EPA/600/2-91/064
December 1991
IDENTIFICATION OF SOURCES OF GROUND-WATER
SALINIZATION USING GEOCHEMICAL TECHNIQUES
by
Bernd C. Richter and Charles W. Kreitler
Bureau of Economic Geology
The University of Texas at Austin
Austin, Texas 78713-7508
Cooperative Agreement No. CR-815748
Project Officer
Bert E. Bledsoe
Extramural Activities and Assistance Division
Robert S. Kerr Environmental Research Laboratory
Ada, Oklahoma 74820
ROBERT S. KERR ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ADA, OKLAHOMA 74820
Printed on Recycled Paper
-------
-------
DISCLAIMER
Although the information in this document has been funded wholly by the United States
Environmental Protection Agency under CR-815748 to The University of Texas at Austin, it does
not necessarily reflect the views of the Agency and no official endorsement should be inferred.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
QUALITY ASSURANCE STATEMENT
All research projects making conclusions or recommendations based on environmentally
related measurements and funded by the United States Environmental Protection Agency are
required to participate in the Agency Quality Assurance Program. This project was conducted
under an approved Quality Assurance Project Plan. The procedures specified in this plan were
used without exception. Information on the plan and documentation of the quality assurance
activities and results are available from the authors.
ii
-------
-------
FOREWORD
EPA is charged by Congress to protect the Nation's land, air, and water systems. Under
a mandate of national environmental laws focused on air and water quality, solid waste
management and the control of toxic substances, pesticides, noise, and radiation, the Agency
strives to formulate and implement actions which lead to compatible balance between human
activities and the ability of natural systems to support and nurture life.
The Robert S. Kerr Environmental Research Laboratory is the Agency's center of expertise
for investigation of the soil and subsurface environment. Personnel at the laboratory are
responsible for management of research programs to: (a) determine the fate, transport and
transformation rates of pollutants in the soil, the unsaturated and the saturated zones of the
subsurface environment; (b) define the processes to be used in characterizing the soil and
subsurface environment as a receptor of pollutants; (c) develop techniques for predicting the
effect of pollutants on ground water, soil, and indigenous organisms; and (d) define and
demonstrate the applicability and limitations of using natural processes, indigenous to the soil and
subsurface environment, for the protection of this resource.
This report deals with geochemical techniques that allow identification of sources of salinity
in the Nation's fresh ground water. Seven major sources of salinity, (1) natural saline ground
water, (2) sea-water intrusion, (3) halite solution, (4) oil- and gas-field activities, (5) agricultural
techniques, (6) saline seep, and (7) road salting, are discussed with respect to occurrence,
geochemical characteristics, and identification techniques. The report is designed to act as a
reference manual that can guide the reader through an investigation of ground-water salinity.
Clinton W. Hall
Director
Robert S. Kerr Environmental
Research Laboratory
-------
-------
TABLE OF CONTENTS
Foreword v
Executive Summary...; xiii
Acknowledgments xiv
1. Introduction 1
1.1. Purpose and Use of this Report 1
1.2. Background 2
2. Geographic Distribution of Major Salinization Sources 6
3. Major Salinization Sources 17
3.1. Natural Saline Ground Water 17
3.1.1. Mechanism 17
3.1.2. Hydrochemistry of Different Sources of Naturally Occurring Salinity 22
3.1.3. Examples of Geochemical Studies of Natural Saline Ground Water 28
3.1.4. Significant Parameters ....35
3.1.5. State-by-State Summary 38
3.2. Halite Solution 56
3.2.1. Mechanism 56
3.2.2. Composition of Halite and other Evaporites 57
3.2.3. Examples of Geochemical Studies of Halite Solution 62
3.2.4. Significant Parameters 75
3.2.5. State-by-state Summary of Halite Occurrences 75
3.3. Sea-Water Intrusion 81
3.3.1. Mechanism 81
3.3.2. Chemistry of Sea Water 86
3.3.3. Examples of Geochemical Studies of Sea-Water Intrusion 88
3.3.4. Reaction Characteristics of Sea-Water Intrusion 103
3.3.5. State-by-state Summary of Sea-Water Intrusion 109
3.4. Oil-Field Brine 115
3.4.1. Mechanism 115
3.4.1 a. Surface disposal 117
3.4.1b. Injection wells 119
3.4.1c. Plugged and abandoned boreholes 119
3.4.2 Oil-Reid Brine Chemistry 122
3.4.3. Examples of Geochemical Studies of Oil- and Gas-Reid Brine Pollution 122
vfi
-------
3.4.4. Significant Parameters 132
3.4.5. State-by-State Summary of Oil- and Gas-Field Contamination 134
3.5. Agricultural Sources 142
3.5.1. Mechanism 142
3.5.2. Water Chemistry 147
3.5.3. Examples of Geochemical Studies of Agricultural Salinization 149
3.5.4. Significant Parameters 156
3.5.5. State-by-state Summary of Agriculturally-Induced Ground-Water Problems 156
3.6. Saline Seep 159
3.6.1. Mechanism 159
3.6.2. Water Chemistry '.. 162
3.6.3. Examples of Geochemical Studies of Saline Seep 163
3.6.4. Significant Parameters 170
3.6.5. State-by-state Summary of Saline Seep Occurrences 172
3.7. Road Salt ....175
3.7.1. Mechanism , 175
3.7.2. Road-Salt Chemistry 179
3.7.3. Examples of Geochemical Studies of Road Salting 180
3.7.4. Significant Chemical Parameters 181
3.7.5. State-by-state Summary of Road-Salt Issues 181
4. Geochemical Parameters 188
4.1. Discussion of Individual Parameters 188
4.2. Summary of Reid Techniques 201
5. Data Availability and Selection 204
5.1. Sources of Data 204
5.2. Selection of Data Criteria 209
6. Graphical and Statistical Techniques 211
6.1. Graphical Techniques 211
6.2. Statistical Techniques 221
7. References 230
FIGURES
1. Map of the United States showing areas of ground water containing more than 1,000 mg/L
total dissolved solids at depths less than 500 ft below land surface 7
2. Map of the United States showing the approximate extent of halite deposits 8
viii
-------
3. Map of the United States showing areas of potential sea-water intrusion 9
4. Map of the United States showing general areas of oil and gas production 11
5. Major irrigation areas in the United States.... 12
6. Saline-seep areas in Montana, North and South Dakota, and Texas ; 13
7. Map of the United States showing amount of road salt used in individual states during
the winters of 1966-1967, 1981-1982,'and 1982-1983 14
8. Composite map of potential salinization sources ,.... 15
9. Relationship between ground-water quality and recharge-discharge areas 19
10. Discharge of regional, saline ground water and mixing with local, fresh ground water 21
11. Relationship between depth of screen and chloride content in two adjacent wells 23
12. Geochemical composition of ground water in a gypsum playa, West Texas 27
13. Separation of local recharge water from nonlocal ground water using stable isotopes 29
14. Use of bromide, iodide, boron, and chloride concentrations in bivariate plots for
identification of salt-water sources 31
15. Separation of different ground-water origins using the strontium-iodide relationship in ground
water T 32
16. Ground-water flow directions and salinities in parts of the Murray Basin, southwest Australia 33
17. Differentiation between fresh-water contamination caused by oil-field brines versus
evaporatively concentrated brine 34
18. Natural salinization of ground water in parts of the San Joaquin Valley of California 36
19. Stable-isotope composition of ground water affected by evaporation 37
20. Solution of halite by circulating ground water, Palo Duro Basin, Texas 58
21. Discharge points of halite-solution brine at land surface in the Rolling Plains of
north-central Texas and southwestern Oklahoma 59
22. Bivariate plots of major ions and Br/CI ratios versus chloride for natural halite-solution brines 61
23. Classification of oil-field/deep-basin waters according to TDS and bromide concentrations 63
24. Modified Piper diagrams of chemical composition of brine springs and surface
waters in southern Manitoba ....64
25. Comparison of Na/CI weight ratios for oil-field brines and brines from salt springs in western
Oklahoma and southwestern Kansas 65
26. Differentiation of oil-field/deep-basin brine from halite-solution brine using bivariate
plots of Br/CI weight ratios versus chloride concentrations 67
27. Grouping of brine analyses according to molar Na/CI ratios and Br/CI weight ratios
typical for halite solution and oil-field/deep-basin brine 69
28. Bivariate plots of chemical constituents in brines collected from salt springs and
shallow test holes in the Rolling Plains of Texas 70
29. Molar ratios of (Ca+Mg)/SO4 versus Na/CI in salt-spring and shallow subsurface brines,
Roling Plains of Texas : 71
30. Relationship between 818O and Cl with depth in shallow brines in parts of the Rolling Plains,
Texas ..., 72
IX
-------
31. Differentiation between halite-solution waters and oil-field/deep-basin brines using
differences in I/CI weight ratios 73
32. Theoretical mixing curves for fresh water and potential salinization sources in Kansas using
sulfate and chloride concentrations > 74
33. Typical ground-water flow patterns along coast lines 83
34. Schematic cross section showing mixing mechanism of sea water with fresh water through
sinkholes and solution openings in a carbonate aquifer ...85
35. Bivariate plots of major ions and of Br/CI ratios versus chloride for sea-water intrusion
samples i 90
36. Increase in chloride concentration with time in coastal wells, Monterey County, California 93
37. Landward movement of sea water mapped by the position of the 500 mg/L chloride
contour in two shallow aquifers, Monterey County, California 94
38. Piper diagram of chemical composition of coastal saline water, Monterey County, California 95
39. Piper diagram of chemical composition of selected saline waters from the Manhattan Beach
area, Caifomia 97
40. Iodide concentrations as indicator of residence time 99
41. Bivariate plot of SOyCI ratios and SCvj. concentrations for coastal saline ground water of
southwest Florida :.. 101
42. Dilution diagrams of major ions with theoretical mixing b'nes between sea water and
local fresh water : ...,....;....; 104
43. Relationship between flow regimes and hydrochemical fades in ground water in a big
coastal plain 105
44. Geochemical processes and changes in ionic composition along the flow path in a
coastal aquifer 106
45. Map of shallow oil fields in the United States , 120
46. Schematic diagram illustrating possible communication scenarios between deep saline
aquifers and shallow fresh-water units through boreholes 121
47. Bivariate plots of major ions and Br/CI ratios versus chloride for oil-field brines 123
48. Monitoring of chloride concentrations in ground water as an aid in identifying salt-water
contamination 125
49. Relationship between soil chloride and depth inside and outside an abandoned
brine-disposal pit 126
50. Bivariate plots of major ions versus chloride from water-supply wells and test holes in the
vicinity of an abandoned brine-disposal pit 127
51. Identification of salinization sources using bar graphs, Stiff diagrams, and contouring 129
52. . Mapping of point source of salinity and the resulting salt-water plume through contouring
of chloride concentrations from water wells.. 130
53. Determination of geochemical constituent criteria that indicate brine contamination of
fresh ground water in parts of Oklahoma 133
54. Transport of salt to discharge areas as a result of irrigation-return flow 144
55. Classification of irrigation water based on sodium-adsorption ratios and conductivity 148
-------
5 6. Increase in TDS concentration in the Arkansas River alluvial aquifer as a result of
irrigation-return flow ..;.. 151
57. Use of Stiff diagrams for identification of water-quaity changes as a result of agricultural
activities ,... 152
58. Salt content in irrigation water, irrigation runoff and in soil during and after three controlled
irrigation experiments.... 154
59. Relationship between TDS, Na, Cl, Ca+Mg, 804 and depth in soil under irrigated and
nonirrigated land 155
60. Diagrammatic cross section of ground-water flow with saline seep in topographically low
areas and in intermediate areas 161
61. Correlation between 864 and TDS concentrations in seep waters from the Colorado Group,
Montana.................. 164
62. Relationship between saturation states and salinity in well waters affected by saline seep,
north-central Montana 165
63. Summary of chemical and transport processes operating in a saline-seep system........ ...166
64. Modified Piper diagram of chemical composition of ground water in saline seeps, Montana 167
65. Piper diagram of shallow ground water in two adjacent counties of West Texas 169
66. Variation in calcium, magnesium, and sulfate concentrations with chloride concentration
in shallow ground water from a two-county area in parts of West Texas 171
67. Rate of growth of saline seep near Fort Benton, Montana 173
68. Salt production and usage of salt for deicing , 177
69. Correlation between increase in salt usage applied to highways and chloride concentrations
in ground water, Massachusetts, 1955-1971 .- .178
70. Bivariate plot of Br/CI ratios versus Cl for selected water samples from northeastern Ohio 182
71. Location map of ground-water stations for which a chloride value is available at
U.S. Environmental Protection Agency's data base STORET ,..-..., 206
72. Location map of water wells in Texas for which a chloride value is stored at the Texas Natural
Resources Information System data bank 207
73. Graphical illustration of chemical analyses by contouring of individual parameters onto
maps and cross sections , , 212
74. Presentation of major ions in form of bar graphs, pie charts, Stiff diagrams, and Schoeller
diagrams 213
75. Presentation of chemical constituents on trilinear (Piper) diagram and classification scheme
of hydrochemical fades 215
76. Bivariate plots of Na versus Cl for halite-solution and deep-basin brines ,.217
77. Use of bivariate plots for identification of mixing trends between fresh ground water and
potential salinization sources in parts of West Texas.... ;...-. :......' 218
78. Modified Schoeller and Piper diagrams using concentration ratios as end points........... 219
79. Calculation of mixing percentages between fresh water and salt water using mixing graphs
of chloride and bromide , 220
80. Bivariate plots of ratios determined by application of Stepwise Discriminant Analysis as the
statistically best ratios to distinguish sea-water intrusion from oil-field brines 226
xi
-------
81. Bivariate plots of ratios determined by application of Stepwise Discriminant Analysis as the
statistically best ratios to distinguish halite-solution brine from oil-field brines 227
82. Bivariate plot of Na/CI ratios versus Ca/CI ratios for Illinois ground water, Illinois oil-field
brines, and ground water contaminated by oil-field brine.. .-.'. 229
TABLES ..... ,
1. Drinking-water standards established for inorganic constituents ....: ....•...;'.. 3
2. Ground-water classification based on IDS ranges .......: :.................... ;'.'• 4
3. Mineral composition of salt from domes in Louisiana and Texas.. ...........;.........:...•.....;../......60
4. Concentration of major and some minor chemical constituents in sea water............. 87
5. Landward changes in chemical composition of Gulf of Mexico water as a result of dilution. 89
6. Changes in chloride concentration between 1977 and 1981 in the Fam'ngtbn aquifer of
New Jersey as a result of sea-water intrusion ,.. 91
7. Changes in ionic ratios due to ion exchange in relation to the position of the
intruding water .. ! .....108
8. Methods of disposal of produced oil- and gas-field brine in the United States
during 1963 ::..........................1... ..116
9. Percentage of saline and sodic areas in seventeen western states and in Hawaii.. 146
10. Soil salinity classification in the United States 150
11. Reported use of road salt and abrasives during the winters of 1966-1967,1981-1982,
and 1982-1983 176
12. Geochemical parameters used for identification of salinity sources 189
13. Approximate costs of chemical and isotopic analyses 192
14. Description of literature data bases 208
15. Listing of constituent ratios that separate best brines from Texas, Louisiana,
Oklahoma, California, Ohio, and Canada, as determined through
Stepwise Discriminant Analysis 225
xii
-------
EXECUTIVE SUMMARY
This report deals with salt-water sources that commonly mix and deteriorate fresh ground
water. It reviews characteristics of salt-water sources and geochemical techniques that can be
used to identify these sources after mixing has occurred.
The report is designed to assist investigators of salt-water problems in a step-by-step
fashion. Seven major sources of salt water are distinguished: (1) Natural saline ground water,
(2) Halite solution, (3) Sea-water intrusion, (4) Oil-and gas-field brines, (5) Agricultural effluents,
(6) Saline seep, and (7) Road salting. The geographic distribution of these sources was mapped
individually and together, illustrating which ones are potential sources at any given area in the
United States. In separate sections, each potential source is then discussed in detail regarding
physical and chemical characteristics, examples of known techniques for identification of mixtures
between fresh water and that source, and known occurrences by state. Individual geochemical
parameters that are used within these techniques are presented in a separate section, followed
by a discussion concerning where and how to obtain them. Also provided is a description of basic
graphical and statistical methods that are used frequently in salt-water studies. An extensive list of
references for further study concludes this report.
xiii
-------
ACKNOWLEDGMENTS
Funding for this project was provided by the U.S. Environmental Protection Agency,
Robert S. Kerr, Environmental Research Laboratory, Office of Research and Development,
under Cooperative Agreement No.CR-815748. We thank reviewers from the U.S. Environmental
Protection Agency for their manuscript reviews. Figures were drafted by Maria Saenz, tari
Weaver, Kerza Prewitt, Jana Robinson, Michele LaHaye, Margaret Koenig, and Joel Lardon under
the direction of Richard L. Dillon. Word processing and typesetting were clone by Susan Lloyd
under the direction of Susann Doenges. The publication was edited by Kitty Challstrom and
designed by Margaret Evans. The efforts of all these people are greatly appreciated.
xiv
-------
1. INTRODUCTION
1.1. Purpose and Use of this Report
The purpose of this report is to summarize geochemical techniques that can be used in studies of
salinization of fresh water. The report is designed to assist investigators through detailed discussion of
potentially useful chemical parameters and techniques, as well as of physical and geographical
characteristics of potential salinization sources. . .
The topic of salt-water contamination has been extensively researched, as evidenced by the long list
of references compiled for this report. No compendium of the overall topic, however, has previously been
compiled. The purpose of this document is not to develop new geochemical techniques for identifying
sources of ground-water salinity, but to summarize known approaches for all different sources into a single
document so that a researcher will have a reference manual that reviews available work.
Salinization of fresh water is perhaps the most widespread threat to ground-water resources. This
document deals with geochemical characteristics of the major known sources of salinity, and as such will
be helpful to investigators of salt-water problems. The extent to which this document will be of help will
depend to a large degree on the background knowledge of the problem and of the investigator. To an
experienced researcher in the field of ground-water quality, this document may serve as a summary of and
reference to some of the known techniques that are being used. To investigators new in this field, we
suggest the following possible methodology of investigation in combination with this report.
Step 1: The general geographic distribution of major potential salinization sources, that is
(1) natural saline ground water, (2) halite solution, (3) sea-water intrusion, (4) oil-and gas-
field brines, (5) agricultural effluents, (6) saline seep, and (7) road salt, is addressed in
Chapter 2 of this report. Through a series of maps that show the distribution of each
source as well as the overlap between these sources, the investigator can get a general
idea which potential salinization source or sources exist at her/his local area of interest at
any given area of the country. There are some limitations to the maps, as discussed in
Chapter 2.
Step 2: After potential sources of salt water have been identified, Chapter 3 should be consulted
for a discussion of the sources. This will provide the researcher with the necessary
background information about the source(s) of interest. Each of the seven sources is
discussed in detail, including mechanisms of mixing with fresh ground water, chemical
characteristics, geochemical case studies, recommended chemical techniques for
identification of salinization caused by these sources, and a state-by-state summary of
occurrences. For each source, a variety of techniques that can be used is presented.
-------
Each section includes a state-by-state summary of known problem cases associated with
individual salinization sources. Before disregarding any source identified in Step 2 as
absent in the area of salinization and of interest, the investigator may want to review all
summaries pertinent to her/his state(s). With the help of references listed in Chapter 7,
extensive background information of the problem can be obtained.
Step 3: After selecting techniques that are useful for the particular problem case, the
geochemical parameters of interest should be reviewed in Chapter 4. This will give the
investigator a general overview of parameter characteristics as well as sampling
techniques and costs of laboratory analyses.
Step 4: Depending on the area of interest, chemical data may or may not be available to the
investigator from published sources, agency files, or computerized data banks. Some of
the techniques selected in Step 3 may be applicable using existing data, but others most
likely will necessitate collection of water samples for parameters that are not determined
on a regular basis (for example, isotopes). Chapter 5 should be consulted for guidelines
of data selection and for a discussion on computerized data banks. This step is crucial, as
existing data can be very helpful but may also be misleading. Chemical analyses that may
be representative of potential salinization sources can be found in the referenced
literature.
Step 5: Once data have been selected from existing sources or collected in the field, evaluation
can be accomplished using techniques selected during Step 2. Useful graphical and
statistical methods are discussed briefly in Chapter 6. Hopefully, the source of salinity will
then be determined.
1.2. Background
All natural waters contain some dissolved minerals through the interaction with atmospheric and soil
gases, mixing with other solutions, and interaction with the biosphere and lithosphere. In many cases,
these processes result in natural waters that contain total dissolved solids (TDS) concentrations above
those recommended for drinking water (Table 1). This deterioration of water quality is enhanced by almost
all human activities through water consumption and contamination.
Salinization, that is, the increase in TDS, is the most widespread form of water contamination. The
effect of salinization is an increase in concentrations of specific chemical constituents as well as in overall
chemical content. A variety of terms have been introduced in the literature to reflect the changing
character of the water as salinity increases, such as saline, moderately saline, very saline, brackish, and
brine (Table 2). For the purpose of this report, we followed the classification of Robinove and others
(1958), which is one of the most widely used. The term "salinization," as it will be used in this report,
-------
Table 1 . Drinking-water standards established for
inorganic consituents (data from Freeze and Cherry,
1979, and U.S. Environmental Protection Agency, 1989).
Recommended
Constituent concentration limit
(mg/L)
Total Dissolved Solids (TDS) 500.000
Chloride (Cl) 250.000
Sulfate (864) . 250.000
Nitrate (NOs) 45.000
Iron (Fe) 0.300
Manganese (Mn) 0.050
Copper (Cu) 1.000
Zinc(Zn) 5.000
Boron (B) 1 .000
Hydrogen Sulfide (^S) 0.050
Maximum permissible
concentration
(mg/L)
Arsenic (As) 0.050
Barium (Ba) 1.000
Cadmium (Cd) 0.010
Chromium (Cr) 0.050
Selenium (Se) 0.010
Antimony (Sb) 0.010
Lead (Pb) 0.050
Mercury (Hg) 0.002
Stiver (Ag) 0.050
Fluoride (F) 1.4-2.4
-------
Table 2. Ground-water classification based on TDS
ranges (in mg/L).
Roblnove and others, 1958:
Fresh 0 - 1,000 TDS
Slightly saline 1,000 - 3,000
Moderately saline 3,000 - 10,000
Very saline 10,000 - 35,000
Briny >35,000
Freeze and Cherry, 1979:
Freshwater 0 - 1,000 TDS
Brackish water 1,000 - 10,000
Saline water 10,000 -100,000
Brine >100,000
-------
indicates an increase in IDS from background levels by any source. As such, salinization may or may not
cause concentrations higher than drinking water standards.
Of the variety of potential sources of salinity, some are natural and others are anthropogenic.
Precipitation interacts with atmospheric gases and particles even before it reaches the earth's surface, as
reflected in often low pH values in areas of high sulfur dioxide content in the atmosphere (formation of
sulfuric acid, "acid rain"). Strong winds carry mineral matter and solution droplets (for example, ocean
spray) that can be dissolved and incorporated into precipitation. Surface runoff dissolves mineral matter on
its way toward a surface-water body, where it mixes with water of different chemical composition. Water that
enters the soil is subject to additional chemical, physical, and biological changes, such as
evapotranspiration tineral solution and precipitation, solution of gases, and mixing with other solution.
Changes in chemical composition continue in ground water along flow paths from recharge areas to
discharge areas. Water-rock interaction and mixing are the dominant processes. Mixing of different waters
is often enhanced by human activities. For example, improper drilling, completion, and final construction
of wells may create artificial connections between fresh-water aquifers and saline-water aquifers. Pumping
of fresh water may change directions of ground-water flow and may cause encroachment of saline water
toward the pumped well; improper waste-disposal activities or techniques may introduce artificial solutions
that contaminate natural ground water.
Some areas of the country experience very little problems regarding salinization of fresh-water
resources, whereas in other areas most of the available ground water is saline, reflecting natural and
human-induced degradation. Where such conditions for salinization of fresh water exist is discussed in
the following chapter.
-------
2. GEOGRAPHIC DISTRIBUTION OF MAJOR SALINIZATION SOURCES
For the purposes of this report, seven major salinization sources were singled out. They are,
(1) natural saline ground water, (2) halite solution, (3) sea-water intrusion, (4) oil-and gas-field brines,
(5) agricultural effluents, (6) saline seep, and (7) road salt. The geographic distribution of these potential
sources and areas of overlap between these sources are discussed in this chapter. A detailed discussion
of each individual source will follow in the next chapter.
Saline ground water (TDS>1,000 parts per million [ppm]) of variable origin underlies approximately
two-thirds of the United States (Feth and others, 1965). It may be encountered in water wells that were
drilled too deep for local conditions and it is a threat to those wells that are pumped at a sufficiently high
rate to induce salt-water flow toward the well. Shown in figure 1 are those areas where TDS concentrations
are greater than 1,000 ppm within 500 ft of land surface. Outside these areas, saline water does occur, but
generally at depths greater than 500 ft below land surface (Feth and others, 1965). Although the potential
of salinization exists in these outside areas wherever wells are pumped at high rates and from great
depths, the cut-off value of 500 ft was adopted from Feth and others (1965) because the search for
usable ground water (and with it drilling activities to greater depths) may be greatest in areas where less
fresh water is available. Conditions may change in the future or may be different locally, as the demand for
ground water increases.
Many sedimentary basins are known to contain large deposits of rock salt in the form of salt beds or
salt domes (Rg. 2). Some of these deposits occur at great depths, such as those in southernmost Florida
that are at greater depths than 10,000 ft below land surface. Others occur close to land surface, such as
those in parts of Utah (Dunrud and Nevins, 1981). Shallow occurrences of salt in Texas, Louisiana,
Alabama, and Mississippi along the Gulf of Mexico are due to salt diapirism. The presence of salt deposits
millions of years old Indicates a relatively high stability, that is, little contact with ground water. Where
ground water comes into contact with salt deposits, often enhanced by heavy drilling and mining activities,
especially in salt-dome areas, solution of salt and salinization of local ground waters will occur. The
shallower the salt deposit, the higher the potential of fresh-water salinization. In this report, all salt deposits
are considered potential salinization sources, regardless of the depth of occurrence.
Where coastal aquifers are interconnected with the open ocean, sea-water intrusion can occur.
Where formation water hasn't been flushed out, where sea-water has intruded or is intruding into coastal
aquifers as a result of high sea-water levels, or where pumping induces landward flow of sea water, the
potential of well-water salinization exists. For the purpose of this report, all coastlines of the country were
considered potential intrusion areas, regardless of the nature of the coastal aquifer (Fig. 3). On a local
scale, some coastal areas can probably be disregarded as a source of salinity, especially where ground-
water pumpage is low.
-------
f'"-^f
\j
QAI7I99
Figure 1. Map of the United States showing areas of ground water containing more than 1,000 mg/L total
dissolved solids at depths less than 500 ft below land surface (data from Feth and others, 1965).
-------
00
300 600 km
QAI7200
Figure 2. Map of the United States showing the approximate extent of halite deposits (data from Dunrud
andNevins, 1981).
-------
300 600 km
Figure 3. Map of the United States showing areas of potential sea-water intrusion.
-------
Associated with the exploration of oil and gas is the creation of avenues for water migration from
great depths into the shallow subsurface. Subsequent production brings huge amounts of brine to the
land surface. These drilling activities and the disposal of these brines are some of the biggest salinization
hazards in the country. Parts of 25 producing states are potentially affected by this hazard, as mapped by
the general distribution of oil and gas fields in the United States (Rg. 4).
Salinization as a result of agricultural activities is found all over the country. Irrigation-return waters
pose a potential threat in the western half of the United States (Rg. 5), where precipitation rates are low
and where evapotranspiration rates and salt contents in soil are high. Another salinization source
enhanced by agriculture is dryland saline seep. Terracing of land and destruction of natural vegetation
caused this phenomenon in several states, resulting in the salinization of soil and ground water (Rg. 6).
Weather conditions favor concentration of road salting in the northeastern part of the country
(Fig. 7). There, millions of tons of salt are applied to roads each winter, imposing a salinization threat to
soil, plants, and surface and ground water in the vicinity of highways.
Mapping of potential salinization sources, as done in figures 1 through 7, is helpful in determining
sources of salinity at any particular area in the country. By overlaying all seven sources, a variety of
combinations between these sources becomes evident. This large variety complicates generic
approaches to salt-water studies, because salt-water characteristics change considerably from area to area
depending on the kind of combination of sources involved. In addition, not only the potential salinization
sources change from area to area, but also the chemical characteristics of individual sources may not be
the same everywhere, greatly increasing the number of potential combinations of possible mixing
between fresh-water and salt-water sources. As the composite map (Fig. 8) of the above-mentioned
potential sources indicates, approximately three-quarters of the country could possibly be affected by two
or less than two of the selected sources. In these areas, identification of an actual salinization source could
be easier than in other parts, where three or more potential sources exist.
There are several limitations to the applicability of the maps (Figs. 1 through 8) presented here.
Limitations that should be kept in mind when selecting or eliminating potential sources at any given area
are (1) only the sources discussed in this report were mapped; (2) the sole presence of a potential source
does not necessarily indicate that the source actively contributes to any salinity problem; (3) effects of
these sources may be felt away from the point of origin; (4) areas of occurrence were generalized with
approximated boundaries; and (5) the known distribution of potential salinization sources may have
changed since originally mapped. Not included on these maps are some other known salinization
sources, such as sewer systems, thermal springs, waste-disposal facilities, or mining areas, all of which
may contribute to ground-water salinization in some areas. These sources were not considered because
large-scate regional mapping would have been more complicated and would have resulted in even more
small areas of overlap. Also, chemical characteristics of these sources vary to a higher degree locally than
the ones discussed here, which would have complicated the discussion even further. The seven
10
-------
300 600 km
QAI7202
Figure 4. Map of the United States showing general areas of oil and gas production (modified from
PennWell Publishing Co., 1982).
-------
300 600 km
QAI7203
Figure 5. Major irrigation areas in the United States (modified from Geraghty and others, 1973).
