical sustainability of the aquifer and the potential for subsidence. Following a brief overview of brackish water resources in the United States, the potential environmental impacts from brackish groundwater withdrawal are discussed in more detail. A brackish aquifer is a geologic deposit of water-bearing permeable rock or unconsolidated materials from which brackish groundwater can be usefully extracted using a well. The processes that generate brackish groundwater depend on the site-specific hydrogeology and geochemistry. In some cases, high levels of dissolved solids are derived from the presence of connate water (i.e., seawater trapped at the time of original deposition), but in most inland brackish water systems these original solutes have long since been flushed away. In arid and semi-arid areas typical of the western United States, the major sources of salinity in groundwater are the evaporative concentration of solutes from precipitation and dissolution of minerals in the subsurface. In the humid east and other areas with higher groundwater recharge rates, major solutes in brackish waters originate from dissolution reactions of the water with minerals (e.g., halite [NaCl], gypsum [CaSO4], anhydrite [CaSO4 •2H2O], calcite [CaCO3], dolomite [CaMg(CO3)2]) present in the aquifer framework. Coastal aquifers form another class of natural brackish water created from mixing of groundwater that is discharging to the ocean. Under natural conditions most groundwater in coastal areas discharges directly to the ocean (Figure 5-1). The processes of molecular diffusion and hydro-
OCR for page 113
Desalination: A National Perspective dynamic dispersion (mixing by movement of fluids through a porous media) create a brackish zone of dispersion or a mixing zone. Coastal groundwater pumping can cause seawater intrusion that increases the thickness of the brackish water zone of dispersion. Brackish water from irrigation return flows can also be utilized as desalination source water, although the quantity and quality typically vary by season and region. In Colorado, some desalination plants use alluvial groundwater with elevated salinity as a result of agricultural land use in the drainage basin (e.g., Platte River). Development of this water source for desalination is site specific as to both quantity and quality. Brackish groundwater exists at less than 305 m (1,000 feet) across much of the United States (Feth, 1965) (see Figure 1-1). This groundwater consists of highly variable concentrations of dissolved solids and ranges from slightly brackish to brines with salt concentrations many times the concentration of seawater. The distribution, volume, and water quality of brackish water aquifers in the United States are largely unknown. Some states, such as New Mexico (Huff, 2004a, 2004b; New Mexico Office of the State Engineer, 2004) and Texas (Brackish Groundwater Manual, 2003), have made brackish water inventories based on existing data. Huff (2004a) estimated that huge quantities of brackish water (16 trillion m3 or 13 billion acre-feet) exist in New Mexico relatively close to the surface, some fraction of which could be desalinated for human use. However, there have been no national assessments, and the current regional assessments exhibit inadequate detail FIGURE 5-1. Diagram showing typical discharge of potable groundwater to the ocean and zone of dispersion (mixing zone). SOURCE: USGS (2007a).
OCR for page 114
Desalination: A National Perspective necessary for water resource management. Detailed site-specific evaluations, such as those conducted by the El Paso and Fort Bliss desalination facilities (see Box 5-2), will be necessary to assess the quantity and quality of water available for a given desalination facility. Nevertheless, a national compilation of existing data and regional evaluations of flow and solute boundary conditions, thickness, extent, and hydraulic conductivity of major brackish aquifer systems in the United States could provide the framework for a potentially greater utilization of brackish groundwater resources. Physical Sustainability Development of a brackish aquifer system for water supply demands an understanding the sustainability and renewability of the aquifer, in terms of both water quality and water quantity. The concept of physical sustainability of a natural resource has been defined in many ways. The United Nations Brundtland Commission (1987) popularized the term in the environmental sense when it defined sustainability as “the ability to meet the needs of the present generation without compromising the ability of future generations to meet their needs.” For the present report, a more conservative viewpoint is taken; a physically sustainable aquifer system is considered to be one in which recharge over human time frames approximately equals withdrawals and discharges from both anthropogenic and natural processes (i.e., renewability). Groundwater withdrawals that exceed the recharge capacity of the aquifer are sometimes referred to as groundwater mining. Under these circumstances, continued withdrawals may deplete the groundwater resource, create subsidence (discussed below), or affect the quality and quantity of adjacent water bodies or aquifers. Because the hydrology of groundwater, lakes, streams, and wetlands are frequently interconnected, the removal of water from one source means less water for one or more of the other sources. In terms of water quality, sustainable aquifers are defined here as those having concentrations that do not change significantly beyond the natural variability over human time frames. Solute concentrations and their ionic ratios are naturally variable throughout all aquifer systems (Hem, 1986), and increased pumping can induce groundwater flow and, thus, solutes from adjacent, underlying, and overlying aquifers or surface waters. Induced groundwater flow will, in most cases, lead to changes in water quality due to chemical reactions and transformations within the aquifer matrix (e.g., ion exchange, dissolution, precipitation). Mineral precipitation and dissolution in the aquifer matrix can potentially alter the hydraulic conductivity of the aquifer over time (Johnson et al., 2005).
