Ground Water in Freshwater-Saltwater Environments of the Atlantic Coast

Ground water is a vital resource for communities and ecosystems of the Atlantic coastal zone. Ground-water withdrawals for public supplies, agriculture, industry, and other uses in coastal counties from Maine to Florida were about 7.7 billion gallons per day in 1995, and in many coastal communities, ground water is the primary or sole source of drinking-water supply. Ground water sustains the flow of coastal streams and rivers and is a source of freshwater to coastal ponds, wetlands, and other coastal ecosystems. The U.S. Geological Survey (USGS) has a long history of conducting scientific studies of ground water in the Atlantic coastal zone. This Circular draws heavily on the body of scientific knowledge developed over several decades of USGS monitoring and studies.

Because of a growing awareness of the critical role of ground water in sustaining coastal populations, economies, and ecosystems, the time is right to review some of the important water-management issues and scientific principles related to ground water in the Atlantic coastal zone, and to identify some of the scientific and management challenges that lie ahead. This Circular describes the occurrence and flow of freshwater and saltwater in ground-water systems of the Atlantic coastal zone; reviews the causes, modes, and management of saltwater intrusion along the Atlantic coast; and illustrates some of the mechanisms of ground-water discharge and contaminant loading to coastal ecosystems. The Circular also illustrates traditional approaches for monitoring and managing saltwater intrusion, and highlights some of the innovative approaches being used to enhance the sustainability of coastal ground-water resources, such as desalination and aquifer storage and recovery systems. As coastal populations and ground-water use increase, new monitoring and research efforts will be needed to characterize the occurrence and movement of saline ground water in different types of coastal terrains and to better understand linkages between ground-water discharge and quality and the health of coastal ecosystems. The USGS looks forward to continued service to the citizens of the Atlantic coastal zone, providing science to help them manage their water resources.

INTRODUCTION

The Nation's coastal regions are economically and environmentally vibrant areas that support major population centers and diverse ecosystems. Although they constitute less than 20 percent of the conterminous land area of the United States, coastal counties are home to more than half the Nation's population and many of its largest cities (National Oceanic and Atmospheric Administration, 1998). In 1995, more than 65 million people from Maine to Florida lived in coastal counties along the Atlantic Ocean and Gulf of Mexico, with large population centers within the Boston, Massachusetts, to Washington, D.C., corridor and in Florida (fig. 1A). Coastal areas include many of the fastest growing counties in the Nation, suggesting that coastal communities will continue to expand in the future.

Coastal populations and industries require a host of natural resources to sustain them, perhaps most important of which is a reliable source of freshwater. On a regional basis, the humid eastern seaboard has an abundant renewable freshwater supply, which has contributed to the growth and prosperity of the region. In 1995, total freshwater withdrawals for public supplies, agriculture, industry, and other offstream uses in coastal counties from Maine to Florida were about 30 billion gallons per day (Bgal/d), of which about 7.7 Bgal/d were supplied by ground water (fig. 1B) (U.S. Geological Survey, 2000). Although ground water supplied only about one-fourth of the total freshwater used, in many coastal communities ground water is the primary or sole source of drinking-water supply, and its use is increasing in many areas (fig. 2).

As ground-water use has increased in coastal areas, so has the recognition that ground-water supplies are vulnerable to overuse and contamination. Ground-water development depletes the amount of ground water in storage and causes reductions in ground-water discharge to streams, wetlands, and coastal estuaries and lowered water levels in ponds and lakes. Contamination of ground-water resources has resulted in degradation of some drinking-water supplies and coastal waters. Although overuse and contamination of ground water are not uncommon throughout the United States, the proximity of coastal aquifers to saltwater creates unique issues with respect to ground-water sustainability in coastal regions. These issues are primarily those of saltwater intrusion into freshwater aquifers and changes in the amount and quality of fresh ground-water discharge to coastal saltwater ecosystems.

Saltwater intrusion is the movement of saline water into freshwater aquifers and most often is caused by ground-water pumping from coastal wells. Because saltwater has high concentrations of total dissolved solids and certain inorganic constituents, it is unfit for human consumption and many other anthropogenic uses. Saltwater intrusion reduces fresh ground-water storage and, in extreme cases, leads to the abandonment of supply wells when concentrations of dissolved ions exceed drinking-water standards. The problem of saltwater intrusion was recognized as early as 1854 on Long Island, New York (Back and Freeze, 1983), thus predating many other types of drinking-water contamination issues in the news.

Coastal ecosystems are sensitive to the salinity and nutrient concentrations of coastal waters. In recent years, scientists, coastal managers, and public decisionmakers have recognized that many of the environmental issues related to coastal ecosystems—red tides, fish kills, loss of seagrass habitats, and destruction of coral reefs—can be attributed to the introduction of excess nutrients (nitrogen and phosphorus) from freshwater discharges (National Research Council, 2000). Ground water can be a significant source of freshwater to some coastal waters, and its role in delivering excess nutrients to coastal ecosystems is of increasing concern because of widespread nutrient contamination of shallow ground water (U.S. Geological Survey, 1999).

This report summarizes the current (2003) understanding of ground water in freshwater-saltwater environments of the Atlantic coast from Maine to Florida (including areas of Florida that are along the eastern Gulf of Mexico). These environments include coastal aquifers, salt marshes, and coastal waters (fig. 3). The report describes the natural occurrence and flow of freshwater and saltwater in coastal aquifers and the causes and modes of saltwater intrusion along the Atlantic coast. The role of ground water in coastal ecosystems—an area of increasing scientific and resource-management interest—also is reviewed. The report makes extensive use of case studies to illustrate the variety of freshwater-saltwater interactions that occur along the Atlantic coast—interactions that take place on time scales that range from daily tidal cycles to millennia and on spatial scales that range from beach faces a few feet wide to regional aquifer systems that extend over thousands of square miles. The case studies also demonstrate the many approaches and tools that are used to detect and monitor saltwater in coastal aquifers and to manage and prevent saltwater intrusion.

Although this report focuses on conditions along the Atlantic coastal zone, saltwater intrusion has been a problem in other coastal areas of the Nation and throughout the world (Bear and others, 1999). Moreover, it is now recognized that ground water can be an important global pathway for transport of materials from the land to coastal ecosystems and the oceans (for example, see Johannes, 1980; Simmons, 1992; Church, 1996; Moore, 1996, 1999; Basu and others, 2001; and Burnett and others, 2002).

CHAPTER 1. OCCURRENCE AND FLOW OF FRESHWATER AND SALTWATER IN COASTAL AQUIFERS

Ground water occurs in pores, fractures, solution cavities, and other openings in geologic formations that underlie the Atlantic coastal zone (fig. 4). The nature of the water-bearing openings within a specific geologic formation depends to a large extent on the mineral composition and structure of the formation and the geologic processes that initially formed and then further modified it. Geologic formations that occur along the Atlantic coast are of three major groups—unconsolidated (soil-like) deposits, semiconsolidated deposits, and consolidated rocks (Heath, 1998; Miller, 1999). Unconsolidated deposits consist of rock fragments and mineral grains that range in size from fractions of a millimeter (clay size) to several meters (boulders). In contrast, consolidated rocks consist of mineral particles that have been welded by heat and pressure or cemented by chemical reactions and precipitation into a solid mass. Consolidated rocks commonly are referred to as bedrock. Semiconsolidated deposits fall between the extremes of unconsolidated deposits and consolidated rocks (Heath, 1998). Although there are numerous types of geologic formations along the Atlantic coast, the most important water-bearing formations are unconsolidated sands and gravels; semiconsolidated sands; and, among the consolidated rocks, carbonates (primarily limestones), sandstones, and fractured crystalline rocks such as granites.

Saturated geologic materials that yield usable quantities of water to a well or spring are called aquifers. An aquifer can consist of a single geologic formation, a group of formations, or part of a formation (Lohman and others, 1972). The capacity of a geologic material to transit water is characterized by the material's hydraulic conductivity, which is commonly referred to as the permeability of the material. Confining units (or confining layers) are geologic units that are less permeable than aquifers. Because of their lower permeability, confining units restrict the movement of ground water into or out of adjoining aquifers. Aquifers and confining units are mapped on the basis of the degree of contrast in hydraulic conductivity among geologic units (Sun and Johnston, 1994). Generally, there is a close correlation between the type of geologic formation and its water-yielding properties. For example, unconsolidated sands and gravels, sandstones, and limestones commonly are major sources of ground-water supplies (aquifers), whereas beds of silt and clay function primarily as confining units (Heath, 1984).

Ground water in the Atlantic coastal zone occurs in confined and unconfined aquifers. Where water completely fills the pore spaces of an aquifer that is overlain by a confining unit, the aquifer is referred to as confined (or artesian). In contrast, where water only partially fills the pore spaces of an aquifer, the upper surface of the saturated zone (which is called the water table) is free to rise and decline, and the aquifer is referred to as unconfined (or as a water-table aquifer). Aquifers within the Atlantic coastal zone vary in size from local-scale aquifers that are a few square miles or less in areal extent to multilayer, regional-scale aquifers that are tens of thousands of square miles in areal extent. In many areas, unconfined aquifers that are close to land surface are underlain by one or more confined aquifers that may be partially or completely isolated from the land surface by confining units (fig. 5). Although these multilayer, regional aquifer systems may be discontinuous locally, they act hydrologically as a single system on a regional scale (Sun and Johnston, 1994).