-------
300 600 km
QAI7204
Figure 6. Saline-seep areas in Montana, North and South Dakota, and Texas (data from Bahls and Wilier,
1975, and Neffendorf, 1978).
-------
3/8 2/87 \ 398/II8/B8
EXPLANATION
4/8.5/9.8 I966-67/ 1981-82 / 1982-83
NR Not reported
OAI7205
Figure 7. Map of the United States showing amount of road salt (in thousands of tons) used in individual
states during the winters of 1966-1967,1981-1982, and 1982-1983 (data from Field and others, 1973;
Salt Institute, undated).
-------
POTENTIAL SOURCES OF SALINITY:
LEGEND: Geogiaphic overlap ol potential salinity souices
tn
• Natural saline giound water
• Sea-waler intrusion
• Hahle dissolution
• Oil- and {jas-lield activities
• Inigalion
• Saline seep
• Road salt
O None
<"> Any one potenlial source
^ Any two potenlial souices
*t Any three 01 more potential sources
Figure 8. Composite map of potential salinization sources as mapped in figures 1 through 7.
-------
potential sources mapped may or may not be active at any area, depending on natural and human-induced
conditions. For example, oil-field activities or the presence of halite in the subsurface don't necessarily
imply that either cause salinity of ground water in a specific area. Also, for reasons of simplicity, only those
states with a tang history of road-salt use (cut-off value of greater than 30,000 tons of road salt use per
year in 1966) was used in the overlap map of figure 8. This should not imply that, for example, a storage
area of road salt in any of the states that applied less than 30,000 tons per year at that time is not a
potential source of salinity, or that increased usage in some other states may not pose a problem now or in
the future. If ground-water salinity is a problem, every potential source in that area should be considered. If
a salinity problem exists in an area where no potential salinizatlon source is indicated on figures presented
here, either transport of salt water from a bordering area or from a source not covered in this report should
be considered. It also should be kept in mind that some of the base maps used in constructing these
maps are up to 30 years old, and may be somewhat outdated regarding the present occurrence of any of
these sources on a local basis. In any case, these maps should probably be used only to get a general
Idea of potential sources involved, after which available maps with more local detail should be consulted.
The scale factor alone should make it imperative that the geographic distribution of salinization sources
should be mapped at the beginning of each salinizatlon study using local maps instead of maps presented
in this report. The sources mapped in this chapter will be discussed individually and in detail in the
following chapter.
16
-------
3. MAJOR SALINIZATION SOURCES
As discussed in the introduction chapter, a variety of salinization sources exists throughout the
country. Some of these are more dominant than others, and individually sources may be active in some
place but not in another, depending on natural conditions and changes induced by man. Individual
salinization sources judged as most important on a regional level are (1) naturally occurring saline ground
water, (2) halite solution, (3) sea-water intrusion, (4) oiMield and gas-field brines, (5) agricultural by-
products and techniques, (6) saline seep, and (7) road salt.
Each section on individual salinization sources is divided into (1) a discussion of the mechanism of
contamination, (2) the chemical composition of the salinity source, (3) examples of geochemical studies
that were conducted to identify the specific salinity source, (4) the most significant chemical parameters
that are commonly used to trace the respective source, and (5) a state-by-state summary of the salinity
source. This way, a researcher is provided with an in-depth review of geochemical methods for identifying
various salinity sources. A cookbook approach for identifying salinity sources is not followed because the
complexity of local contamination and hydrogeology precludes a step-by-step approach. After reviewing
the information contained in a section, a researcher needs to develop his/her own hydrochernical criteria
based on the type of salinity, the hydrogeotogic setting, and the type of data and budget available.
Although comprehensive, the following discussion of salinization sources cannot be complete.
Local sources of reference for the area of interest should be incorporated by the researcher, who should
have a general understanding of ground-water conditions in the area.
3.1. Natural Saline Ground Water
3.1.1. Mechanism
Natural saline ground water, as used for this manual, is regionally occurring saline ground water that
underlies fresh-water aquifers. This chapter deals with natural discharge of such saline ground water, with
pumping-induced mixing between saline ground water and fresh water, and the upward migration of saline
ground water along boreholes drilled through the fresh-water section into the salt-water section. Not
discussed is salinization associated with solution of halite (see chapter 3.2), with sea-water intrusion (see
chapter 3.3), and with oil-field brine production (see chapter 3.4), all of which also deal with natural saline
ground waters. Chemical characteristics of deep-basinal formation brines are similar or identical to most
brines produced from oil and gas reservoirs. The nature of occurrence in salt-water problems is different,
however, as contamination by oil-field brines involves pumping and disposal of brine or creation of artificial
migration pathways, whereas contamination by deep-basin brings involves subsurface migration as a
result of natural or pumping-induced conditions. This difference in mechanisms of contamination but
17
-------
similarity in chemical characteristics creates some overlap in the discussion of natural saline ground water
in this section and the discussion of oil-field brines in chapter 3.4.
Saline ground water underlies the country at variable depths. Some areas don't contain any fresh
ground water, whereas in other areas thick fresh-water aquifers overly saline ground water. For example,
little or no saline water is known to occur within 1,000 ft of land surface in most of Nebraska and Missouri,
whereas in some other states, such as Indiana, Ohio, and North Dakota, saline water is encountered at
less than 500 ft below land surface throughout almost the entire state (Rg. 1). Where plenty of fresh water
overlies shallow saline water, no major problems of salt-water mixing with fresh water may occur, such as in
Iowa (Atkinson and others, 1986). Other areas may not be so fortunate, such as New Mexico, where an
estimated 75 percent of all ground water is too saline for most uses (Ong, 1988).
The occurrence of saline ground water is dependent on a variety of factors, including distribution
and rates of precipitation, evapotranspiration and recharge rates, type of soil and aquifer material,
residence time and flow velocities, or nature of discharge areas. The origin of natural saline ground water
can be residual (connate) water from the time of deposition in a saline environment, solution of mineral
matter in the unsaturated and saturated zones, concentration by evapotranspiration, intrusion of sea
water, or any mixture of the above. Residual saline water is not found very often within the shallow
subsurface because of the normal flushing of formation water by precipitation through time. Relatively
young coastal aquifers may still contain pockets of connate water where hydraulic gradients are low and
where hydraulic conductivities are tow. Typically, natural salinity in ground water increases with depth
below land surface as chemical reactions with aquifer material, resident time, and mixing of different waters
increase. Ground water in discharge areas typically is of lower quality than ground water in recharge areas
because of water-rock interactions and possible mixing with saline water along the flow path (Rg. 9).
In some settings, especially in the western half of the United States, salts may be concentrated in the
shallow subsurface due to evaporation rates that exceed precipitation by up to one order of magnitude
(Geraghty and others, 1973). Significant recharge pulses may dissolve this salt and flush it into ground
water. Evaporation is enhanced by transpiration of water by plants, which is a serious problem in several
southwestern states. Woessner and others (1984) reported that phreatophytes can cause TDS increases
from 2,000 mg/L to 11,000 mg/L in a single growing season in parts of Arizona and Nevada. The salt that
accumulates in the soil during the growing season, from 403 mg/kg to 28,177 mg/kg, is flushed toward
discharge areas during major recharge periods.
Many ground-water basins in the western United States are closed. In those basins, natural recharge
along the surrounding highlands flows toward the basin centers. Along the flow path, ground water
dissolves mineral matter, resulting in a general increase in TDS content from recharge areas to discharge
areas. Evaporation in the basins and especially in salt flats in the center of the basins is the most influential
process in the development of the chemical composition of the shallow, saline ground water in these
settings (Boyd and Kreitler, 1986). Evaporation and mineral precipitation concentrate the ground water to
18
-------
RECHARGE
,Woter tobie
DISCHARGE
Zone of woter low
in dissolved solids
Ground - water
flow lines
Sond and gravel
deposits
Zone of water
moderately high in
dissolved solids
7**+*;
Zone of water +
high in chloride*
Figure 9. Relationship between ground-water quality and recharge-discharge areas (from Kantrowitz,
1970). Solution of aquifer material and evapotranspiration in discharge areas commonly cause water-
quality deterioration along the flow path.
19
-------
the composition of a Na-Mg-SCXt-CI-type brine, noted for Deep Springs Lake (Jones, 1965), Death Valley
(Hunt and others, 1966), and Sierra Nevada Basin (Garrels and MacKenzie, 1967) in California, for Abert
Lake in Oregon (Jones and others, 1969), for Teels Marsh in Nevada (Drever, 1982), and for the northern
Salt Basin of West Texas (Boyd and Kreitler, 1986). Surface water that accumulates at topographically low
areas in the center of the basins is often concentrated by evaporation. Infiltration of these concentrated
waters can reach the water table and cause severe pollution. In the western United States, many areas
which are now drained by major river systems may have been under closed conditions in the past, during
which high concentrations of residual salt were left behind in low-permeable, lacustrine deposits. Because
of low permeabilities, these salts may have been preserved in these settings under natural conditions.
Overpumpage of good-quality ground water can lead to inflow of these saline ground waters, as fresh-
water sources are depleted and waters are drained out of low-permeable units.
Natural contamination of fresh water by saline ground water occurs where salt water from saline
aquifers (a) discharges at land surface or (b) mixes with fresh water in the subsurface, (a) Natural saline
springs have been reported at many localities in the United States, such as in New York by Grain (1969), in
Oklahoma by Ward (1961), in Texas by Richter and Kreitler (1986a,b), or in Arizona by Fuhriman and
Barton (1971). Some of these natural springs contain TDS of up to 300,000 mg/L as a result of solution of
halite in the shallow subsurface (see also chapter 3.2.). Geothermal springs associated with fault zones or
volcanic activity are often mineralized. In other areas, saline-water discharge occurs as nonpoint
contamination, such as along the Colorado River, where more than 50 percent of the average annual 10.7
million tons of salt is contributed by diffuse seepage (U.S. Department of Agriculture, 1975; Atkinson and
others, 1986). (b) Mixing of saline ground water and fresh recharge water occurs in discharge areas of
regional and local aquifer systems, such as in the outcrop areas of Permian formations in north-central
Texas (Core Laboratories, Inc., 1972) and of Mississippian formations in central Missouri (Rg. 10) (Banner
and others, 1989). Fresh-water and salt-water facies are generally separated by a zone of mixing of
variable thickness. The position of this mixing or transition zone may vary in response to changes in either
flow component, which, on a nongeological time scale, is mostly in response to heavy pumpage of fresh
water. Drawdown of the water table or the potentiometric surface in some aquifer systems is so severe that
this interface moves to within the cone of depression of individual wells or well fields, resulting in
contamination of wells after some time of pumping. Mixing between fresh water and saline water can also
occur through any type of boreholes that penetrate and are open to both types of waters. If heads in the
fresh-water unit are higher than in the salt-water unit, fresh-water will drain into the saline formation. If salt-
water heads are higher than fresh-water heads, contamination of an entire aquifer in the vicinity of the well
may occur.
Intermittent pumping of wells can lead to changes in water quality associated with chemical reactions
within the cone of depression. Oxidation of pyrite within the cone during pumping of the well and the
subsequent solution of iron and sutfate during the recovery of water levels to prepumping levels when the
20
-------
West
Eost
OAI7208
Figure 10. Discharge of regional, saline ground water and mixing with local, fresh ground water. Meteoric
recharge in the Front Range of Colorado acquires high salt contents from Permian evaporites under the
High Plains aquifer along its flow path to recharge areas in central Missouri, where it mixes with local
meteoric, fresh ground water (modified from Banner and others, 1989).
21
-------
well is not pumped, often leads to high iron and sulfate concentrations in well waters (Custodio, 1987).
Changes In water quality during pumping may also depend on the depth of the well. For example, in a salt-
water study in Arkansas, Morris and Bush (1986) sampled two adjacent wells of different total well depths
(Fig. 11a). Over time, chloride content increased in the shallow well and decreased in the deep well (Fig.
11 b). The increase in the shallow well suggested inflow of deep saline water, whereas the decrease in
chloride in the deep well suggested inflow of fresh water (Morris and Bush, 1986).
3.1.2. Hydrochemistry of Different Sources of Naturally Occurring Salinity
Nearly all geologic environments may contain naturally-occurring saline water resulting from
geochemical processes within each geologic setting. Natural, highly saline waters typically have chloride
as the dominant anion and sodium as the dominant cation. Exceptions are waters associated with saline
seep and some salt flats, which often have sulfate as major anion. Calcium concentrations are sometimes
very high in deep-basin brines, as IDS concentrations approach several hundreds of thousands mg/L.
These brines are comprised almost entirely of NaCI and CaCl2. Natural brines associated with mineral
deposits often contain unusually high concentrations of ions that are normally not concentrated in most
other brines, such as Cu, Zn, Ni, Co, Mb, Pb, or Ag.
The origin of the chemical composition of brines in sedimentary basins is widely discussed and
sometimes disputed. Depending on the hydrologic environment of each respective basin, the chemical
composition of saline water may differ, as outlined by Kreitler (1989):
Basin waters often are referred to as either connate, meteoric, or a mixture of both. The term
"connate" is defined for this paper as water that was trapped at the time of sedimentation. The term
'meteoric" indicates ground water that originated as continental precipitation. By definition, the age
of connate waters coincides with the age of the host sediments. Basinal waters of meteoric origin are
younger than the host sediments. Though most basins are composed predominantly of marine
sediments, formation waters generally do not resemble sea water in either chemical composition or
concentration. Many basins contain waters with total dissolved solid concentrations significantly
greater than sea water concentrations and as high as 400,000 ppm. Maximum salinities in the Tertiary
section of the Gulf Coast Basin are approximately 130,000 ppm; in the East Texas Basin, 260,000
ppm; in the Palo Duro Basin, 250,000 ppm; in the Illinois Basin, 200,000 ppm; in the Alberta Basin,
300,000 ppm; and in the Michigan Basin, 400,000 ppm (Bassett and Bentley, 1983; Manor, 1983).
Two different types of brines are generally found, a Na-CI brine and a Na-Ca-CI brine, neither having
chemical composition ratios similar to sea water.
In recent years, three mechanism have been used to explain the high ionic concentrations and the
chemical composition of the brines (Manor, 1983) (1) the brines originated as residual bittern brine
solutions left after the precipitation of evaporites; (2) basinal waters have dissolved halite that was
22
-------
(a)
WelM78 Well 177
Land surface
Confining bed
(b) 3500-
3000-
_ 2500-
-1
O
. 2000-
1500-
•1000-
500-
Well 177 depth 130ft
Well 178 depth 85 ft
0500 1000
July 27, 1975
1500 2000 0100 -0600 , 1100
Time (hours) July 28, 1975
QA17209C
Figure 11. Relationship between depth of screen and chloride content in two adjacent wells (from Morris
and Bush, 1986). (a) Pumpage-induced changes in flow directions cause upward migration of saline water
into fresh water in well No. 178 and downward migration of fresh water into saline water in well No. 177,
resulting in (b) water-quality deterioration and improvement, respectively.
23
-------
present as either bedded or domal salt; and (3) basinal waters have been forced through low-
permeability shales (reverse osmosis), leaving a brine on the high-pressured side of the membrane.
At present, there is no consensus about the relative importance of each of these three mechanisms
to the origin of brine.
The theory of salt sieving (reverse osmosis) was first introduced by DeSitter (1947). During
compaction, and in response to flow through tightly packed clays, anions are repelled from negative
charges on clay-mineral surfaces (membrane effect), whereas positively charged ions may pass through
the clay layers, maybe in a stepwise fashion moving from one cation exchange site to the next (Atkinson
and others, 1986). This way, chloride as well as large, multJvalent cations and heavy isotopes tend to be
enriched on the high-pressure side of the membrane. Schwartz (1974) proposed that this process may
also be active at relatively shallow depths in parts of Canada, involving residual effective stress conditions
across till layers. According to Land (1987), opponents of the salt-sieving theory note that (a) sufficient
pressure gradients may not be generated in nature, (b) the chemistry of brines in shale-rich sections
doesn't appear to conform to expected patterns, and (c) a wide variety of formation-water types are
observed in similar settings. Instead of the salt-sieving theory, Land (1987) suggested that the primary
source of salinity are evaporite deposits. Evaporites have always been deposited through geologic time
but have become increasingly rarer in progressively older rocks, suggesting dissolution. Garrels and
MacKenzie (1971) estimated that present evaporites have been dissolved and precipitated 15 times in
the last three billion years, resulting in an average cycle of 200 million years. Dissolution of salt in the
subsurface may be the best explanation for (a) this relatively short cycle, (b) the missing salt in older rock
formations, and (c) the abundance of sedimentary basin brines (Land, 1987). In addition to evaporite
solution, mineral equilibria control the overall composition of brines. Sodium and calcium are by far the
most dominant cations in almost all brines because equilibrium conditions between Na-feldspar (albite)
and K-feldspar result in sodium dominance over potassium, and equilibrium conditions between calcite
and dolomite result in calcium dominance over magnesium (Helgeson, 1972; Land, 1987). To explain
other constituents in brines, Land (1987) proposed two types of rock-water interaction, one occurring in
the sedimentary basin itself and the other occurring in the basement underlying the sedimentary basin.
The massive destruction of detrital feldspar releases significant amounts of calcium, potassium,
strontium, barium, Uthium, etc., to solution. Feldspar equilibria prove that large amounts of potassium
cannot remain in solution until temperatures reach very high values, and potassium is commonly
consumed in the formation of diagenetic illitic day from a smectitic precursor (Schmidt and McDonald,
1979). The strontium and barium contained in brines are known to be vastly more abundant than
predicted simply by the evaporation of seawater, and the 87Sr/86Sr ratio of the brine is commonly
elevated. 87Sr is produced by the decay of 87Rb, an element characteristic of silicate phases,
especially K-feldspar. Thus, significant involvement of silicate phases in determining brine chemistry
is proven. Other isotope ratios, such as 18o/16O (Clayton and others, 1966) and D/H (Yeh, 1980)
24
-------
are shifted from the values expected for surficial seawater-derived brines, providing additional
evidence for extensive rock-water interaction in the subsurface.
It is well known that fluid inclusions in minerals of hydrothermal, metamorphic, or igneous origin are
commonly very saline (Roedder, 1984). Metamorphism is accompanied by devolatilization, during
which carbon dioxide and water are released, presumably to overlying sediments. The formation of
slaty cleavage during low-grade metamorphism apparently requires the loss of large volumes of the
rocks themselves (Buetner and Charles, 1982). The loss of appreciable volumes of insoluble
components, such as SiO2 and AI2O3 means that they must be transported into overlying strata. In
addition to such large-scale material transport, the protons which are bound in alumino-silicates
during weathering at the Earth's surface are progressively replaced by cations during metamorphism.
For example, the H/Na of a solution in equilibrium with albite and kaolinite increases nearly two orders
of magnitude over the temperature interval 25 to 200°C (Helgeson, 1972). In solutions dominated by
1 chloride, weak HCI is thus produced. Acid water lost from metamorphic reactions into the overlying
sediments will be neutralized by minerals like calcite, generating Ca-enriched solutions and CO2- In
fact, such a reaction sequence is one possible reason why the CC>2 content of natural gasses
increases with increasing depth (Lundegard and Land, 1986). The 13C/12C ratio in natural gas also
tends to become enriched in 13C with increasing CC>2 content as more carbon is apparently derived
from inorganic as opposed to organic sources.
Occurrences of brines in igneous rocks may be of multiple origin, as suggested by a long list of
possible sources suggested in the literature. Edmunds and others (1987) list the following modes of
possible origin: (1) marine transgression and subsequent concentration (for example, Frape and Fritz,
1982), (2) migration of sedimentary basin brines (for example, Fritz and Frape, 1982), (3) residual
hydrothermal fluids (for example, Alderton and Sheppard, 1977), (4) dissolution of grain-boundary salts
(for example, Grigsby and others, 1983), (5) silicate mineral hydrolysis and related water-rock interactions
(for example, Edmunds and others, 1984, 1985; Frape and others, 1984), (6) breakdown of fluid
inclusions in quartz and other minerals (for example, Nordstrom, 1983), and (7) radiolytic decomposition of
water during a-series decay (Vovk, 1981). Some of these processes are believed to have caused brine
concentrations of up to 550,000 mg/L TDS (Vovk, 1981). Edmunds and others (1987) added an
extensive list of chemical reactions to the above list to explain the origin of saline ground water in a granite
of Cornwall, United Kingdom. Eliminating ancient sea water as a potential source (518O arid SD indicated a
local, meteoric origin), chemical reactions along the flow paths down to approximately 4,000 ft and at
temperatures reaching 131°F (55aC) can explain these saline waters having TDS of up to 19,300 mg/L.
Some results of these chemical reactions are: (a) acid hydrolysis of plagioclase and biotite as the prinicipal
origin of ground-water salinity, (b) enriched Ca/Na ratios by selective reaction of the more calcic centers of
zoned plagioclase, (c) high lithium concentrations related to biotite reaction, and (d) high Cl
25
-------
concentrations as a result of hydroxyl exchange for Cl in the biotite interlayer (Edmunds and others,
1987).
Evaporation from a shallow water table (within three or four feet of land surface) can lead to high salt
concentrations in soils. This is known to have occurred in the San Joaquin Valley of California. There,
ground-water salinity in Coast Range alluvium can be correlated with high concentrations of selenium,
molybdenum, vanadium, and boron (Deverel and Gallanthine, 1989). Evaporation is the major process
that accounts for high salinities in some closed basins in the western half of the United States. Weathering
processes and selective mineral precipitations modify the chemical composition of closed-basin waters
further. Major inorganic reactions in such settings are silicate hydrolysis, uptake of CC*2 from the
atmosphere and/or of sulfate from oxidation of sulfides, and precipitation of alkaline earth compounds
(Jones, 1965). Concentration ratios of major chemical constituents are much less uniform in closed-basin
brines than in most deep-basin brines, the latter being nearly exclusively Na-Ca-CI dominated. Closed-
basin brines may be (1) chloride dominated, such as those in the Bonneville Basin, (2) carbonate
dominated, such as those in Alkali Valley, Oregon, or (3) sulfate dominated, such as those of the Mojave
Desert (Jones, 1965). These differences are due to differences in inflow characteristics and precipitation
reactions although, in most instances, increased salinization (evaporation) is associated with a trend
toward Cl dominance until halite precipitation is reached. Closed-basin brines start out by simply dissolving
readily soluble mineral compounds, such as halite. Leaching of absorbed ions or of trapped interstitial
fluids may be another process that provides dissolved minerals to the water. Silicate hydrolysis and
subsequent evaporative concentration is the source of some of the high carbonate contents. Other
processes that subsequently change the composition of the waters are: mixing with other water types,
CO2 addition by organic activity at lake bottoms or irrterstitially in lacustrine sediments, anaerobic decay of
organic matter, and loss of sulfate (Jones, 1965). As the waters increase in concentration due to
evaporation, precipitation reactions induce major changes to concentrations of individual ions. Calcium
carbonate precipitates first, followed by gypsum, and finally by alkali salts. These processes can happen
simultaneously, as described by Hunt (1960) and Jones (1965) for the Death Valley salt pan. There,
carbonates precipitate in the outermost areas, sulfates in the intermediate zones, and chlorides at the
center.
Ground-water evolution in closed basins may be similar to any other evolution along the flow path,
from a tow-TDS, Na-Ca-HCOs recharge water to a high TDS, Na-CI water, as a result of reactions such as
calcite dissolution and precipitation, cation exchange on clay minerals, and evaporation in discharge areas
near the center of the basins. In discharge areas, mixing between this recharge water and brine may
induce radical changes in the water type. Boyd and Kreitler (1986) demonstrated the geochemical
evolution of a Ca-SO4 recharge water to a Na-CI brine in a gypsum playa in West Texas (Rg. 12). Brine
evolution included precipitation of calcite, gypsum, and dolomite.
26
-------
Waters from
• Peripheral bedrock
a Alluvial fans and outer
basin fill
x Salt flats
Cation diagram
Na-Mg- Co
Anion , diagram
CI-S04-HC03
BO 100
OAI72IO
Figure 12. Geochemical composition of ground water in a gypsum playa, West Texas. Ground water
evolves from a Ca-SC>4 type in recharge areas to a Na-CI type in discharge areas (salt flat) (from Boyd and
Kreitler, 1986).
27
-------
In many areas of the United States, hydrothermal waters are found close to or at land surface. In some
cases, these waters are high in mineral content as a result of increased mineral reactions at elevated
temperatures and at large depths from which these waters originate. Stable isotope compositions in most
geothermal systems have been found to reflect recharge of local meteoric waters, modified by one or
more physical or chemical processes (Truesdell, 1976; in Welch and Preissler, 1986). One of these
processes may be evaporation at land surface prior to infiltration, as reported by Welch and Preissler
(1986) for Bradys Hot Springs, Nevada.
3.1.3. Examples of Geochemical Studies of Natural Saline Ground Water
In a case of salt-water intrusion from a single potential source of salt water, Sproul and others (1972)
used chloride concentrations and temperature values to identify deep, artesian salt water as the source of
intrusion in parts of Lee County, Florida. Upward flow in that area is possible through either abandoned,
open boreholes or along natural fault and fracture systems. Similarly, Wait and Gregg (1973) identified
inflow of saline water in the Glynn County area of Georgia by an increase in chloride concentration from
average background levels of 23 mg/L to several hundreds of mg/L. There, heavy pumpage from shallow
fresh-water zones was the cause of intrusion. Sowayan and Allayla (1989) identified progressive
concentration (evaporation) as the source of saline water in a part of Saudi Arabia by plotting sodium
concentrations versus chloride concentrations. This was indicated by a slope of near unity and the general
absence of other possible salt-water sources, such as brines, evaporite deposits, or geothermal springs.
In parts of West Texas, natural and anthropogenic salinization sources are common. Some of the
anthropogenic sources are associated with oil- and gas-field operations or agricultural techniques while
natural salinization is associated with shallow evaporite deposits, discharge of saline formation water, and
high evapotranspiration rates. Richter and Kreitler (1986a,b) and Richter and others (1990) used major
cations and anions, the minor constituents Br and I, and the isotopes oxygen-18, deuterium, and
sulfur-34 to distinguish between these sources. Oxygen-18 and deuterium concentrations separated
local recharge from nonlocal recharge (Fig. 13) and Br/CI and Na/CI ratios separated natural halite solution
from deep-basin discharge. These two sources also were differentiated clearly by other constituent ratios,
such as I/CI, Mg/CI, K/CI, Ca/CI, and (Ca+Mg)/SO4 (see also chapter 3.2). The same constituents have
been used in Oklahoma and Kansas to differentiate natural halite solution from oil-field brines (for
example, Ward, 1961, Leonard, 1964, Whittemore and Pollock, 1979, Gogel, 1981, and Whittemore,
1984, 1988). The concept of local and nonlocal recharge reflected in oxygen and hydrogen stable
isotopes was also used by Banner and others (1989). Isotoptc concentrations in saline waters of central
Missouri are much lighter than typical for Missouri rain water, which suggested a source further to the west.
According to Banner and others (1989), the Front Range of Colorado is the most likely area of recharge for
these waters. This recharge water picks up high salt contents from Permian salts (indicated by Br/CI ratios)
28
-------
20-
0:
-20-
o
2
in
°-4d-
-60-
-80
Cl = 98,000 mg/L
r**V +
4- "*"
mg/L
•—Cl = 2,700 mg/L
A
ROLLING
PLAINS
EXPLANATION
£> Shallow ground water/High Plains &
o Salt springs/ Rolling Plains
• Test hole/ North Croton Creek
+ Deep-basin brines /Palo Duro Basin
Rolling Plains
NEW MEXICO
-10
-8
-6
-4 -2
18 0(SMOW)%o
QA287I
Figure 13. Separation of local recharge water from nonlocal .ground water using the stable isotope ratios
518O and 5D. Locally recharged ground water plots along the meteoric water line at the composition of
local precipitation (open dots and open triangles), whereas nonlocally derived ground water typically is
either enriched or depleted in isotopes compared to local precipitation (solid dots) (from Richter and
Kreitler, 19865).
29
-------
underlying Kansas and finally mixes with fresh water near discharge areas in central Missouri. Ancient sea
water or intensive interaction of ground water with aquifer material were precluded by these authors as
possible sources of salinity because isotopic compositions of 87Sr/86Sr and 144Nd (Neodymium) in
ground water were significantly different from those in Paleozoic sea water and in local host rock. Bivariate
plots of Br/CI were also used by Morris and Bush (1986) in combination with plots of I/CI and B/CI to
identify deep formation water as the source of salt-water contamination in parts of Arkansas (Rg. 14).
Lloyd and Heathcote (1985) demonstrated the use of strontium and iodine concentrations for
differentiating saline ground water in the Lima Basin alluvial aquifer, Peru. Saline water derived from sea-
water intrusion plotted on a dilution line between ocean water and alluvial water (Fig. 15). Waters that plot
off this dilution line are associated with inflow from saline aquifers. One of these is characterized by Sr
enrichment and the other is associated with iodine enrichment.
Saline playas and gypsum flats are natural discharge areas of closed basins in many parts of the
western United States (Boyd and Kreitler, 1986). They also occur elsewhere, such as in the Murray Basin
of southeastern Australia. Saline ground water in that area, ranging from 20,000 mg/L to 50,000 mg/L
IDS, appears to be relict sea water with changes in concentrations reflecting local recharge/discharge
conditions (Macumber, 1984). Concentrations increase due to evaporation at three major discharge
points (Rg. 16a). Under these lake basins, ground water salinity also increases as the heavy brines
recharge from the lakes. The similarity to sea water is preserved in these waters with little changes in
concentration ratios. Brine under these lakes will continue to accumulate and spread beneath the less
saline water in these basins until a massive inflow of fresh water changes the concentrations of the brine
lakes (Macumber, 1984). Similarly, dissolved solids concentrations in saline springs and lakes on the
Southern High Plains of Texas and New Mexico are the result of evaporation of shallow ground water
rather than discharge of saline water from deep brine aquifers, as indicated by constituent ratios of Cl/Br,
Na/K, and CI/SCX* being consistently smaller in evaporated waters than in deep-basin brines (Wood and
Jones, 1990). Salinities increase in lakes and springs as surface water evaporates, and percolation water is
recycled from the lakes to the springs (Fig. 16b). Locally, ground water in the Ogallala Formation of the
Southern High Plains is affected by these evaporated waters, but also by oil-field brines. Salinization by
these two sources can be differentiated through the use of salinity diagrams (Fig. 17), as evaporative
waters are characterized by Na-SC>4 waters in contrast to high-TOS, Na-CI fades typical for oil-field-brine
affected water (Nativ, 1988).