OCR for page 115
Desalination: A National Perspective Box 5-2 Kay Bailey Hutchison Desalination Plant A desalination project was proposed for the El Paso, Texas, area after the 50-year Texas State-Wide Water Resource Management Plan for Far West Texas Region indicated that the projected future population growth of the El Paso area would experience water demands in excess of supplies. This proposed shortfall would occur in spite of the already large consumption decline associated with conservation and rate-structure change practices that resulted in a per capita decline from 0.870 m3/day (230 gallons per day) in 1977 to 0.518 m3/day (137 gallons per day) in 2006. At the same time the nearby U.S. Army facility at Fort Bliss was expanding its mission and sought to increase its water supply. Looking at all possible water sources, El Paso Water Utilities (EPWU) studies indicated that desalinated water would be less costly than using reclaimed sewage or importing additional water but approximately twice as expensive as current groundwater and surface water supplies. A proposal was made for a cooperative effort between the city of El Paso and the U.S. Army to develop a 10,400 m3/day (27.5 million-gallon-per-day) desalination facility using brackish groundwater from the Hueco Bolson. The capital cost of the source water, desalination facility, and management of by-product concentrate was to be $87 million (2005 U.S. dollars). Previous pilot plant operation by EPWU in 1993-1994 suggested the likely success of using reverse osmosis technology with Hueco Bolson water. The source water would have TDS ranging between 1,200 and 1,500 mg/L TDS and produce a final blended water of 700 to 800 mg/L TDS, comparable to existing water quality. The desalination facility would require 17 wells for source water and 16 wells for blending. The addition of antiscalant and mineral acid to the source water is designed to inhibit mineral precipitation on the membranes, transmission pipes, and walls of the injection wells. A MODFLOW numerical model was constructed with cooperation by the U.S. Geological Survey and Fort Bliss, the International Boundary and Water Commission, and the City of Juarez Mexico Water Utility to evaluate the influence of removing brackish water on the aquifer system. This peer-reviewed model found that the relatively low projected rate of withdrawal for the desalination and blending from this large aquifer system would be environmentally benign and was sustainable through the projected life of the facility. Disposal of byproduct fluids by well injection was determined to be cost-effective compared to both passive and enhanced evaporation procedures. Geophysical investigations and test wells indicated the Fusselman dolomite, a relatively high-permeability formation located 29 miles northeast of the desalination facility, would be hydrologically suitable for disposal. A disposal permit was obtained from the State of Texas to inject desalination effluent into this aquifer. The injected water was lower in TDS than the native water in the Fusselman formation and only exceeded drinking water standards for arsenic and selenium. SOURCE: Ed Archuleta, EPWU, personal communication, 2006. Ion exchange in certain instances may also change the physical properties of the ion-exchanger clay mineral and result in loss of permeability of the aquifer (Civan, 2000). These impacts to water quality and quantity can be anticipated through hydrogeologic and geochemical analyses and the use of model
OCR for page 116
Desalination: A National Perspective simulations (see Box 5-3). The challenge to hydrogeologists is to identify and quantify all hydrologic boundaries and transport conditions under different scenarios of water supply development as well as any chemical reactions that add, subtract, or change solutes or ratios under varying flow conditions. Each well field is unique with respect to these properties and requires site-specific information. Subsidence When groundwater withdrawals exceed recharge rates, the hydraulic head (related to fluid pressure) will gradually be reduced and may result in land subsidence in areas with unconsolidated sediments (Figure 5-2). Subsidence results from the compression of the skeletal framework caused by reduced hydraulic head and rearrangement of grains in the aquifer matrix. Under equilibrium conditions, total stress (a function of the mass of water and rock) acting downward on a plane is balanced by a combination of the geologic framework acting upward (the skeletal framework resisting compression, or effective stress) and fluid pressure (hydraulic head). The reduction in hydraulic head associated with withdrawal of fluids increases the effective stress on the geologic framework and causes compaction and reduction in elevation of the land surface (Tergazi and Peck, 1967). Excessive development of a brackish water resource, particularly one that possesses thick sections of at-risk lithologies (i.e., clay, silt, organic material), may create the potential for subsidence. Land subsidence resulting from removal of groundwater has affected areas in 45 states (Figure 5-3) and ranges from regional lowering to ground failure and collapse (Galloway et al., 1999; NRC, 1991). Galloway et al. (1999) estimate that more than 80 percent of the identified subsidence in the United States has been caused by overexploitation of groundwater resources. Parts of Texas, California, and Nevada have experienced tens of meters of surface decline (Leake, 2007). Regional lowering will increase the probability of flooding in coastal areas, while local settling may damage engineered structures such as buildings, roads, and utilities. Subsidence also has the potential to trigger earthquakes and activate faults; for example, the Gulf Coast basin is a region where subsidence and fault activation are common around large, mature oil and gas fields (USGS, 2007b).