The general pattern of fresh ground-water flow in coastal aquifers is from inland recharge areas where ground-water levels (hydraulic heads) typically are highest to coastal discharge areas where ground-water levels are lowest. This pattern of flow is illustrated by the ground-water flow paths in figures 3 and 5. Hydraulic head (often simply referred to as "head") is a measure of the total energy available to move ground water through an aquifer, and ground water flows from locations of higher head (that is, higher energy) to locations of lower head (lower energy). The distribution of hydraulic head within an aquifer is determined by measuring ground-water-level elevations in observation wells that are open to a small interval of the aquifer. The ground-water level (elevation) at each well most often is reported as feet (or meters) above or below sea level. Note that the upward direction of some of the ground-water flow paths near and beneath the ocean in the aquifer system shown in figure 5 indicates that ground-water heads in the deeper part of the flow system are above sea level near the coast.

Fresh ground water comes in contact with saline ground water at the seaward margins of coastal aquifers. The seaward limit of freshwater in a particular aquifer is controlled by the amount of freshwater flowing through the aquifer, the thickness and hydraulic properties of the aquifer and adjacent confining units, and the relative densities of saltwater and freshwater, among other variables. Because of its lower density, freshwater tends to remain above the saline (saltwater) zones of the aquifer, although in multilayered aquifer systems, seaward-flowing freshwater can discharge upward through confining units into overlying saltwater (fig. 5). As used in this report, saltwater is defined as water having a total dissolved-solids concentration greater than 1,000 milligrams per liter (mg/L) (Box A). Seawater has a total dissolved-solids concentration of about 35,000 mg/L, of which dissolved chloride is the largest component (about 19,000 mg/L). Concentrations of chloride in fresh ground water along the Atlantic coast are typically less than about 20 mg/L, so there is a large contrast in chloride concentrations between freshwater and saltwater.

A. Chemical Characteristics and Sources of Saltwater

All water contains dissolved chemical materials called "salts." When the concentration of these dissolved materials becomes great, the water is referred to as "saltwater" or as "salty", "brackish", or "saline." These terms commonly are used interchangeably, which can be confusing because they can mean different things to different people. As a result, different classification methods have been developed to differentiate between freshwater and saltwater and to define the degree of salinity (that is, the total dissolved-solids concentration) of the water.

The classification method used in this report defines freshwater as having a total dissolved-solids concentration of less than 1,000 mg/L; waters with a total dissolved-solids concentration greater than 1,000 mg/L are considered to be saltwater (or saline). This somewhat arbitrary upper limit of freshwater is based on the suitability of the water for human consumption. Although waters with total dissolved-solids concentrations of greater than 1,000 mg/L have been used for domestic supply in some areas of the United States where water of lower dissolved-solids concentration is not available (Krieger and others, 1957; Feth and others, 1965), water containing more than 2,000 to 3,000 mg/L total dissolved solids is generally too salty to drink (Freeze and Cherry, 1979). Brackish waters can be defined as those having a total dissolved-solids concentration of 1,000 to 35,000 mg/L. The upper concentration limit for brackish water is set at the approximate concentration of seawater (35,000 mg/L). The average concentrations of the major dissolved constituents of seawater (those with concentrations exceeding 1.0 mg/L) are given in table A–1; chloride, sodium, sulfate, and magnesium have the largest concentrations. Water with a dissolved-solids concentration exceeding that of seawater is called a brine. Although there are different types of brines in terms of chemical composition, the largest number represent concentrated seawater containing mostly sodium chloride (Krieger and others, 1957).

Table A–1. Average concentrations of major dissolved constituents of seawater (from Hem, 1989, p. 7)

Constituent	Concentration (milligrams per liter)
Chloride	19,000
Sodium	10,500
Sulfate	2,700
Magnesium	1,350
Calcium	410
Potassium	390
Bicarbonate	142
Bromide	67
Strontium	8
Silica	6.4
Boron	4.5
Fluoride	1.3

The U.S. Environmental Protection Agency (USEPA) has established secondary maximum contaminant levels (SMCLs) for total dissolved solids, chloride, and sulfate in drinking water. Unlike maximum contaminant levels (MCLs) that have been established to protect the public against drinking-water contaminants that present a risk to human health, the secondary contaminant levels have been established as guidelines to assist operators of public-water systems in managing the aesthetic qualities of the water such as taste, color, and odor. The SMCL set by the USEPA for total dissolved solids is 500 mg/L, whereas those for chloride and sulfate are each 250 mg/L (U.S. Environmental Protection Agency, 1992, 2002a). Concentrations of chloride in fresh ground water along the Atlantic coast are typically less than about 20 mg/L. Because of the high concentration of chloride in seawater (19,000 mg/L), less than a 2-percent contribution of seawater mixed with fresh ground water would render the water unsuitable for public supply using the USEPA guideline. The USEPA also has established a drinking-water advisory for sodium. The advisory recommends reducing sodium concentrations in drinking water to concentrations of 30 to 60 mg/L to avoid adverse effects on taste; a health-based guidance level of 20 mg/L in drinking water also is recommended by the USEPA for individuals on a very low sodium diet (less than 500 milligrams per day) (U.S. Environmental Protection Agency, 2002b). In some instances, States along the Atlantic coast have established standards for dissolved constituents in drinking water that differ from the levels established by the USEPA.

Two additional characteristics of water that are important in ground-water systems are density and viscosity, both of which are dependent on the type and amount of solutes dissolved in the water. As noted by Reilly (1993), density is important because it is part of the driving force that defines the direction and rate of fluid movement through a ground-water system; moreover, the density and viscosity of the water affect the hydraulic transmitting properties (permeability and hydraulic conductivity) of the ground-water system, which influence the rate of fluid movement.

Although seawater in the ocean and estuaries that laterally bound the eastern seaboard is by far the primary source of saline water to coastal ground-water systems, a number of other sources can affect coastal ground-water quality (Krieger and others, 1957; Feth and others, 1965; Task Committee on Saltwater Intrusion, 1969; Custodio, 1997; Richter and others, 1993; Jones and others, 1999). These sources include:

Precipitation: Oceans are the largest single source of salts in the atmosphere, and sodium and chloride are the most abundant ions in air masses over the sea (Feth, 1981). Chloride and sodium concentrations, therefore, are high in air masses near sea coasts but decrease rapidly with increasing distance inland. These airborne salts are delivered to coastal watersheds by precipitation. Chloride concentrations in precipitation, however, are relatively small compared to seawater. Average chloride concentrations in precipitation measured at about two-dozen atmospheric deposition monitoring stations along the Atlantic coast in 2000, for example, ranged from less than 0.2 to 2.8 mg/L (National Atmospheric Deposition Program, 2001). Concentrations of sodium and chloride can be increased in soils, shallow surface waters (such as tidal lagoons), and ground water by evaporation and evapotranspiration.
Sea-spray accumulation, tides, and storm surges, which can be local sources of increased ground-water salinity in low-lying coastal areas.
Entrapped fossil seawater in unflushed parts of an aquifer: Such water either was trapped in sedimentary formations when they were deposited (connate water) or flowed into the formations during periods of relatively high sea levels when seawater flooded low-lying coastal areas.
Dissolution of evaporitic deposits such as halite (rock salt), anhydrite, and gypsum (Manheim and Horn, 1968; Meisler, 1989).
Pollution from various anthropogenic sources including sewage and some industrial effluents, oil- and gas-field brines brought to the land surface during exploration and production, road deicing salts, and return flows of irrigation water.

The freshwater and saltwater zones within coastal aquifers are separated by a transition zone (sometimes referred to as the zone of dispersion) within which there is mixing between freshwater and saltwater (figs. 6 and 7). The transition zone is characterized most commonly by measurements of either the total dissolved-solids concentration or of the chloride concentration of ground water sampled at observation wells. Although there are no standard practices for defining the transition zone, concentrations of total dissolved solids ranging from about 1,000 to 35,000 mg/L and of chloride ranging from about 250 to 19,000 mg/L are common indicators of the zone. In this report, the term "transition zone" implies a change in the quality of ground water from freshwater to saltwater, as measured by an increase in dissolved constituents such as total dissolved solids and chloride.

Within the transition zone, freshwater flowing to the ocean mixes with saltwater by the processes of dispersion and molecular diffusion. Mixing by dispersion is caused by spatial variations (heterogeneities) in the geologic structure and the hydraulic properties of an aquifer and by dynamic forces that operate over a range of time scales, including daily fluctuations in tide stages (fig. 8), seasonal and annual variations in ground-water recharge rates, and long-term changes in sea-level position (figs. 9 and 10). These dynamic forces cause the freshwater and saltwater zones to move seaward at times and landward at times. Because of the mixing of freshwater and saltwater within the transition zone, a circulation of saltwater is established in which some of the saltwater is entrained within the overlying freshwater and returned to the sea, which in turn causes additional saltwater to move landward toward the transition zone (fig. 6).