The distribution of natural salinity in shallow ground water reflects the geomorphology and hydrology
of the* system, such as in alluvial fans in the western San Joaquin Valley of California (Deverel and
Gallanthine, 1989). Lowest salinities were found in the upper and middle fan areas and highest salinities
(greater 5,000 mg/L TOS) in the lower fan areas and at the margins of the fans where fine-grained soils are
poorly drained. Historically, soils associated with ephemeral streams received less water and less flushing
than larger intermittent streams and, therefore, were more saline. Leaching of soil salinity by irrigation
30
-------
1000 =
100.
o> =
E, •
m 1i
0.1s
0.01-
A Alluvial aquifer a
« Sparta aquifer a
a Nacatoch aquifer
D
I I I Hill I I I Illll I I I HUH I I I Illll \ I 111 HI
1 10 100 1000 10,000 100,000
Cl (mg/L)
1000:
100
0.1
0.01.
i A Alluvial aquifer
i > Sparta aquifer
i a Nacatoch aquifer o
i D
AA"
A 4 A A
i 11uHI i i mini r i iiiim i i nun rTTirnn
1 10 100 1000 10,000 100,000
Cl (mg/L)
—
—
10,000 =
-
5"
g> 1000 =
E =
CD -
"" "
100-
—
z
1 A
1 X
1 D
ill
1 1
£'"
1
" 1
1
1
A ' A
A ' «AS^
1 A
A A &
I
1 1 1 mill
Alluvial aquifer
Sparta aquifer g
Nacatoch aquifer Q
D
A,
x ^
If-
A
A
i 1 1 nun i i i nun i 1 1 i mi
10 100 1000 10,000 100,000
Cl (mg/L) QA17212C
Figure 14. Use of bromide, iodide, boron, and chloride concentrations in bivariate plots for identification of
salt-water sources. Mixing between fresh water and salt water is suggested by linear trends between
potential endmembers of mixing (from Morris and Bush, 1986).
31
-------
9-
8-
7-
6-
5* 5-
t
3-
2-
1 -
Pacific Ocean water
i
0.5
I
10
I
20
I
50
I (mg/L)
100
QA17211C
Figure 15. Separation of different ground-water origins using the strontium-iodide relationship in ground
water. Data shown are from the Lima Basin alluvial aquifer, Peru: Group I water is associated with alluvium;
Group II water enters the alluvium from Jurassic sediments; Group III water enters the alluvium from
granodiorites; Group IV water results from sea-water intrusion (from Lloyd and Heathcote, 1985).
32
-------
(a)
Groundwoter
divide
:-:Cloy
(b)
.Lond surface
•——.
Water table
Ogallolo S Blockwater Draw
EXPLANATION
V Water table
irir Cretaceous ir^rrir^
J-ir-I ground - water r£Hrf
_-i^-r-_- flow ---r-_j^:
Lacustrine and
evaporite sediments
Saline water -
( Not to scale)
Triossic
OAI72I3
Rgure 16. Ground-water flow directions (arrows) and salinities (in percent) in parts of (a) the Murray Basin,
southwest Australia, (from Macumber, 1984) and (b) at saline lakes, Southern High Plains of Texas (from
Wood and Jones, 1990). Evaporation at major discharge sites and mixing causes IDS increases along the
flow path.
33
-------
(a)
500n
100-
CT
a>
E
c
a>
o
o
O
50-
10-
5-
Saline-lake contamination of
Ogallala water
(b)
Ca * Mg
Na
Cl
S04"2 HC03"
500-.
Oil-field brine contamination
of Ogallala water
100-
Ca+2 Mgt2
S04"2 HC03'
EXPLANATION
EXPLANATION
•Ogallala (well 24-06-201)
•Bull Lake (Reeves, 1970)
•Illusion Lake (Reeves, 1970)
Mean values of water in the
Ogallola,Southern High Plains
Ogallala (well 27-19-411)
• Brine (Clear Fork Formation,
Robertson Field, Burnitt
and others, 1963)
• Mean values of water in the
Ogallala,Southern High Plains
QA 4960
Figure 17. Salinity diagrams of Ogallala water contaminated by (a) saline lake water and (b) by oil-field
brines (from Nativ, 1988).
34
-------
water was the primary process that initially formed shallow, saline ground water. Subsequently,
evaporative concentration caused further increases in salt content. This evaporative trend is illustrated in a
plot of 818O and 8D, which shows that samples collected at the lower fan areas are the most enriched
ones (Fig. 18a). Waters with the highest salinities in the area are of the Na-SC>4 type (Fig. 18C), with sulfate
concentrations controlled by gypsum precipitation, as suggested by saturation indices (Fig. 18b).
Geothermal spring water in Churchill County, Nevada, is of local, meteoric origin as indicated by SO
and 50 values (Rg. 19) (Welch and Preissler, 1986). Salinization caused by evaporation is suggested by a
shift toward oxygen and hydrogen isotope enrichment. This evaporation trend was also indicated by a
uniform Br/CI ratio. Of the minor constituents Ba, B, Br, F, Pb, Li, Mn, and Sr tested, bromide is largely
controlled by evaporative concentration, whereas barium, fluoride, lead, and manganese are controlled by
mineral phases such as barite, fluorite, cerussite, and rhodochrosite, respectively. Boron, lithium, and
strontium may be indicators of the presence of hydrothermal waters, as concentrations are higher in the
higher temperature samples (Welch and Preissler, 1986).
Saline ground water trapped in portions of aquifers since deposition or since subsequent salt-water
intrusion generally does not contain any tritium because of the tritium isotope's short half-life of only
12.4 years. Anthropogenic sources of salt, such as road salt, in contrast, may contain measurable
amounts of tritium, as these salts are dissolved and flushed into aquifers by modern precipitation. This
relationship was used by Snow and others (1990) to distinguish trapped sea water from road-salt
contamination in the coastal wells of Maine.
3.1.4. Significant Parameters
Most of the salinity sources described in this report occur naturally at some place or another, where
they mix with fresh ground water. In other cases, mixing of naturally saline water with fresh water is initiated
or facilitated by anthropogenic activities, such as heavy pumpage of fresh water, drilling through fresh-
water and salt-water bearing zones, or disposal of produced water. In most instances, chemical
characteristics will not differ significantly between natural mixing of fresh and salt water and artificial mixing
of the same salt water with fresh water. Therefore, significant parameters for identification of natural
salinization are the same as those for any individual source discussed in the following sections.
Salinization is generally indicated by an increase in chloride concentration. If this increase is
substantial and occurs suddenly and is localized, a nonnatural mechanism and source should be
suspected. However, if the change is subtle and of regional scale, a natural mechanism or source may be
responsible.
The stable isotopes Oxygen-18 and deuterium are generally useful to distinguish between local
precipitation water and water that is derived from a nonlocal source and identify evaporation of local
recharge water. Molar ratios of major chemical constituents, such as Na/CI, Ca/CI, and Mg/CI, can be used
35
-------
(a)
(b)
o -
. —
C. .50 -
o
CO
-100-
Sample altitude above >X
sea level (m) x
A 50-58 ^9^
a 59 - 70 -fx
• 71 - 100 *X^
\^
V3£|H» A |l* *-
-^- ^*"*~ Evaporative
s-S-y^A trend line
X*
1 1
O
CO
?2
O~
in
T3 4 -
CD
>
O
I/)
-------
-100-
-110-
Q
CO
Possible local
meteoric water
-120-
-130-
518 O (%o)
GA17215C
Figure 19. Stable-isotope composition of ground water affected by evaporation, causing enrichment in
oxygen-18 and deuterium as salinity increases. Chloride ranges: A <50 mg/L; 910
-------
to differentiate an evaporation trend (1:1 slope) from a mixing trend (typically not a 1:1 slope). Mixing
trends can be evaluated best using the most conservative constituents dissolved in ground water, that is,
chloride and bromide. These constituents are often useful to not only identify the mixing source of salinity
(see also chapters 3.2 through 3.7), but also to estimate the mixing ratio (see also chapter 6.1).
Mixing of fresh water with naturally saline ground water or evolution of ground water toward higher
salinities, as opposed to mixing with road-salt solutions, mixing with brine along boreholes, or disposal of
produced oil-field brine, can be expected to be a relatively slow process during which the water has time to
react extensively with the aquifer matrix. Therefore, saline ground water in its natural environment will
reflect conditions of chemical equilibrium more closely than artificially induced mixtures of fresh water and
saline water. This may be used to distinguish natural mixing or evolution from induced mixing.
3.1.5. State-by-State Summary
Natural saline ground water underlies nearly every state at some variable depths (Fig. 1). Where
fresh-water aquifers are plentiful or where ground-water usage is low, little may be known about the base
of fresh water. In other areas, in contrast, lack of usable water and, with it, heavy drilling in an attempt to find
water has helped to define the interface between fresh and saline water remarkably well. The following
sections taken entirely from miscellaneous published sources deal with the occurrence of saline ground
water on a state-by-state basis.
Alabama: Saline ground water is found at shallow depths in the north, north-central, and
southwestemmost parts of the state. TDS exceeding 3,000 ppm occur in southern Lowndes County,
southwest of Montgomery, and in western Clarke and northeastern Washington Counties, including a well
reported to contain between 10,000 and 35,000 ppm in southeastern Choctaw County (Feth and others,
1965). Depth to saline ground water exceeds 1,000 ft in the remainder of the state, including the Gulf
Coast (Feth and others, 1965). Natural salinization in southwest Alabama is due to brine migration along
faults, affecting all major aquifers in that area (Slack and Planert, 1988). In Green County, wells in the
Gordo and McShan Formations furnish water ranging from 4 to 3,700 mg/L Cl and 3.8 to 2,560 mg/L,
respectively. Many of these wells are flowing at land surface.
Arizona: Shallow saline ground water (TDS>1,000 ppm) is found in the northeastern and
southwestern part of the state, including TDS greater than 35,000 ppm at depth of 500-1,000 ft in the
Puerco River drainage east of Holbrook. TDS in ground water between 1,000 and 10,000 ppm occur
along the Gila River, and between 1,000 and 3,000 ppm along the Colorado River of southwestern
Arizona at depths less than 500 ft below land surface (Feth and others, 1965).
Water quality in many streams deteriorates in downstream direction. Often, mineral springs
contribute large amounts of dissolved solids to the streams, such as those near the mouths of Chevelon
and Clear Creeks near Winstow, where TDS concentrations of up to 4,000 ppm are recorded in the Little
38
-------
Colorado River. Concentration greater than 6,000 ppm have been measured in the Gila River, in part due
to inflow of saline tributary water, such as those from San Francisco River, which in turn receives saline
water of up to 9,000 ppm from Cjifton Hot Springs (Krieger and others, 1957). Over 200 tons of dissolved
solids are discharged daily into the Salt River by three major springs that originate in the Mpgpllon Rim in
central and east-central Arizona (Fuhriman and Barton, 1971). At Phoenix, the Salt River carries
approximately 500,000 tons of salt per year (Skibitzke and others, 1961).
Most of the ground-water basins in the Salt River Valley are overdeveloped and yield water which is
mineralized, with TDS exceeding 3,000 mg/L in many wells. Highest salt concentrations are often found in
the center of heaviest ground-water usage; maximum values exceed 7,000 mg/L The deeper aquifers
usually are of better water quality than the shallow aquifers. Improperly abandoned deep wells can lead to
pollution of these deep fresh waters by shallow saline water depending on the head difference between
them. Mineralization is high in the San Pedro, Willcox, and Safford subbasins within the Upper Gila River
and adjoining Mexican drainage. TDS as high as 5,000 ppm have been reported from wells in the Palomas
Plains. Highly mineralized ground water in the Colorado River Basin of southwestern Arizona can be found
in the Ranegrass Plain (TDS up to 4,000 ppm) and in some wells in the Yuma area (TDS up to 5,000 ppm)
(Fuhriman and Barton, 1971). In the Safford Valley, ground-water salinity has increased from 1940
through 1972 as a direct result of predominantly pumping-induced aquifer leakage; other factors
contributing to the salinity are natural recharge to the water table by saline water, vertical flow of artesian
ground water through saline lacustrine beds, and agricultural recharge waters (Muller, 1974).
Natural mineralization in areas of low precipitation and naturally restricted drainage is the most
common ground-water quality problem in Arizona. Leaching of highly saline soils and rocks accounts for
most of the salinity problems (Smith, 1989). Of the total irrigable area of 1,565,000 acres, approximately
25 percent (398,830 acres) were considered saline or alkaline in 1960 (Fuhriman and Barton, 1971).
Arkansas: Saline ground water (TDS>1,000 ppm) underlies Arkansas at shallow depth (<500 ft
below land surface) within a narrow zone that extends from the northeastern to the southwestern part of
the state. To the east of this zone, saline ground water is restricted to depths mostly greater than 1,000 ft
below land surface. Another shallow zone of saline water is located in westernmost Arkansas including all
of Sebastian and Crawford Counties and parts of Logan and Franklin Counties (Feth and others, 1965).
Lateral movement of saline water into fresh water by updip migration resulting from pumping was
detected in areas of eastern and in the Sparta Aquifer in southern Arkansas (Newport, 1977; U.S.
Geological Survey, 1984). In the Brinkley area, approximately 56 mi2 of the alluvial aquifer have been
contaminated by salt water from underlying saline formations (Morris and Bush, 1986). Artesian upward
movement of salt water in this area is enhanced by irrigation and water-supply pumping. A chloride
concentration greater than 50 mg/L was considered indicative of salt-water contamination and was used
for mapping of contaminated areas by Morris and Bush (1986).
39
-------
California: Isolated areas of shallow saline ground water (TDS>1,000 ppm at depth less than 500 ft
below land surface) occur in many parts of the state, including major areas such as the San Joaquin Valley,
the Colorado River Basin, and most of Imperial County. TDS in excess of 10,000 pprn are reported from
closed basin lakes in the southeast (Feth and others, 1965). Saline lakes in California include from north to
south (located mostly along the Catifomia-Nevada/Arizona state lines): Surprise Valley, Honey Lake,
Pyramid Lake, Lake Tahoe, Mono Lake, Saline Valley, Owens Lake, Seartes Lake, Soda Lake, Bn'stol Dry
Lake, Cadiz Lake, Danby Lake, and Satton Sea (Hardie and Eugster, 1970).
Approximately 33 percent (3,745,000 acres) of California's total irrigable land was considered saline
or alkaline in 1960. Maximum TDS concentrations in mg/L at selected locations throughout California as
reported through 1965 were: Ukiah Valley, 1,280; Santa Rosa, 560; East Bay (San Francisco), 4,100;
South Bay (San Francisco), 1,750; Livermore Valley, 4,700; Petaluma Valley, 19,760 (sea-water
intrusion); Napa Valley, 1,840; Sonoma Valley, 660; Suisum-Fairfietd Valley, 2,560; Pajaro Valley, 1,310
(TDS>5,000 in areas of sea-water intrusion); Gilroy-Hollister Basin, 1,480; San Luis Obispo Hydrologic
unit, 3,024; Carrizo hydrotogic unit, 10,460; Santa Maria-Cuyama hydrologic unit, 5,088; San Antonio
hydrotogto unit, 4,070; Salinas Valley, 3,134; Carmel Valley, 729; Paso Robles Basin, 3,280; Santa Ynez
hydrologic unit, 21,800; Santa Barbara hydrologic unit, 2,487; Mission Basin, 13,930; San Dieguito
Valley, 27,402; Tfa Juana Valley, 4,680; Oxnard Plain area, 33,180; West Coast Basin, 41,397; East
Coastal Plain, 41,800; Main San Gabriel Valley, 1,140; Chino Basin, 1,417; Perns Valley, 11,620; San
Joaquin County, 3,840; San Joaquin Basin, 6,400 (connate water?); and Tulare Basin, 6,450 (Fuhriman
and Barton, 1971).
Mineralized hot springs contribute to the salinity in the North Lahontan Basin in the Bridgeport area,
whereas intrusion of brackish water in gravel-packed wells open to zones of variable water quality may
pose a problem in some wells in the San Joaquin Valley (Fuhriman and Barton, 1971). Dissolved solids in
shallow ground water from alluvial deposits along the San Joaquin River in Fresno County vary between
less than 1,000 mg/L to greater than 15,000 mg/L. Natural salinization of soils occurred by evaporation of
ground water at times when the water table was close to land surface. These salts were subsequently
dissolved by Irrigation waters, causing highest concentrations of TDS in those areas that have been
Irrigated only in the last 40 years (Deverel and Gallanthine, 1988).
To compensate for extensive overdraft of ground water in Orange County during the 1940's, large
quantities of Colorado River water were imported and recharged along the Santa Ana River channel. This
resulted in salinization of a substantial portion of ground water in the county due to the high TDS content
of the imported water (U.S. Environmental Protection Agency, 1973).
Saline water intrusion in southern Alameda County is due to the following mechanism: direct
movement of bay waters through natural "windows" (that is, high-transmissive zone in an otherwise tow-
transmisslve clay layer), spilling of degraded ground waters through windows in day layer, slow percolation
40
-------
of salt water through reservoir roof (clay layer), and spilling or cascading of saline waters or degraded
ground water through deep wells (U.S. Environmental Protection Agency, 1973).
Colorado: Saline ground water (TDS>1,000 ppm) occurs at shallow depth (<500 ft below land
surface) in the western third and in southeastern Colorado. Ground waters containing TDS greater than
3,000 ppm underlie the state at depths of >1,000 ft in the eastern and northeastern parts (Feth and
others, 1965).
Approximately 35 percent (982,000 acres) of Colorado's irrigable land was considered saline or
alkaline in 1960 (Fuhriman and Barton, 1971).
Approximately 37 percent of the salt load in the upper part of the Lower Colorado River Basin is
contributed by irrigation-return flow. The other 63 percent is contributed by interaction of water with
naturally saline soils and aquifer material overlying a marine shale that contains numerous lenses of salt.
Within an area comprising 1,876 ha, approximately 51,000 metric tons of salt are added to the river system
by these two sources (Skogerboe andi Walker, 1973).
Delaware: Saline ground water (TDS> 1,000 ppm) underlies Delaware at depth >1,000 ft except for
the coastal strip in the southeast, where saline ground water may be found at depth <500 ft below land
surface (Feth and others, 1965). The inland limit of saline ground water is located along the coastline in
Cretaceous rocks, falls along the western and northern borders of Sussex County in Tertiary rocks, and
stretches all the way north to the city of Dover in Pleistocene formations (Miller and others, 1974).
Brackish water occurs locally in shallow water-table aquifers in coastal areas. Some salt water is
contained in the Chesapeake Group (Miocene) in Sussex County, and the interface between fresh and
saline water in the Piney Point Formation (Eocene) is located just north of Milford, extending to the
northeast across the state. Brackish water also exists in the Magothy and Potomac Formations
(Cretaceous) a few miles south of Middletown. Heavy pumpage near these brackish-water zones may
cause water-quality deterioration due to inflow of saline water toward the wells. High chloride
concentrations between 6,000 mg/L and 17,000 mg/L were measured in core samples obtained from
wells along the Atlantic Coast of southeastern Delaware. The depths of wells from which core samples
were obtained ranged from 5 ft to 60 ft below land surface (Woodruff, 1969).
Florida: A 30-mile-wide stretch of shallow saline ground water (TDS>1,000 ppm at depth <500 ft
below land surface) extends from St. Augustine in the north to Vero Beach in the south along the Atlantic
Coast of Florida. An isolated occurrence of shallow saline ground water is also reported for westernmost
Florida. In contrast, saline ground water generally is found at depths between 500 and 1,000 ft in
southern Florida and deeper than 1,000 ft in the remainder of the state. In southern Florida, large
quantities of moderately saline water with TDS less than 5,000 mg/L can be found in the Avon Park
Limestone Formation at depths greater than 1,200 ft below land surface. Less saline water at shallower
depths, between 300 and 1,100 ft below land surface, occurs in the Hawthorn, Tampa, and Suwannee
Formations (Meyer, 1971). Individual wells along the western Gulf Coast in Sarasota and Manatee
41
-------
Counties as well as wells along the southeastern shoreline may exhibit salinities in excess of 3,000 mg/L
TDS (Feth and others, 1965).
The Florida State Geological Survey collected information on 1,800 uncontrolled flowing wells
between 1955 and 1959. Chloride concentration in approximately 50 percent of the wells exceeded the
drinking water limit of 250 mg/L. At present, there are an estimated 2,000 to 3,000 wells that flow poor-
quality water in southwest Florida alone. At La Belle, upward leakage through five deep artesian wells that
were not in use and lateral movement through a subsurface water-distribution system were the probable
sources of contamination of the shallow aquifer over a broad area (Miller and others, 1977).
The chemical composition of ground water in the Floridan aquifer of coastal southwest Florida
changes from a fresh Ca-Mg-HCOs type water in the upper, upgradient areas to a similar fresh Ca-Mg-SO4
type water In the deeper, more permeable, dotomitic zone, and finally to a highly saline Na-Mg-CI type
water in the downgradient areas. Water-quality degradation can occur in areas of high pumpage, such as
the Verna well field in northern Sarasota County (Steinkampf, 1982).
Vertical migration of saline water along a deep, abandoned irrigation well at Highland Estates caused
chloride increases in the shallow aquifer and in some water wells from background levels of 20 mg/L to
contamination levels of 590 mg/L (Boggess, 1973; Miller and others, 1977).
Georgia: Saline ground water (TDS>1,000 ppm) underlies the southern half of the state at depths
exceeding 1,000 ft (Feth and others, 1965).
Locally, salinization occurs due to flowing, abandoned wells (Clarke and McConnell, 1988). Intrusion
of saline water at Brunswick is caused by upward migration of brines through faults and fractures in
response to the daily withdrawal of 105 million gallons by industry and municipalities. In one test well,
chloride concentration increased from 4,400 to 7,000 mg/L during the past 10 years (U.S. Geological
Survey, 1984).
Idaho: Isolated occurrences of shallow saline ground water (TDS >1,000 ppm at depth <500 ft) are
reported from wells in the Snake River Basin of southern Idaho and in the southern part of Oneida County
in southeast Idaho (Feth and others, 1965).
Thermal water occurs at many locations at depths greater than 400 ft below land surface. Maximum
chloride concentration in nonthermal valley-fill aquifer is 3,900 mg/L (Partiman, 1988).
Illinois: Saline ground water of TDS>1,000 ppm underlies central and southern Illinois at depths
less than 500 ft below land surface (Feth and others, 1965). Saline water from the Mt. Simon aquifer may
leak into shallow fresh-water aquifers due to overpumping (Atkinson and others, 1986).
Indiana: The entire state of Indiana is underlain by saline ground water (TDS >1,000 ppm) at
depths less than 500 ft below land surface. Individual wells have been reported supplying water
containing between 3,000 and 10,000 ppm of TDS (Feth and others, 1965).
Ample rainfall in the state prevents overpumping of shallow aquifers which reduces the potential of
salt-water intrusion. However, a leaky aquifer in Knox County has caused salinization of fresh ground water
42
-------
(Atkinson and others, 1986). In the Mt. Vernon-West Franklin area, upward flow of saline water through
fault zones into fresh water was caused by overpumping of the fresh-water aquifer (Newport, 1977). Near
Vincennes, saline water has moved up through bedrock fractures into the bottom of the glacial outwash
aquifer near the municipal well field, endangering future water supplies if pumpage increases more than
fourfold (Shedtock, 1978).
Iowa: Except for the northeastern part, all of Iowa is underlain by saline (IDS between 1,000 to
3,000 ppm) to very saline (IDS between 10,000 to 35,000 ppm) ground water at depths less than 500 ft
below land surface (Feth and others, 1965). Despite this shallow occurrence of saline ground water, Iowa
does not currently experience salt-water intrusion, but the potential of contamination from shallow saline
water in response to intensive pumping exists (Newport, 1977; Atkinson and others, 1986). In glacial drift
aquifers underlying Wayne County; TOS concentrations range up to 3,600 mg/L in wells as little as 100 ft
deep (Cagle, 1969). „ ,
Kansas: The eastern half of the state is underlain by ground water containing IDS greater than
1,000 ppm at depths less than 500 ft below land surface. Concentrations in excess of 10,000 ppm occur
locally, and are widespread in central and south-central Kansas (Feth and others; 1965). Natural
salinization of Kansas' surface water is derived principally from salt springs, salt marshes, and direct contact
with saliferous geologic formations. Major sources of saline water occur in a 150-mile-wide band that
stretches across the central part of the state from the southern to the northern border (Krieger and others,
1957). Salt-water intrusion has been reported in Sedgwick, Reno, Harvey, McPherson, Saline, Dickinson,
Seward, and Meade Counties. Salinization problems are especially serious in outcrop areas of the
Wellington Formation along the Solomon and Smoky Hill Rivers (Atkinson and others, 1986).
Natural intrusion of mineralized water is occurring in central Kansas from Permian Redbeds, the
Wellington Formation, and the Dakota and/or Kiowa Formations. A prominent example of chloride and
sulfate intrusion from the Dakota and or Kiowa Formations is along the Smoky Hill and Saline Rivers in
Russell County, The Wellington Formation contains abundant beds of gypsum and a thick bed of halite
which pose problems from Ottawa-Dickinson Counties to the south. The salt bed (Hutchinson Salt
Member, up to 350 ft thick) is dissolved easily by circulating ground water. The most damaging effects
from salt dissolution are along the Smoky Hill and Solomon Rivers in eastern Saline and western Dickinson
Counties and the Belle Plain area in Summer County on the Ninnescah and Arkansas Rivers and the Slate
and Salt Creeks. Certain zones of the Permian Red Beds contain salt and gypsum which is easily
dissolved by circulating ground water. Surface streams and ground water contaminated by this source are
found along the lower reaches of the Rattlesnake Creek in Stafford and Reno Counties and along the
South Fork Ninnescah River in eastern Pratt and western Kingman Counties. Ten minor areas of salt-water
intrusion occur along (1) the Solomon River in Mitchell and Cloud Counties (for example, Waconda Spring
west of the city of Betoit), (2) Salt Creek in Mitchell and Lincoln Counties, a tributary of the Solomon River,
(3) Buffalo Creek in Jewell and Cloud Counties, (4) Salt Creek and Marsh Creek in Republican County,
43
-------
(5) the Cottonwood River in Marion County, (6) the Medicine Lodge River in Barber County, (7) the Salt
Fork Arkansas River in Comanche and Barber Counties, (8) the Cimarron River in Meade County, (9) the
Crooked Creek in Meade County, and (10) Cow Creek in Reno and Rice Counties (Hargadine and others,
1979).
In the area south and west of Salina, fresh water moves down through collapsed and brecciated
shales to recharge the Wellington aquifer, dissolving halite within the Hutchinson Salt Member. The
resulting brine then moves northward beneath the Smoky Hill River Valley to the Saline River Valley.
Upward discharge of brine occurs through collapse structures in an area from about three miles east of
Salina to just east of Solomon, where the potentiometric surface of the Wellington aquifer is above the
water table in the alluvial aquifer. According to landowners, the interface between fresh water and saline
water in the alluvial aquifer along the Smoky Hill River in northern Saline County is about 35 ft below land
surface. Chloride concentrations of 48,000 mg/L at 94 ft depth and of 180,000 mg/L in gypsum cavities at
112 ft have been reported. A relatively impervious shale separates the shallow fresh-water unit from the
deeper brine unit except for areas of collapse and brecciation caused by salt and gypsum dissolution
(Gillespie and Hargadine, 1981). Major rivers affected by these saline ground-water discharges include the
Smoky Hill near Salina, the Arkansas, Ninnescah, Saline, and Solomon Rivers. Increased pumpage from
the principal alluvial aquifer in central Kansas ("Equus beds") could cause inflow of mineralized water from
the Arkansas River into the aquifer (U.S. Geological Survey, 1984).
Approximately 24 percent (102,000 acres) of all the irrigable land,in Kansas was considered saline or
alkaline in 1960 (Fuhriman and Barton, 1971).
Kentucky: Most of Kentucky, with the exception of its easternmost part, is underlain by shallow
saline water, that is, ground water containing TDS concentrations greater than 1,000 ppm underlie the
state at depths less than 500 ft below land surface (Feth and others, 1965).
Louisiana: Shallow saline ground water (TDS>1,000 ppm) can be found along the Gulf Coast and
In northeast and central Louisiana (Feth and others, 1965). Sea-water intrusion has occurred all along the
coastal shores of Louisiana. In addition, the cities of Lake Charles, Baton Rouge, and New Orleans have
experienced severe cases of salt-water intrusion due to high pumpage of ground water. Shallow saline
water appears to be associated with salt-dome provinces. Shallow saline water also occurs along the
northeastern part of the state in the Mississippi River alluvial aquifer (Atkinson and others, 1986).
During periods of low flow tidal water invaded fresh-water aquifers in the Vermillion River area
(Newport, 1977). In southwestern Louisiana, saline water is moving landward and updip in the Chicot
aquifer, and, in the Baton Rouge area, saline water is moving across faults toward municipal wells
(Whiteman, 1979; U.S. Geological Survey, 1984). Shallow salt water also occurs in Pleistocene delta
sediments that may not yet be flushed.
44
-------
Maine: Overpumping has resulted in intrusion of tidal estuary waters into local aquifers south of
Augusta, in Kennebec and Sagadahoc Counties. Several domestic water wells have been affected by
salt-water intrusion whereby the problem fluctuates with time over the year (Atkinson and others, 1986).
A 300-ft-deep well producing from the bedrock aquifer near the town of Bowdoinham, Sagadahoc
County, was contaminated by salt water from the tidal reach of the Kennebec River, resulting in
abandonment of the well (Miller and others, 1974).