OCR for page 117
Desalination: A National Perspective BOX 5-3 Tools for Addressing Brackish Groundwater Physical Renewability Several modeling tools are available to assess the potential impacts of a proposed brackish groundwater desalination facility on the physical renewability of an aquifer. Once appropriate site-specific information is obtained, numerical groundwater models, such as variations of MODFLOW (Harbaugh, 2005), can be used to test various scenarios of development and effects on water quantity. If density contrasts between water bodies are significant, then codes such as SEAWAT-2000 (Langevin et al., 2003) and SUTRA (Voss and Provost, 2003) may be more appropriate codes to use. Other models, such as PHRQPITZ (Plummer et al., 1988) and PHREEQC (Parkerhurst, 1995), address water quality impacts. MODFLOW-2005 simulates steady and nonsteady flow in an irregularly shaped flow system in which aquifer layers can be confined, unconfined, or a combination of confined and unconfined. Flow from external stresses, such as flow to wells, areal recharge, evapotranspiration, flow to drains, and flow through riverbeds, can be simulated. Specified head and specified flux boundaries can be simulated. In addition to simulating groundwater flow, MODFLOW-2005 incorporates related capabilities such as solute transport and groundwater management. SEAWAT-2000 is the latest release of the SEAWAT computer program for simulation of three-dimensional, variable-density, transient groundwater flow in porous media. SUTRA is a model for saturated or unsaturated, variable-density groundwater flow with solute or energy transport. SUTRA version 2D3D.1 includes both two- and three-dimensional simulation capability. Water quality changes and aquifer transformation in brackish or brine conditions can be assessed using the model PHRQPITZ, which addresses chemical reactions (e.g., mineral precipitation, dissolution) within brackish aquifers. PHRQPITZ is a computer code that permits calculations of geochemical reactions in brines and other highly concentrated electrolyte solutions using the Pitzer virial-coefficient approach for activity-coefficient corrections. Reaction-modeling capabilities include calculation of (1) aqueous speciation and mineral-saturation index, (2) mineral solubility, (3) mixing of aqueous solutions, (4) irreversible reactions and mineral-water mass transfer, and (5) reaction path. PHREEQC (version 2), which adds the Pitzer coefficients for dealing with brackish water, is a computer program that is designed to perform a wide variety of low-temperature aqueous geochemical calculations. PHREEQC has capabilities for (1) speciation and saturation-index calculations, (2) batch-reaction and one-dimensional (1D) transport calculations, and (3) inverse modeling. Transport capabilities involve reversible reactions (e.g., aqueous, mineral, gas, solid-solution, surface-complexation, and ion-exchange equilibria) and irreversible reactions (e.g., kinetically controlled reactions, mixing of solutions, temperature changes). PHREEQC version 2 includes new capabilities to simulate dispersion (or diffusion) and stagnant zones in 1D-transport calculations and to model kinetic reactions with user-defined rate expressions.3 3 For more information on these models, see http://water.usgs.gov/software/lists/ground_water/.
OCR for page 118
Desalination: A National Perspective FIGURE 5-2. Land subsidence from groundwater withdrawal. SOURCE: Galloway et al. (1999). FIGURE 5-3. Areas where subsidence has been attributed to groundwater withdrawal. SOURCE: Galloway et al. (1999).