The horizontal (or lateral) width of the transition zone can be narrow, such as the approximately 1,500-foot (ft) width shown for the Biscayne aquifer in figure 7, or very wide, such as in parts of the New Jersey Coastal Plain where the transition zone spans several miles. The vertical thickness of the transition zone also varies among aquifers, but generally is much smaller than the horizontal width and is limited by the total thickness of the aquifer. For example, the thickness of the transition zone shown for the Biscayne aquifer ranges from about 50 to 70 ft (fig. 7). For the convenience of illustrating freshwater-saltwater interactions as simply as possible and facilitating simplified scientific analysis of these interactions when possible, the freshwater and saltwater zones often are assumed to be separated by a sharp boundary that is referred to as the freshwater-saltwater interface, such as those shown in figure 5. Although the depth to this interface is quite variable along the Atlantic coast, it can be estimated approximately under some circumstances by using a technique known as the Ghyben-Herzberg relation (Box B).

The variety of geologic settings, aquifer types, and hydrologic conditions along the Atlantic coast has resulted in many patterns of freshwater-saltwater flow and mixing in coastal aquifers. The four case studies that follow illustrate some of the important freshwater-saltwater environments that exist in aquifers along the Atlantic coast and highlight the many variables that control the natural occurrence and flow of freshwater and saltwater in coastal aquifers. The case studies progress from the glacial aquifer of Cape Cod, Massachusetts, which is representative of shallow, single-layer aquifers, to two of the most productive regional aquifer systems in the United States—the Northern Atlantic Coastal Plain aquifer system that extends from Long Island, New York, through North Carolina, and the Floridan aquifer system that extends from South Carolina to Alabama. These are thick, multilayered aquifer systems that underlie thousands of square miles.

B. How Deep is it to Saltwater?

Many who live along the coast often wonder how deep it is to salty ground water, particularly those searching for freshwater supplies. Although it is difficult to estimate the thickness of the freshwater zone in the absence of water-quality data, there is a relatively simple equation that has been used successfully in many areas to estimate the depth to saltwater and the thickness of freshwater in a water-table (unconfined) aquifer. The equation relates the elevation of the water table to the elevation of the boundary of the interface between the freshwater and underlying saltwater zones of an aquifer (fig. B–1), and is based on the balance of the height of two columns of fluids of differing density (Reilly and Goodman, 1985). In the equation, the thickness of the freshwater zone above sea level is represented as h and that below sea level is represented as z, as shown in figure B–1. The two thicknesses are related by

Equation (1)

where

ρ_f	is the density of freshwater, and
ρ_s	is the density of saltwater.

This equation often is called the Ghyben-Herzberg relation after two European scientists who derived it independently in the late 1800s.

Freshwater has a density of about 1.000 grams per cubic centimeter (g/cm³) at 20 °C, whereas that of seawater is about 1.025 g/cm³. Although the difference between the density of freshwater and seawater is small, equation 1 indicates that this density contrast results in 40 ft of freshwater below sea level for every 1 ft of freshwater above sea level, that is:

Equation (2)

The total thickness of the freshwater zone is the sum of the freshwater zones above and below sea level: h + z. An example of the application of equations 1 and 2 is shown in figure B–2.

The Ghyben-Herzberg method of calculating the thickness of the freshwater zone in a coastal water-table aquifer is based on a number of simplifying assumptions. Most importantly, it is assumed that hydrostatic conditions exist within the aquifer, which implies that there are no vertical gradients in ground-water levels (heads) within the aquifer. This assumption incorrectly causes the thickness of the freshwater zone to be represented as zero at the shore where the elevation of the water table is zero. In reality, the freshwater zone must have some thickness for freshwater to discharge to the ocean. It also is assumed that there is an abrupt (or sharp) boundary between the freshwater and saltwater zones at the interface between the two types of water. However, as shown earlier, freshwater and saltwater mix at the boundary between the two waters, which results in a zone of dispersion at the interface.

Figure B–1. Simplified freshwater-saltwater interface in a coastal water-table aquifer.

Figure B–2. Example application of equations 1 and 2 for estimating the thickness of freshwater in a coastal water-table aquifer: A freshwater-supply well near the coast has deteriorated with time and has not been pumped for a number of years. The owner of the well would like to know the total thickness of freshwater that might be available for development at the well site and, as a first approximation, has decided to estimate the total thickness by using equations 1 and 2. A topographic map of the area indicates that land surface on the property is about 20 feet above sea level. The depth to the water table measured in the unpumped well is about 10 feet. The elevation of the water table (h) therefore is about 10 feet above sea level, and the estimated thickness of the freshwater zone below sea level (z), estimated by use of equation 2, is 400 feet. The total thickness of freshwater at the site (h + z) therefore is estimated to be about 410 feet.

Patterns of Freshwater-Saltwater Interaction in a Single-Layer Aquifer, Cape Cod, Massachusetts

Detailed studies of the freshwater-saltwater transition zone in the ground-water flow system of Cape Cod, Massachusetts, illustrate how the configuration of freshwater and saltwater zones can be affected by the distribution of geologic materials within an aquifer and by the distribution of coastal marshes and embayments that bound an aquifer. The Cape's ground-water flow system consists of a single-layer aquifer that is composed of extensive layers of glacially derived sand and gravel interbedded in places with layers and lenses of clay, silt, and unsorted till (LeBlanc and others, 1986). These unconsolidated sediments range in thickness from about 100 ft at Cape Cod Canal to more than 1,000 ft in the town of Truro. The ground-water flow system consists of six separate water-table mounds (fig. 11); ground water moves radially outward from these mounds toward the ocean, saltwater bays, tidal inlets, and coastal wetlands and streams.

Chloride-concentration profiles for 29 monitoring sites distributed across the Cape indicated that the transition between freshwater and saltwater was relatively thin, ranging from about 20 to 80 ft in thickness (LeBlanc and others, 1986). At many of the monitoring sites, such as site BHW 221 (fig. 11), the location and shape of the transition zone was similar to that typically found in relatively thin and homogeneous aquifers; that is, a single transition zone located close to the coastline separated overlying freshwater from underlying saltwater. However, at a number of monitoring sites along Cape Cod Bay, two transition zones have been formed by the presence of very fine sand, silt, and clay deposits (see site A1W 318). At these sites, a shallow transition zone lies above the fine-grained sediments, and fresh ground water discharges directly to the bay; a lower transition zone lies beneath the fine-grained sediments and is displaced offshore. Fresh ground water in the deeper part of the aquifer flows below the fine-grained sediments and then discharges upward into overlying saltwater. At site BMW 47, which is in a saltwater marsh, brackish water having a chloride concentration of about 10,000 mg/L overlies freshwater in the aquifer. The source of the shallow brackish water is periodic flooding by seawater from the bay.

Water-quality data collected at the 29 monitoring sites were used with other hydrogeologic information to develop a conceptual understanding of ground-water flow throughout the Cape Cod aquifer (fig. 12). In the central part of Cape Cod (fig. 12A), the freshwater flow system is truncated by bedrock and fine-grained sediments, whereas on the eastern, outer limb of Cape Cod, the thick deposits of unconsolidated sediments and the generally narrower width of the peninsula cause the freshwater flow system to be underlain everywhere by saltwater (fig. 12B).

Effects of Sea-Level Fluctuations on the Freshwater-Saltwater Transition Zone, Northern Atlantic Coastal Plain Aquifer System

The Northern Atlantic Coastal Plain encompasses a land area of about 50,000 square miles (mi²) extending from Long Island, New York, southward to the North Carolina-South Carolina border (fig. 13). The Coastal Plain is underlain by a seaward-thickening wedge of predominantly unconsolidated sediments that increases in thickness from the Fall Line, which is the inland limit of the Coastal Plain, eastward toward the Atlantic Ocean. The Fall Line is so named because of the prevalence of falls and rapids in streams that cross the contact between the hard rocks of the Piedmont Plateau to the west and the less-resistive sediments of the Coastal Plain. The sediment wedge reaches a maximum onshore thickness of about 10,000 ft at Cape Hatteras, North Carolina, but exceeds 7.5 miles in thickness offshore from New Jersey and the Delmarva Peninsula (Trapp and Meisler, 1992). The sediments are mostly gravel, sand, silt, and clay, and have been subdivided into an aquifer system that consists of a vertical sequence of highly permeable aquifers separated by less permeable confining units. Ground-water withdrawals from the aquifer system total more than a billion gallons per day, making it one of the most productive aquifer systems in the United States.

Saltwater underlies freshwater in eastern parts of the regional aquifer system. The transition zone between freshwater and saltwater was delineated throughout the aquifer system in the early 1980s by using geochemical and geophysical data collected at more than 500 locations (Meisler, 1989). The transition zone was defined as the zone of water with chloride concentrations from 250 mg/L to 18,000 mg/L. Generally, chloride concentrations increase in the seaward direction of each aquifer and with depth from the shallowest to the deepest aquifers. Waters within the transition zone probably were produced by the mixing of fresh ground water with either seawater or highly concentrated brines. In the area from Virginia to New Jersey, some of the water samples showed the presence of chloride at concentrations greater than those in seawater, suggesting that the transition zone in that area is largely a mixture of freshwater with a brine. For example, the chloride-concentration profile at well Virginia 57 (fig. 13) shows a maximum chloride concentration of nearly 27,000 mg/L, which is about 8,000 mg/L greater than that of seawater. The most likely source of the brines appears to be the leaching of ancient evaporite deposits of probable early Jurassic age beneath the Continental Shelf and Slope (Meisler, 1989; Knobel and others, 1998).