Maryland: Shallow saline ground water (TDS>1,000 ppm) occurs in the southern half of the
Chesapeake Bay peninsula (Feth and others, 1965). The inland limit of saline ground water in coastal plain
aquifers are along the coast for the Cretaceous aquifer, along a line from Dorchester to the southwestern
state-line corner with Delaware in the Tertiary aquifer, and along a line parallel to the Chesapeake shore
along the shore in the south and approximately half way between the shore and the Maryland-Delaware
state line in the north, crossing the state line just north of 39 degrees latitude (Miller and others, 1974).
Downdip portions of coastal aquifers contain natural salt water (Wheeler and Maclin, 1988).
Michigan: A zone of moderately saline ground water (TDS between 3,000 and 10,000 ppm)
underlies most of the eastern part of the state along Lake Huron to Lake Erie. Saline ground water
(TDS>1,000 ppm) underlies the remainder of the Michigan peninsula at variable depths of less than 500 ft
to more than 1,000 ft below land surface. Saline ground water is also encountered at shallow depth along
Lake Michigan and the western shore of Lake Superior (Feth and others, 1965). Higher TDS in ground
water occurs in the eastern part of the state, where more water is produced from bedrock aquifers. The
average depth to saline ground water along Lake Erie, Lake St. Glair, Lake Huron, and near Saginaw Bay is
200ft.
Pumpage has resulted in upward intrusion of saline water from deep bedrock into glacial aquifers at
various places throughout the state. This problem has been aggravated locally by leaky well casings
(Newport, 1977). Saline water in predominantly shale and silly shale of the Saginaw Formation underlies
glacial deposits at shallow depths in the area of Bay County, in the east-central part of Michigan's Lower
Peninsula. Chloride concentrations increase markedly to several thousand mg/L in many wells deeper
than 100 ft below land surface. Abandoned coal mines in the area may contribute to mixing of deep, saline
water and shallow, fresh water, but do not constitute a major source of water-quality deterioration (Twenter
and Cummings, 1985).
Many salt seeps and salt springs occur in the state (Krieger and others, 1957). Natural brines are
produced from rocks of the Detroit River Group at Manistee and Ludington in the western part of the state,
and at Midland, St. Louis, and Mayville in the eastern part of the state (Sorensen and Segal), 1973).
Minnesota: Shallow saUne ground water (TDS>1,000 ppm) has been reported along the western
border, from the northeastern part (along the shore of Lake Superior), and the southwestern and
southeastern parts of the state (Feth and others, 1965; Albin and Breummer, 1988). In northwestern
Minnesota, mineralized ground water discharges from Qrdovician and Cretaceous bedrock. Pumping from
45
-------
glacial aquifers in this area has caused upward intrusion of saline water from these deep bedrock aquifers
(Newport, 1977). Natural discharge occurs from the Red River-Winnipeg aquifer, degrading water quality
in overlying alluvial deposits and in the Red River of the North. Chloride concentrations of up to
46,000 mg/L have been reported from wells along the shore of Lake Superior (Albin and Breummer,
1988).
Mississippi: Saline ground-water generally underlies the state at depths of more than 1,000 ft
below land surface (Feth and others, 1965). Geohydrologic data indicate that most of the principal aquifers
of the state previously contained salt water. This salt water was later partially replaced by fresh water but still
occupies downdip portions of these aquifers (Bednar, 1988).
Missouri: Saline ground water containing more than 1,000 ppm TDS underlies the northern one-
third of the state at depths less than 500 ft below land surface. This includes an area of very saline water
(TOS from 10,000 to 35,000 ppm) that stretches from Clay County in the west to Marion County in the east
(Feth and others, 1965). The intrusion of natural salt-water from saline formations into shallow fresh water
is spreading from the northwestern part of the state toward the south (Atkinson and others, 1986).
Intrusion of brackish water in response to ground-water withdrawal has occurred in Bates, Barton, and
Vemon Counties (Carpenter and Darr, 1978).
The history and flow path of saline ground water in central Missouri has been suggested to be
(1) meteoric recharge in the Front Range of Colorado, (2) dissolution of Permian halite in the subsurface
of Kansas, (3) interaction with predominantly silicate mineral assemblages in Paleozoic strata, (4) dilution
and migration to shallow aquifer levels in central Missouri, and (5) mixing with local meteoric recharge
through Mississippian carbonates (Banner and others, 1989).
Montana: Most of the eastern half of the state is underlain by saline water (TDS> 1,000 ppm) at
depths less than 500 ft below land surface. Locally, TDS concentrations exceed 10,000 ppm (Feth and
others, 1965). Maximum TDS concentrations in the glacial-deposits aquifer are 30,000 mg/L and in the
Virgelle aquifer 5,100 mg/L (Taylor, 1983).
Nebraska: Saline ground water (TDS> 1,000 ppm) underlies the state at 500 to 1,000 ft below land
surface in the southeast and at greater than 1,000 ft below land surface in the remainder of the state. TDS
increase to values greater than 3,000 ppm in the southwest (Feth and others, 1965).
Salt water appears to be no serious threat to ground water in the state except for the easternmost
part where TDS up to 3,500 mg/L are reported from the Dakota Aquifer and Paleozoic rocks (Engberg and
Druliner, 1988). Taylor (1983) reported maximum TDS concentrations in the Dakota aquifer of
30,000 mg/L.
According to Atkinson and others (1986), local occurrences of salt-water intrusion have been
reported in Saunders, Lancaster, and Saline Counties. Saline surface water occurs in some of the
sandhills lakes in western Nebraska and in some localities in eastern Nebraska (Krieger and others, 1957).
46
-------
Approximately 24 percent (290,000 acres) of Nebraska's total irrigable acreage were considered saline
and alkaline in 1960 (Fuhriman and Barton, 1971).
Nevada: Saline surface water and shallow saline ground water has been reported locally
throughout central and western Nevada. TDS concentrations vary between 1,000 ppm and 35,000 ppm
(Feth and others, 1965). As much as 42 percent (475,000 acres) of the state's irrigable acreage is saline or
alkaline (in 1960).
Evaporation of ground water in the many closed basins is the principal salinization mechanism in
Nevada. There are two places where TDS in ground water exceeds 10,000 mg/L; these are point sources
in the Tonopah Basin in Soda Spring Valley and Clayton Valley a few miles south and east of Walker Lake.
Seepage flows from two springs contain TDS of 15,000 and 30,000 mg/L. Water from most aquifers is
saline in the Smoke Creek Basin, the Desert Creek Basin, and the Black Rock Deserts in the lower Quinn
River Basin, as well as in the Lovelock area in the lower Humboldt River basin. Long residence time of
ground water in sediments containing large amounts of salts is the primary reason for TDS concentrations
greater than 1,000 mg/L, and whereby mineral content increases toward the center of the basins. The
areas of Carson Sink, Walker Lake, and east of Pyramid Lake contain about,50 to 60 percent of all the
mineralized ground water in the state. North of Lake Mead is an area of a few thousand acres that is
underlain by saline water ranging in TDS from 1,000 to 3,000 mg/L; soluble salts in the aquifers are the
most likely source of the high salt content (Fuhriman and Barton, 1971).
Closed-basin lakes of high salinity include Big Soda Lake, Pyramid Lake, Walker Lake, Winnemucca
Lake, Carson Sink, Rhodes marsh, and a closed-basin sump northeast of Fernly. The primary sources of
salt in some of these lakes probably are unconsolidated sediments in the Lahontan Valley Group, from
which highly concentrated ground water discharges into rivers and lakes (Whitehead and Feth, 1961).
Ground water in geothermal areas frequently exceeds 1,000 mg/L TDS (Thomas and Hoffman,
1988). Evaporative concentration of meteoric water before recharge and mineral-rock interactions govern
the chemical composition of thermal waters from springs and wells in the Bradys Hot Springs geothermal
area of Churchill County. The hottest water sampled at Bradys Hot Springs contained 2,600 mg/L but
concentrations of greater than 6,000 mg/L have been reported from the Desert Peak area (Welch and
Preissler, 1986).
New Hampshire: Tidal waters have intruded aquifers in the Portsmouth area (Newport, 1977).
New Jersey: Saline ground water (TDS> 1,000 ppm) underlies most of the southern and
southeastern parts of the state at depths at or greater than 1,000 ft below land surface. Depth to saline
water decreases toward the coast in the south and is generally less than 500 ft below land surface along
the entire New Jersey coastline (Feth and others, 1965). Two major water bodies of salty ground water are
present along the New Jersey Atlantic Coast: a shallow one in Pleistocene deposits and a deep one,
generally below the 800-ft sand. The position of the salt-water/fresh-water boundaries in those water
bodies depends on the head distribution of fresh water in the respective units (Upson, 1966). Salt water
47
-------
from tidal estuaries and bays has intruded into water table and artesian aquifers due to pumping and
dredging in the areas of Sayreville (Baritan River), Gibbstown-Paulsborough, Newark (Passaic River),
Rahway, Camden (Delaware River), and Salem.
New Mexico: With the exception of isolated areas in the north-central and southwestern parts of
the state, saline ground water (TDS>1,000 ppm) underlies New Mexico at shallow depth (<500 ft below
land surface). TDS concentrations vary considerably with local well waters exceeding 35,000 ppm in
Chaves, Sierra, and Otero Counties (Feth and others, 1965). Almost all aquifers in the state contain fresh
as well as saline water. Seventy-five percent of New Mexico's ground water is too saline for most uses
(Ong, 198S). Approximately two-thirds of the shallow ground water in storage in the eastern Tularosa
Basin is slightly saline (TDS between 1,000 and 3,000 mg/L). In the entire Tularosa Basin, possibly 180
million acre-ft of brine (TDS>35,000 mg/L) are in storage, compared to only approximately 1A to
2.1 million acre-ft of fresh water (Orr and Myers, 1986). More than 22 percent (191,000 acres) of New
Mexico's irrigable area were considered saline or alkaline in 1960 (Fuhriman and Barton, 1971). High
salinities in valley-fill aquifers are the result of recharge from saline base flow from rivers and the result of
irrigation return flows (Ong, 1988). In parts of south-central New Mexico, the potential yield ratio of usable
ground water to impotable ground water is smaller than 1:1,000 (Scalf and others, 1973).
Heavy pumpage has caused salt-water intrusion from deep, saline bedrock formations into
producing aquifers at various places throughout the state (Newport, 1977), such as in the Roswell Basin
of southeastern New Mexico. This problem is anticipated in the future for the Tularosa and Estancia
ground-water basins (Smith, 1989). With the exception of recharge areas in the west, ground water in the
Tularosa Basin is saline, whereby concentrations generally increase from west to east. In the Pecos River
Basin, saline water from deeper strata mixes with less saline water in shallow units (Ong, 1988). Water in
the Pecos River below Alamogordo Dam is slightly saline and of the calcium-sulfate type. Spring inflow in
the Malaga Bend area (TDS up to 270,000 ppm) and irrigation-return flow from Roswell southward
increases salinity in the river and changes the water type to a sodium-chloride type. The resulting brine
discharges at the surface along the Pecos River south of Carlsbad. (Scalf and others, 1973). Ground-
water withdrawal in the Rio Grande drainage basin exceeds annual recharge by far; therefore, ground
water in storage is being depleted steadily, with an accompanying deterioration in quality (Kelly, 1974).
Naturally occurring brine mixes with fresh water in the shallow subsurface in Eddy County before
discharging into the Pecos River. To alleviate surface-water-quality deterioration, brine was pumped from
the aquifer into Northeast Reservoir (Havens and Wilkins, 1979). Brine leakage from this reservoir added
chloride to the river and caused gradual chloride increases from an average of 140 tons per day between
1952 and 1963; 167 tons per day between 1963 and 1966, and 256 tons per day between 1967 and
1968 (Havens and Wilkins, 1979).
New York: Most of the central and southern parts of New York are underlain by saline ground water
(TDS>1,000 ppm) at variable depths between <500 ft below land surface and >1,000 ft below land
48
-------
surface. Saline water appears to be shallowest along Lake Ontario and adjacent to the St. Lawrence River
in northwestemmost New York (Feth and others, 1965).
Natural saline water occurs at shallow depth in western and central New York. In northwestern
Cattaraugus County, brine moves from bedrock into shallow unconsoiidated aquifers (U.S. Geological
Survey, 1984). Oil and gas seeps have been known to occur in western New York since historic time, as
documented by stories about Indians collecting oil from the Seneca Oil spring near Cuba, New York. Other
sites of oil springs are at Freedom, Allegany County, and around Canandaigua Lake. This discharge of gas
or oil at land surface probably is too small to have contaminated water supplies; an exception to this may be
Oil Creek near Cuba. However, salt springs that issue from Silurian rocks containing halite in western New
York have contaminated surface and shallow ground water. Only a few of these salt-water springs appear
to be associated with oil- and-gas bearing units (Grain, 1969). Some of the Paleozoic sedimentary rocks
underlying the glacial debris contain lenses of salt, contributing to naturally saline ground water in
Paleozoic rocks, in the overlying glacial debris, and in local springs (Fairchild, 1935; Diment and others,
1973). -..
Saline water in the Jamestown area may be due to upward migration of deep saline water (from 1,500
to 2,000 ft below mean sea level) along natural breaks in the rocks and/or along abandoned oil and gas
wells that have not been adequately plugged. Salt beds and highly mineralized water underlie areas of
Chemung County in western New York. Old abandoned gas wells are conduits for the upward migration of
these saline waters. Fresh-water aquifers in the area generally contain less than 10 mg/L; wells
contaminated by brines from deeper strata contain between 100 and 500 mg/L. Similar conditions may
exist in the Susquehanna River Basin (Gass and others, 1977).
North Carolina: Saline ground water containing TDS concentrations greater than 1,000 ppm
occur at depths less than 500 ft along the Atlantic Coast. The depth to saline water generally increases
inland (Feth and others, 1965). The three major aquifers of northeast North Carolina all contain salt water in
their eastern portions (Wilder and others, 1978), imposing a threat if heavy pumpage of fresh water should
occur.
North Dakota: The entire state is underlain by saline ground water (TDS>1,000 ppm) at depths
less than 500 ft below land surface. Concentrations of greater than 10,000 ppm occur in the northeastern
portion of the state and in northern Burke County (Feth and others, 1965).
Artesian pressure on deep saline aquifers and saline seep are major salinization hazards. In the Red
River Valley, heads of deep saline aquifers are above land surface, resulting in the surface discharge of
saline water (Atkinson and others, 1986). In addition, pumping of fresh ground water from overlying
aquifers has induced upward migration of saline ground water from deep aquifers in the Red River Valley
(Newport, 1977).
Closed basins, such as the Devils Lake chain, contain slightly saline water or brine. The total surface
area of Devils Lake has decreased from 90,000 acres in 1867 to 6,500 acres in 1940, transforming the
49
-------
water to a shallow body of stagnant brine. Salt concentrations in the lake range from 6,000 to 25,000 ppm.
In the East Stump Lake, IDS ranges from 19,000 to 106,000 ppm (records from 1889-1923 and 1948-
1952) (Krieger and others, 1957). Maximum TDS concentrations in the states aquifers are: Glacial
deposits, 5,000 mg/L; Fort Union, 7,000 mg/L; Dakota, 11,000 mg/L; and Madison, 350,000 mg/L
(Taylor, 1983).
With the exception of the eastern state-border area, all land areas of North Dakota can be considered
potential saline-seep areas (Bahls and Miller, 1975). As of 1960, approximately 31 percent (817,000
acres) of the state's irrigable land was considered saline or alkaline (Fuhriman and Barton, 1971).
Ohio: The entire state of Ohio is underlain by saline ground water (TDS>1,000 ppm) at depths less
than 500 ft below land surface (Feth and others, 1965). Nevertheless, upconing and intrusion of natural
salt water has not yet been a major problem (Atkinson and others, 1986).
Oklahoma: Almost the entire state of Oklahoma is underlain by shallow saline ground water
(TDS>1,000 ppm at depths less than 500 ft below land surface). Elevated TDS concentrations in excess
of 3,000 ppm can be found in the southwestern part of the state and in spring waters in Blaine, Harper,
and Alfalfa Counties (Feth and others, 1965).
Natural salinity and oil-field brines constitute problems in the Cimarron Terrace from Woods County
southeast to Logan County (Atkinson and others, 1986). Salt springs and seeps issuing from underlying
salt beds increase the salinity of the Cimarron River near Mocane. The salinity of the river water is
increased further in the lower reaches by natural salt beds and oil-field brines (Krieger and others, 1957).
Naturally occurring salt water in Permian rocks also accounts for ground-water salinization in an area near
Dover, Kingfisher County (Oklahoma Water Resources Board, 1975). Overdevelopment of ground water
has resulted in salt-water intrusion at many localities throughout the state. One example is the
overdevelopment from the Garber-Wellington aquifer in central Oklahoma (Atkinson and others, 1986).
Approximately 23 percent (194,000 acres) of the state's irrigable land was considered saline or alkaline in
1960 (Fuhriman and Barton, 1971).
Oregon: Local areas of shallow saline ground water (TDS>1,000 ppm at depths <500 ft below land
surface) are reported for western Lewis, southeastern Columbia, southwestern Clark, western Multnoma,
northwestern Clackamas, central and northern Yamhill, most of Washington, and central Hamey Counties.
In addition, isolated springs and wells in the western, southern, and eastern portions of the state may
exhibit TDS concentrations greater than 3,000 ppm (Feth and others, 1965). Shallow saline ground water
can be found in valley sites near principal streams in many areas of western Oregon underlain by
sedimentary and volcanic rocks. Locally, saline ground water discharges to streams (U.S. Geological
Survey, 1984).
The majority of salt-water problems due to intrusion have occurred in the northwestern part of
Oregon. Upconing and/or lateral intrusion has been reported near Portland in Multnomah County and
near Salem in Yamhill, Marion, and Polk Counties (Atkinson and others, 1986).
50
-------
Saline lakes of Oregon include Alkali Valley, Abert Lake, Hamey Lake, and Summer Lake (Hardie and
Eugster, 1970).
Approximately 7 percent (103,000 acres) of the state's irrigable area was considered saline or
alkaline in 1960 (van der Leeden and others, 1975).
Pennsylvania: Saline ground water containing >1,000 ppm IDS underlie the western half of the
state at depths <500 ft below land surface (Feth and others, 1965).
South Carolina: Saline ground water (TDS>1,000 ppm) underlies the eastern part of the state
within a belt reaching from the coast approximately 100 miles inland in the south and 30 miles inland in the
north. Depths to saline water generally exceeds 500 ft with the exception of coastal areas in the southeast
and the northeast, where depths to saline water are less than 500 ft (Feth and others, 1965). The zone of
salt water, which extends through the center of the state from the southwest to the northeast, is
associated with salt deposits (Speiran and others, 1988).
Heavy pumpage has caused upward and downward intrusion from layered saline aquifers and lateral
intrusion from the Atlantic Ocean in the Beauford and Charleston areas (Miller and others, 1977). It has
lowered water levels in the Black Creek aquifer to more than 100 ft below land surface, threatening fresh
ground-water sources with the potential of lateral and upward intrusion of saline water (U.S. Geological
Survey, 1984).
Background concentrations of sodium and chloride in the Black Creek aquifer of Horry and
Georgetown Counties are less than 280 mg/L and less than 40 mg/L, respectively. Concentrations
greater than those may indicate mixing of sea water and fresh water. There is no indication of sea-water
intrusion caused by pumping, suggesting that deterioration of water quality is associated with vertical
migration of salt water. Residual sea water may not be flushed from downdip portions of aquifers, resulting
in a fairly wide zone of dispersion and diffusion (Zack and Roberts, 1988).
South Dakota: Ground water of highly variable quality containing between 1,000 and 10,000 ppm
TDS underlies the state at depths between less than 500 ft and greater than 1,000 ft below land surface
(Feth and others, 1965).
Upconing of saline water due to pumpage of fresh water aquifers has been reported at various
places throughout the state, including in the Black Hills area of southwestern South Dakota (Newport,
1977). But the major salt-water problem in the state is associated with the occurrence of saline seeps,
which exist throughout most of the northern, central, and northeastern parts of the state (Atkinson and
others, 1986).
In 1950 it was estimated that approximately 12,000 to 15,000 artesian wells within the state leak
water into aquifers above them. Inadequately plugged test holes drilled for oil, gas, and uranium may
permit upward migration of saline water even in areas where no production is occurring. Maximum TDS
concentrations in the state's aquifers are: glacial deposits, 10,000 mg/L; Dakota, 8,000 mg/L; Inyan Kara,
51
-------
10,000 mg/L; Sundance, 7,600 mg/L; Minnelusa, 4,300 mg/L; Madison, 120,000 mg/L; Red River,
130,000 mg/L; and Deadwood, 40,000 mg/L (Taylor, 1983).
More than 70 percent (1,196,000 acres) of the state's irrigable land surface was considered saline or
alkaline in 1960 (Fuhriman and Barton, 1971).
Tennessee: The western two-thirds of the state are underlain by ground water containing more
than 1,000 ppm TDS. The depths to the interface between saline and fresh water is <500 ft in the center
of the state but generally more than 1,000 ft in the western part of the state. In addition, a large area of
extremely saline water, with TDS between 3,000 ppm and 35,000 ppm, underlies the midsection of the
state at depths less than 500 ft below land surface (Feth and others, 1965). No salt-water problems have
been reported in Tennessee (Atkinson and others, 1986).
Texas: Much of Texas is underlain by saline ground water (TDS> 1,000 ppm) at depths <500 ft
below land surface. Depth to saline water is somewhat greater in the southern part of West Texas, in
south-central and east-central Texas, in East Texas, and along the Gulf Coast. The occurrence and
availability of subsurface saline water in Texas were summarized for all major aquifers by Snyder and others
(1972). Included in their listings are waters with TDS greater than 3,000 ppm. Brine springs and shallow
saline ground water with TDS concentrations in excess of 3,000 ppm occur in the Rolling Plains of north-
central Texas. Other high-saline areas are along the Pecos River and the Rio Grande in West Texas.
Ground-water quality in the Salt Basin of Trans-Pecos Texas deteriorates in a northward direction/with
TDS ranging from 1,550 mg/L to more than 6,000 mg/L in the heavily pumped areas of the Dell City area
(Davis and Gordon, 1970). High-saline areas also occur locally throughout South Texas (Feth and others,
1965). In Central Texas, a zone of saline water borders the Edwards aquifer, which is a major source of
fresh water in the area, to the east of the so-called "bad-water yne" (Senger and others, 1990).
Leakage of salt water and gas from fault zones has been detected in Wood County. As a direct result
of heavy pumpage sait-water intrusion has occurred in the Wintergarden area southwest of San Antonio,
and in East Texas along the Gulf Coast.
Tributaries of the Red River contain high salt loads. The Salt Fork of the Red River and Mulberry
Creek contain high proportions of calcium sulfate derived from gypsum. The Prairie Dog Town Fork of the
Red River is highly saline with common salt and the Pease River has high sulfate and chloride
concentrations. The low flows of the Salt Fork and the Double Mountain Fork of the Brazos River are at
times slightly to moderately saline. Saline surface water is most common in western and northwestern
Texas where rainfall is low, evaporation is high, and where rock formations at the surface contain large
amounts of readily soluble minerals. Salt springs contribute to the salinity of the Colorado River above
Colorado City. Saline surface waters in the eastern and southern parts of the state often originate in the
west or northwest or are due to oil-field brine pollution. Along the Gulf Coast sea water mixes with surface
water in the tidal reaches of the rivers, such as in the Calcasieu River channel, Catoasieu Lake, Sabine
Lake, and Lower Sabine River (Krieger and others, 1957).
52
-------
The Ogallala aquifer is contaminated by salt water downstream from every one of about 30 large
pluvial lake basins that contain saline water or saline lacustrine sediments. In some instances, saline
ground water may cover an area of several hundreds of square miles (Reeves, 1970). Saline ground water
and brine lakes occur in parts of Terry and Lynn Counties. The source of the salt water is concentrated
brine which exists in the Tahoka, Ogallala, and Cretaceous rocks of the area. Irrigation pumpage caused a
migration of the interface between fresh water and salt water, resulting in salinization of water wells and
soils in topographically low areas where evaporation of water-logged soils occurs. Chloride as well as
sulfate concentrations in closed drainage basins of the area range as high as 170,000 mg/L. Disposal of
brine and brine spills at a sodium-sulfate solution mining operation has contributed locally to the
salinization of soils and ground water, but is not believed to play a major role as salinization mechanism
(Bluntzer, 1982).
Localized chloride contamination of Pliocene sands in Texas may be caused by casing corrosion of
active or inactive water wells, permitting flow of highly mineralized water from shallow strata to deeper
aquifers (Sayre, 1937, in Gass and others, 1977). Serious regional contamination problems may have
occurred in areas where highly pressured brine aquifers occur, such as the Rustler Formation of
southwest Texas, the Coleman Junction Limestone in west-central Texas, and the deep Miocene brine
aquifers of the Gulf Coast. Plugging of boreholes often was done by simply putting wood, mud, or rocks
into the hole and dry holes were often left uncased (McMillion, 1965).
East Texas towns, pumping from the Woodbine Formation more than 20 miles from the outcrop,
often experience TDS of 1,500-4,000 ppm in their drinking-water supplies. For example, water samples
from the town of Anna, Collin County, contained 4,112 ppm TDS (Sundstrom and others, 1948). The
source of saline water may be intrusion of brine from the Tyler Basin onto the North Texas Shelf (Parker,
1969).
Heavy pumpage of irrigation water has caused intrusion of highly mineralized river water into ground
water in the Pecos River Valley, especially in Reeves and Pecos Counties. By far ground-water withdrawal
in the Rio Grande drainage basin exceeds annual recharge; therefore, ground water in storage is being
depleted steadily with accompanying deterioration in quality (Kelly, 1974). The same mechanism of
overpumpage is responsible for salt-water intrusion into the Beaumont clay aquifer in Brazoria County, the
Chicot aquifer in Orange County, the Seymour Formation in Knox County, and into alluvium in Ward
County (Scalf and others, 1973). Winslow and others (1957) estimated that the interface between fresh
and saline ground water in Harris County is moving toward the centers of heavy pumping at a speed of a
few hundred feet per year.
Utah: Salinization in the state can in part be attributed to vertical movement of saline water from
saline aquifers (Waddell and Maxell, 1988). Many areas in the western and southeastern parts of the state
are underlain by slightly to very saline water, at depths less than 500 ft below land surface (Feth and others,:
1965). Extensive areas of low-quality ground water can be found in Uintah, Emery, and Grand Counties,
53
-------
and in the Price, San Rafael, and the Dirty Devil River basins. Saline ground water in the Bear River Basin
may be caused by (a) long residence time of water in fine-grained deposits containing soluble salts, (b)
evapotranspiration, (c) recharge with saline irrigation-return water, and/or (d) lateral movement of saline
water from areas of large thermal springs. Highly mineralized ground water also underlies areas west of
Great Salt Lake (Fuhriman and Barton, 1971). Highly saline water can be found in closed-basin lakes and
their tributaries, such as Bonneville Salt Flat, El Monte Hot Spring, Great Salt Lake, Hooper Hot Spring,
Locomotive Spring, Stinking Spring west of Corinne, and Utah Hot Springs. Bedded halite and gypsum
underlying the Sevier River Basin contribute to the high content of dissolved solids carried by the Sevier
River (Whitehead and Feth, 1961). Mineral springs contribute highly mineralized water to the Virgin River
below La Verkin (Krieger and others, 1957). Although these springs contribute only approximately ten
percent of the water in the Virgin River, contribution of salt amounts to 60 percent of the river's total salt
load (Thome and Peterson, 1967).
In the western part of the state and in the Great Salt Lake area the potential of pumpage-induced
intrusion of salt water is high (Newport, 1977). Recharge from precipitation in the Northern Great Salt Lake
Desert is approximately 200 times higher than surface outflow to the Great Salt Lake (Stephens, 1974).
High evaporation rates in this area account for briny conditions in the shallow aquifer composed of salt and
lake-bed deposits. A second aquifer in the area, which is made up of surfidal and buried alluvial fans along
mountain flanks, yields fresh to moderately saline water. However, the most extensive aquifer underlying
the area yields brine from 1,000 to 1,600 ft below land surface in the Bonneville Salt Flats area (Stephens,
1974). Of all the shallow, recoverable ground water in the area, approximately 75 percent are highly saline
or briny (Stephens, 1974).
Streams that contribute to the salinity of the Colorado River include the Duchesne, Price, San
Rafael, Dirty Devil, Escalante, Paria, and Virgin Rivers. The principal sources of the salt content are
irrigation-return waters, natural runoff from soils developed on shale, and saline springs discharging from
the shale and other sail-bearing formations (U.S. Geological Survey, 1984).
Mineral and thermal springs occur in fault zones along the Wasatch mountain range, along the
Hansel Valley fault in northwestern Utah, on the east side of Promontory Point, near Howell, Grantsville,
Callao, and Gandy, near Timpie at the northwestern tip of Stansbury Mountains, and in Skull Valley,
(Fuhriman and Barton, 1971).
Virginia: Saline ground water at depths less than 500 ft below land surface can be found along the
Atlantic Coast in southeastern Virginia. Depth to saline water increases to greater than 1,000 ft below land
surface in the northeastern coastal area and in the westernmost parts of the state (Feth and others, 1965).
Historic changes in chloride content of water wells are local rather than regional phenomenons. In
general, a wedge of salt water coincident with the mouth of Chesapeake Bay extends into the York-James
and southern Middle Neck Peninsula, where the greatest chloride increase (175 mg/L) was measured.
Possible sources of salt water in ground water of the Virginia Coastal Plain are natural or induced sea-water
54
-------
intrusion, incomplete flushing of marine sediments, solution of evaporite deposits, or concentration by
movement of water through clay-rich sediments (Larson, 1981).
Washington: Some of the coastal saline waters in the state are probably relict sea water or connate
water (van der Leeden and others, 1975). Saline lakes in the state include Hot Lake, Lake Lenore, and
Soap Lake (Hardie and Eugster, 1970). Of the total irrigable land area in the state, approximately 12
percent (266,000 acres) was considered saline or alkaline in 1960 (Fuhriman and Barton, 1971).