OCR for page 136
Desalination: A National Perspective Elevated salinity is stressful to many freshwater organisms. Sarma et al. (2005) studied freshwater crustaceans (anostracans) that inhabit ephemeral water bodies in which the water level decreases due to evaporation, increasing the salt concentration. They found that increased salinity resulted in decreased survivorship. Females showed several peaks of reproduction at 0 and 1 ppt salinity, whereas at 4 or 8 ppt there were fewer peaks. The highest reproductive rate was in 0 ppt of salt, while the lowest was at 8 ppt. Average lifespan, life expectancy, gross and net reproductive rates, generation time, and the rate of population increase were inversely related to the salt concentration. Freshwater species from environments that do not normally experience increased salinity would likely be much more susceptible than these anostracans. To minimize environmental effects, concentrate discharge to rivers (and sewers) needs to be coordinated with background water quality, the composition of the concentrate, discharge rates, blending characteristics, and local water quality standards. Sewer discharge of concentrate upstream of the wastewater treatment plant should also be managed so as not to exceed the capacity of the treatment plant or to adversely impact its biological processes with excessive salinity. Evaporation Ponds. Potential environmental impacts from the use of evaporation ponds include leakage of the concentrate and degradation of underlying aquifer systems or adjacent freshwater resources. Engineered low-permeability barriers are used to reduce the likelihood of leakage from the pond. Other factors that affect environmental water quality include sufficient basin storage volume to prevent overflow in case of major precipitation events, and location of sites topographically above long-term flood reoccurrence intervals of nearby water sources. The elevated salinity and trace constituents in evaporation ponds may be problematic for breeding and migrating birds, as was seen with the selenium effects on birds at the Kesterson National Wildlife Reserve (Hannam et al., 2003; Hoffman et al., 1988; NRC, 1989). Land Application. As described in Chapter 4, the allowable salinity for land application depends on the tolerance of target vegetation, percolation rates, and the ability to meet the groundwater quality standards and is, therefore, more viable for lower salinity concentrate. In many cases, additional dilution water is needed for land application to be feasible. It may be possible to genetically engineer better salt-tolerant plants in the future and utilize these plants for animal fodder (Grattan et al., 2004; Grieve et al., 2004). However, if transpiration from the plants exceeds precipitation to the soil, over time any salts not taken up by the plants will accumulate in the soil. If the source water, and thus the concentrate, contains contaminants of concern such as arsenic, nitrate, or other harm-
OCR for page 137
Desalination: A National Perspective ful trace metals, the potential environmental impacts could include uptake of these contaminants by the plants or leaching of these contaminants into the soils or groundwater. Currently, in arid and semi-arid environments (generally west of the 100 Meridian in the United States), land application is not a physically sustainable method for disposal of desalination concentrate, because it is likely to exacerbate an already a large worldwide problem of soil salinization (NRC, 1993). Injection Wells. Disposal of concentrates through injection wells is required to meet criteria established by the EPA for its Underground Injection Control Program (EPA, 2007b; see Box 5-1) to ensure that well injections do not endanger aquifers supplying drinking water by allowing the injected concentrate to enter the aquifer and degrade the resource. It should be understood that the amount of aquifer storage in a typical confined aquifer injection environment is small—about 1 m3 per 10,000 m3 of aquifer material—and thus, there will be displacement of current aquifer fluids. If disposal occurs in a depleted oil reservoir or an unconfined aquifer, storage would be much larger—about 1 m3 per 10 m3 of aquifer material. To prevent adverse impacts to surrounding aquifers, the volume, location, and solute composition of any displaced fluids and how they might influence the water quality of surrounding aquifers or surface waters should be well understood. This involves quantifying all flow boundaries and simulating groundwater flow dynamics using appropriate three-dimensional numerical transport and flow models (see Box 5-3). Concentrate injection in artesian aquifer systems, which are typical of most formations used for deep-well injection, locally causes increase in fluid pressure and vertical expansion of the aquifer framework, which may be expressed as a rise in land surface. This increase in fluid pressure can also trigger earthquakes in certain geologic environments. Deep injection wells have caused several large-magnitude earthquakes (5 or greater on the Richter scale) and several thousand smaller ones in areas that are structurally stressed, such as the Rocky Mountains in Colorado (Evans, 1966; Hsieh and Bredehoeft, 1981) and Rangely oil field, Colorado (De la Cruz and Raleigh, 1972). Thus, proposed injection sites need to consider the potential for this condition if the target formation is deep and in an area that has experienced tectonic activity in the relatively recent geologic past. Landfilling. One potential disposal option is to convert the concentrate from a liquid to a solid (or a dense slurry) and then dispose of the waste material in a suitable landfill (see Thermal Evaporation in Chapter 4). It requires a great deal of energy, however, to remove and recover the liquids from the concentrate and then to transport the wastes to a landfill, and these requirements may have significant financial, social, and envi-