Two striking features of the transition zone within the regional aquifer system are its large vertical thickness and substantial horizontal width. The thickness of the transition zone ranged from 400 to 2,200 ft, whereas the width of the transition zone was as much as 40 miles in some areas (fig. 14). The development of the broad transition zone has been attributed to the cyclic movement of saltwater caused by global sea-level fluctuations that resulted in repeated advance and retreat of the freshwater-saltwater interface during at least the last 900,000 years (Meisler and others, 1985). As the sea level rose, saltwater invaded the aquifer sediments and mixed with freshwater. As the sea level declined, the fresher water advanced seaward, and the process of mixing continued. Repeated advance and retreat of the saltwater produced a broad zone of mixed waters in which saltwater predominates in the deeper and seaward parts, and freshwater predominates in the shallower and landward parts (Meisler and others, 1985).

The depth to the top of the transition zone is shallowest in North Carolina and deepens northward, reaching its greatest depths—as much as 2,800 ft below sea level—in Maryland and along the coast of New Jersey. Moreover, ground water containing chloride concentrations of less than 5,000 mg/L has been found as much as 55 mi from the New Jersey coast, but extends progressively shorter distances from the coast southward to Virginia and North Carolina (Meisler, 1989). The occurrence of the transition zone at great depths in New Jersey and Maryland and the occurrence offshore of water considerably fresher than seawater have been attributed to long periods when sea levels generally were lower than at present (Meisler, 1989). Overall, the average sea level during the past 900,000 years is estimated to have been about 150 ft lower than the present sea level. It has been hypothesized that, at least in some areas, the transition zone may not be in equilibrium with the present-day sea level, but may still be moving landward and upward to adjust to the present sea level (Meisler and others, 1985; Pope and Gordon, 1999).

Virginia's Inland Saltwater Wedge and the Chesapeake Bay Impact Crater—The Importance of Geologic Control on Coastal Ground-Water Systems

For many years, scientists have been puzzled by an anomalous wedge of saltwater that extends inland from the mouth of Chesapeake Bay into the Coastal Plain aquifers beneath the southeastern parts of the Middle and York-James Peninsulas of Virginia. This inland wedge is shown on regional maps of the freshwater-saltwater transition zone in the Northern Atlantic Coastal Plain aquifers and, in greater detail, on a map of dissolved-solids concentrations for the upper Potomac aquifer of Virginia (fig. 15). Several hypotheses have been proposed to explain the occurrence of this saltwater wedge. As early as 1911, scientists suggested that the wedge was the product of incomplete flushing of ancient seawater that had invaded the aquifers from above during high stands of the sea. The incomplete flushing of the saltwater was attributed to poor circulation of ground water caused by a decrease in the permeability or by a pinching out of the aquifers in the area of the wedge. Others hypothesized that the wedge was the result of relatively low ground-water levels in the area of major ground-water discharge to lower Chesapeake Bay. Scientists also have noted that chloride concentrations in the wedge often are greater than those in seawater, which has led to speculation that the source of the saltwater was a brine.

During the early 1990s, scientists studying the geologic structure and history of the Atlantic Coastal Plain made an important discovery that is dramatically altering the understanding of Virginia's inland saltwater wedge. Through the collection and analysis of deep sedimentary cores, geophysical data, and fossil assemblages, geologists have identified a large impact crater formed by a meteorite near what is now the mouth of Chesapeake Bay (figs. 16 and 17) (Poag and others, 1994; Poag, 1998, 1999, and 2000; Powars and Bruce, 1999; Powars, 2000). The crater is three times larger than any other U.S. crater and is the sixth largest crater known on Earth. Discovery of the crater helps to explain several of the previously anomalous geologic and hydrologic conditions in southeastern Virginia and is an excellent example of the important role of geologic control on ground-water-flow systems and ground-water quality.

Approximately 35 million years ago, a large meteorite struck the shallow waters of the Continental Shelf of what is now southeastern Virginia, blasting through the water column and thousands of feet of unconsolidated sediment and underlying granitic crust. Sea level was high everywhere on Earth at the time of the impact, and the ancient shoreline in the Virginia region was somewhere in the vicinity of present-day Richmond (fig. 16). The meteorite measured 1 to 2 miles in diameter and was traveling about 50,000 miles per hour (Poag, 2000). The impact likely vaporized seawater and the crystalline granitic bedrock immediately beneath the impact site and created a supersonic shock wave, tsunamis (great sea waves traveling at very high speeds), and a huge atmospheric vapor cloud. The impact produced an enormous, 56-mile-wide crater that reaches about 6,000 ft deep—a depth comparable to the Grand Canyon of the Colorado River. The crater was immediately filled with a chaotically mixed, sandy rubble bed known as breccia (fig. 17). The breccia contains hand-size to person-size chunks (clasts) of clay, limestone, and sand (Poag, 1998, 1999). These clasts are fragments of the geologic beds that were removed by the impact and small pieces of the underlying granitic basement rocks. Along the crater's outer rim, large-scale blocks of sediments (megaslump blocks) collapsed into the crater. The entire impact event, from initial atmospheric penetration to the end of breccia deposition, lasted only a few hours or days. The crater then was buried by additional sedimentary beds that accumulated during the following 35 million years.

The Chesapeake Bay impact crater is the most significant discontinuity in the sedimentary strata of the Atlantic Coastal Plain. Its discovery has led to a re-interpretation of the hydrogeologic framework and inland saltwater wedge of southeastern Virginia. Previous views of Coastal Plain sediments in the area were of a layered sequence of alternating aquifers and intervening confining units similar to the structure of aquifers and confining units throughout most of the Northern Atlantic Coastal Plain aquifer system (see, for example, fig. 5). The new interpretation, however, is that these aquifers and confining units were truncated and excavated by the impact, which left in their place rubbly breccias. The poorly sorted breccias are thought to be of lower permeability than the surrounding aquifers and, therefore, likely affect ground-water flow patterns near and within the crater.

The spatial association of the inland saltwater wedge with the configuration and location of the outer rim of the meteorite crater has led scientists to propose that the impact crater may have been the cause, or at least part of the cause, of the saltwater wedge. The association between the saltwater wedge and crater is shown by lines of equal dissolved-solids concentrations drawn for one of the aquifers that abuts the crater (the upper Potomac aquifer; fig. 15). Although it currently (2003) is not known for certain whether the crater is the ultimate cause or one of the causes of the saltwater wedge, it is clear that the impact crater affects the framework of the aquifers and the ground-water conditions in southeastern Virginia. Moreover, geophysical surveys across the crater show many faults that cut the sedimentary beds above the breccia and extend upward toward the floor of Chesapeake Bay (Poag, 1998, 1999). These faults are a result of the compaction subsidence of the breccia that has occurred since the impact. Some of the faults appear to completely breach the confining unit over the saltwater wedge, which could allow upward flow of saltwater.

Continuing investigations of the geology and hydrology of the impact crater are underway to better understand how the crater affects ground-water resources of the Virginia Coastal Plain. During the summer of 2000, a 2,084-ft-deep core hole was drilled through the crater at the NASA Langely Research Center in Hampton, Virginia (fig. 17). This core hole is the first of several that are planned to be drilled in and around the crater by a multiagency collaboration that was formed to continue scientific investigations of the crater (Gohn and others, 2001). These investigations could provide information on the hydrogeologic framework, ground-water flow system, and water quality in and near the crater. One of the planned uses of this information will be the revision of a computer model of the ground-water system of the Virginia Coastal Plain. This model was first developed in the early 1980s and is used by the State of Virginia to help manage the more than 100 million gallons per day (Mgal/d) of ground water that are withdrawn from Coastal Plain aquifers in the State.

U.S. Geological Survey scientist collects sediment core at a site within the Chesapeake Bay impact crater, Hampton, Virginia, during the summer of 2000 (top). Drilling into the Chesapeake Bay impact crater at NASA Langley Research Center, Hampton, Virginia, during the summer of 2000 (bottom).

Saltwater in the Floridan Aquifer System, South Carolina to Florida

The Floridan aquifer system is one of the most productive aquifers in the world. It consists of a thick sequence of carbonate rocks (limestones and dolomites) that underlie all of Florida, southern Georgia, and small parts of adjoining South Carolina and Alabama; a total area of about 100,000 mi² (fig. 18). An estimated 4.0 Bgal/d of water was withdrawn from the aquifer system in 2000 (R.L. Marella, U.S. Geological Survey, written commun., 2003), and, in many areas, it is the sole source of freshwater (Johnston and Bush, 1988). In addition to water supply, the Floridan is being used increasingly for aquifer storage and recovery systems, in which freshwater is injected into more saline zones of the aquifer and stored for later use. Moreover, in several places where the aquifer system contains saltwater, such as along the southeastern coast of Florida, treated sewage water and industrial wastes are injected into it.

The aquifer system generally thickens seaward from a thin edge near its northern limit to a maximum of about 3,500 ft in southwestern Florida. In most places, the system consists of the Upper and Lower Floridan aquifers separated by a less permeable confining unit (the "middle confining unit") that restricts movement of water between the two aquifers (fig. 19). Much of the aquifer system is overlain by an upper confining unit that, where present, limits the amount of recharge to the system. Where the upper confining unit is thin or absent, recharge is plentiful and ground-water circulation is high. In these areas of high recharge and vigorous circulation, ground water readily dissolves the carbonate rocks that make up the aquifer system, creating large and highly permeable conduits that store and transmit tremendous volumes of ground water. These large conduits are the cause for the many first-magnitude springs—those with a flow of 100 cubic feet per second or more—that issue from the aquifer system.