West Virginia: With the exception of the northeasternmost part of the state, West Virginia is
underlain by saline water at depths ranging from less than 500 ft below land surface in the west to
northwest to greater than 1,000 ft in the south and east (Feth and others, 1965). Salt-water contamination
has not been reported in the state but there is a potential for movement of residual saline water toward
pumping centers (Newport, 1977).
Historic records report salt and oil springs and shallow brine occurrences at various localities in West
Virginia. Among these are Campbells Creek, Kanawha County, and Bulltown, Braxton County. At some
localities, salt water occurs significantly higher than fresh water, suggesting salt-water movement upward
through uncased holes and laterally into fresh-water aquifers. The town of Walton, Roane County, was
without potable water supplies for an extended period of time due to salt-water contamination (Bain,
1970).
Natural salt springs in the Kanawha River valley occur near the mouth of Campbells Creek at Maiden,
Kanawha County. In most areas of the western half of the state, salty ground water below major stream
valleys occurs at depths approximately 100 to 300 ft below land surface. Ground water turns from salty to
brine (Cl>30,000 mg/L) at depths of 1,000 ft to 5,000 ft. In the last few decades, salt-water migration
toward the land surface has been caused by vertical leakage along hundreds of unplugged wells and test
holes. These wells had been drilled during exploration for brine, oil, gas, and coal, and commonly were
abandoned uncased or improperly plugged. In Fayette County, chloride concentration of ground water
increased from 53 mg/L to greater than 1,900 mg/L within 51/2 years due to fresh-water pumpage and
inflow of salt water from abandoned boreholes. Heavy pumpage of fresh ground water from bedrock
aquifers at Charleston caused salt-water intrusion and chloride increases from less than 100 mg/L to
greater than 1,000 mg/L (Wilmoth, 1972).
Wisconsin: Saline ground water occurs at depths less than 500 ft below land surface along the
shore of Lake Superior and in some areas along the shore of Lake Michigan (Feth and others, 1965).
Heavy pumpage has caused lateral intrusion of this saline water into fresh water at various locations in the
state (Newport, 1977).
Wyoming: Saline ground water underlies the state at depths ranging from less than 500 ft below
land surface in the southwest, center, north, and northeast to greater than 1,000 ft below land surface in
the southeast and scattered areas in the center (Feth and others, 1965)
55
-------
Maximum TDS concentrations in the state's aquifers are: Wind River, 6,500 mg/L; Mesaverde,
16,300 mg/L; Frontier, 20,000 mg/L; Cleverly, 24,000 mg/L; Sundance, 4,700 mg/L; Nugget,
4,900 mg/L; Casper, 9,600 mg/L; and Tensleep, 280,000 mg/L (Taylor, 1983). The average TDS toad of
the Green River has increased from 410,000 to 612,000 tons per year (49 percent) at a time during which
the average streamflow has increased by only 5 percent. Saline-water seeps and irrigation-return flows in
the Big Sandy River account for the increase in salinity (U.S. Geological Survey, 1984).
About 22 percent (280,000 acres) of the state's irrigable land was considered saline in 1960
(Fuhriman and Barton, 1971).
3.2 Halite Solution
3.2.1. Mechanism
Many sedimentary basins in the United Sates contain thick layers of rock salt (halite) (Rg. 2). In most
instances these layers occur as beds but in some cases salt has been deformed and is present now in the
form of salt diapirs or salt domes. Dome provinces are found predominantly along the Gulf Coast, with
more than 300 individual domes in Texas, Louisiana, Mississippi, and Alabama. The depth to salt beds and
salt domes varies widely, from more than 10,000 ft below land surface in Florida to a few hundred ft below
land surface in parts of Texas and Louisiana, or at land surface at places in Utah (Dunrud and Nevins,
1981). Although halite is highly soluble, some of these deposits have been stable for hundreds of millions
of years with little or no dissolution going on, indicating that they are not in contact with circulating fresh
water. Other salt occurrences are located within local or regional ground-water flow systems and are being
dissolved, predominantly along the tops and margins, causing salinization of ground water. For example,
the Canadian River and Pecos River Valleys of Texas and New Mexico overly areas where as much as
200 m of salt have been dissolved by circulating ground water (Gustavson and others, 1990). Besides
this natural solution of salt, solution mining associated with the recovery of salt, oil, gas, or sulfur has been
practiced in the country for more than 100 years. In 1956, salt production amounted to more than
24 million tons, with approximately 75 percent produced by salt dissolution and 25 percent produced as
rock salt (Pierce and Rich, 1962). Natural and man-made removal of salt has caused subsidence and
collapse where salt occurs at shallow depths.
Johnson and others (1977) summarized the requirements necessary for dissolution of halite as
being (1) a supply of water unsaturated with respect to halite, (2) a deposit of salt through which or against
which the water flows, (3) an outlet that will accept the resultant brine, and (4) energy (such as a hydrostatic
head) to cause flow of water through the system. If those requirements are met and if the outlet of the
resultant brine is at land surface or becomes part of an underground fresh-water flow system, mixing of
brine and fresh water will occur. For example, such process was described by Hitchon and others (1969) in
the Mackenzie River Basin of western Canada, by Gillespie and Hargadine (1981) in the Kansas Permian
56
-------
Basin, and by Gustavson and others (1980) on the eastern flank of the Texas Permian Basin (Fig. 20).
Discharge of the resulting brine at land surface or in the shallow subsurface at those sites has been
documented by Hitchon and others (1969) in Canada, by Whittemore and Pollock (1979) in Kansas, and
by Richter and Kreitler (1986a,b) in Texas. In .each case, local meteoric water infiltrates the ground and
dissolves halite in the shallow subsurface on its way to regional discharge areas at topographically low
areas. . ,
Discharge of halite-solution brine and further evaporation of the salt water at land surface can lead to
the development of salt flats. A series of such salt flats, covered by gypsum and halite crusts, occurs in the
Rolling Plains of north-central Texas and southwestern Oklahoma, as described by Ward (1961) and by
Richter and Kreitler (1986a,b). Salt-water discharges at those areas (Rg. 21) ranges from a few thousand
milligrams per liter to greater than 150,000 mg/L dissolved chloride, affecting surface-water quality for
hundreds of miles downstream from salt-emission areas.
3.2.2. Composition of Halite and other Evaporites
Halite occurs in the subsurface in form of bedded or domal salt. Depending on the depositional
history, halite deposits may be associated with other chloride salts (for example, carnallite, KMgCl3*6H2O
or sylvite, KCI), with sulfates (for example, polyhalite, ^CagMg [SO4]4«H2O; anhydrite, CaSC^; or
gypsum, CaSO4'2H2O), orwith carbonates (for example dolomite, CaMg(CC>3)2; or limestone, CaCOs),
which contribute to the overall salinity of ground water in contact with halite. Salt domes are composed
mainly of pure halite, with minor amounts of anhydrite, gypsum, and limestone. Analyses of salt from the
Gulf Coast salt-dome province consist of more than 92 percent NaCI with varying amounts of other mineral
constituents and insoluble matter (Table 3). The insoluble or less-soluble fractions stay behind when the
tops of shallow salt domes are dissolved by circulating fresh waters, forming what is known as the cap rock.
Cap rock is often missing from deeper domes but can be several hundreds of ft thick in shallow domes.
For example, in a mined salt dome in Harris County, Texas, limestone caprock extends to 76 ft, gypsum
caprock from 76 to 107 ft, and anhydrite caprock from 107 to 1,010 ft, at which depth salt is encountered
(Pierce and Rich, 1962).
Solution of halite by fresh water or dilution of halite brines produces water of relatively uniform
chemical composition, as indicated by plots of chemical constituents in halite-solution samples from
Texas, Kansas, and Canada (Fig. 22). Solution of mined rock salt deviates from the well-defined mixing
(dilution) trends indicated on figure 22 because of its nearly pure NaCI composition. Natural solutions of
halite, in contrast, often are in contact with other evaporites, such as gypsum and anhydrite, which
accounts for somewhat higher concentrations of other major cations and anions, such as Ca and 804.
57
-------
piOOO
EXPLANATION
SonGitor*/
I \ Muesioi
i t Krtiym . . j docoliy converted to gypsum)
AL io
-------
KHAM
EXPLANATION
Brine-emission area
Salt spring
Self Crii*
^~xJ
ARMSTRONG!
[
IBRISCOE «C JHALL
[FLOYD ? TMOTLEY
Caprock
Escarpment
- ' r •-'%—• -'••:/
^^
I "V* aro* -'•** I
_GA_RZA | K£NT X^ | ...^-^STQNEWALL J
[-—1
0 10
i i
30 50 km
OA 2866
Figure 21. Discharge points of halite-solution brine at land surface in the Rolling Plains of north-central
Texas and southwestern Oklahoma (from Richter and Kreitler, 1986a,b).
59
-------
Table 3. Mineral composition of salt from domes in Louisiana and Texas (data from Pierce and Rich, 1962).
Coastal domes, Louisiana
Interior domes,
Texas
Sodium Chloride (NaCI)
Calcium sulfate (CaSC-4)
Magnesium chloride(MgCl2)
Magnesium carbonate (MgCOa)
Sodium carbonate (N32CO3)
Sodium suffate (N32SO4)
Calcium carbonate (CaCOa)
Calcium chloride (CaCIa)
Iron and aluminum oxide
Insoluble matter
1
92.750
0.201
0.067
0.837
1.804
0.000
0.500
3.325
2
96.405
0.694
0.012
3
99.252
0.694
0.012
4
95.720
3.950
0.008
0.226
0.025
0.059
0.042
0.140
0.012
0.030
5
98.883
1.099
Trace
0.008
0.010
Trace
6
98.926
1.041
0.023
0.100
60
-------
1000000
100000 -
o>
10000 •!
1000
1000
10000 100000
Cl (mg/L)
1000000
10000 -3
•5.
E, 1000 -
5
100
1000
10000 100000
CI(mg/L)
1000000
1000 -
"&
E, 100 -
a
S
10 -
.:.V*
$L°^§y^^ o °
"**
10000 -a
1000
10000 100000
Cl (mg/L)
1000000
1000 -
E 100 -I
10 -
1000
coo
10000 100000
Cl (mg/L)
1000000
i uvuuu
10000 -
i ;
I :
cf
W 1000 r
10
°»^«rfb^t*^*§
CfeS*o^ ° C
o o
o
00 10000 100000 100C
. I •
.01 1
!§" •
"a
£ .001 -.
•?
" .0001 •
CD
.00001 -
WOO 10
9 %^g|og J|
O H "O ^°8
8
00 10000 100000 100C
KX30
Cl (mg/L) Cl (mg/L)
QA17216C
Figure 22. Bivariate plots of major ions and Br/CI ratios versus chloride for natural halite-solution brines
from Canada (data from Hitchon and others, 1969), Kansas (data from Whittemore and Pollock, 1979), and
Texas (data from Dutton and others, 1985, and Richter and Kreitler, I986a,b; open circles). Relatively little
scatter in these plots suggests little variation between halite-solution brines from different areas. The
composition of laboratory solutions of mined halite (solid squares; data from Whittemore and Pollock,
1979) deviates from natural solution of salt beds because of the absence of associated evaporites.
61
-------
3.2.3. Examples of Geochemical Studies of Halite Solution
As with other salinization sources, identification of halite solution as source of salinity in absence of
any other possible source is indicated simply by an increase in IDS and all or most other individual
chemical constituents. Constituent ratios, however, have to be used to distinguish halite solution from
other potential sources, such as oil-field brine. Described below are several studies where ionic ratios
were used to effectively differentiate halite solution from other salt sources.
Hitchon and others (1969) determined the source of salinity in salt springs and saline seeps of
northeastern Alberta to be halite, based on a variety of chemical and isotopic constituents. The K/Na ratio
in salt spring samples was much lower than in typical deep-basin formation water in the area and the
relative concentrations of bromide and potassium were too low to represent residual evaporative brine.
Low bromide concentrations associated with high IDS contents indicate halite solution as the source of
salinity, as suggested from the brine-classification system established by Rittenhouse (1967). Within this
classification system, IDS contents greater than that of sea water in combination with bromide contents
less than would be expected from simple concentration of sea water, suggest solution of halite (Fig. 23).
Hydrogen-and oxygen-stable isotopes show a local meteoric origin, further supporting the argument that
local dissolution instead of regional deep-basin discharge of formation water is the source of salt water
(Hitchon and others, 1969). Dissolution of halite is also documented in southern Manitoba, Canada,
where saline springs and saline seeps contribute to chloride levels of up to 600 mg/L in Lake
Winnipegosis and Lake Manitoba (van Everdingen, 1971). Identification of halite solution as opposed to
discharge of deep formation water was done by using the ratios of (Ca+Mg)/(Na+K) and SO4/CI for
samples with chloride concentrations greater than 10,000 mg/L. At lower concentrations, dilution with
local ground water masked the salinization source. Similarly, the ratios of (Na+K)/CI and Ca/Mg were of little
use in this study (apparently because of overlapping values in the endmembers). This mixing, however,
was suggested from modified Piper plots, in which brackish waters plot on the theoretical mixing line
between salt-spring samples and fresh runoff samples (van Everdingen, 1971) (Fig. 24).
Leonard and Ward (1962) were among the first to use the Na/CI ratio to distinguish halite-solution
brine from oil-field brine in Oklahoma. One type of brine, derived from salt springs in western Oklahoma,
typically shows a Na/CI weight ratio in the range of 0.63 to 0.65, which suggests that nearly pure halite
(Na/CI weight ratio of 0.648) is the source of sodium and chloride in those brines. Another type of brine in
the same area consistently has Na/CI weight ratios less than 0.60, whereby the ratio decreases with
increase in chlorinity (Rg. 25). This type of brine was derived from oil wells. Based on this difference in
ratios between the two potential sources, Leonard and Ward (1962) determined which of the two
endmembers is the contaminating source in several streams of western Oklahoma, that is, halite-solution
brine contributes to salinity in the Cimarron River (#5, #6), and oil-field brines contribute to salinity in the
Little River (#4) and the Arkansas River (#7, #8). Surface-water degradation of streams in Kansas was
62
-------
1000 q
100 =
Cf)
Q
10 :
Concentration of sea water.
Group m
Group V
ia water
Group I
i i i 111 in
i i i 111 in
I 1 I Mill
10
100
Br (ppm)
1000
10,000
QA17217C
Figure 23. Classification of oil-field/deep-basin waters according to TDS and bromide concentrations, as
suggested by Rittenhouse (1967). Brine samples from saline seeps in northeastern Alberta (solid dots)
(from Hitchon and others, 1969) plot within the halite-solution group (Group III). Other groupings are:
Group I, simple concentration or dilution of sea water; Group II, same as one with additional bromide,
possibly added during eahy diagenesis; Group IV, diluted Group III waters or dilution of Groups I, II, or V
with low-Br water; Group V, possibly altered bitterns left after salt deposition (from Rittenhouse, 1967).
63
-------
Na Cl
HC03
OA17218C
Figure 24. Modified Piper diagrams (rotated axes) of chemical composition of brine springs and surface
waters in southern Manitoba. Mixing trends suggest solution of halite as the source of salinity in brackish
ground water of the area (modified from van Everdingen, 1971).
64
-------
«:uv,uuu -
180,000-
160,000-
140,000-
•g- 120,000-
Q.
Q
O 100.000-
80,000-
60,000-
40,000-
20,000-
\
\ •
• !•
\
\
0j
i
t*
3*
X
2
>
v «
\
\
i\ . •
\
\
v •
• v
\
v •
\
1 V- x'
i
A
16
A
AA
i
A A
40/100 50/100 60/100
Na/CI (weight ratio)
70/100
QA17219C
Figure 25. Comparison of Na/CI weight ratios for oil-field brines (dots) and brines from salt springs
(triangles) in western Oklahoma and southwestern Kansas (from Leonard and Ward, 1962). Weight ratios
of approximately 0.65 suggest halite solution as the source of salinity in the salt springs and in some
surface waters (crosses; Nos. 5 and 6), whereas ratios of less than 0.60 suggest oil-field contamination in
other surf ace waters. '
65
-------
studied by Gogel (1981) using the same technique. Samples from the Ninnescah River alluvium ranged
from 0.65 to 0.67 in Na/CI weight ratios, indicating that halite solution in the underlying Wellington aquifer
is the source of salinity. Measurements along Slate Creek showed the same ratio range for samples having
chloride concentrations greater than 500 mg/L but a ratio of 0.91 for a sample having a chloride
concentration of only 78 mg/L. This sample was obtained upstream from the saline-water inflow area and
represents uncontaminated background concentrations. Very tow Na/CI ratios, ranging from 0.54 to 0.28,
along Salt Creek, in contrast, suggest oil-field contamination (Gogel, 1981). The Na/CI ratio was also used
by the Oklahoma Water Resources Board (1975) to distinguish halite-solution brine from oil-field
contamination in parts of Oklahoma. Weight ratios of 0.66 were indicative of halite solution in the Dover
area, whereas ratios of 0.38 were indicative of oil-field pollution in the Crescent area.
Whittemore (1984) pointed out that halite solution brines in Kansas usually have lower Ca/CI and
Mg/CI ratios and higher SO^CI ratios than oil-field brines. The same relationship was found for halite-
dissolution brines and deep-basin brines in Texas by Richter and Kreitler (1986a,b). However, these ratios
appear to work best as tracers when little chemical reactions occur after mixing of the respective salt-water
source and fresh water. To avoid this change of chemical constituent ratios by mechanisms other than
mixing or dilution, Whittemore and Pollock (1979) suggested the use of minor chemical constituents,
such as bromide, iodide, and lithium, that are relatively conservative in solution. Of those, bromide
concentrations and Br/CI weight ratios are the most widely used tracers because bromide is similarly
conservative as chloride, and because a significant difference in Br/CI ratios between most halite-solution
brines and oil-field waters could be established (for example, Whittemore and Pollock, 1979; Whittemore,
1984,1988; Richter and Kreitler, 1986a,b). In Kansas, halite-solution brines typically have Br/CI ratios less
than 10 x 10"4, whereas oil-field brines typically have Br/CI ratios greater than 10 x 10"4 (Whittemore,
1984). This difference can be used to calculate mixing curves between fresh water composition and the
range of values for either endmember. By superimposing values from test holes in the Smoky Hill River
area, Whittemore (1984) was able to show that halite solution is the dominant mechanism of salinization in
that area (Rg. 26a). In contrast, in the Stood Orchard area south of Wichita, all of the observation-well
samples indicate mixing of fresh water with oil-field brines (Fig. 26b). Note in figure 26b that the sample
obtained from the Arkansas River (square) suggests halite solution; this chemical signature was derived
from areas upstream from the Blood Orchard area. Also, some testhole samples plot between the two
mixing fields, suggesting mixing of fresh water with both halite-solution brine and oil-field brine.
Richter and Kreitler (1986a,b) made use of the Na/CI and Br/CI ratios in an investigation of salt
springs and shallow subsurface brines in parts of north-central Texas and southwestern Oklahoma. Using
ratio ranges established by Ward (1961) for Na/CI in Oklahoma brines and Br/CI ratio ranges established by
Whittemore and Pollock (1979) for Kansas brines, Richter and Kreitler (1986a,b) grouped 120 chemical
analyses of saline waters into three groups: (1) a halite-solution group (Group A) with Na/CI molar ratios
greater than 0.95 (Na/CI weight ratio >0.62) and Br/CI weight ratios less than 4 x 10"4, (2) a deep-basin
66
-------
11111) I I I I I I I II 1 1 1 1 I I I 11 I 1 1 1 I 11 II
Cl (mg/L)
(b)
100
.9
n!
'jc.
O)
'
-------
group (Group C) with Na/CI ratios less than 0.95 and Br/CI ratios greater than 25 x 10"4, and (3) a mixing
group (Group B) with Na/CI ratios greater than 0.95 and Br/CI ratios greater than 4 x 10"4, or Na/CI ratios
less than 0.95 and Br/CI less than 25 x 10"4 (Fig. 27). After this initial grouping, other ratios, such as
Ca/CI, Mg/CI, K/CI, and I/CI, supported the grouping by generally showing higher values for deep-basin
brines than for halite-solution brines (Fig. 28). This suggests that all of these ratios can be used for
differentiating between halite-solution brines and deep-basin brines in this area. Molar ratios of
(Ca+Mg)/SC*4 and of Na/CI are close to unity in halite-solution samples, which suggests halite and gypsum
to be the main sources of dissolved constituents and distinguishes this water from deep-basin brines (Fig.
29). The third group (Group B) plotted intermediate in all these ratios. Group differences were also
reflected in the isotopic composition of 8180 and 8D. Little difference was observed between local, fresh
ground-water values and salt-spring values, suggesting dissolution of halite by local, meteoric ground
water (Fig. 13). Testhole samples, in contrast, formed a trend from a light isotopic composition at low
chloride concentrations to a heavy isotopic composition at high chloride concentrations, suggesting the
mixing of local, fresh ground water with deep-basin brines. This mixing was also documented in the depth
relationship of chloride and 818O for testhole samples (Rg. 30).
In addition to the ratios discussed above, Whittemore and Pollock (1979) pointed out the usefulness
of I/CI ratios in differentiating between halite-solution and oil-field brines. Halite-solution brines typically
have I/CI weight ratios less than 1 x 10~5, whereas oil-field brines in Kansas have ratios greater than
2 x 10~5 (Fig. 31). A similar relationship in I/CI ratios between halite-solution brine and deep-basin brine
was found by Richter and Kreitler (1986a,b) for samples from salt springs and shallow test holes in north-
central Texas (Rg. 28).
Dissolution of salt-dome halite in parts of the upper Gulf Coast of Texas accounts for high salinities in
geopressured Frio Formation waters (Morton and Land, 1987). Besides the high salinity
(TDS>105,000 mg/L), halite dissolution is also suggested as the source by low Br/CI ratios (<22 x 10"4),
and by low potassium and calcium concentrations (<500 mg/L and <5,000 mg/L, respectively) (Morton
and Land, 1987).
Mast (1982) suggested the use of mixing diagrams for differentiating between mixing of fresh water
with naturally saline ground water and mixing of fresh water with oil-field brine in parts of Kansas.
Contaminated waters were obtained from an aquifer that is in a producing oil field and is underlain by
natural halite and gypsum beds from which tt is separated by a shale of low permeability. Plotted on mixing
graphs of 804 versus Na+K, 804 versus TDS, and 804 versus Cl (Fig. 32) are the endmembers of
mixing, that is (1) uncontaminated fresh water, (2) naturally saline (halite-solution) water, and (3) produced
(oil-field) water. Between the endmembers are drawn theoretical mixing lines that were calculated using
mean values of the respective endmembers. Superimposed on the plot are all samples to be tested,
allowing visual inspection of the most likely salinization source, if any, as well as an estimate of the relative
percent contribution of each endmember (Fig. 32).
68
-------
(a) (b)
1.0-
0-90-
\
o
Z
£ .80-
.70-
60"
{
x All samples
. X
k , , xx » , • ,
•fc '; • I . * , V ,>jf
i Kfc X X j, y —
x' »"» <
4»> X * *
It
x ,
* «
>
x x' * /
X X /
* » X
" > ' X «
X X
) 1 23451
1.0-
o 90-
o
Z
E.80-
.70-
> (
" Group A
X
&"* >" * ' X '** '^
»*', .'
) 1 2 3 4 5 (
J All samples
1
80-j "
c
•^60-
o
\
m40-
-
20-
-
0
> (
x x .* . x ' x „
t x x x "x
X * * , X ,
X XX
X
x x
** x
X
X x x 'id*
»»»^u<>y xx,* x> jf x, ,, , v"i
1 ' 1 ' l' ' 1 ' 1 '
3 1 L 3 < 5 (
1 UW
80-^
a
* 60—
u
£ 40-
-
20-
-
Group A
X*8W,»« >to, „ , ,, ,, „ ,L,
0— ' — ~ • ' • ' 1 — • ' 1 —
; c : - -• f. ' t
CI(mol/L) CI(mol/L) Clltnol /L) Climol/L!
0) '-°-
<0
,
G.90-
E
'"Si
O
Z
E.80-
.70-
fin
•DU
Group B
x
x
« -* m
« " X » * XX«
xx x r
X ' .
n
t.o-
D-90-
E
•N.
O
Z
E.80-
.70-
1 1 1 1 1 1 1 1 ' 1 1 .DU-
0123456
Group C
X
*
«"»
* *
/ K
*^* y i k
* XX
1 * „ t
n
" K
R - * X *
* * X «
" X
0 I 2 34 5
1 W
60-
260-
. — .
U
£*°~
•20-
0,
6
Group B
if
* X " X ' ,1*
, * ' » • . x" »*r
i > i i j » j • r i
1-2 34 5
1W^
80-
,,
o
- 60-
^>
D
•***
£ 4O-
20-
Group C
X
" I
x^ fc ' > * r >
X x x >V *"*
x » J , • ' i
X •-'"•' X „
X
X
X >
X
0-) — i — i — i — ; — ' — | — i | i — : 1 — i
; c . • 2 ; •« e €
CI(mol/L) -CI(mol/L) Climol/L) - " Cumol/L)
Figure 27. Grouping of brine analyses according to (a) molar Na/CI ratios and (b) Br/CI weight ratios typical
for halite solution and oil-field/deep-basin brine. Group A: Na/CI >0.95 and Br/CI <25 x 10~4 (halite
solution); Group B: Na/CI >0.95 and Br/CI >4 x KT4, or, Na/CI <0.95 and Br/CI Br/CI <25 x 1Q-4 (mixing of
A and C?); Group C: Na/CI <0.95 and Br/CI >25 x 10~4 (deep-basin brine) (from Richter and Kreitler,
1986b).
OAI722I
-------
- Group A
.60 -j
1
J
1
.H
^ 1
!10-
** i
C '
" 201
.,oJ
\— •* ""' ', '• ', •« • A
C 1 2 3 4 5 6
CKmol/L)
:• W
,60-
.60-
ftO-
i-
•3,20-
.10-
0~v
Group C
' "
»
X
X *
* X * * X
** t*\ * * t
%*l i * "
0123456
CKmol/L)
J05-]
.04-
3-03-
1 '
x .02-
.0:-
•
Group A t
x x* x x
X "f X x^
^i,U A* « *
0123451
CKmol/L)
.04-
"
3X13-4
Group C
X
X
X
< i
|
x" .02-
"
JOI-
X
x * * x
* X
X
* * X X
c 1 *' l" ,
0 1 2 3 4 5 «
Clfmol/L)
4-
"e
£
» 9 _
s -2
.1 -
Group A
i
1
"'*">,""'
0— ^ ^ , , 1 . , r -
0'2 J « 5 6
CKmol/L)
.4-
3 -3-
1 •
f .2-
•
•'"
(
Group C
,
. -.' • 1
K *
* X
V
* X
" J* *
X X K * y
It
«» i
3123456
CKmol/L)
,20-
100-
80-
o
x €0-
D
^40-
20-
Group A
*
X "^ Xx"»«^
D 1 2 3 4 5
6
CKmol/L)
-
:00-
"
80-
to
O
; 60-
P
— 40-
20-
,», , > Group C
"
X X
» 'x « » « X
X X X X ,
X
, * * '
X
K
X
'x.
01 2345
6
CI(mol/L) OAI7222
Figure 28. Bivariate plots of chemical constituents in brines collected from salt springs and shallow test
holes in the Rolling Plains of Texas. Grouping of brines according to molar Na/CI ratios and Br/CI weight
ratios (fig. 27} is supported by other chemical constituents, such as Ca, Mg, K, and I (modified from Richter
and Kreitler, 1986a,b).
70
-------
20,0-
"o
O>
5
i
•5
15.0-
_i
e io.o-
0*
Eicn
~£
o
5.0-
•
l .O1
C
ZO.O-
GROUP A
/
/
'•
,
•
••
1.5-
5-
4
1
\
'x
6
X
X
o
J
O"
+_
"o
—"
;.-..-. _,,-- o
x
15.0-
_l
^
o
e 10.0 •
0*
to
. . ; '5.0 -
;• • . ^.--'
.'--'''
Lw. --'' in.
%"
) .5 1.0 1.5 , 2.0
No(mol/L)
C
.
^ •
1 •
9
GROUP C
•
!
. ' *,•!.. .-. • •
"^
1
) .5 1.0 1.5 2.0
No(mol/L)
CKmol/L) ' CI(mol/L) • ?aJ86«
Rgure 29. Molar ratios of (Ca+Mg)/SC>4 versus Na/CI in salt-spring and shallow subsurface brines, Rolling
Plains of Texas (from Richter and Kreitler, 1986a,b). Molar ratios close to one suggest chemical control by
halite and gypsum in Group A waters. Large deviations from unity are typical for deep-basin brines and
suggest such a source in Group C waters.
71
-------
(a)
6-
12-
18
CL
CD
Q
24-
30-
36
-10
samples from same well
-8 -6 -4-2 0 2
S'8 O(SMOW)
rO
-25
-50.
a.
a>
-75
-100
(b)
6-
12-
Q.
a>
a
24-
t 30-
36'
10
samples from same well
f I I I I I I |
100
r i i i i T ) T i i i i i n~j I i
1,000 10,000
CKmg/L) - -
100,000
-0
-25
-50.
a.
-------
•
-
£ 100-
2 :
S I
o>
o
X
5 10.-
.
-
1 -
1
eP
0
•*. ° ° °° ft 0%
^X^ * ° ooftS*?^
* ^^ « & "^ fo ^0»'• ^
• *« ^^^^ %
m Fresh water, saline formation water, * CQO ,.,otQr ^ >.
• and halite-solution brines aeawaier ^^
o Oil-field brines
0 100 1,000 10,000 100,000 1,000,000
C\ (mg/L)
QA17223C
Figure 31. Differentiation between halite-solution waters and oil-field/deep-basin brines using differences
in I/CI weight ratios (from Whrttemore and Pollock, 1979).