OCR for page 138
Desalination: A National Perspective ronmental ramifications (see Greenhouse Gases in this chapter). Because most landfills eventually leak, there are also potential future environmental impacts to groundwater near the landfill. WATER QUALITY ISSUES IN DESALINATED PRODUCT WATERS4 Because desalination processes employ advanced water treatment techniques, it is commonly assumed that desalinated water is devoid of contaminants. In reality, although desalination technologies remove various constituents to a large extent, not all constituents are fully removed and some species are removed to a lesser extent than others. In RO, a small fraction of ions, especially monovalent ions such as sodium and chloride, and dissolved organic molecules (e.g., some pesticides or herbicides) can pass through to the permeate water. Desalinated product water quality depends on the raw water quality, the treatment technology selected (e.g., RO, distillation, electrodialysis), and within membrane technologies, by the specific membranes employed and the implementation of second-pass RO. Boron and bromide are two inorganic constituents associated with water quality concerns in RO desalination, and these challenges along with approaches to mitigate these concerns are described next. Boron occurs in the oceans at an average concentration of 4.5 mg/L (Weast et al., 1985). Although thermal desalination removes boron, the rejection of boron in RO desalination is dependent on the pH. Rejection increases with pH, although the single-pass RO process is operated at a low pH to avoid scaling. Single-pass RO desalination processes do not remove the majority of boron in the raw water at typical operating pH ranges; thus, boron (occurring as borate or boric acid) can be found at milligram-per-liter levels in the finished water. Implementation of a second pass through RO membranes with a pH adjustment to place boric acid in its negatively charged borate form can provide effective boron removal (Karry, 2006; Magara et al., 1998). Second-pass RO installation and operation, however, have significant cost implications and historically are not routinely included in desalination projects. 4 Minor changes have been made to this section, a related conclusion at the end of the chapter, and an associated research recommendation in Chapter 8 after release of the prepublication version to incorporate data on boron toxicity and exposure levels from the 2000 IOM report, Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc.
OCR for page 139
Desalination: A National Perspective Although boron is recognized to have a beneficial role in some physiological processes in some species, higher exposure levels may cause adverse health effects (IOM, 2000). No human health effect data are available on adverse health effects from ingestion of large amounts of boron from food and water, although data are available on the human health effects of large and small doses of boric acid or borax. However, most of the boron toxicity data come from studies in laboratory animals. High-dose boron exposures had the greatest effect on developing fetuses and on testes and led to reduced fertility in experimental animals (IOM, 2000). The EPA concluded that there is inadequate data to assess the human carcinogenicity of boron (EPA, 2006). Some research on the environmental effects of boron has shown that boron at milligram-per-liter levels also can adversely impact crops and grass species (Yermiyahu et al., 2007). Boron was considered in the EPA’s second Drinking Water Contaminant Candidate List (CCL2).5 In the CCL2, the EPA defined a reference dose (or the level of lifetime exposure at which no adverse health effects are expected) of 0.2 mg/kg/day, conservatively estimated based on developmental effects in rats as well as applied uncertainty factors based on the extrapolation of data from animals to humans (EPA, 2006). Using the same toxicity data, but slightly different assumptions about uncertainty factors, the Institute of Medicine (2000) recommended a modestly higher exposure level of 0.32 mg/kg/day (or 20 mg/day for adults; 17 mg/day for adolescents, and exposures ranging from 3 to 11 mg/day for children ages 1-13). Translating these exposure guidelines into recommended limits for boron concentrations in water requires assumptions regarding other possible sources of boron. According to IOM (2000), airborne boron contributes very little to the daily exposure of the general population. For humans not taking dietary supplements, diet is the major source of boron intake followed by drinking water. In the U.S., the median intake of dietary and supplemental boron was estimated to be approximately 1.0 to 1.5 mg/day for adults (IOM, 2000).6 Based on its calculated boron reference dose and an assumption that 20 percent of total daily boron consumption would come from drinking water, the EPA developed a health reference level for drinking water as 1.4 mg/L boron. With different assumptions of the total amount of boron exposures from drinking 5 See http://epa.gov/safewater/ccl/ccl2.html#chemical. 6 Ninety-fifth percentile dietary intakes of boron in the U.S. are approximately 2.3 mg/day and 1.6 to 2.0 mg/day for men and women, respectively; 2.7 mg/day and 4.2 mg/day for vegetarian men and women, respectively (Rainey et al., 1999). The average intake of supplemental boron at the ninety-fifth percentile is approximately 0.4 mg/day for adults (IOM, 2000).