Ground-water withdrawals have resulted in long-term regional water-level declines of more than 10 ft in three broad areas of the flow system: (1) coastal Georgia and adjacent South Carolina and northeast Florida; (2) west-central Florida; and (3) the Florida panhandle (fig. 18). In these and a number of other coastal areas, ground-water withdrawals have reversed the generally seaward direction of ground-water flow, creating the potential for saltwater intrusion from the Gulf of Mexico or Atlantic Ocean or from deep parts of the aquifer that contain saltwater.

The transition between freshwater and saltwater in the Floridan aquifer system is illustrated by the distribution of chloride in water in the Upper and Lower Floridan aquifers (fig. 20). Although large areas of the Upper Floridan aquifer contain water with a chloride concentration less than 250 mg/L (fig. 20A), much of the Lower Floridan aquifer contains water with chloride concentrations that exceed the 250 mg/L drinking-water limit (fig. 20B), which has limited the aquifer's use for water supply. In general, chloride concentrations in the Upper Floridan aquifer are related to ground-water-flow conditions and proximity to the coast. In areas where the upper confining unit is thin or absent, fresh ground-water circulation rates are high and chloride concentrations tend to be low (less than 250 mg/L). Where the flow system is tightly confined, flow is more sluggish and chloride concentrations in the aquifer are higher. This is the case in Florida south of Lake Okeechobee (fig. 20A), where the Upper Floridan aquifer is extensively confined and ground-water flow is quite sluggish. Because of the slow movement of ground water in the area, it is thought that residual seawater that entered the aquifer during the Pleistocene when sea level was higher than its current level has not been completely flushed out by modern freshwater (Sprinkle, 1989; Johnston and Bush, 1988; Reese, 1994, 2000; Reese and Memberg, 2000). The anomalously high concentrations of chloride along the St. Johns River and the eastern coast of Florida are thought to be the result, in varying amounts, of two processes (Sprinkle, 1989): (1) incomplete flushing by the modern-day freshwater flow system of residual seawater that invaded the aquifer during high sea-level stands of the Pleistocene; and (2) upward flow of brackish water from the underlying Lower Floridan aquifer along fracture zones in the aquifer system. The high chloride concentrations in the western panhandle of Florida also may have resulted from incomplete flushing of residual seawater from Pleistocene highstands.

The generally low chloride concentrations in the Upper Floridan aquifer along the Georgia coast (fig. 20A) have been attributed to the thick confining unit that overlies the Upper Floridan aquifer in that area. The confining unit has created relatively high ground-water heads that have kept the freshwater-saltwater interface offshore. In fact, freshwater flow has been observed to extend as far as 50 mi offshore of southeast Georgia (Johnston and others, 1982). However, along the coast in South Carolina and extending to Tybee Island, Georgia, high chloride concentrations in water from the aquifer system are attributed to intrusion of offshore saltwater caused by large ground-water withdrawals from the Upper Floridan aquifer in the Savannah, Georgia, and the Hilton Head Island, South Carolina, areas. Saltwater most likely enters the aquifer system by lateral intrusion from offshore areas combined with some downward vertical leakage of seawater to the Upper Floridan aquifer where the overlying confining unit is thin or absent (Krause and Clarke, 2001).

Ground-water temperature and geochemical data from the Lower Floridan aquifer in the south Florida area suggest that a geothermally driven circulation of ground water occurs in the Lower Floridan aquifer in that area (Kohout, 1965; Meyer, 1989; Sanford and others, 1998). In this circulation, cold, dense ocean water enters the Lower Floridan aquifer where it is in direct contact with the ocean. As the seawater moves landward, it is warmed by geothermal heat generated naturally below the base of the thick Floridan aquifer system that underlies the Florida peninsula. The geothermal heating lowers the density of the seawater, causing it to rise where it is diluted and transported back to the ocean with seaward-flowing water of lower salinity.

In 1979, an oil exploration well located about 55 miles offshore of St Marys, southeast Georgia, detected freshwater in the upper part of the Upper Floridan aquifer (Johnston and others, 1982).

CHAPTER 2. CAUSES, MODES, AND MANAGEMENT OF SALTWATER INTRUSION

The natural balance between freshwater and saltwater in coastal aquifers is disturbed by ground-water withdrawals and other human activities that lower ground-water levels, reduce fresh ground-water flow to coastal waters, and ultimately cause saltwater to intrude coastal aquifers. Although ground-water pumping is the primary cause of saltwater intrusion along the Atlantic coast, lowering of the water table by drainage canals has led to saltwater intrusion in a few locations, notably southeastern Florida. Other hydraulic stresses that reduce freshwater flow in coastal aquifers, such as lowered rates of ground-water recharge in sewered or urbanized areas, also could lead to saltwater intrusion, but the impact of such stresses on saltwater intrusion, at least currently (2003), likely is small in comparison to pumping and land drainage.

The variability of hydrogeologic settings, sources of saline water, and history of ground-water withdrawals and freshwater drainage along the Atlantic coast has resulted in a variety of modes of saltwater intrusion across the region. Saltwater can contaminate a freshwater aquifer through several pathways, including lateral intrusion from the ocean; by upward intrusion from deeper, more saline zones of a ground-water system; and by downward intrusion from coastal waters. A few of these pathways are illustrated on figures 21 and 22. Some authors have used the term saltwater encroachment to refer to lateral movement of saltwater within an aquifer and the term saltwater intrusion to refer to vertical movement of saltwater. This distinction is not made in this report because there are often both lateral and vertical components to a particular saltwater-intrusion problem, and because the net result—contamination of a freshwater aquifer—is the same for either type of saltwater movement. Another term that has been used to describe a specific type of vertical saltwater intrusion is saltwater upconing, which refers to the movement of saltwater from a deeper saltwater zone upward into the freshwater zone in response to pumping at a well or well field (Reilly and Goodman, 1987). An example of saltwater upconing that occurred on Cape Cod, Massachusetts, is shown in figure 22.

Saltwater intrusion reduces freshwater storage in coastal aquifers and can result in the abandonment of freshwater supply wells when concentrations of dissolved ions exceed drinking-water standards. Saltwater intrusion has been documented throughout the Atlantic coastal zone for more than 100 years (Barlow and Wild, 2002), but the degree of saltwater intrusion varies widely among localities and hydrogeologic settings. In many instances, the area contaminated by saltwater is limited to small parts of the aquifer and has little or no effect on wells pumped for ground-water supply (Box C). In other instances, contamination is of regional extent and has substantially affected ground-water supplies. For example, in Cape May County, New Jersey, more than 120 supply wells have been abandoned since 1940 because of saltwater contamination (Lacombe and Carleton, 1992). The extent of saltwater intrusion into an aquifer depends on several factors, including the total rate of ground water that is withdrawn compared to the total freshwater recharge to the aquifer, the distance of the stresses (wells and drainage canals) from the source (or sources) of saltwater, the geologic structure and distribution of hydraulic properties of the aquifer, and the presence of confining units that may prevent saltwater from moving vertically toward or within the aquifer. Moreover, the time required for saltwater to move through an aquifer and reach a pumping well can be quite long. Depending on the location and lateral width of the transition zone, many years may pass before a well that is unaffected by saltwater intrusion suddenly may become contaminated.

Irrigation well pumping from the Floridan aquifer system near Tampa, Florida. Large ground-water withdrawals from the Floridan aquifer system have caused water-level declines and saltwater intrusion in some areas of South Carolina, Georgia, and Florida.

C. Saltwater Intrusion in a Fractured Crystalline-Rock Aquifer, Harpswell, Maine

Large parts of the northeastern Atlantic coastal zone are underlain by crystalline igneous and metamorphic rocks (often referred to as bedrock) that generally yield only small amounts of water to wells. However, because these rocks extend over large areas, large volumes of ground water are withdrawn from them, and in many places, they are the only reliable water sources (Miller, 1999). Crystalline-rock aquifers are geologically complex systems in which the movement of ground water is strongly dependent on the presence of fractures, joints, faults, and other openings in the rock mass. The potential for saltwater intrusion into these aquifers exists where the openings extend to the coast or to depths at which they become saturated with saltwater. Although incidences of saltwater intrusion into crystalline-rock aquifers of the northeastern United States have been reported, the problem does not appear to be widespread. In a review of the coastal bedrock aquifers of Maine, Caswell (1979a, b) noted that most coastal wells do not pump saltwater, even though the ocean nearly surrounds the numerous islands and narrow peninsulas of Maine's irregular coastline. This may be partly the result of the generally low pumping rates of wells that tap the bedrock aquifers, but Caswell suggested that the absence of extensive saltwater intrusion is probably the result of the particular orientation of the water-bearing fractures of crystalline rocks along the coastline, which prevents, or at least limits, the landward migration of saltwater. For example, the principal water-bearing fractures of the bedrock aquifer on High Head peninsula in Harpswell, Maine (fig. C–1), are nearly vertical and are parallel to the length of the peninsula and to the coastline. Of more than 60 bedrock wells that had been drilled on the peninsula as of the mid-1970s, only 3 had been contaminated by saltwater (Richard, 1976; Caswell, 1979a, b). These three wells were in an area where the principal water-bearing fractures have been cut by a number of large fractures that lie perpendicular to the length of the peninsula and the dominant fracture direction (fig. C–1). These cross-fractures appear to make the rock very permeable perpendicular to the coastline, and to provide a conduit for saltwater to move readily toward the wells.