73
-------
10.000
1,000
o
tn
100-
10
10
Average
halite-solution
end member\
Average
fresh-water
end member
Average
oil-field brine
end member
100
1,000 10,000
Cl (ppm)
100.000
1,000,000
QA17224C
Figure 32. Theoretical mixing curves for fresh water and potential salinization sources in Kansas using
sulfate and chloride concentrations (from Mast, 1982).
74
-------
3.2.4. Significant Parameters
Halite solution produces some of the lowest Br/CI ratios found in natural salt waters. Ratios typically
are less than approximately 10 x 10~4 in halite-solution brines and greater than 10 x 10~4 in oil-field
brines and formation brines (Whittemore and Pollock, 1979; Whittemore, 1984; Richter and Kreitler,
1986a,b). Ratio differences between these two potential mixing endmembers with fresh water are
generally big enough to allow differentiation of the respective source in brackish water down to chloride
concentrations of a few hundreds of milligrams per liter, although identification is best at high
concentrations. Sea water also has a much higher Br/CI ratio than halite-solution brine, which could allow
differentiation between halite solution and sea-water intrusion in coastal salt-dome areas.
The ratio of Na/CI works well to distinguish halite-solution brine from oil-field brine at high chloride
concentrations. Sodium and chloride occur in halite at equal molar concentrations (Na/CI molar = 1, Na/CI
weight = 0.648). Brines that originate from halite solution within a shallow ground-water flow system will
exhibit a similar ratio as long as concentrations are high enough so that the Na/CI ratio is not affected
appreciably by ion exchange reactions. In most oil-field brines molar Na/CI ratios are much less than unity.
Exchange of calcium and magnesium for sodium on clay mineral surfaces and alteration of feldspar may
account for the low ratios in formation brines and oil-field brines. The Na/CI ratio is also much smaller in sea
water (mNa/mCI» 0.85) than in halite-solution brine.
Ratios of I/CI in halite-solution brines are typically small and less than oil-field/deep-basin brines,
which allows separation between these two major sources of salt water.
Halite deposits are often associated with abundant beds of gypsum and anhydrite. Dissolution of
these beds is reflected in molar (Ca+Mg)/SO. ratios close to unity, which is much smaller than the
respective ratio in oil-field brines (»1) and in sea water (2.3).
3.2.5 State-by-State Summary of Halite Occurrences
This section provides a state-by-state summary of some of the halite-solution occurrences in the
United States (Fig. 2), as compiled from published sources.
Alabama: Bedded and diapiric occurrences of halite are known in the southwestern part of the
state. All major aquifers in Marango County are affected by upward migration of brine along faults, which is
probably related to salt-dome movement (Slack and Planert, 1988). Solution mining near Mclntosh,
Washington County, is from a salt dome where evaporites occur at 400 ft below land surface (Dunrud and
Nevins, 1981).
Arizona: Large quantities of salt are discharged in the state from mineral springs and thermal
mineral springs. For example, three major springs in the Mogollon Rim area of central and east-central
Arizona discharge daily TDS toads of 200 tons (Fuhriman and Barton, 1971). Clifton Hot Springs
discharges saline water of up to 9,000 ppm into the San Francisco River, and mineral springs near the
75
-------
mouths of Cheveton and Clear Creeks near Winslow contribute saline water to the Little Colorado River
(Krieger and others, 1957). Five areas in the state are known to be underlain by halite, the largest of which
is the Holbrook Basin in the eastern part of the state. There, depth to the top of the halite-bearing Supai
Formation ranges from 650 to 800 ft below land surface (Pierce and Rich, 1962). Other areas where halite
is known in the subsurface are the Red Lake and Detrital Wash areas in the western and Luke and Picacho
in the south-central parts of the state. In the Red Lake and Detrital Wash areas, salt occurs at depths of
420 to 600 ft below land surface (Pierce and Rich, 1962). Top of evaporite deposits at Luke, where
solution mining is practiced, is at 875 ft below land surface (Dunrud and Nevins, 1981).
Arkansas: Halite occurs in the Buckner and the Eagle Mills (Louann salt) Formations of southern
Arkansas.
Colorado: Halite occurs at shallow depth in several parts of the state. Selected wells show depth to
top of salt ranging from 395 ft to 6,485 ft below land surface (Pierce and Rich, 1962). An estimated
205,000 tons of salt from salt-solution enters the Dolores River of southwest Colorado annually through
springs and seeps along the channel bottom (Jensen, 1978).
Florida: Halite occurs in the deep subsurface of southwestemmost Florida, at depths greater than
10,000 ft (Dunrud and Nevins, 1981).
Idaho: Halite occurs at or near land surface in the Crow Creek area (Pierce and Rich, 1962).
Kansas: Shallow Permian salt underlies most of the central and southwestern parts of the state.
Salt springs, salt marshes, or direct contact with salt-bearing strata have contaminated many surface
waters, such as the Smoky Hill, Arkansas, Ninnescah, Saline, and Solomon Rivers (Krieger and others,
1957; U.S. Geological Survey, 1984). Much of the saline ground water is discharged from the Wellington,
Dakota, and Kiowa Formations, and the Permian Red Beds. The most widespread zone of salt is the
Hutchinson salt member of the Wellington Formation. The salt bed, which is up to 350 ft thick, is easily
dissolved by circulating ground water. The most damaging effects from salt dissolution are along the
Smoky Hill and Solomon Rivers in eastern Saline and western Dickinson Counties and the Belle Plain area
in Sumner County on the Ninnescah and Arkansas Rivers and the Slate and Salt Creeks (Gogel, 1981).
Other areas contaminated by salt in Permian Red Beds are found along the lower reaches of the
Rattlesnake Creek in Stafford and Reno Counties and along the South Fork Ninnescah River in eastern
Pratt and western Kingman Counties. A well-documented example of chloride and suit ate intrusion from
the Dakota and or Kiowa Formations is along the Smoky Hill and Saline Rivers in Russell County
(Hargadine and others, 1979).
Ten minor areas of mineral intrusion occur along (1) the Solomon River in Mitchell and Cloud
Counties (for example, Waconda Spring west of the city of Beloit), (2) Salt Creek in Mitchell and Lincoln
counties, a tributary of the Solomon River, (3) Buffalo Creek in Jewell and Cloud Counties, (4) Salt Creek
and Marsh Creek in Republican County, (5) the Cottonwood River in Marion County, (6) the Medicine
Lodge River in Barber County, (7) the Salt Fork Arkansas River in Comanche and Barber Counties, (8) the
76
-------
Cimarron River in Meade County, (9) the Crooked Creek in Meade County, and (10) Cow Creek in Reno
and Rice Counties (Hargadine and others, 1979).
Hargadine and others (1979) studied six major areas of natural salt-water intrusion throughout the
center of the state from the Oklahoma state line to the Nebraska state line. These intrusions are
associated with the Dakota and/or Kiowa Formations in north-central Kansas, with Permian Red Beds in
south-central Kansas, and with the Wellington Formation in east-central and south-central Kansas. Within
the Wellington Formation, halite dissolution is prominent in the western part, whereas gypsum dissolution
is dominant in the east.
In the area south and west of Salina, fresh water moves down through collapsed and brecciated
shales to recharge the Wellington aquifer, dissolving halite within the Hutchinson Salt Member. The
resulting brine then moves northward beneath the Smoky Hill River valley to the Saline River valley.
Upward discharge of brine occurs through collapse structures in an area from about three miles east of
Salina to just east of Solomon, where the potentiometric surface of the Wellington aquifer is above the
water table in the alluvia! aquifer (Gillespie and Hargadine, 1981). Between New Cambria and Solomon,
approximately 369 tons of chloride enter the Smoky Hill River (J. B. Gillespie, personal communication,
1981; in Qogel, 1981). According to landowners, the interface between fresh water and saline water in the
alluvial aquifer along the Smoky Hill River in northern Saline County is about 35 ft below land surface.
Chloride concentrations of 48,000 mg/L at 94 ft depth and of 180,000 mg/L in gypsum cavities at 112 ft
have been reported. A relatively impervious shale separates the shallow fresh-water unit from the deeper
brine unit except for areas of collapse and brecciation caused by salt and gypsum dissolution (Gillespie
and Hargadine, 1981).
Some of the brine originating from salt dissolution in Kansas does not discharge in Kansas but
discharges instead in neighboring Missouri to the east. According to Banner and others (1989), regional
ground-water flow that starts with meteoric recharge in Colorado picks up high loads of dissolved solids
from halite-bearing formations in the subsurface of Kansas and finally discharges in central Missouri.
Most of the old solution-mining operations that were started as early as the 1880's have been
abandoned. This includes the operations near Great Bend and Pawneee Rock in Barton County, near
Anthony in Harper County, near Kingman in Kingman County, near Nickerson in Reno County, near
Sterling in Rice County, and near Wellington in Sumner County (Dunrud and Nevins, 1981).
Louisiana: The entire state is underlain by halite with major salt-dome provinces along the coast
and in the northwest. Some of the salt domes are close to land surface, such as Weeks Island, Jefferson
Island, and Avery Island, where tops of salt are at 97,69, and 15 ft below land surface, respectively (Pierce
and Rich, 1962). All three domes are mined for salt. Bedded halite generally does not occur within
approximately 5,000 ft below land surface (Dunrud and Nevins, 1981). Solution mining has occurred in
the past in the northwestern dome province but is now restricted to domes along the Gulf Coast. Top of
77
-------
evaporite deposits in mined domes ranges from few ft to 6,000 ft below land surface, with most production
going on since the early 1900's (Dunrud and Nevins, 1981).
Michigan: Some of the oldest salt deposits (Silurian) in the country underlie the Michigan
peninsula. Combined salt thickness ranges from 1,800 ft in the basin center, where the salt is
approximately 6,000 ft deep, to 500 ft at the basin margins, where the salt may be found at 500 ft below
land surface (Pierce and Rich, 1962). Old and modem solution mining is concentrated along the basin
margins. According to Pierce and Rich (1962), salt production in 1955 amounted to more than 5 million
short tons, most of which was produced by solution mining of salt. Rock salt production in 1953 was
approximately 1 million tons, mined from a salt bed 1,020 ft deep in Wayne County, near Detroit. Natural
brines are produced from rocks of the Detroit River Group at Manistee and Ludington in the western part
of the state, and at Midland, St. Louis, and Mayville in the eastern part of the state (Sorensen and Segall,
1973). Most of the current solution operations are at depths greater than 1,300 ft below land surface and
have been going on for approximately 100 years (Dunrud and Nevins, 1981).
In Manistee County, the combined production of salt from salt beds in 1898 was approximately
435 million pounds (Childs, 1970). Production is from salt beds in the top of the Detroit River Group
between 1,900 and 2,050 ft below land surface. Another producing horizon is at 3,700 ft. Modem
operations require 600,000 to 800,000 gal/day of water to run economically. Ground water in and around
the salt plants is contaminated with chlorides and is not a source of potable water (Childs, 1970). Backflush
fluids from salt galleries, which were dumped into open pits, are part of the problem. Since the 1980's salt
brines have also been continuously disposed of into Lake Manistee (Childs, 1970).
Mississippi: All of the southern part of the state is underlain by salt bed and salt domes. Solution
mining has been practiced at Petal, Forrest County, and Richton, Perry County, but these operations
have been abandoned (Dunrud and Nevins, 1981).
Montana: Salt beds within the Williston Basin of northeastern Montana generally occur at depths
greater than 4,000 ft below land surface (Pierce and Rich, 1962).
Nebraska: Halite beds occur in southwestern Nebraska at depths greater than 3,200 ft in parts of
the Northern Denver Basin (Pierce and Rich, 1962; Dunrud and Nevins, 1981).
Nevada: In the southeastemmost part of the state some domelike occurrences of salt are found in
the Muddy Creek Formation along the Virgin River. These occurrences used to be exposed at land
surface but are now covered by the Overton arm of Lake Mead (Pierce and Rich, 1962).
New Mexico: Halite underlies the state along its eastern state line, where salt thickness in the
Permian Basin is greatest (up to 2,800 ft aggregate thickness) (Pierce and Rich, 1962). Dissolution of
shallow salt on this western edge of the Permian Basin causes salinizatton of the Pecos River and ground
water in that portion of the state. At Malaga Bend, inflow of spring water to the Pecos River contains up to
270,000 ppm of TDS. Salt occurs in the Guadalupe and Ochoa series, where it is associated with gypsum
78
-------
and anhydrite. In southeastern New Mexico, potassium salt with an average thickness of 4 to 5 ft are
included in salt within the Salado Formation (Pierce and Rich, 1962).
New York: Halite underlies most of the western part of the state with a maximum aggregate salt
thickness of 800 ft just southeast of Seneca Lake (Pierce and Rich, 1962). From there, salt thins in all
directions. Associated with the salt are traces of sylvite, polyhalite, and camallite, all potassium salts (Ailing,
1928). Salt beds and highly mineralized water underlie areas of Chemung County (Gass and others, 1977)
and Cattaraugus County (U.S. Geological Survey, 1984), arid salt-water springs are common throughout
western New York in the outcrop areas of Silurian rocks, which contain halite deposits in the subsurface.
According to Grain (1969), most of these springs are not associated with deep formation waters or oil-field
brines. Brines from deeper saline sources may also contaminate ground water in the Susquehanna River
Basin (Gass and others, 1977). Most of the solution-mining operations in the state, starting as early as in
the 1880's, have been abandoned. According to Dunrud and Nevins (1981), mining continues at Tully,
Onandaga County, at Watkins Glen, Schuyler County, at Ithaca, Thomkins County, and at Dale and Silver
Springs, Wyoming County. Depths to evaporite strata in those mining areas range from 1,200 to 2,500 ft
(Dunrud and Nevins, 1981). In the Jamestown area, halite beds at 1,500 to 2,000 ft below land surface are
penetrated by numerous abandoned gas wells, which are suspected to allow vertical migration of salt
water into shallow fresh-water aquifers (Grain, 1966; Miller and others, 1974).
North Dakota: The western part of the state that is known as the Williston Basin is underlain by
halite at depths greater than 3,600 ft below land surface (Pierce and Rich, 1962). Dunrud and Nevins
(1981) list one solution-mining operation in the state, located near Williston in Williams County, where the
top to evaporite deposits is 8,250 ft.
Ohio: Halite deposits are thickest along the eastern state line, where the combined thickness of all
beds approaches 300 ft (Pierce and Rich, 1962). Salt deposits thin toward the west. Depth to salt is
approximately 1,000 ft below land surface, with solution mining having been practiced at several locations.
According to Dunrud and Nevins (1981), only two of the original seven mining operations are still active in
the state. Those active operations are located near Akron in Summit County and near Rittman in Wayne
County. Depth to evaporites in those areas is approximately 2,700 ft (Dunrud and Nevins, 1981).
Oklahoma: All of the west-central and northwestern parts of the state, including the Oklahoma
Panhandle, are underlain by halite, gypsum, and anhydrite. Halite dissolution is evidenced in salt springs
and in shallow saline ground water in those areas, as documented in areas such as the Cimarron River near
Mocane (Krieger and others, 1957), Perkins (Leonard and Ward, 1962), or Dover in Kingfisher County
(Oklahoma Water Resources Board, 1975). Halite lenses in the Flowerpot Shale within 600 ft of land
surface have in the past been dissolved at most places by percolating ground water, but occasionally salt
deposits still exist, such as those in the Little and Big Salt Plains at the northwestern end of the Cimarron
Terrace (Oklahoma Water Resources Board, 1975).
79
-------
Solution mining at 1,500 ft below land surface has been practiced on and off at Sayre, Beckham
County. Pumping of salt water, from a producing interval of 12 to 558 ft at the site of a former brine spring,
is practiced in Harmon County. Richter and Kreitler (1986a,b) reported a chloride concentration of 58,000
mg/L for the produced brine, which is only about 30 percent of the chloride concentration measured in
two natural brine springs in the area.
Pennsylvania: All of the western and northern parts of the state are underlain by halite at depths
greater than 1,000 ft. Salt beds are up to 200 ft thick, with aggregate thickness of up to 650 ft, along the
western state line, but thin toward the east (Pierce and Rich, 1962).
Texas: Many areas of the state are underlain by halite deposits, with major dome provinces along
the Gulf Coast, In northeast Texas, and in South Texas. Solution mining occurs predominantly in those
dome provinces where depths to salt are relatively small. For example, depth to salt in Brooks Dome,
Smith County, is 220 ft; in Grand Saline Dome, Van Zandt County, 212 ft; and in Palestine Dome,
Anderson County, only 140 ft below land surface (Pierce and Rich, 1962). According to Dunrud and
Nevins (1981), solution mining is still practiced at 12 sites in the state, ten of which produce from salt
domes in Brazoria, Chambers, Duval, Fort Bend, Harris, Jefferson, Matagorda (all Gulf Coast), and Van
Zandt (northeast Texas) Counties, and two produce from salt beds in Ward and Yoakum Counties.
Dissolution of salt under the High Plains Escarpment causes brine discharge in the Rolling Plains of
north-central Texas and degrades surface-water quality in the Brazos and the Red River for hundreds of
miles downstream (Richter and Kreitler, 1986a,b). Other affected surface streams in the area include the
Pease and Wichita Rivers, the Double Mountain Fork of the Brazos, and the Prairie Dog Town Fork, and
Salt Fork of the Red River. More than 3,500 tons of sodium chloride are discharged from salt seeps and
springs to these rivers every day (Gustavson, 1979). Some tributaries to these rivers originate in salt flats,
such as at Dove Creek and Haystack Creek in King County, Short Croton and Hot Springs in Kent County,
and Jonah Creek in Childress County (Richter and Kreitler, 1986a). Gypsum and anhydrite beds are
generally associated with halite beds, contributing to the overall salinization of ground water in those
areas. At least some of the high salt content in the Pecos River of West Texas is due to discharge of halite-
solution brine from Permian formations in New Mexico.
Mining of salt domes for commercial use (for example, brine, oil, gas, and waste disposal) has been
and is being done in many salt domes along the Gulf Coast and in East Texas. Locally, these activities
have caused subsidence and formation of sinkholes, such as at Boling Dome and Orchard Dome along
the Gulf Coast (Mullican, 1988) and at Palestine Salt Dome in East Texas (Fogg and Kreitler (1980). At
Palestine Dome, ground water recharges at highland areas, moves downward to the dome, dissolves
halite, and discharges in topographic tows, such as a lake and nearby sinkholes (Fogg and Kreitler, 1980).
Saline ground water at other domes, such as southeast of Mount Sylvan Dome and east of Whitehouse
Dome and Bullard Dome in East Texas, may be associated with processes from the geologic past, such as
previous salt-dome solution, or they may represent depositional, marine waters but not current solution of
80
-------
domal halite (Fogg, 1980). Under certain conditions, such as at Oakwood Dome, East Texas, dissolution
of salt may be stow due to a protective layer of caprock that prevents circulating water from contact with
halite and tow permeabilities that prevent significant ground-water circulation and with it significant removal
of salt (Fogg and others, 1980). Ground-water salinity caused by natural mixing of fresh ground water with
halite-solution water may be enhanced by drilling and brine disposal activities, as suggested by Hamlin and
others (1988) for the tower Chicot aquifer south and west of Barber Hill Salt Dome (Chambers County).
Since 1956, an estimated 1.5 billion barrels of brine have been injected into porous zones into this
dome's caprock, with the present rate being 1 to 5 million barrels per month (Hamlin and others, 1988).
Utah: The Sevier River Basin in central Utah is underlain by halite and gypsum, which have
contributed to the high TDS content in the Sevier River (Whitehead and Feth, 1961). Near Redmond salt
is mined in open pits from the Arapien shale in the Sevier River Valley (Pierce and Rich, 1962). In the
southeastern part of the state, in the Paradox Basin, depth to top of salt in selected .wells ranges from 883
to 12,200 ft below land surface. There, sylvite and camallite are associated with some of the halite
deposits (Pierce and Rich, 1962). .
Virginia: A shallow occurrence of salt exists in southwestern Virginia. Pierce and Rich (1962)
specify a depth below land surface of 800 to 2,000 ft for this occurrence, whereas Dunrud and Nevins
(1981) specify a depth of 165 to 3,300 ft. Between 1895 and 1972, solution mining, which caused local
subsidence, was practiced at this shallow occurrence of salt (Dunrud and Nevins, 1981).
West Virginia: Halite that underlies the northern part of the state is generally restricted to depths
greater than 5,000 ft below land surface (Pierce and Rich, 1962). Nevertheless, salt is mined at four sites
located near Moundsville and Natrium in Marshall County, near Bens Run in Tyler County, and near New
Martinsville in Wetzel County (Dunrud and Nevins, 1981). All of these operations are relatively recent
(post-1942), which is in contrast to other, mostly much older, salt-mining operations in the country.
Wyoming: Halite occurs in the subsurface in the northeast (Powder River Basin), in the southeast
(Northern Denver Basin) and in the southwest. Near the Idaho/Utah border, salt was encountered in one
well at 125 ft below land surface (Pierce and Rich, 1962). In the other areas, depth to top of salt is much
greater, being more than 2,700 ft in the Northern Denver Basin and more than 6,000 ft in the Powder
River Basin (Dunrud and Nevins, 1981).
3.3 Sea-Water Intrusion
3.3.1. Mechanism
Sea-water intrusion has been reported from almost all coastal states in the country. In many
instances, ground-water contamination from sea-water intrusion has forced abandonment of water wells or
entire well fields, and much money has been spent to prevent further intrusion. Florida is the most
81
-------
seriously affected state, followed by California, Texas, and New York (U.S. Environmental Protection
Agency, 1973).
Under natural conditions, ground-water flow in unconfined and shallow-confined coastal aquifers is
toward the oceans because of flow potentials driven predominantly by topography. Generally this flow has
caused flushing of saline water from coastal aquifers that may have occupied the aquifer since deposition
since the last high stand of sea level, or since the last major storm flooded the coast. For example, lenses
of fresh water extend as far as 75 miles from shore or to the edge of the shelf off the Florida Peninsula
(Manheim and Paull, 1981). These lenses, which shield coastal ground water from saline encroachment,
originate from fresh-water recharge during the exposure of the continental shelf during the Pleistocene
glacial maximum (Manheim, 1990). Along the Gulf Coast of southwestern Louisiana, saline water was
flushed gulfward during the Pleistocene, but moved landward through highly transmissive zones as sea
level rose again (Nyman, 1978). The degree of flushing depends on many factors, such as the amount of
recharge, the hydraulic head, and the permeability of the aquifer. For example, in the Houston area of
Texas, fresh water has flushed the original salt water out of the aquifer to a depth of more than 2,200 ft,
whereas just 50 miles toward the coast in the Galveston area salt water in the shallow aquifer has been
flushed out to a depth of only 150 ft (Jorgensen, 1977). This process of flushing may have been more
effective in the past during lower stands of sea level (higher fresh-water heads), as sea level fluctuated
between present stand and 300 ft below present stand during the past 900,000 years (Meisler and
others, 1985).
Under natural conditions, flow in unconfined and shallow-confined aquifers responds to slow
changes in sea level relatively quickly. If fresh-water heads are high enough, all salt water in an aquifer
open to the sea may be flushed out and a fresh-water spring may exist where this aquifer unit crops out at
the sea floor. If fresh-water heads are lower than salt-water heads, a reversal in flow occurs, with salt water
replacing fresh water. If fresh-water heads are not large enough to replace the salt water, a dynamic
equilibrium will exist between the sea water and the fresh water (Hubbert, 1940). Kohout (1960)
demonstrated that the dynamic equilibrium involves flowing sea water in the Biscayne aquifer of Florida,
whereby sea water is returned to the sea. A transition zone, or zone of diffusion of variable thickness,
exists between the two bodies of fresh water and salt water (Fig. 33). Tidal action and its impact on salt-
water movement, as well as changes in the fresh-water potentiometric surface due to variation in fresh-
water recharge or discharge, are the principal mechanisms governing the position and shape of this
transition zone (Cooper and others, 1964).
Conditions may be somewhat different in deep confined aquifers, with regional ground-water flow
reacting more slowly to changes in sea-level elevations. Such a case of slow equilibration, and the
resulting present landward flow of saline water, was proposed by Meisler (1989) for deep aquifers in the
Northern Atlantic Coastal Plain. Meisler and others (1985) estimated that the interface reflects a sea-water
level between 50 and 100 ft below current level and that the interface adjusts to present sea level by
82
-------
Lond surfoce
;OAI7225
Figure 33. Typical ground-water flow patterns along coast lines, with circulation of salt water from the sea
and back to the sea within a zone of diffusion (from Cooper and others, 1964).
83
-------
moving inland at a rate of approximately one foot per year. Depths to top and the thickness of the interface
or transition zone varies widely. For example, along the coast of the Northern Atlantic Plain, the transition
zone of salt water containing between 250 mg/L and 18,000 mg/L is between 400 ft and 2,200 ft thick,
reflecting long-term variations in the position of the fresh-water and salt-water zones (Meisler, 1989). This
zone is thinnest in areas of fresh-water discharge (along coastal rivers and bays) because of low fresh-
water heads in those areas. The top of the transition zone is shallowest in the southern part of the
Northern Atlantic Plain, with the 250 mg/L contour at 200 to 400 ft below sea level. In the northern part of
the Atlantic Plain, the same contour is at 2,000 ft below sea level, as fresh water extends farther onto the
continental shelf in the north than in the south (Meisler, 1989).
The transition zone between fresh water (Cl <250 mg/L) and sea water (Cl >18,000 mg/L)
In the Biscayne Aquifer at Miami, Florida, is only approximately 50 ft thick, reflecting the influence of tidal
fluctuations and fresh-water discharge fluctuations as well as hydraulic connection between sea water and
fresh water (Kohout, 1960). Within the transition zone some of the sea water may again be discharged to
the sea (Fig. 33). Kohout (1960) suggested that under certain conditions approximately 20 percent of the
intruding sea water discharges back to the sea through the transition zone in the Biscayne aquifer of
Florida. Because of this flow in the transition zone, sea water must constantly intrude the aquifer (Cooper
and others, 1964). Under certain conditions such as low fresh-water heads and inland flow of sea water
along sinkholes and solution openings in carbonate aquifers, this mixing between fresh water and salt
water may be encountered far enough inland to cause discharge of the brackish transition-zone water
along the coast (Rg. 34) (Stringfield and LeGrand, 1969; Cotecchia and others, 1974).
Sea-water intrusion is not confined to the lower parts of aquifers, but also can occur in the upper,
shallower sections of an aquifer, where strong storms cause flooding of coastal zones. Such a scenario is
known along the coastal dunes aquifer of the northwestern United States, where winter storms create a
several foot thick sea-water zone overriding fresh ground water (Magaritz and Luzier, 1985). The natural
position of the interface between ground water and sea water is affected by tidal action, with fluctuations
on a sandy beach vertically near 5 ft and horizontally 200 ft (Urish and Ozbilgin, 1985). This can create a
mounding of sea water and consequently an effective mean sea level of close to 2 ft higher than local
mean sea level (Urish and Ozbilgin, 1985).
The natural equilibrium of shallow coastal aquifers can be disturbed when the flow of fresh water
decreases either in response to heavy pumping from water wells or in response to decreased recharge to
the aquifer. Increased urbanization along United States coastlines have caused both of these
mechanisms to occur in many places. Large areas have been eliminated as recharge points and today
much water that used to recharge aquifers ends up in storm sewers. At the same time, pumpage of
ground water from coastal aquifers in many areas has lowered water, levels from originally above sea level
and flowing conditions to sometimes tens or hundreds of feet below sea level, resulting in a reversal of
ground-water flow and the potential of salt-water intrusion in the pumped area if heads are below the
84
-------
£i"Vn
2oo-*-6o
15 km
QAI7226
Figure 34. Schematic cross section showing mixing mechanism of sea water with fresh water through
sinkholes and solution openings in a carbonate aquifer (from StringfiekJ and LeGrand, 1969).
85
-------
submarine outcrop of the pumped unit or if the transition zone between fresh water and salt water falls
within the cone of depression. For example, the city of Galveston used to produce fresh water from a
800-ft-deep well on Galveston Island but had to abandon the well in 1896 because of salinization. A new
well field had to be established 15 miles inland (Turner and Foster, 1934). This early occurrence of sea-
water intrusion may not be typical, but as ground-water withdrawals have increased in response to
expanded demands, forced abandonments of water wells due to salinization have become widespread in
many coastal areas. It is estimated that one to two public wells and 20 to 30 domestic wells are lost each
year in Delaware as a result of salinization due to sea-water intrusion (Atkinson and others, 1986).
Increased domestic, agricultural, and industrial consumption of surface water from coastal streams
can also lead to intrusion of sea water upstream. Where pumpage from river alluvium is high, subsequent
intrusion of salt water into ground water can occur where the waterway is a losing stream. Dredging of
waterways can enhance this process whenever a formally sealed-off permeable unit, such as permeable
river alluvium, is encountered during dredging (Bruington, 1972). Such a case has been described by
Barksdale (1940) for the Parlin area, New Jersey, where pumpage has decreased heads from originally
flowing conditions to as much as 70 ft below land surface. Dredging of a new canal for shortening of the
navigation route along the South River allowed salt water to enter through the canal bottom into the local
aquifer, where it advanced at a rate of approximately one mile per six years (Barksdale, 1940; Schaefer,
1983). The sea-water Intrusion problem can be aggravated by leaky or corroded well casings through
which saline water can migrate to fresh-water aquifers (Miller and others, 1974; Monterey County Flood
Control & Water Conservation District, 1989).
Another mechanism that adds salinity preferentially to coastal aquifers is sea spray during violent
storms. This mechanism causes chloride concentrations in the tens of milligrams per liter in precipitation
and deposits large amounts of sea salt along coastlines. For example, approximately 2.7 million tons of sea
salt are deposited annually over the entire area of Japan (Tsunogai, 1975).