OCR for page 140
Desalination: A National Perspective water and other sources, this water quality guidance can vary. The State of California has adopted a notification level for boron at 1 mg/L (California Department of Public Health, 2007). The current World Health Organization (WHO, 2004) guideline for boron in drinking water is 0.5 mg/L, but this is due to be reconsidered under the rolling revision of the guidelines (WHO, 2007). A recent draft WHO report notes that the revised health-based guideline (anticipated in 2008) might be 1 mg/L or higher (WHO, 2007). Because boron is not likely to be found at levels of concern in surface waters and groundwater, the EPA also made a preliminary determination not to regulate boron with enforceable drinking water standards (see Box 5-1; EPA, 2007a). Instead, the EPA encouraged states with public water systems that have boron at concentrations higher than 1.4 mg/L to evaluate site-specific protective measures and to consider whether state-level guidance or regulation is appropriate. Although boron is not specifically regulated in product water in the United States, consumer expectations may pressure desalination planners to design future seawater plants to follow these current guidelines. Treatment to these levels will increase the cost of new seawater desalination plants. Additional analysis of the human health effects of boron in drinking water, considering other sources of boron, are needed to support firm state-level water quality guidance for seawater desalination process design that is suitably protective of public health. If seawater desalination becomes a significant source for drinking water supply in the United States, additional regulatory attention or national guidance may be needed. Because it is difficult for RO technologies to meet current boron guidelines in single-pass operations if there is boron in the feedwater, membranes and processes are being developed to reduce the level of boron in the product water. In some areas, specific resins combined with a small-scale RO are used to reduce the amount of boron. Boron can also be removed by optimization of RO, such as via multistep desalination or by coprecipitation with hydroxides (Cotruvo, 2005; Hyung and Kim, 2006). A technique for boron removal through reacting seawater with fly ash and coal materials has also been developed (Vengosh et al., 2004). Future seawater desalination projects should consider boron treatment options early in their planning efforts when considering the various end uses of the water produced. Bromide is another water quality consideration for membrane desalination projects. Bromine is formed by the reaction between bromide and free chlorine, which is often used as a biocide to control biological growth in the intake and pretreatment systems for seawater desalination plants. Bromine in its uncharged form (HOBr) passes through RO membranes and is found in permeate water. Bromine participates in the
OCR for page 141
Desalination: A National Perspective formation of disinfection by-products (e.g., bromoform, dichlorobromomethane, dibromochloromethane) when it reacts with natural organic matter (Laine et al., 1993; Singer, 1999; Summers et al., 1994). These by-products may have adverse human health effects (Richardson, et al., 2007) and are regulated through the SDWA (see Box 5-1) Disinfection By-product Rules. RO membranes have relatively low rejection capability for trihalomethane disinfection by-products; thus, some of these compounds pass through the membranes and reside in the permeate. Brominated by-products may also be formed if the desalination product water containing bromine is blended with water from other traditional sources containing natural organic matter. Bromide can also adversely affect the stability of chloramine in finished waters (Duirk and Valentine, 2007). To minimize disinfectant by-product formation, chlorine is generally used only intermittently during pretreatment, but, in some cases of high organic loading, this may not be possible. At the Tampa Bay Seawater Desalination Plant, chlorine dioxide is utilized to control biological growth in the pretreatment process. The plant had previously used free chlorine but the disinfectant was changed due to elevated disinfection byproduct formation. During the post-treatment process, monochloramines are formed as a secondary disinfectant to further reduce the disinfection by-product formation. Most utilities that have switched from a free chlorine residual to a monochloramine residual have done so primarily to reduce disinfection by-product formation to comply with current and future disinfection by-product regulations (Dyksen, 2007). During the formation of monochloramines, the ammonia can also combine with bromine to form bromamines (Bousher et al., 1989). Bromamines are not recognized by the EPA as an approved drinking water disinfectant. In summary, boron, bromide, and disinfection by-products can affect product water quality. All are controllable through treatment optimization, but that treatment could aversely affect the cost of desalination. GREENHOUSE GAS EMISSIONS Water resource management currently uses significant amounts of electrical and natural gas energy to capture, treat, and transport water. The California Energy Commission (2005a) estimates that capture, transportation, and treatment of water uses approximately 5 percent of the electrical energy consumed in the state. Because of climate, geology, topography, and long water conveyance routes, the energy use for capture, transportation, and treatment in California is higher than the national average of 3.5 percent of electrical energy consumed (U.S. Department of Energy, 2006). Desalination is an energy-intensive process
OCR for page 142
Desalination: A National Perspective that would add more demand. A comparison of energy use for different water sources (Table 5-2) suggests that seawater RO requires about 10 times more energy than traditional treatment of surface water (Cohen et al., 2004). Concerns over anthropogenic climate change have spurred interest in the energy requirements of desalination. Although the percentage of statewide energy use is likely too small for desalination, planners will need a clear understanding of the energy and climate implications of desalination relative to other water supply alternatives as the nation takes steps to address the issue of greenhouse gas emissions. Energy sources other than fossil hydrocarbons can provide energy for desalination and thus avoid or significantly reduce greenhouse gas emissions. Technologies such as nuclear (19.3 percent of electrical power in the United States), hydroelectric (6.5 percent), wind (<1 percent), and solar photovoltaic (<1 percent) are providing input to the electrical grid (Edison Electrical Institute, 2005) and are not associated with the generation of greenhouse gases. Other alternative energy sources such as biofuels (1.6 percent) are nearly neutral in terms of greenhouse gas emissions (Adler et al., 2007), and closed-loop geothermal systems can be essentially greenhouse gas-free. As discussed in TABLE 5-2. Comparison of Energy Use for Different Water Sources in California. Water Source Energy Used per Cubic Meter of Water (kWh/m3) Pumping groundwater 120 ft 0.14 Pumping groundwater 200 ft 0.24 Treatment of surface water 0.36 Brackish water desalination ~0.3 to 1.4 Water recycling (no conveyance) ~0.3 to 1.0 Conveyance of water (examples): Colorado River Aqueduct to San Diego 1.6 San Francisco Bay Delta to San Diego 2.6 Seawater desalination (no conveyance) ~3.4 to 4.5 NOTE: Numbers reflect cited case-study examples and are not statewide averages. SOURCE: Cohen et al. (2004), reprinted with permission from the National Resources Defense Council.