Fractures and other openings in crystalline rocks such as these along the rocky coast of Acadia National Park, Maine, can provide conduits for saltwater to enter coastal aquifers.

Figure C–1. High Head peninsula, Harpswell, Maine, where saltwater contamination of three domestic wells in a fractured crystalline-rock aquifer was documented in the mid-1970s.

Many States and communities along the Atlantic coast are taking actions to manage and prevent saltwater intrusion to ensure a sustainable source of ground water for the future. These actions can be grouped broadly into three general categories: engineering techniques, regulatory (legislative) approaches, and scientific monitoring and assessment (van Dam, 1999). Often, several actions are taken simultaneously or as part of a more comprehensive strategy for managing both the ground-water and surface-water supplies of a coastal community. The State of Georgia, for example, has established a multicomponent approach for managing saltwater intrusion that restricts withdrawals in some coastal areas, encourages water conservation, and relies on hydrologic studies and water-quality monitoring to better understand saltwater movement in the State's aquifers and to evaluate alternative sources of freshwater. The regional scale of many of the aquifers along the Atlantic coast means that several communities and political jurisdictions often share a single aquifer or aquifer system, and, therefore, ground-water development in one community can affect the water resources of neighboring communities.

A common approach for managing saltwater intrusion has been to reduce the rate of pumping from coastal wells or to move the locations of withdrawals further inland. Reductions in coastal withdrawals allow ground-water levels to recover from their lowered (or stressed) levels, and fresh ground water to displace the intruded saltwater. In New Jersey, for example, State-mandated reductions in ground-water withdrawals in some coastal counties have resulted in ground-water-level increases in aquifers that have been affected by saltwater intrusion (Box D). An alternative to reducing ground-water withdrawals is to artificially recharge freshwater into an aquifer to increase ground-water levels and hydraulically control the movement of the intruding saltwater. Artificial recharge can be accomplished through injection wells or by infiltration of freshwater at the land surface. In either case, the recharged water creates hydraulic barriers to saltwater intrusion. Perhaps the most prominent example of the use of artificial recharge to control saltwater intrusion on the east coast of the United States is in southeastern Florida. In that area, an extensive network of surface-water canals is used to convey freshwater from inland water-conservation (storage) areas during the dry season to coastal areas, where the water is recharged through the canals to the underlying Biscayne aquifer to slow saltwater intrusion in the aquifer.

In addition to more conventional methods, innovative approaches are now being used to manage saltwater intrusion along the Atlantic coast. These include aquifer storage and recovery systems, desalination systems, and blending of waters of different quality. Aquifer storage and recovery (ASR) is a process by which water is recharged through wells into a suitable aquifer, stored for a period of time, and then extracted from the same wells when needed (Pyne, 1995). Typically, water is stored during wet seasons and extracted during dry seasons. ASR systems have been developed in Wildwood (Cape May County), New Jersey, the town of Chesapeake, Virginia, and at several locations in Florida.

Desalination refers to water-treatment processes that produce freshwater by removing dissolved salts from brackish or saline waters (Buros, 2000). Desalination systems are becoming more widespread in the United States as desalination technologies improve, costs decrease, and new sources of freshwater become more difficult to develop. One of the interesting aspects of the increased use of desalination systems is that it changes the perspective on brackish or saline water from that of a potential water problem (a contaminant) to that of a potential water source. Past studies have highlighted the importance of saline ground water as a potential resource (Kohout, 1970) and estimated the saline ground-water resources of the conterminous United States (Feth and others, 1965). Despite these efforts, substantial research remains to be done to characterize the distribution and quality of brackish and saline ground waters that might be used for desalination systems (Alley, 2003).

Scientific monitoring and assessment provide basic characterization of the ground-water resources of an area, an understanding of the different pathways by which saltwater may intrude an aquifer, and a basis for management of water supplies. Numerous water-level and water-quality monitoring networks have been established along the Atlantic coast to monitor changes in ground-water levels, ground-water quality, and movement of the freshwater-saltwater interface. Water-quality monitoring networks are particularly important in serving as early-warning systems of saltwater movement toward freshwater supply wells; their use along the Atlantic coast is described in greater detail in the following chapter.

Six case studies are provided in the remainder of this chapter to illustrate specific causes and modes of saltwater intrusion along the Atlantic coast and to describe several approaches that have been used to manage and prevent saltwater intrusion in this coastal region. The six examples are from the States of New Jersey, Georgia, and Florida, which, with the possible exception of Long Island, New York (fig. 23), have experienced the most widespread problems of saltwater intrusion along the Atlantic coast. The case studies demonstrate several factors that must be considered in the analysis and management of a saltwater-intrusion problem, specifically: the source (or sources) of saltwater to the aquifer, the physical structure and dynamics of the aquifer or aquifer system into which the saltwater intrudes, and the types and locations of stresses to the aquifer (supply wells and drainage canals). The first two case studies are drawn from areas within the Northern Atlantic Coastal Plain aquifer system, which consists of unconsolidated to semiconsolidated sand and gravels. The last four case studies describe examples of saltwater intrusion into carbonate aquifers and aquifer systems of the southeastern States.

The North Reverse Osmosis (RO) Water Plant is part of the water-supply system for Dare County, which is on the Outer Banks of North Carolina. The RO system has been in operation since August 1989 and at the time of its initiation was the largest RO system outside the State of Florida. The plant is supplied with brackish ground water that is pumped from the Yorktown aquifer by 10 wells at the facility, each of which can produce as much as 500 gallons per minute (Robert W. Oreskovich, Dare County Water Department, Kill Devil Hills, North Carolina, June 2003). The treatment system (bottom photograph) consists of three reverse osmosis units, each of which produces 850,000 gallons per day of treated water. The treated water, which is referred to as the permeate, then is mixed with 150,000 gallons per day of raw brackish ground water, for a total production of 1.0 million gallons per day of potable water for each RO unit.

D. Water-Level Response to Mandated Decreased Withdrawals in the New Jersey Coastal Plain

Ground-water withdrawals from the 10 major confined aquifers in the New Jersey Coastal Plain (fig. D–1), which is part of the Northern Atlantic Coastal Plain, began in the late 1800s and were approximately 210 Mgal/d in 1997 (Lacombe and Rosman, 2001). These withdrawals have created several regions of large ground-water-level declines, with declines of more than 200 ft below sea level in some areas. Since 1978, the USGS, in cooperation with the New Jersey Department of Environmental Protection (NJDEP), has measured ground-water levels every 5 years in more than 700 monitoring wells to provide a "snapshot" of ground-water conditions in each of the confined aquifers. In 1985–86, the NJDEP used these water-level measurements to establish two "Water Supply Critical Areas" in the State—areas where excessive water usage threatens the long-term sustainability of the water-supply source.

Water levels in the four aquifers underlying Critical Area 1 in parts of Middlesex, Monmouth, and Ocean Counties—specifically, the Wenonah-Mount Laurel (fig. D–2), Englishtown, and the Upper and Middle Potomac-Raritan-Magothy aquifers—have been lowered below sea level as a result of ground-water withdrawals, causing saltwater to flow landward toward the aquifers. In response to these lowered ground-water levels, NJDEP mandated that withdrawals from aquifers underlying the Critical Area be reduced after 1988, and that surface water from reservoirs located in Monmouth County be used as a substitute. Upon completion of the Manasquan Reservoir in 1991, water users in Critical Area 1 began to withdraw less water from the confined aquifers.

As a result of the mandated reduced withdrawals, ground-water levels in aquifers underlying Critical Area 1 rose by as much as 120 ft between the 1988 and 1993 water-level measurements, and there was a reduction in the overall breadth of the area of lowered ground-water levels (Lacombe and Rosman, 1997). For example, water levels in observation well 25–486 in eastern Monmouth County rose from about 180 ft below sea level to about 90 ft below sea level during the 5-year period (fig. D–3). Additional water-level increases of as much as 47 ft were measured in the aquifer between the 1993 and 1998 water-level measurements (Lacombe and Rosman, 2001).

Figure D–1. Diagrammatic hydrogeologic section of the New Jersey Coastal Plain. Line of section shown on figure D–2.

Figure D–2. Ground-water-level contours for the Wenonah-Mount Laurel aquifer, (A) 1988 and (B) 1993.

Figure D–3. Water-level hydrographs for observation wells screened in the Wenonah-Mount Laurel aquifer, 1984–94. Seasonal water-level fluctuations of 5 to 15 feet occurred in well 25–486 in response to seasonal fluctuations in ground-water withdrawal rates.

Development of a Desalination System in Response to Saltwater Intrusion, Cape May City, New Jersey

Cape May County, New Jersey, is on a peninsula that is surrounded by the salty waters of Delaware Bay and the Atlantic Ocean (fig. 24). The county is a popular vacation area that has experienced substantial increases in resident and summer tourist populations. Withdrawals from five freshwater aquifers that underlie the peninsula (fig. 25) have lowered ground-water levels as much as 100 ft in some areas and have caused saltwater to intrude each of the aquifers. Prior to the start of ground-water withdrawals, water levels in each of the aquifers stood above sea level, and ground water flowed radially outward from inland recharge areas to low-lying streams, tidal wetlands, Delaware Bay, and the Atlantic Ocean. Saltwater intrusion in the county began about 1890 after the first deep wells were pumped and fresh ground-water levels declined to below sea level for the first time (Lacombe and Carleton, 1992, 2002; Spitz, 1998).