3.3.2. Chemistry of Sea Water
Sea water contains approximately 35,000 mg/L of dissolved solids, with chloride and sodium
combining for 84 percent of the total concentration (Table 4). Salinity is somewhat higher in the Atlantic
Ocean (36,900 mg/L at latitude 25°N) than in the Pacific Ocean (33,600 mg/L at latitude 40°N), depending
on local circumstances such as continental influence, degree of evaporation, and oceanic currents
(Custodio, 1987). Much larger differences can be found elsewhere. For example, inflow of fresh water
reduces the salinity of the Baltic Sea to between 3,000 to 8,000 mg/L, whereas evaporation increases
salinity in some areas of the Mediterranean and the Red Sea as high as 45,000 mg/L (Custodio, 1987).
The influence of fresh-water inflow on water quality is illustrated by samples taken at increasing distance
86
-------
Table 4. Concentration of major and significant minor chemical
constituents in sea water (from Goldberg and others, 1971;
Hem, 1985) (concentrations in mg/L).
Constituent
Chloride, Cl
Sodium, Na
?"ifate, SO4
Magnesium, Mg
Calcium, Ca
Potassium, K
Bicarbonate, HCOa
Bromide, Br
Strontium, Sr
Silica, SiO2
Boron, B
Fluoride, F
Iodide, I
Concentration
19,000
10,500
2,700
1,350
410
390
142
67
8
6.4
4.5
1.3
0.06
Percent
54
30
8
4
1
1
0.4
0.2
0.02
0.02
0.01
0.003
0.00017
54
84
92
96
97
98
98.4
98.6
98.6
98.6
98.6
98.7
98.7
87
-------
from the shore along the Texas Gulf Coast (Table 5). Chloride concentrations increase from 3,200 mg/L at
the mouth of the Houston Ship Channel at Galveston Bay, to 13,000-14,000 mg/L three miles offshore
and to 18,000 mg/L 25 miles offshore (Jorgensen, 1977).
As in most natural waters, the sum of all major cations and anions in sea water amounts to more than
98 percent of the total concentration of dissolved constituents (Table 4). However, the percentage of
individual constituents is very different from that in fresh water and many other natural waters. In sea water,
sodium is the dominant cation and chloride is the dominant anion, whereas in most fresh waters calcium is
the dominant cation and bicarbonate or sulfate are the dominant anions. Even more striking is the
difference in relation between calcium and magnesium concentrations. In sea water, the weight ratio of
Ca/Mg is approximately 0.3. In contrast, in most fresh waters the Ca/Mg weight ratio is greater than 1.0.
Also, potassium concentrations in most fresh waters are smaller than calcium concentrations by one order
of magnitude or more, whereas they are nearly the same in sea water (Table 4).
Water samples from various sea-water intrusion sites suggest two mechanisms that alter the
composition of the intruding sea water (Fig. 35). On bivariate plots of Ca versus Cl, K versus Cl, and Na
versus Cl, cation exchange is suggested by some of the samples, as calcium content is greater and
sodium and potassium contents are smaller than in the well-defined mixing trends indicated by other
samples and by other constituent plots (Fig. 35). Mixing trends are less well defined at chloride
concentrations of less than 1,000 mg/L because local fresh-water variations dominate over the relatively
uniform sea-water composition at low concentrations.
3.3.3. Examples of Geochemical Studies of Sea-Water Intrusion
Whenever sea-water intrusion is the only source of salt water in a given area, identification of this
salinlzation source in an affected well is indicated by an increase in total dissolved solids and possibly all
major cations and anions. This recognition poses little difficulty. The most-used tracer of simple sea-water
intrusion scenarios is the chloride ion, which is the most conservative, natural constituent in water once it
is in solution. After establishing the base-line (background) chloride concentration in a given area from
historical data or from wells that are not yet affected by salt water, periodic sampling of monitoring wells and
analysis of chloride is being done along many areas along the United States coast where the potential of
sea-water intrusion is feared. The importance of knowledge of background levels is illustrated in
sometimes subtle chloride increases, such as those measured in samples from Middlesex County, New
Jersey, where intrusion has been observed for the past 40 years. During the period of 1977 to 1981,
chlorinity increased in some wells already affected by intrusion as well as in others that had shown
background levels of less than 10 mg/L dissolved chloride in 1977 (Table 6) (Schaefer, 1983). Increases
in chloride content from background levels of less than 5 mg/L to greater than 600 mg/L also indicated
sea-water intrusion in response to heavy pumpage in some coastal wells of Monmouth County, New
88
-------
00
to
Table 5. Landward changes in chemical composition of Gulf of Mexico water as a result of dilution (data from Jorgensen, 1977).
Dissolved constituents in mg/L
Source of Water
Houston Ship Channel,
at Morgans Point
Gulf of Mexico, 3 miles
offshore of Sabine Pass
Gulf of Mexico, 3 miles
offshore of Sabine Pass
Gulf of Mexico, 25 miles
offshore of Sabine Pass
Date SIO2 Ca Mg Na K HCOa 804 Cl IDS Remarks
7/5/73 9.8 89 220 1,800 66 104 500 3,200 6,050 near surface
10/9/74 0.5 280 840 7,100 330 140 1,800 13,000 23,400 at 1-ft depth
10/9/74 0.4 300 880 7,900 250 140 2,000 14,000 25,400 at 24-ft depth
10/9/74 0.1 370 950 9,500 330 146 2,500 17,000 30,700 at 1-ft depth
Gulf of Mexico, 25 miles 10/9/74 0.2 380 940 9,800 320 148 2,400 18,000 31,900 at 42-ft depth
offshore of Sabine Pass
-------
100000
10000 -
£. 1000 •:
£
100 •!
10
10
i i>i*| T < i i Tm|- ( T-r—ntwrf
100 1000 . 10000
Cl (mg/L)
100000
10000
1000 •;
I* 100
o
10
• • • /!
10
100 1000
Cl (mg/L)
10000 100000
10000
1000 -
I
,§ 100 •:
f
10 '.
- "
10 100 1000 10000
Cl (mg/L)
100000
10000
1000
100
10
10
100
1000
Cl (mg/L)
• • "I ''•
10000 100000
10000 -a
1000 -,
J. 100
M
10 •:
,»
10 100 1000 10000
Cl (mg/L)
100000
f
.1 •?
.01-
O .001
ffl
.0001
10 100 1000 10000 100000
Cl (mg/L)
QA17227C
Figure 35. Bivariate plots of major ions and of Br/CI ratios versus chloride for sea-water intrusion samples
from California (data from Brennan, 1956), Israel (data from Arad and others, 1975), Hawaii (data from Mink,
1960), Spain (data from Price, 1988), and Texas (data from Jorgensen, 1977). Little scatter at chloride
concentrations greater than approximately 1,000 mg/L indicates little variation between sea-water
intrusion samples, with the exception of changes by cation-exchange reactions.
90
-------
Table 6. Changes in chloride concentration between
1977 and 1981 in the Farrington aquifer of New Jersey
as a result of sea-water intrusion (data from
Schaefer, 1983).
Clorlde concentration
in mg/L
Well name 1977 1981
PerthAmboyWDIA 6.3 32.0
Perth Amboy WD 2 49.0 54.0
South River BORO WD 2 7.2 12.0
South River Boro WD 5-77 7.0 12.0
South River Boro WD obs. 12.0 26.0
Sayreville Boro WDM 100.0 *190.0
Thomas and Chadwick 1 16.0 *42.0
El duPont - Parlin 1 7.5 45.0
El duPont - Parlin 3 47.0 96.0
Duhemal Water System 60F 680.0 1,300.0,
NL Industries 3 7.6 *55.0
* Sampled in 1979
91
-------
Jersey (Schaefer and Walker, 1981). There, intrusion was caused by heavy pumpage as hydraulic heads
declined to 45 ft below sea level in the center of the cone of depression.
In the Southampton, New York, area, background chloride concentration of fresh water is less than
20 mg/L. Concentrations of 200 mg/L to 13,890 mg/L indicate mixing of fresh water with sea water
(Anderson and Berkebile, 1976). A similar approximate limit, in some cases less than 20 mg/L in other
cases less than 50 mg/L of chloride, above which sea-water intrusion is suspected, was suggested by
Lusczynski and Swarzenski (1966) for Long Island, by Tremblay (1973) for Prince Edward Island, New
York, by Wilson (1982) for the Floridan Aquifer of west-central Florida, by Kohout (1960) for the Biscayne
aquifer of southeast Florida, and by Zack and Roberts (1988) for the Black Creek Aquifer of Horry and
Georgetown Counties in South Carolina. On the other hand, rainwater near coastlines may contain
chloride concentrations between 10 and 40 mg/L, and thus approach these background concentrations
in some areas (Custodio, 1987). During hurricanes, ocean spray may affect coastal zones many miles
inland, as documented during a 1938 storm event in New England, during which plants were damaged by
spray as far as 45 miles inland from the coast (Hale, 1973). Background chloride concentrations may be
higher in other areas, such as in the Monterey County area of California, where chloride levels range from
100 to 200 mg/L in uncontaminated wells (Fig. 36). Knowledge of background levels is useful in
identifying the movement of an advancing intrusion front. For example, between 1977 and 1981 the
continuous chloride increases in the Middlesex County area, New Jersey (Table 6) indicated a 0.2 to
0.4 mile inland migration of the transition zone between fresh water and salt water in just those four years
(Schaefer, 1983).
In case sea-water intrusion is not the only potential source of salinity in a given area, differentiation of
the respective salinization sources is more complicated and requires additional tracers or other tracers
than chloride. In the Castroville area of Monterey County, California, two sand and gravel aquifers (180-
and 400-ft aquifers) are affected by sea-water intrusion due to heavy pumpage (Monterey County Flood
Control & Water Conservation District, 1989). Sea water has intruded the shallow unit up to five miles
landward beginning in 1944 and the deep unit up to one mile landward beginning in 1959, causing
chloride concentrations in water wells to increase from background values of 100-200 mg/L to greater
than 1,000 mg/L (Rg. 36). Rgure 37 illustrates this landward advancement in a plot of distance of the 500
mg/L chloride contour from the coastline over time for the two aquifer units. Leakage of salt water can
occur from the shallow aquifer to the deep aquifer along wells penetrating both units. Using Piper
diagrams, this local downward leakage of brackish water can be distinguished from regional sea-water
intrusion (Monterey County Flood Control & Water Conservation District, 1989). The exchange of calcium
and sodium between aquifer matrix and intruding sea water is reflected by a curve (Rg. 38) in the diamond-
shaped field of the Piper diagram. This ion exchange proceeds in the intruding water until the aquifer
sediments are saturated with sodium (Rg. 38 a1), after which no additional calcium is released from the
matrix (Rg. 38 a2). Therefore, sea water that intrudes the aquifer behind this initial front will not exhibit the
92
-------
Q.
P
900-
800-
700-
600-
500-
400-
300-
200-
100-
0-
^ /
•' \
I
,'.'.
'/*
_^^A^\ 1 \'''tf""---S
^ **•**•*'*•' ^^ ^
~*"^
1 1 1 1 < 1 1 I 1 1 I 1 1 1 1 1 1 1 1 1 1 1 I 1 I 1 J 1 f 1 1 1 1 1
8 S .8
O> O) O> ' '
Year
QA1722BC
Figure 36. Increase in chloride concentration with time in coastal wells in Monterey County, California,
suggesting sea-water intrusion (from Monterey County Flood Control & Water Conservation District.
1989).
93
-------
25.
o
o
X 15-
•+• 180 ft aquifer
• 400 ft aquifer
71 I I I | l I i t i I l I I | I I I i I i I I ( | I I I I I I i I I | i l ( I I I I
to <£ r** co
Year
QA17229C
Rgure 37. Landward movement of sea water mapped by the position of the 500 mg/L chloride contour in
two shallow aquifers, Monterey County, California. Since 1944, sea water has advanced approximately
20,000 ft in the 180-ft aquifer to a position 25,000 ft inland from the coast (from Monterey County Flood
Control & Water Conservation District, 1989).
94
-------
Sea water
x 0.1
QA17230C
Figure 38. Piper diagram of chemical composition of coastal saline water, Monterey County, California.
Regional sea-water intrusion is characterized by ion exchange (at) and mixing (32), whereas local mixing of
intruded sea water with fresh water along boreholes is characterized by simple mixing (b) (modified from
Monterey County Flood Control & Water Conservation District, 1989).
95
-------
increase in calcium and decrease in sodium as observed in the advancing front. Often, this stage is not
reflected in readily available water-chemistry data because wells producing this kind of water most likely will
have ceased to be used unless the wells function as monitoring wells. According to Piper and others
(1953), sodium-saturated conditions occur in Southern California aquifers when IDS levels reach
approximately 2,000 ppm (Cl of approximately 1,000 ppm). This stage has been reached in some parts of
the shallow aquifer in the Monterey area. Leakage water from these parts in the shallow aquifer into the
deep aquifer are characterized by a simple mixing line between sea-water composition and fresh-water
composition (Fig. 38b). The same curve-shaped relationship in the diamond field was shown by Brennan
(1956) for waters affected by sea-water intrusion in the Manhattan Beach area of California (Rg. 39). In this
case, oil-field brines, the second potential source of salinization, were ruled out in favor of sea-water
intrusion because salinity in ground water increased toward the coast but not toward nearby oil fields. Oil-
field brine was also ruled out because the chemical composition did not correspond to mixing lines in the
diamond field and the triangular fields (Fig. 39).
Flushing of formation water may be less effective in some areas than in others or some coastal wells
may be drilled into saline water that represents old instead of modem sea-water intrusion. With respect to
water-resource management it may be of importance to be able to differentiate between modern sea-
water intrusion and past sea-water intrusion or between modem sea-water intrusion and formation water.
Howard and Lloyd (1983) pointed out that major chemical constituents are frequently inconclusive in
differentiating between these possible sources because of the chemical similarities of these sources to
present-day sea water. Different residence time of the salt water within the aquifer may hold the key for
distinguishing between these sources. Tritium content in recent continental water exceeds a few tritium
units (TU) whereas deep sea water has almost a zero TU content because of the long residence time of
sea water (Custodio, 1987). Therefore, mixing of continental fresh water with sea water should be
reflected in a lowering of the tritium content in the mixing water. Age dating of the waters using carbon
isotopes is another possibility for differentiating between modern and old sea water. Sea water has a
carbon-14 isotopic value close to that of modem organic matter (>80 percent modern; S14C = 0 permille
for recent, pro-bomb ocean water), whereas old sea water, connate or intruded, has a low carbon-14
content (Custodio, 1987). Hanshaw and others (1965) reported a carbon-14 value of +285 permille for
ocean water at Jekyll Island, Georgia, and a range from +4.2 permille (±7.6 permille) for surface samples to
-54.3 permille (±11.5 permille) for deep water masses from various parts of the North Atlantic Ocean.
These values are very different to values of fresh ground water (-965 to -987 permille) and for deep
saline ground water in the Claibome Group (-968 to -981 permille) in the Brunswick, Georgia, area. The
isotopic composition of contaminated ground water (-970 to -980 permille) in the area was identical to
that of fresh water and deep saline ground water, which suggests that recent sea water is not the source
of salinization in the contaminated water (Hanshaw and others, 1965).
96
-------
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
.15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
.30
A
B
15
22-26
12, 19-21, 27
, 14, 16-18
ppm
Cl
67
111
172
212
365
303
644
760
829
1,150
'1,580
3,400
6; 840
5,330
4,840
3,000
2,200
3,500
4,200
5,070
10,500
14,400
14,900
15,100
17,100
18,400
18,100
18,600
18,510
18,332
41
59
QA 1 7263c
Figure 39. Piper diagram of chemical composition of selected saline waters from the Manhattan Beach
area, California (from Brennan, 1956). Cation exchange is suggested by the curve in the center field, with
fresh water (No. 1) and ocean water (No. 28) as end points. Linear trends in the triangles also have ocean
water as the high-CI end member, as opposed to oil-field samples (No. 29, No. 30).
97
-------
Age dating in combination with I/CI ratios allowed Howard and Lloyd (1983) to differentiate three bodies of
saline water as three separate intrusions into the Chalk limestone of eastern England. Modern radiocarbon
ages were associated with minor iodide enrichment (I/CI * 4 x Kr6, as compared to I/CI = 3 x Kr6 for sea
water) and were interpreted as modem and active saline intrusion. Moderate iodide enrichment (I/CI = 3 x
10~5) was typical for shallow saline water related to Middle/Late Flandrian marine transgression and
radiocarbon dates of 5,000 to 8,000 years. Deep saline water, dated at 7,000 to more than 21,000 years
old and related to the Ipswichian irrterglacial stage, were characterized by high iodide enrichment (I/CI =
10~4). In this setting, the presence of iodide is an indicator for ground-water residence time as more
iodide is leached out of the sediments overtime (Rg. 40) (Lloyd and others, 1982). The amount leached
from the sediments also depends on the sediment type, as iodide concentrations in sandstones are
typically much lower (0.1-1 ppm) than those in argillaceous shales and limestones
(» 1 ppm); highest iodide concentrations are often associated with organic-rich marine deposits as well as
with evaporites and caliche deposits (Lloyd and others, 1982). The ratio of I/CI in sea water is
approximately 500 to 1,000 times less than the value typically measured in rain samples from the Hawaiian
atmosphere (Duce and others, 1965) and also one order of magnitude less than typical oil-field/deep-
basin brine ratios and typical fresh water ratios (I/CI >10~5; Whittemore and Pollock, 1979). This difference
may be useful to differentiate a sea water source from an oil-field/deep-basin brine source. The I/CI ratio in
sea-water is similar to the ratio found in many halite-dissolution brines (I/CI <10~5; Whittemore and Pollock,
1979; Richter and Kreitler, 1986a,b), which suggests that this ratio is not a good tracer for differentiating
between these two sources. However, in combination with boron and barium, iodide was a useful tracer to
distinguish sea-water intrusion from connate brine in a contamination study conducted by Piper and
others (1953) in California.
Modem and old sea water or formation water may also differ in their stable isotope composition of
oxygen-18 and deuterium as long as different climatic conditions existed during intrusion of the water
(Custodio, 1987). By convention, sea water has a composition close to the zero permille value because
•standard mean ocean water" (SMOW) is taken as a reference value against which all other samples are
measured. Fresh ground water is typically isotopically lighter because of fractionation processes that occur
during evaporation of sea water and condensation leading to precipitation. Water that has undergone
evaporation before entering local ground water, however, may be heavier than sea water. Also, rock-water
interactions at elevated temperatures tend to isotopically enrich the oxygen-18 content of the water
(Custodio, 1987). These differences may allow differentiation of salinization sources.
In a similar fashion, the value of the stable carbon isotope 13C differs between sea water and
continental water. Mook (1970) reported 813C concentrations between 1.5 and 2.0 permille for sea water,
whereas continental waters are isotopically lighter with 13C of about -13 permille. According to B. B.
Hanshaw (personal communication, 1976; in Jorgensen, 1977), a 813C value of -1 permille would be
expected in sea water and a 813C range of -12 to -25 permille would be expected for ground water in the
98
-------
10'4-
D)
I.
^ 10'5'
10-6-
o o
00
Old saline
ground
o waters
o o
0 o
o
Iodine
enrichment
Modem saline
ground waters
North Sea water-dilution line
i i i i i i i r ii i i i
2 A 6 8 10 12
Cl (g/L)
OA17231C
Figure 40. iodide concentrations as indicator of residence time. The longer the residence time, the more
leaching of iodide from sediments can occur, giving rise to higher concentrations than in waters with small
residence times (modified from Lloyd and others, 1982).
99
-------
Houston area. Testhole samples, ranging from -9.4 to -12.9 permille, appear to be inconclusive
concerning a sea-water source (Jorgensen, 1977). Age dating, using tritium, and the use of the stable
isotope ratios 518O,62H, S^S, and 513C were also inconclusive with respect to the most likely source of
salinization of ground water along the Houston Ship Channel. Instead, major cations and anions, depicted
in Piper diagrams and Stiff diagrams, appeared to best reflect the potential sources of salinity in that area
(Houston Ship channel in the shallow subsurface, formation water in deeper units). Although water
samples had been analyzed for many minor and trace elements such as Br, I, Li, Fe, B, K, F, and SiCL,
Jorgensen (1977) did not make use of or discuss these constituents as possible tracers.
Another ratio indicative of residence time may be the ratio of Ca/Mg. The Ca/Mg weight ratio of 0.3 in
sea water is very small when compared to the characteristic value of >1 found in most fresh waters and
brines. Ion exchange between intruding sea water and fresh-water aquifer matrix results in an increase in
the Ca/Mg ratio when compared to the ratio of sea water. This increase may be more pronounced in older
waters than in modern waters, which led Sidenvall (1981) to propose that saline water in the Uppsala,
Sweden, area is fossil ground water representing the last sea transgression.The ratio of CI/SO4 was used
by Snow and others (1990) to distinguish sea-water intrusion from road-salt contamination in Maine, based
on the fact that sutfate concentrations and CI/SC>4 ratios are substantially lower in road-salt affected well
waters than in sea-water affected well waters. This ratio can also be used to distinguish modem sea-water
intrusion from previous intrusion in waters having chloride concentrations greater than 500 mg/L, as done
by Pomper (1981) in the Netherlands. As a result of sulfate reduction, older saline waters are
characterized by Cl/SC-4 ratios that are higher than those found in modem seawater. In a similar fashion,
Martin (1982) explained the source of saline water in the Santa Barbara, Caifomia area as ocean water that
had undergone ion exchange and sulfate reduction. In addition to the Cl/SO4 ratio, the ratios of B/CI and
Ba/CI appeared to indicate that sea water is the source of the deep saline ground-water in that area (Martin,
1982). The SO4/CI ratio also seems to indicate mixing between sea water and recharge water in the
Ftoridan aquifer of southwest Florida, as shown by samples that plot close to the theoretical mixing line in a
plot of SCyCI ratios versus 804 concentrations (Fig. 41) (Steinkampf, 1982). However, gypsum-
anhydrite solution causes sulfate concentration to increase downgradient in some of the samples, limiting
the use of the SO4/CI ratio as a tracer of mixing. Because of the generally high degree of water-rock
interaction in this carbonate-dominated aquifer which governs concentrations of calcium, magnesium, and
bicarbonate and masks the source of salinity, the conservative nature of the chloride and bromide ions
was used by Steinkampf (1982) to demonstrate that dilution of marine-like ground water is a significant
mechanism in the evolution of satine ground water in this aquifer. Past marine inundations, after saline
interstitial waters had been flushed from the sediments, are the probable source of salinity (Steinkampf,
1982).
The Br/CI weight ratio in sea water is approximately 3.3 x 10"3, which is comparable to most fresh
waters. Therefore, mixing of sea water and fresh water may cause only a slight change in Br/CI ratio in the
100
-------
100:
10.0:
.0
2
?
.§. 1.0:
o ;
0* :
cn
0.1 =
0.01-
0.1
Sea water
1.0 10.0
SO4 (mmol/L)
OA17232C
Figure 41. Bivariate plot of SO^CI ratios and 864 concentrations for coastal saline ground water of
southwest Florida. Sea-water intrusion is Indicated in those samples that plot close to the theoretical
mixing line between sea water and fresh water (modified from Steinkampf, 1982).
101
-------
mixing water when compared to the fresh-water value. For example, Brennan (1956) reported Br/CI ratios
of 3.4 x 10"3 for Pacific Ocean water and a range of 3.2 x 10~3 to 3.7 x 10~3 for sea-water intrusion
samples in the Manhattan Beach area, California. However, because of the conservative nature of Br and
Cl, this ratio can often be used to differentiate between salt-water sources. For example, halite-dissolution
brine is typically characterized by Br/CI ratios smaller than 5 x 10~4, which is one order of magnitude
smaller than the value in sea water. On the other hand, oil-field brines often have ratios of Br/CI that are
significantly higher than sea-water and halite-dissolution brine values (Whittemore and Pollock, 1979;
Richter and Kreitler, 1986a,b; Whittemore, 1988). Therefore, the Br/CI ratio may be useful to differentiate
between sea-water intrusion and oil-field contamination in coastal settings that include oil fields. In settings
that include storage and application of salt for road deicing, such as along the coast in the northeastern
part of the United States, the Br/CI ratio can also be used to distinguish sea-water intrusion from road-salt
contamination. Saline ground water originating from past or current intrusion of sea water contains
detectable bromide with Br/CI ratios similar to sea water, whereas saline ground water originating from road
saft typically contains little bromide with Br/CI weight ratios less than 1 x 10"3 (Snow and others, 1990).
Ion exchange and carbonate dissolution are major chemical reactions that alter the composition of
intruding sea water. This is the case even in aquifers that contain only small amounts of clay and
carbonates, as shown in a study of rock-water interactions and seawater-freshwater mixing in coastal dune
aquifers of Oregon. Magaritz and Luzier (1985) found that of all the major ions, chloride was the only one
to act conservatively although the aquifer consisted of rather uniform quartz sand with only minor amounts
of sift, clay, and some organic matter. Sulfate reduction, oxidation of plant material, formation of authigenic
K-feldspar, and Ca-Na base exchange on clay minerals cause the composition of the saline water to differ
considerably from the theoretical composition of diluted sea water in the mixing zone. Nadler and others
(1981) summarized processes occurring in the transition zone within sandy sediments as being (a) dilution
of sea water, (b) Ca-Mg exchange, (c) Na-Ca or Na-Mg base exchange, and (d) sutfate reduction. The same
was shown by Mink (1960), who studied concentration changes in sea water as the water moves through
calcareous and alluvial deposits on the ocean bottom before entering the basaltic aquifer of southern
Oahu, Hawaii. Even at a very small degree of dilution, indicated by a chloride concentration of 17,600 ppm
in the intruded water as compared to 19,000 ppm in open sea water, changes in major cations are very
noticeable, with the calcium concentration increasing by more than 100 percent and the potassium
concentration decreasing by more than 50 percent (Mink, 1960). Similarly, sea-water intrusion samples in
the coastal zone of Fuji City, Japan, do not reflect the theoretical mixing composition of sea water and
fresh water. Instead, ion exchange and solution-precipitation reactions cause (1) higher Ca content,
(2) slightly higher SO^ content, and (3) lower Na, K, and HCO3 content than expected from simple mixing
(Ikeda, 1967). The increase in Ca content with time was used by Jacks (1973) to distinguish old sea water
from young sea water, based on the assumption that old water had more time for ion exchange reactions
to occur and, therefore, would be characterized by higher Ca contents than younger sea water. Under
102
-------
certain circumstances, manganese concentrations may be indicative of ion exchange. This is the case in
sea-water intrusion samples from Manhattan Beach, California, in which sodium ions of the intruding sea
water are exchanged for manganese in the sand and gravel aquifer. Largest manganese concentrations
appear to be associated with Mg/Ca ratios close to or slightly greater than one (Brennan, 1956).
Dilution diagrams were used by Howard and Lloyd (1983), who investigated the relationship of three
groups of saline ground water in east-central England (Rg. 42). On these plots, data points close to the
theoretical mixing line between sea water and local ground water indicate simple mixing, whereas data
points away from mixing lines indicate chemical reactions, such as mineral dissolution and precipitation, ion
exchange, and sulphate reduction. It is interesting to note that Howard and Lloyd (1983) did not at first
attempt differentiating saline waters by the use of major chemical constituents but instead used isotopes
and the I/CI ratio to distinguish three groups. Only because of this grouping did dilution diagrams allow an
explanation of the hydrochemical evolution of the individual groups that otherwise would have only
resulted in a big cluster of data points.
3.3.4. Reaction Characteristics of Sea-Water Intrusion
The chemical composition of sea water changes as it intrudes a fresh-water aquifer. Changes occur
in response to mixing and chemical reactions as summarized in Figures 43 and 44 and below. These
changes are most pronounced within the initial sea-water front that mixes with fresh water. Subsequent
intrusion deviates little from sea-water composition.
Mixing: Mixing of fresh water and sea water occurs within a transition zone, which is characterized by
chloride concentrations from just above background concentration values to just less than sea-water
concentration. The front of this transition zone is characterized by ion exchange as discussed below.
Behind the ion-exchange front, simple dilution characterizes deviation of the brackish water from sea-
water composition. This can easily be identified on trilinear plots in the straight-line relationship between
data points. On bivariate plots of major cations and artions versus chloride, data points plot close to the
theoretical mixing fine between local fresh water and sea water.
Ion Exchange: Clay minerals, especially morrtmorillonite, have free negative surface charges that are
occupied by cations in direct proportion to the abundance of cations in the water and to the sorption
characteristics of the cations and the minerals. In a typical fresh-water aquifer, these sites are saturated
mainly with calcium ions, whereas in a typical salt-water aquifer, the sites are occupied mainly by sodium
ions. Whenever the relation of calcium to sodium in the water changes, for example in response to sea-
water intrusion into a fresh-water aquifer, ton exchange will occur whereby sodium will be taken out of
solution and calcium will be released from mineral exchange sites. Magnesium and potassium may also be
exchanged for calcium but the Na-Ca exchange is the most significant one. For example, Howard and
Lloyd (1983) attributed more than 96 percent of the base exchange in the Chalk aquifer of east-central
103
-------
500
400
^ 300-
o
200
100
100
500
1,000 5,000 10,000
Cl (mg/L)
100 500 1,000 5,000 10,000
Cl (mg/L)
100 500 1.000 5.000 10.000
Cl (mg/L)
(a) Dilution with fresh ground waters around Grimsby
(b) Dilution with fresh ground waters east of Louth
SALINE WATER TYPES
o East of Louth (Zone lll)~~"*|
• (Grimsby (Zone II) >— Type C
• Holderness (Zone IV) J
A Immingham-Pyewipe (Zone I) Type A
500 1,000 5,000 10,000
Ci (mg/L)
200
150
^
100
50
100 500 1,000 5,000 10,000
Of (mg/L)'
QA17233C'
Figure 42. Dilution diagrams of major ions with theoretical mixing fines between sea water and local fresh
water. Simple mixing is indicated for samples plotting close to theoretical mixing lines. Deviations from
these lines suggest additional chemical reactions (from Howard and Lloyd, 1983).