OCR for page 143
Desalination: A National Perspective Chapter 4, thermal desalination plants can utilize low-grade or waste heat resources and substantially reduce their prime energy demands. Commercial applications of alternative energy sources to power desalination remain somewhat limited. A 125,000 m3/day membrane desalination facility in Perth, Australia (Water Corporation, 2007), that began operation in 2007 is the first example of using alternative energy to power desalination at a large scale. The Perth wind farm is not a dedicated stand-alone power source; rather it feeds into the power grid from which the desalination plant contracts to withdraw its electrical power. Waste heat from Japanese nuclear plants has been used to generate boiler water for the plants’ own use, but no dedicated nuclear power plants have yet been developed for the purpose of powering water desalination (IAEA, 2005; Minato and Hirai, 2003; Pankratz, 2005). A review of the potential for alternative energy for desalination (European Commission, 1998) and discussions of alternative energy for remote offgrid areas (García-Rodríguez, 2003; Tzen and Morris, 2003) suggest that several alternative energy sources hold promise. A variety of alternative energy sources have been proposed for various locations, depending on local conditions. These include photovoltaic (Richards and Schäfer, 2003) and heat-driven processes, such as direct solar evaporation (Trieb et al., 2003), closed geothermal (Bourounia et al., 1999; Karytsasa et al., 2004), ocean thermal energy conversion, and salinity-gradient solar ponds (Lu et al., 2001). Solar-powered desalination coupled with water reuse is a centerpiece of Masdar, an initiative in the United Arab Emirates to build the world’s first carbon-neutral city. Proposed mechanical-driven alternative energies for desalination include wind power (Liu et al., 2002), wave power, tides, and hydrostatic head. Thus, there are numerous alternative energy technologies available, and these technologies may be able to provide the right quality of energy for desalination while reducing overall greenhouse gas emissions. More research, however, is needed to analyze the alternatives for coupling desalination with alternative energy sources in both inland and coastal areas. Climate Change and Desalination There seems to be no question that climate change will significantly impact the water resources sector and, as such, will indirectly impact desalinization. A rise in sea level over tens of years may have adverse impacts on coastal aquifers from increased seawater intrusion. Direct impacts of rising ocean levels may over the lifetime of the project have some minor effect on desalination structures built adjacent to coastlines because current sea-level rise is approximately 2 mm/year (United Na-
OCR for page 144
Desalination: A National Perspective tions Intergovernmental Panel on Climate Change, 2007). Furthermore, storms associated with climate warming may be of either higher frequency or higher intensity. Depending on the location of the intake, the temperature of the water may increase slightly, requiring small changes to the desalination process. Although these direct impacts to desalination structures and processes appear to be small, they should be clearly understood prior to the design of a major desalination facility. CONCLUSIONS AND RECOMMENDATIONS Knowledge of the potential environmental impacts of desalination processes is essential to water supply planners when considering desalination among many water supply alternatives. All components of the water-use cycle should be considered, including source water impacts, the likely greenhouse gas emissions from the energy requirements of the desalination process, potential impacts from concentrate management approaches, and environmental health considerations in the product water. Ideally, these considerations should be compared against equally rigorous environmental impact analyses of water supply alternatives. The role of science and engineering is to clearly articulate the environmental impacts in a transparent manner so that society can make an informed decision after comparing the full economic costs—including environmental costs—and benefits among the various water supply alternatives (as discussed in Chapter 6). Because of the limited amount of long-term research, there is presently a considerable amount of uncertainty about the environmental impacts of desalination and, consequently, concern over its potential effects. A variety of environmental impacts are possible with desalination. Seawater desalination can cause impingement and entrainment of marine organisms and create ecological impacts from concentrate discharge. Desalination of inland brackish groundwater sources could lead to groundwater mining and subsidence, and improper concentrate management practices can negatively affect drinking water aquifers and freshwater biota. Site-specific information necessary to make detailed environmental conclusions on the ecological impacts of both source water withdrawal and concentrate management associated with desalination is lacking. The limited studies to date suggest that the environmental impacts may be less detrimental than many other types of water supply, but definitive conclusions cannot be made until more research is done.