The area affected by saltwater intrusion is along the populated coast of southern Cape May peninsula, including Cape May Point, Cape May City, the Wildwood Island communities, Lower Township, and Middle Township (Lacombe and Carleton, 1992, 2002). Saltwater intrusion has forced the closure of at least 20 public- and industrial-supply wells and more than 100 domestic-supply wells since the 1940s. State and local water-resource planners recognize that saltwater intrusion toward the Cape May peninsula is likely to continue if withdrawals from the aquifer system are maintained at current rates or are increased to meet projected increases in water demands. In response to these concerns, alternative sources of water supply and new approaches for water-supply management at county and local levels are being evaluated and implemented.

One of the communities most affected by saltwater intrusion is Cape May City, where the history of saltwater contamination has been well documented (Gill, 1962; Lacombe and Carleton, 1992, 2002; Blair and others, 1999). The primary source of freshwater for the city has been ground water withdrawn from five wells that tap the Cohansey aquifer (fig. 26). Pumping from these wells has led to saltwater contamination in four of the wells (fig. 27). The city's first two wells were installed in 1940 and 1945 and were located within 3,400 ft of the Atlantic Ocean shoreline (Blair and others, 1999). However, by 1950, saltwater contamination had forced the first well to be removed from service, and a third well was installed further inland to replace the contaminated well. Chloride concentrations in water from well 2 increased steadily through the 1950s, and by 1963, the chloride concentration in the well was 200 mg/L. Two additional wells (4 and 5) were installed in 1965 landward of the first three wells as the quality of water in well 2 deteriorated and as demand for additional water increased. Chloride concentrations in water from well 3 increased slowly from about 20 to 50 mg/L from 1950 to the mid-1960s, but then increased to more than 100 mg/L in the 1970s after wells 1 and 2 had been removed from service and their withdrawals shifted to well 3. With the increased use of wells 4 and 5 since the 1970s, chloride concentrations have increased at well 4 from about 20 mg/L in 1965 to more than 100 mg/L beginning about 1985.

From 1985 to 1998, the city depended on well 5 for the majority of its year-round water supply. Withdrawals from well 3 were restricted to periods of peak water demand, and the water that was withdrawn from the well was blended with that from wells 4 and 5 to produce water of acceptable quality (Blair and others, 1999). In addition, since 1994 the city has operated a program of aquifer storage and recovery at well 4, in which water purchased from the Lower Township Municipal Utilities Authority is injected into the well during the off-season and withdrawn from the well during the summer tourist season.

During the mid-1990s, engineers working with the city had determined that projected population growth would result in a peak water demand that exceeded available supply by the year 2000 (Metcalf & Eddy, Inc., 1996). In response, the city sought an economical alternative to the existing water-supply system that would allow further expansion of tourism, would protect the Cohansey aquifer from saltwater intrusion, and could be implemented in a reasonable time period (Blair and others, 1999). The alternative that was judged most viable was one that involved the combined use of existing sources with a new source derived from desalinated brackish ground water. In this alternative, as much as 2 Mgal/d of brackish water would be pumped from two new supply wells installed in the Atlantic City 800-ft sand aquifer, which underlies the Cohansey aquifer and is brackish in the Cape May City area. The desalinated water would be used with freshwater obtained from continued operation of the aquifer storage and recovery system at well 4 and continued use of well 5 during periods of peak demand. The desalinated water is anticipated to reduce the city's reliance on wells 3 and 5 in the Cohansey aquifer and on water imported from Lower Cape May Township. Use of the brackish-water wells also will limit withdrawals from the Cohansey aquifer to periods of peak demand, which should reduce the rate of saltwater intrusion into the aquifer.

The desalination system began operation in 1998 after installation of wells 6 and 7 (fig. 26) and construction of a 2 Mgal/d-capacity desalination plant at the city's Canning House Lane Water Works facility. Brackish ground water is desalinated by use of reverse-osmosis membrane-filtration systems designed to treat brackish water containing up to 2,000 mg/L dissolved solids. Currently (2003), the systems treat about 0.5 Mgal/d on a typical day (David A. Carrick, City of Cape May Water Utility, New Jersey, oral commun., June 2003).

Saltwater Intrusion from the Delaware River During Drought—Implications for the Effects of Sea-Level Rise on Coastal Aquifers

Many shallow aquifer systems along the Atlantic coast are in direct hydraulic connection with streams and other surface waters that are often sources of recharge to the underlying ground-water system. This is the case in the Camden area of New Jersey, where the shallowest parts of the area's aquifer system are in connection with the Delaware River. Freshwater flowing seaward in the river mixes with saltwater from Delaware Bay in a transition zone that is characterized by an increase in the river's chloride concentration similar to the transition zone between freshwater and saltwater in coastal aquifers. The location in the river where the chloride concentration is 250 mg/L is called the "salt line" or "salt front" and is monitored by the Delaware River Basin Commission (Delaware River Basin Commission, 2001). The salt front's position is not stagnant, but moves upstream and downstream in response to tides, rainfall, and freshwater releases from reservoirs within the Delaware River Basin. During periods of average rainfall and streamflow, the position of the salt front is near Wilmington, Delaware (fig. 28A), about 30 mi downstream from Camden; however, during the summer and other periods of reduced flow in the river, the salt front migrates further upstream toward Philadelphia and Camden.

Ground-water withdrawals provide most of the potable water supply for the Camden area. Before these withdrawals began, ground water from the aquifers discharged to the river. However, because of the withdrawals, the direction of flow between the aquifer system and river has been reversed, and in many places river water now is drawn into the aquifer system. Under normal, or average, flow conditions, water in the river is fresh in the Camden area. However, during a period of severe drought that occurred in November and December 1964, freshwater flows to the river were extremely low, and the salt front migrated above Camden to its maximum recorded upstream position (fig. 28A). Ground-water withdrawals in the Camden area during the 1964 drought are thought to have drawn a slug of saltwater from the temporarily saline reaches of the river into the aquifer system (Navoy, 1991; Navoy and Carleton, 1995). Soon after the saltwater-contamination event occurred, higher-than-normal chloride concentrations were measured in ground water pumped from supply wells near the river (fig. 28B). Chloride concentrations at the wells generally peaked in 1965 or later, but because the sampling frequency was low, the peak concentration and timing of contamination at the wells could not be clearly established. The highest chloride concentration measured was about 80 mg/L, whereas the background chloride concentration of water in the aquifer system near the Delaware River is about 10 to 20 mg/L.

Although chloride concentrations in the Camden area supply wells remained below potability standards during and after the short drought-related saltwater-intrusion event of 1964, contamination of the wells demonstrates the vulnerability of coastal aquifers to saline surface water (Navoy, 1991; Navoy and Carleton, 1995). Sea-level rise could have a similar effect on coastal aquifers because as the sea level rises, saltwater would be pushed landward and further upstream into coastal estuaries. The potential consequences of sea-level rise on coastal aquifers is of concern because global (eustatic) mean sea level has increased at an average rate of 1 to 2 millimeters per year (mm/yr) during the 20th century (a total of about 4 to 8 inches), and this increase has been attributed to an increase in the Earth's temperature during the same period (Intergovernmental Panel on Climate Change, 2001). Global warming drives sea-level rise by thermal expansion of seawater and by melting of ice caps, ice fields, and mountain glaciers. Moreover, recently developed climate models suggest that global sea level may increase from 0.09 to 0.88 meter (m) (0.3 and 2.9 ft) from 1990 to 2100 (Intergovernmental Panel on Climate Change, 2001).

Vertical land movements in the earth's crust locally alter the rate of sea-level change from the global value, but usually by no more than a factor of 2 or 3; along the conterminous United States, local rates of sea-level rise average about 2 to 3 mm/yr, with rates of as much as 6 mm/yr or more in some specific areas (Douglas and others, 2001). The variability in the local rate of land movement is reflected in different rates of relative sea-level rise (that is, relative to a fixed datum on land) measured at individual gaging stations along the Atlantic coast. At the Boston, Massachusetts, tide gage, for example, relative sea level rose at a rate of about 2.5 mm/yr from 1921 to 2000 (fig. 29).

Sea-level rise could affect coastal aquifers in several ways (Navoy, 1991; Nuttle and Portnoy, 1992; Ayers and others, 1994; Navoy and Carleton, 1995; Oude Essink, 1999; Sherif, 1999; Sherif and Singh, 1999; and Douglas and others, 2001). Perhaps most fundamentally, a landward movement of seawater would push saltwater zones in coastal aquifers landward and upward, which could accelerate rates of saltwater intrusion into aquifers already experiencing saltwater contamination. Rising sea levels also might cause upstream migration of saltwater in coastal estuaries, inundation of low-lying areas including wetlands and marshes, and submergence of coastal aquifers. In some areas, sea-level rise would erode beaches and bluffs, leading to shoreline retreat, narrowing of aquifers, and diminished areas of aquifer recharge. Sea-level rise also might cause increases in coastal ground-water levels, because of the overall rise in the position of the freshwater-saltwater interface. For example, McCobb and Weiskel (2003) report an increase of 2.1 mm/yr in the average ground-water level measured from 1950 to 2000 at an observation well on the outer part of Cape Cod, Massachusetts (well TSW1 shown in fig. 11). Because this increase is similar to the relative sea-level rise measured at the Boston tide gage from 1921 to 2000 (2.5 mm/yr), McCobb and Weiskel (2003) hypothesize that the ground-water-level increase may reflect sea-level rise.