OCR for page 145
Desalination: A National Perspective Site-specific assessments of the impacts of source water withdrawals and concentrate management should be conducted and the results synthesized in a national assessment of potential impacts. Adequate understanding of impacts of source water withdrawals and concentrate management results only from site-specific assessments. The ecological effects of concentrate discharge into the ocean appear to vary widely and depend on the site-specific environment, the organisms examined, the amount of dilution of the concentrate, and the use of diffuser technology. The ecological impacts of surface water intakes (i.e., impingement and entrainment) have been well studied for power plants but not for desalination plants, and these impacts will likely vary from place to place. General information on potential impacts from groundwater withdrawal and injection are available from decades of hydrogeologic studies for other purposes, but site-specific analyses are necessary to understand the impacts from a proposed facility. Once a number of rigorous site-specific studies are conducted, this information should be synthesized to develop an overarching assessment of the possible range of impacts from both seawater and brackish water desalination in the United States. A characterization of the volume, hydraulic properties, flow boundary conditions, and solute chemistry of the nation’s brackish groundwater resources and a characterization of the spatial distribution, thickness, and hydraulic properties of aquifer systems suitable for concentrate injection, relying heavily upon existing data, would assist the financial and environmental planning process for inland desalination facilities. Longer-term, laboratory-based assays of the sublethal effects of concentrate discharge should be conducted. Except for a few short-term lethality studies that do not give insight into long-term effects, research on the impacts of concentrate discharges on organisms in receiving waters has been minimal. Longer-term laboratory-based biological assays, running from weeks to months in duration, should evaluate impacts of concentrate on development, growth, and reproduction using a variety of different organisms, including those native to areas where desalination plants are proposed. These results should be put into a risk assessment framework. Monitoring and assessment protocols should be developed for evaluating the potential ecological impacts of surface water concentrate discharge. Adequate site-specific studies on potential biological and ecological effects are necessary prior to the development of desalination facilities because biological communities in different geographic areas will have differential sensitivity. For large desalination facilities, environmental data should be collected for at least 1 year in the area of
OCR for page 146
Desalination: A National Perspective the proposed facility before a desalination plant comes online so that sufficient baseline data on the ecosystem are available with which to compare postoperating conditions. Once a plant is in operation, monitoring of the ecological communities (especially the benthic community) receiving the concentrate should be performed periodically and compared to reference sites. Water quality guidance, based on an analysis of the human health effects of boron in drinking water and considering other sources of exposure, is needed to support decisions for desalination process design. There are concerns about boron in product water from seawater desalination because the boron levels after single-pass RO commonly exceed current WHO health guidelines and the EPA health reference level. A range of water quality levels (0.5 to 1.4 mg/L) have been proposed as protective of public health based on different assumptions in the calculations. The EPA has decided not to develop an MCL or health-based MCLG for boron because of its lack of occurrence in most groundwater and surface water and has encouraged affected states to issue guidance or regulations as appropriate. Therefore, most U.S. utilities lack clear guidance on what boron levels in drinking water are suitably protective of public health. Boron can be removed through treatment optimization, but that treatment could aversely affect the cost of seawater desalination. Further research and applications of technology should be carried out on how to mitigate environmental impacts of desalination and reduce potential risks relative to other water supply alternatives. For example, intake and outfall structures could be designed to minimize impingement and entrainment and encourage improved dispersion of the concentrate in coastal discharges. Research could also explore beneficial reuse of the desalination by-products and develop technologies that reduce the volume of this discharge. There are numerous alternative energy technologies available, and these technologies may be able to provide the right quality of energy for desalination while reducing overall greenhouse gas emissions; however, research is needed to analyze the alternatives for coupling desalination with alternative energy sources in both inland and coastal areas. Additional research investments should be able to clarify the potential risks of desalination and develop approaches to substantially mitigate the environmental impacts. Nevertheless, desalination efforts do not need to be halted until this research is done and uncertainties removed.