Although sea-level rise could increase saltwater intrusion into coastal surface and ground waters, landward saltwater movement also will depend in part on changes in precipitation, runoff, and recharge that may occur within coastal watersheds. For example, increased freshwater runoff could counterbalance the landward movement of saltwater. Moreover, should saltwater intrusion into coastal aquifers occur in response to sea-level rise, it is likely that some aquifers may require hundreds to thousands of years to re-equilibrate to changes in sea level, such as has occurred in parts of the Northern Atlantic Coastal Plain aquifer system where freshwater-saltwater interfaces appear to be still responding to sea-level increases that began at the end of the last ice age.

Multifaceted Strategy for Managing Saltwater Intrusion in Coastal Georgia

Ground water withdrawn from the Upper Floridan aquifer is the principal source of water supply for 24 counties of coastal Georgia (fig. 30), an area of rapid population growth, increased tourism, and sustained industrial activity. The Upper Floridan aquifer is an extremely permeable, high-yielding aquifer that was first developed in the 1800s and has been used extensively in the area ever since. Total ground-water withdrawals in coastal Georgia were nearly 350 Mgal/d in 1997 (fig. 31); most of the water was used for industry, public supply, and irrigation. These withdrawals have resulted in substantial water-level declines in the coastal zone (fig. 30), decreased fresh ground-water discharge to springs and other surface-water features, and saltwater contamination at the northern end of Hilton Head Island, South Carolina, and in Brunswick, Georgia (figs. 32 and 33). This contamination is restricting further development of the Upper Floridan aquifer in coastal Georgia and adjacent parts of South Carolina and Florida and has created competing demands for the limited freshwater supply (Krause and Clarke, 2001). In response to the contamination and related water-supply issues, the Georgia Environmental Protection Division (GaEPD) has developed an interim strategy for managing saltwater intrusion in the Upper Floridan aquifer in the 24 coastal-area counties. The State's objective is to stop the intrusion of saltwater before municipal supply wells on Hilton Head Island and Savannah are contaminated and to prevent the existing saltwater contamination problem in Brunswick from worsening (Georgia Environmental Protection Division, 1997).

The multifaceted management strategy calls for a mix of regulatory and nonregulatory approaches for controlling saltwater intrusion. As part of the interim strategy, the GaEPD has restricted permitted withdrawals of water from the Upper Floridan aquifer in parts of the coastal area (including the Savannah and Brunswick areas) to 1997 rates, has restricted additional permitted pumpage in all 24 coastal counties to 36 Mgal/d above 1997 rates, and has encouraged and promoted conservation and reduced ground-water use wherever feasible throughout southeast Georgia. In addition, each of the coastal counties is required to develop comprehensive water-supply plans to manage the Upper Floridan aquifer, including identifying possible alternative sources of water to the Upper Floridan.

Flowing wells such as this one were common throughout the Georgia coastal area prior to large-scale ground-water development.

An important element of the interim strategy has been a program of scientific and engineering studies to support development of GaEPD's final strategy to protect the Upper Floridan aquifer from saltwater contamination, which has been proposed for implementation in January 2006. The Coastal Georgia Sound Science Initiative is a collaborative effort among governmental, university, and private organizations to better understand the processes of ground-water flow and saltwater intrusion in the coastal zone of Georgia and adjacent parts of South Carolina and Florida, to monitor saltwater conditions in the coastal zone, and to evaluate alternative sources of water to the Upper Floridan aquifer. Some of the activities of the Sound Science Initiative are:

Drilling offshore of Tybee Island, Georgia, is helping to delineate the location of the freshwater-saltwater interface in the Upper Floridan aquifer as well as areas where saltwater is entering the Floridan aquifer system. The offshore data collection from a U.S. Army Corps of Engineers drill rig is part of the Coastal Georgia Sound Science Initiative.

E. Numerical Modeling of Coastal Aquifers

Throughout this report, references are made to the development of numerical models that simulate coastal aquifers. Numerical modeling has emerged over the past 40 years as one of the primary tools that hydrologists use to understand ground-water flow and saltwater movement in coastal aquifers. Numerical models are mathematical representations (or approximations) of ground-water systems in which the important physical processes that occur in the systems are represented by mathematical equations.¹ These equations are solved by mathematical techniques (such as finite-difference or finite-element methods) that are implemented in computer codes. The primary benefit of numerical modeling is that it provides a means to represent, in a simplified way, the key features of what are often complex systems in a form that allows for analysis of past, present, and future ground-water flow and saltwater movement in coastal aquifers. Such analysis is often impractical, or impossible, to do by field studies alone.

Numerical models have been developed to simulate ground-water flow solely or ground-water flow in combination with solute transport (the movement of chemical species through an aquifer). For a number of reasons, numerical models that simulate ground-water flow and solute transport are more difficult to develop and to solve than those that simulate ground-water flow alone. Coastal aquifers are particularly difficult to simulate numerically because the density of the water and the concentrations of chemical species dissolved in the water can vary substantially throughout the modeled area. To address these difficulties, one of two approaches generally is used to simulate freshwater-saltwater interactions (Reilly and Goodman, 1985; Reilly, 1993). In the first approach, the freshwater and saltwater zones are assumed to be immiscible (that is, they do not mix) and separated by a sharp interface. In the second approach, the freshwater and saltwater are considered to be a single fluid having a spatially variable salt concentration that influences the fluid's density; this approach is referred to as density-dependent ground-water flow and solute-transport modeling.

A recent example of the use of a numerical model to evaluate a coastal ground-water system is one developed to quantify the rates and patterns of submarine ground-water discharge to Biscayne Bay in southeastern Florida (fig. E–1). The bay is a coastal barrier-island lagoon that relies on substantial quantities of freshwater to sustain its estuarine ecosystem. The two most important sources of freshwater to the bay are thought to be drainage canals and ground-water discharge from the Biscayne aquifer (Langevin, 2001). Although canal discharges are routinely measured, few studies have attempted to quantify the contribution from ground water. To better understand the interaction between the Biscayne aquifer and Biscayne Bay, a numerical model was developed to simulate conditions during 1989 to 1998.

The numerical model was developed on the basis of field investigations that provided information on the structure and hydraulic properties of the aquifer; interactions of the ground-water system with the bay and drainage canals; and hydraulic stresses to the aquifer, which include recharge from precipitation, evapotranspiration, and ground-water withdrawals. The computer code that was used in the study, which is called SEAWAT (Guo and Bennet, 1998; Guo and Langevin, 2002), is based on the second approach for simulating coastal freshwater-saltwater interactions (density-dependent ground-water flow and solute transport). One of the outputs of the model was a calculated distribution of ground-water salinity within the modeled area (fig. E–1). In general, the model reasonably simulated the inland extent of saltwater within the aquifer, particularly south of the Miami Canal. The average rate of fresh ground-water discharge to the bay calculated by the model for 1989 to 1998 was about 53 Mgal/d. Overall, ground-water discharge to the bay was only about 6 percent of the measured canal discharge to the bay for the same 10 years.

Figure E–1. Simulated values of ground-water salinity at the base of the Biscayne aquifer.

¹Several textbooks and articles are available on numerical modeling of ground-water flow systems, including those by Anderson and Woessner (1992) and Konikow and Reilly (1999), which were used as references for this discussion. A review of several computer codes that have been developed for saltwater-intrusion modeling is provided by Sorek and Pinder (1999).

Vertical Migration of Saline Water Along Preferential Flow Conduits in the Floridan Aquifer System—Implications for Saltwater Monitoring Networks

The Floridan aquifer system is the major source of water supply in southeastern Georgia and northeastern Florida. Ground-water withdrawals from the aquifer system in the three northeastern-most coastal counties of Florida alone were 217 Mgal/d in 1995 (Marella, 1999). These withdrawals have caused long-term water-level declines in several areas of the region; in northeastern Florida, for example, water levels in the Floridan aquifer system have declined during the past 50 to 60 years at an average rate of about one-third to one-half foot per year (Spechler, 1994, 2001). Ground-water development in the region has resulted in a complex process of saltwater contamination by upward migration of saline water from the deepest zone of the aquifer system (called the Fernandina permeable zone) to overlying freshwater zones along structural anomalies in the geologic framework of the aquifer system that provide preferential conduits for ground-water flow (figs. 34 and 35). This type of contamination has occurred at Brunswick, Georgia (fig. 33), at Fernandina Beach, Florida, and near Jacksonville in Duval County, Florida (fig. 36).

The Fernandina permeable zone is an extremely permeable, partly cavernous water-bearing zone in the Lower Floridan aquifer. Chloride concentrations of water pumped from the zone have been reported to be as high as 16,800 mg/L in St. Johns County, Florida, and 33,000 mg/L in the Brunswick area of Georgia (Spechler, 1994; Krause and Clarke, 2001). The Fernandina is overlain by a low-permeability confining unit that, in most places,

Circular 1262

Ground Water in Freshwater-Saltwater Environments of the Atlantic Coast

by Paul M. Barlow

The PDF for the report is 10,886 kb

Table of Contents

Boxes

Figures

Tables

Conversion factors, water-quality units, and vertical datums

FOREWORD