104
-------
Typical flow regimes
Local flow
Estuary
Shallow regime
Last chemical reactions
slow chemical reactions
XXXXXX\XX\X\XXXXXX\XXXXXX
Increasing dissolved solids
Typical hydrochemical facies
Flow velocity = 5 m/year
Increasing calcite saturation
Decreasing calcite
saturation
Calcite
oversaturation
xVx\Ca-Mg-Na-HCp3
Bedrock/x'x'x'x'x'x'x'x'x-VxVxV
A Feldspar hydrolysis; clay alteration; pyrite oxidation where lagoons exist, pools, and sewage discharge
B Gas generation; ionic exchange; mineral weathering
C Estuary environment; local and regional sea-water encroachment; heavy-metal mobilization
D Solution and precipitation of carbonates; ionic exchange; sulfate reduction: evaporite solution; peat and
lignite decomposition
E Mineral alterations; cement deposition; dolomitization
F Man-made effects where deep injection of waste water occurs
OA17235C
Figure 43. Relationship between flow regimes and hydrochemical facies in ground water in a big coastal
plain (from Back, 1966, and Custodio, 1987).
105
-------
SO,2'
Oceanx
HCO,
Ocean„
EXPLANATION
£30) 613C%.
NaH
•s \ S. \ X S
OA17236C
Figure 44. Geochemical processes and changes in Ionic composition along the flow path in a coastal
aquifer (from CustocHo, 1987).
106
-------
England to Na-Ca exchange. This exchange is assumed to be instantaneous (Kafri and Arad, 1979).
Intrusion of fresh water into a salt-water aquifer will cause the opposite ion exchange. Table 7 summarizes
major changes on ionic ratios as a result of these changes in water fades. The chloride concentration is
not affected by ion exchange, which makes the Na/CI ratio a potential tracer of intrusion. If sea water
intrudes a fresh-water aquifer, Na/CI ratios will decrease from ratios often greater than one to ratios often
less than the value in sea water. In contrast, if fresh water replaces marine water or washes out marine
sediments, very high Na/CI ratios can result (Custodio, 1987). Figure 39 illustrates this process of ion
exchange in the early part of the intruding sea-water front as plotted on a Piper diagram. Ion exchange
between calcium and sodium characterizes the cluster in the cation triangle (samples 1 through 12,
representing the lowest chloride values), whereas mixing characterizes the straight line in the anion
triangle as chloride content increases. The diamond-shaped field reflects the slight increase in Ca+Mg,
the matching decrease in Na+K, and the high increase in CI+SO4 percentages (Rg. 39).
No changes in the Na/CI ratio will occur in water that intrudes behind the front of ion exchange
because all the exchange sites are already occupied. Therefore, the Na/CI ratio should approach the ratio
of sea water (0.85 molar ratio), which differs from the typical ratio of halite-dissolution brines (0.64 molar
ratio) and from the small ratio characteristic for many oil-field/deep-basin brines (<0.50 molar ratio). The
degree of change that occurred because of ion exchange may not only indicate the position within the
intruding front (Table 7), but also the timing of the intrusion. Recent sea-water intrusion would be
expected to be associated with data points predominantly showing ion exchange, whereas old sea-water
intrusion would be expected to include many data points with little or no evidence of ion exchange
(Howard and Lloyd, 1983).
Carbonate Dissolution: Mixing of fresh water and sea water, both saturated with respect to calcium
carbonate, can result in a mixing water that is undersaturated with respect to calcium carbonate (Hanshaw
and others, 1971; Back and others, 1979; Custodio, 1987). This mixing water can dissolve carbonates
and thus calcium and bicarbonate concentrations will increase. Additional calcium carbonate dissolution
may occur in the presence of sulfate reduction of organic-rich sediments because of the associated
change in the pH and CO2 contents of the water.
Sulfate Reduction: Sea water is relatively high in dissolved sulfate content (Table 4). Under reducing
conditions in ground-water systems, and with the presence of compounds that can be oxidized (that is,
organic matter) as well as of reaction catalysts (for example, bacteria or isolated enzymes), sulfate will be
reduced according to the equation CH2O +1/2 SO42~ -»> HCC^- +1/2HS + 1/2H+ O""1" HS~ + H+ —>
H2S at pH <7) Freeze and Cherry, 1979). This results in a decrease in sulfate concentration relative to the
sea-water composition.
Bromide: Sea-water intrusion may lead to bromide concentrations greater than a few milligrams per
liter in coastal aquifers. This has a big potential impact on future drinking-water resources because
elevated bromide concentrations make water treatment more difficult. Traditional chloride treatment or
107
-------
Table 7. Changes in ionic ratios due to ion exchange relative to the position
of the intruding water.
Fresh-water
Recent Salt-water intrusion Intrusion
(a) (b)
Behind
Advancing advancing
front front
Ca/Na increase no change decrease
A|Ca+Mg|/A|Na| constant no change constant
Na/CI decrease no change increase
108
-------
ozonation of high-bromine water causes the formation of bromoform and other brominated
trihalomethanes, which may, at bromide concentrations greater than 2 mg/L in the water, exceed allowable
concentration in future (1992) standards (McGuire and others, 1989, U.S. Water News, 1990a).
Minor Constituents: As sea water passes through muds on estuary or ocean bottoms, it may become
enriched in those minor constituents that are typically concentrated in those muds, such as iodide,
strontium, and fluoride (Lloyd and Heathcote, 1985).
3.3.5. State-by-Sate Summary of Sea-Water Intrusion
Sea-water intrusion is not a new phenomenon and has been reported to occur in every coastal state
of the United States. For example, the city of Galveston had to abandon one of their water-supply wells as
early as in 1896 (Turner and Foster, 1934) and in Florida, intrusion has been experienced along most of
the Gulf and Atlantic Coast (Atkinson and others, 1986). The following section lists the published
occurrences for sea-water intrusion for all the states with marine shorelines. This discussion includes sea-
water intrusion due to anthropogenic activities but also mentions the occurrence of saline ground water in
coastal aquifers related to natural processes, such as inclusion since deposition or past intrusions during
times of higher sea-water levels. This compilation is a result of iterature review.
Alabama: Sea-water intruded fresh-water aquifers from Fort Morgan to Gulf Shores (Atkinson and
others, 1986), and overdevelopment of a well field in the Mobile-Gulf Coast region caused lateral intrusion
of saline water from the Mobile River into Pleistocene sand-gravel aquifers (Miller and others, 1977). Sea-
water contamination of wells in Baldwin County has been caused by overproduction of a fresh-water zone,
sea-water flooding and natural sea spray, and leakage from salt-water ponds (Chandler and others, 1985).
California: In southern California, the potentiometric surfaces of coastal aquifers were above sea
level at the turn of the century but gradually decreased in the following decades. Because of heavy
developments in the 1920's and 1940's water levels have declined in some areas to as much as 70 ft
below sea level (Brennan, 1956). In the Manhattan Beach area of Los Angeles, most of the original water
wells were abandoned after 1940 when salinity of ground water increased due to sea-water intrusion
(Poland and others, 1959). Banks and Richter (1953) summarized: known sea-water intrusion areas (A),
threatened sea-water intrusion areas (B), and potential sea-water intrusion areas (C) in California as being:
(A) Mission, San Luis Rey, Santa Margarita, Coastal Plain Orange County, Coastal Plain Los Angeles
County, Malibu Creek, Trancas Creek, Morre Bay, Salinas, Coastal strip Salinas Valley and Pajaro Valley,
Pajaro, Santa Clara, and Sacramento-San Joaquin Valley between Pittsburgh, and Antioch; (B) Tia Juana,
Otay, Santa Clara River, Carpenteria, Goleta, Arroyo Grande; and (C) Sweetwater, San Dieguito, San
Onotre, San Mateo, San Juan, Ventura River, Santa Ynez River, Santa Maria River, Carmel, Ygnacio
Clayton, Napa-Sonoma, Petaluma, Eel River, Mad River, and Smith River. Of those, the most serious sea-
water intrusions have occurred in the West Coast Basin of Los Angeles County, East Coastal Plain
109
-------
Pressure Area of Orange County, Petaluma Valley in Sonoma County, Napa-Sonoma Valley, Santa Clara
Valley in the San Francisco Bay area, Pajaro Valley in Monterey and Santa Cruz Counties, Salinas Valley in
Monterey County, Oxnard Plain Basin in Ventura County, and Mission Basin in San Diego County
(Fuhriman and Barton, 1971). In 1975, the 14 known (A) and 14 suspected (B) areas of sea-water
intrusion were (A) Eel River Valley, Petaluma Valley, Napa-Sonoma Valley, Santa Clara Valley, Pajaro
Valley, Elkhom slough area, Salinas Valley pressure area, Morro Basin, Chorro Basin, Los Osos Basin,
Oxnard Plain Basin, West Coast Basin (Los Angeles County), San Luis Rey Valley, and Mission Basin, and
(B) Russian River Basin, Drakes Estero Basin, Bolinas Lagoon Basin, San Rafael Basin, Suisum-Fairfield
Valley, Sacramento-San Joaquin Delta, Tonitas Creek Basin, and San Diego River-Mission Valley Basin
(Smith, 1989). Farrar and BertokU (1988) identified areas of salt-water intrusion around San Francisco Bay
and the delta area, around Monterey Bay, Morro Bay and at the mouths of Petaluma, Sonoma, and Napa
Valleys, where the salt water infiltrated from tidal channels. Diversion of stream flow from the Sacramento
and San Joaquin Rivers has caused inland migration of salt water in the areas around Fairfield and
Pittsburgh, whereas fresh-water pumping has caused sea-water intrusion and land subsidence in
Alameda County. Pumping of water for irrigation purposes has led to sea-water intrusion near the mouths
of the Pajaro and Salinas Rivers (Farrar and Bertoldi, 1988).
Connecticut: Lateral sea-water intrusion from harbors and tidal river estuaries due to heavy
pumpage has contaminated many industrial and municipal wells in the Long Island Sound coastal area
including the cities of New Haven and Bridgeport (Miller and others, 1974).
Delaware: Salt-water intrusion has been reported all along the Delaware coastline and is
responsible for the abandonment of one to two public wells and 20 to 30 domestic wells each year
(Atkinson and others, 1986). During dry periods, sea water has migrated up the Delaware River nearly as
far as Philadelphia (U.S. Environmental Protection Agency, 1973). Lateral and vertical intrusion of salt
water from the Delaware River and Delaware Bay and from the Atlantic has contaminated many well fields
due to heavy pumpage from shallow aquifers and the dredging of impermeable soils (Miller and others,
1974). High chloride concentrations between 6,000 mg/L and 17,000 mg/L were measured in core
samples obtained from wells along the Atlantic Coast of southeastern Delaware. The depths of wells from
which core samples were obtained ranged from 5 to 60 ft below land surface (Woodruff, 1969).
Florida: The Florida problem of sea-water intrusion stems from a combination of permeable
limestone aquifers, a lengthy coastline, and heavy pumpage in coastal areas. Intrusion was reported (in
1953) at 28 specific locations and some 18 municipal water supplies have been adversely affected since
1924. Interior drainage canals which lowered the water table and permitted sea water to advance inland by
tidal action contributes to the problem (Bruington, 1972; U.S. Environmental Protection Agency, 1973).
Sea-water intrusion is a permanent threat to the Biscayne Aquifer because the aquifer is unconfined, is
hydraulically connected to the sea, is heavily pumped, and is extensively cut by a network of canals. Heavy
pumpage has resulted in water levels below sea level near some well fields (Johnston and Miller, 1988).
110
-------
Large cones of depression and sea-water intrusion due to heavy pumpage have been reported in many
areas, such as Jacksonville, Tampa, and Miami. Intrusion on a smaller scale has been experienced along
most of the Atlantic and Gulf Coasts (Atkinson and .others, 1986). Miller and others (1977) listed some of
the affected counties, including Escambia County (heavy pumping near Bayou Chico caused intrusion
into sand and gravel aquifer and reversed flow gradient and induced salt-water migration from the
Escambia River), Bay County (saline surface water leaked downward from bays contaminating two wells),
Martin County (encroachment of water into shallow aquifer from St. Lucie River due to overpurnpage of
city wells), Pasco County (sea-water intrusion due to overpurnpage of coastal wells), Charlotte County
\
(sea-water intrusion from Gulf and estuaries and intrusion due to poor well construction and improper
abandonment), Palm Beach County (inland flow of saline water through canals and intracoastal waterways
during dry period caused contamination of Biscayne aquifer), Broward County (salt-water migration into
Biscayne aquifer in coastal areas due to heavy pumpage and construction of canals), Dade County (dry
period caused inland flow of salt water up the Miami canal and contaminated four water wells), and Miami
(contamination of wells due to inland flow of salt water during dry periods and inflow of saline ground water
due to heavy pumpage).
In 1943 in the Miami area, sea-water intrusion was restricted to wells within two miles of the coast
except for some areas along tidal canals where intrusion was observed somewhat farther inland (Love,
1944). In Citrus County sodium and chloride concentrations exceed acceptable levels twofold and
threefold, respectively, as a result of high pumpage of fresh water and seepage of salt water from the Gulf
(U.S. Water News, 1990b). In the Ftoridan aquifer of west-central Florida, the transition zone between
fresh water and saline water may be as far as 50 miles inland from the coast. Future ground-water
withdrawal could lead to further landward movement of the salt-water front in that area at an average rate of
approximately 0.35 ft per day (Wilson, 1982). In Pascola County, the interface between fresh and saline
ground water, at 100 ft below sea level, is located approximately one to two miles inland (Reichenbaugh,
1972). Three encroachment mechanisms are responsible for that position: (1) lateral inflow through
permeable limestone where the aquifer is in contact with sea water offshore, (2) leakage from tidal streams
and canals in which sea-water intruded, and (3) upward movement of salt water in response to lowered
hydraulic heads caused by pumpage of fresh water (Reichenbaugh, 1972).
Sea-water intrusion into fresh-water reaches of the Catoosahatchee River occurs between La Belle
and Olga during low-flow periods as repeated injections through the lock chamber at W. P. Franklin Dam.
In 1968, the upstream limits of water containing less than 250 mg/L of chloride were 11.4 miles from the
dam in the deeper parts of the river and approximately five miles from the dam at shallow depth (Boggess,
1970).
Georgia: Sea-water intrusion has been reported for Chatham County (Savannah) and Glynn
County (Atkinson and others, 1986).
111
-------
Louisiana: Sea-water intrusion has occurred all along the coastal shores of Louisiana. In addition,
the cities of Lake Charles and New Orleans have experienced severe cases of salt-water intrusion due to
high pumpage of ground water (Atkinson and others, 1986). During periods of low flow, tidal water from
the Gulf of Mexico invades coastal streams and fresh-water aquifers, such as the one in the Vermillion
River area (Newport, 1977). , > .
Maine: Overpumping has resulted in intrusion of tidal estuary waters into local aquifers south of
Augusta, in Kennebec and Sagadahoc Counties.; Several domestic water wells have been affected by
salt-water intrusion, the problem fluctuating at times over the year (Atkinson and others, 1986). A 300-ft-
deep well producing from the bedrock aquifer near the town of Bowdoinham, Sagadahoc County, was
contaminated by salt water from the tidal reach of the Kennebec River, resulting in abandonment of the
well (Miller and others, 1974). ,
Maryland: Sea-water intrusion has been reported in St. Mary's, Anne Arundal, Harfork, Dorchester,
and Somerset Counties as well as on Kent Island. Overpumping, leaky well casings, and dredging
activities appear to be responsible for ground-water contamination. Salty water from the Patapsco River
estuary has intruded shallow fresh-water aquifers in the harbor district of Baltimore. Heavy pumpage and
leaky casings also induced the inflow of salt water from Chesapeake Bay into wells at Joppatowne, Harford
County, and Westover, Somerset County. Lateral and vertical intrusion of salt water from tidal river
estuaries, enhanced by casing leaks in abandoned wells, has been reported in the areas of Cambridge,
Dorchester County, Annapolis, Anne Arundal County, and the Solomons-Patuxent River, St. Mary's
County. The inland limit of saline ground water in coastal plain aquifers are along the coast for the
Cretaceous aquifer, along a line from Dorchester to the southwestern state-line comer with Delaware in
the Tertiary aquifer, and along a line parallel to the Chesapeake shoreline in the south and approximately
half way between the shore and the Maryland-Delaware state line in the north, crossing the state line just
north of 39 degrees latitude (Miller and others, 1974).
Massachusetts: Sea-water intrusion has been reported in several areas in Massachusetts,
including Bristol, Plymouth, and Barnstable Counties (Atkinson and others, 1986) as well as
Provincetown, Scituate, and Somerset Counties (Newport, 1977).
Mississippi: Heavy pumpage and the resulting decline in water levels induced lateral intrusion of
saline water from the Gulf and the Pascagoula River estuary at Moss Point into Miocene and the Citronelle
Formations (Miller and others, 1977). Sea-water intrusion has also been reported in Hancock and Jackson
Counties (Atkinson and others, 1986).
New Hampshire: Tidal waters have intruded aquifers in the Portsmouth area (Newport, 1977) and
pumpage has Induced intrusion in Rockingham County (Atkinson and others, 1986).
New Jersey: Intrusion of sea-water is a major problem that has been monitored since 1923 (Ayers
and Pustay, 1988). In 1977, the salt-water monitoring network included more than 400 wells (Schaefer,
1983). Overpumping, leaky casings, and dredging have led to intrusion in Salem, Gloucester, Cape May,
112
-------
Middlesex, Monmouth, Ocean, and Atlantic Counties (Miller and others, 1974; Newport, 1977; Schaefer,
1983). The most significant intrusions have occurred in the areas of Sayreville, where salt-water intrusion
has been observed for more than 40 years, in the Keyport-Union Beach area, where the original well field
had to be abandoned in 1976, and in Cape May City (Schaefer, 1983). Even areas 20 miles or more
inland, such as in Gloucester and Atlantic Counties, have experienced sea-water intrusion due to
pumping and corroded well casings. At Somers Point, Atlantic County, a wedge of salt water has moved
3,000 ft inland into the Cohansey aquifer. At Artificial Island, Salem County, high pumpage at a nuclear
generating plant has induced salt water (Miller and others, 1974; Newport, 1977). Since the early 1970's,
well waters in the Old Bridge aquifer in the Boroughs of Keyport and Union Beach, Monmouth County,
have shown high chloride concentrations, indicating sea-water intrusion. Concentrations increased from
background levels of less than 5 mg/L to greater than 600 mg/L in some coastal wells (Schaefer and
Walker, 1981). In the Sayreville area, the transition zone between fresh water and salt water has migrated
0.2 to 0.4 miles inland between 1977 and 1981 (Schaefer, 1983). A listing of previous salt-water studies
in New Jersey was presented by Schaefer (1983).
New York: Sea-water intrusion has been documented on Long Island (Lusczynski and
Swarzenski, 1966), in the Town of Southampton, the Eastport-Remsenburg-Spoenk-Westhampton area,
and the North Haven-Sag Harbor area (Anderson and Berkebile, 1976). Coastal plain aquifers have been
contaminated with salt water in the Port Washington area, where sea water was used in settling ponds.
Long-term use of the salt water in sand and gravel pits has raised chloride levels in nearby shallow and
deep wells from a normal level of less than 20 mg/L to greater than 1,000 mg/L in an area of more than two
square miles (Swarzenski, 1963; Miller and others, 1974).
Heavy pumping and reduced natural recharge have caused lateral intrusion of ocean water into
producing aquifers on Long Island (Newport, 1977). Sea-water intrusion in southern Nassau and
southeastern Queens Counties, Long Island, occurs as one wedge of salt water in shallow glacial deposits
and two more wedges in the upper and lower portions of the underlying artesian aquifer (Miller and others,
1974).
Salt-water intrusion may be occurring in two areas in the Town of Southampton, the East port-
Remsenburg-Spoenk-Westharnpton area and the North Haven-Sag Harbor area, Long Island. Salt water
underlies a lense of fresh water in the area, possibly allowing salt-water intrusion both laterally and
vertically. The depth to salt water in three testholes was found to be at 350, 620, and 1,060 ft below sea
level, respectively. Chloride concentrations in affected wells range from 200 to 13,890 mg/L (Anderson
and Berkebile, 1976).
North Carolina: Sea-water intrusion has been documented in Carteret, Pamlico, Beaufort, Hyde,
Dare, and Tyrell Counties (Atkinson and others, 1986). Tidal action resulted in intrusion of saline water into
ground-water sources through drainage canals on former marsh land in the Coastal Plains of eastern North
Carolina. This forced abandonment of large areas of cropland in parts of Tyrell, Dare, Hyde, Beaufort,
113
-------
Pamlico, and Carte ret Counties (U.S. Geological Survey, 1984). Heavy pumpage could also induce
recharge of salt water from streams affected by sea water resulting from tidal action and/or decreased
fresh-water flow.
Oregon: Sea-water intrusion has been documented in Eugene (Lane County), North Bend (Coos
County), in the Willamette Valley in the West Cascades, and in Clatsop and Tillamook Counties (Atkinson
and others, 1986). Sea-water overrides fresh-water aquifers in the Horsfall Beach area as a result of winter
storms (Magaritz and Luzier, 1985).
Pennslyvanla: Dredging of tidal rivers and heavy pumpage from wells near tidal rivers have caused
sea-water intrusion near Philadelphia in the eastern part of the state (Atkinson and others, 1986).
Rhode Island: Heavy pumpage from wells near tidal rivers has caused sea-water intrusion near
Warwick in Bristol and Kent Counties and near Providence in Providence County (Atkinson and others,
1986).
South Carolina: Sea-water intrusion has occurred at several locations along the state's coast,
including areas in Charleston, Beauford, and Horry Counties (Miller and others, 1977; Atkinson and
others, 1986). Background concentrations of sodium and chloride in fresh ground water within the Black
Creek aquifer of Horry and Georgetown Counties are less than 280 mg/L and 40 mg/L, respectively.
Concentrations above that may indicate mixing of fresh water with sea water (Zack and Roberts, 1988).
The most extensive encroachment at present occurs in an area that extends from the Beaufort Basin to
the Cape Fear Arch (Siple, 1969). The upper zones of Eocene limestones and the sub-sea contact of
Eocene and Oligocene deposits are subject to considerable salt-water encroachment, with maximum
chloride concentrations of up to 8,500 ppm (Siple, 1969)
Texas: Sea-water intrusion is occurring in the Galveston, Texas City, Houston, and Beaumont-Port
Arthur areas and around Corpus Christ! (Atkinson and others, 1986). Along the Gulf Coast, sea water
mixes with surface water in the tidal reaches of the rivers, such as in the Catoasieu River channel, Calcasieu
Lake, Sabine Lake, and lower Sabine River (Krieger and others, 1957). Sea-water intrusion occurs also
along the lower 36 miles of the Neches River and 3 miles of Pine Island Bayou (Port of Beaumont) during
6 months of the year. Salt-water barriers, which divert fresh water into a canal system upstream, cause this
unhindered inflow of salt water; tidal action flushes the salt water back and forth below the barriers causing
it to become Increasingly concentrated. Deterioration of water quality is aggravated by waste disposal into
the waterway (Harrel, 1975). Sea-water intrusion from the Houston Ship Channel has been reported from
shallow wells between Baytown and Houston. Large withdrawals of ground water (525 million gal/d),
resulting in a decline of artesian pressure equivalent to a 400 ft drop of hydraulic head, has caused this
intrusion over an area up to four miles wide at Baytown at a rate of several hundred feet per year
(Jorgensen, 1981).
Virginia: Sea-water intrusion due to the overpumping of fresh ground water in areas of shallow
saline ground water has occurred in Northhampton and Isles of Wight Counties (Atkinson and others,
114
-------
1986) and in the areas of Newport News and Cape Charles (Newport, 1977). According to Larson (1981),
historic changes in chloride content of water wells are a local rather than regional phenomena. In general, a
wedge of salt water coincident with the mouth of Chesapeake Bay extends into the York-James and
southern Middle Neck Peninsula, where the greatest chloride increase (175 mg/L) was measured.
Washington: Several cases of sea-water intrusion have been documented in the Puget Sound
areas and along the Pacific Coast. Affected areas include Island County, the areas around Tacoma and
Olympia, Kitsap County, and northern Jefferson County (Kimmel, 1963). While sea-water intrusion
appears to be a more widespread problem in Island and San Juan Counties, it is of only local significance in
Clallam, Jefferson, Pierce, Thurston, and Whatcom Counties (Dion and Sumioka, 1984).
3.4 Oil-Reid Brine
3.4.1 Mechanism
This chapter deals with salt water that is produced with oil and gas. This water is naturally-occurring,
deep-basin formation water that differs from the type discussed in chapter 3.1 more through its
mechanism of mixing with fresh water than through its water chemistry. The major difference in the mixing
mechanism is that formation brines unassociated with oil and gas are normally separated from fresh ground
water by a transition zone of slightly to very saline water, which normally lessens the degree of natural or
induced salinization. Anthropogenic contamination of fresh water by oil-or gas-field brine, in contrast, is
not associated with a transition zone but instead brings concentrated brine into direct contact with fresh
water. Therefore, salinization of fresh ground water by oil-and gas-field brine is often very abrupt,
characterized by large increases in dissolved solids within relatively short time periods and short distances.
There are currently 25 major oil-and gas-producing states in the country (Fig. 4). In those states,
more than one million holes have been drilled in search for oil and gas. According to Newport (1977), this
drilling for oil and all other contamination hazards associated with the oil and gas industry are major
contributors to salt-water intrusion in the inland part of the United States.
Contamination hazards associated with the oil and gas industry stem largely from the huge amount of
salt water that is produced with oil and gas. In Texas, approximately 2.5 barrels of salt water were produced
with every barrel of oil in 1961 (McMilfion, 1965). Miller (1980) estimated that 4 barrels of salt water are
produced with every barrel of oil. Others estimate ratios of salt water to brine of up to 20:1, with ratios
generally increasing as the production of oil from a field decreases. In 1956, salt production at oil fields in
the United States totaled 125 million tons, which is equivalent to approximately one-half of the stream
discharge in the United States (Thome and Peterson, 1967). The total amount of salt-water production
increases with time, as documented by early data reported by Collins (1974) (7.7 billion barrels per year)
and more recent data reported by Michie & Associates (1988) (22 billion barrels in 1986). Production data
for 1963, listed state-by-state in Table 8 (Miller, 1980), identify Texas, Kansas, and Oklahoma as the three
115
-------
Table 8. Methods of disposal of produced oil- and gas-field brine in the United States during 1963 (data from Miller, 1980).
o>
Current volumes Injection
State
Alabama
Arizona
Arkansas
California
Colorado
Florida
Georgia (A)
Illinois
Indiana
Kansas
Kentucky
Louisiana
Michigan
Mississippi
Montana
Nebraska
Nevada (A)
New Mexico
New York (A)
North Dakota
Ohio (A)
Oklahoma
Pennsylvania
South Dakota
Texas
Utah
West Virginia
Wyoming (A)
Totals
produced
2,493
100
539,132
2,740,850
202,194
600
876,712
81,797
5,011,400
123.287
2.785.000
149,587
340,079
50,000
121,907
356,624
31,000
3,751.911
191.780
68
6.127.671
81.634
115.068
23,682.022
water flooding
89,082
445,768
131,500
50,960
800,000
73,973
184,000
40,000
10,000
17.329
55.176
23,500
3,160.577
2,736.755
2,981
7,821,601
Injection
disposal only Lined pits
1,397
340,734
208.665 3.127
5,000 70
600
14.724
4,200,000 1,800
35,616 2,740
1,762,000
147.849
203,836 8,219
31.400
7.567
165.423
583.280 5.370
1.472.954
9,182.173 21,326
Surface Disposal
Unllned pits Streams
493 603
100
7,444 101,871
399,933 501
65.624
15.132
9.600
5,480 5,480
698,000
982
74.329
8,600
97,011
136,025
7.500
2,685
191.780
68
1.262.719 615.566
4,862
115,068
2.796.587 1,030.869
Other methods
1.682.855
876.712
982
141.000
756
13.699
39.677
73.790
2.829.471
Documented
brine pollution
(B) (C)
(B) (C)
(B)
(B) (C)
(B) (C)
(B) (C)
(C)
(B) (C)
(B) (C)
(C)
(B) (C)
(B) (C)
(C)
(C)
(B) (C)
(B) (C)
(C)
(B) (C)
(B) (C)
(B) (C)
(C)
(B) (C)
(C)
(B) (C)
(C)
(A) Brine data not available; (B) from Miller. 1980; (C) see section 3.4.5
-------
largest producers in the country, making up more than 50 percent of the total production. In 1986 the total
volume of brine disposal at oil and gas fields in the United States amounted to approximately 60 million
barrels per day. Of this, 42 million were injected into producing oil and gas formations, 17 million were
injected into salt-water formations, three million were released into surface waters, and two million
percolated into the ground where an underground source of drinking water is not present (primarily in the
San Joaquin Basin of California) (Michie & Associates, 1988). Assuming a conservative average IDS
concentration of 50,000 mg/L in those waters, the disposal equals a total salt toad of 8.6 million metric tons
per year deposited into surface waters and of 5.8 million tons per year percolated into the ground.
Contamination of ground water and surface water can occur where the disposal of this produced
water is done in a way that allows mixing between brine and fresh water. This potential hazard is highly
dependent on the disposal method. Surface disposal and brine ponding in unlined surface pits have
been used widely in the past, which may have caused high potentials of fresh-water contamination in
Texas, Louisiana, and California, where the largest amounts of brine were disposed of by this method
(Table 8). According to Miller (1980), at least 17 oil-and gas-producing states (Table 8) have experienced
water pollution from disposal of oil-field brine, with a high likelihood that contamination has taken place at
one time or another in all oil-producing states.
The following sections will describe mechanisms that allow mixing of oil-and gas-field brine with fresh
ground water.
3.4.1 a. Surface disposal
Discharge of oil-field waters into coastal waterways, bayous, estuaries, or into inland streams, creeks,
and lakes directly pollutes surface waters. Where these surface-water bodies are interconnected with
ground water, ground-water pollution will also occur. Spillage from disposal and drilling pits has the
potential to contaminate surface waters and leakage through the bottom of a pit has the potential of
contaminating the vadose zone and ground water underlying the pit.
At earlier times with little regulation of the oil and gas industry, the often indiscriminant disposal of
brines into surface waters caused severe contamination in many areas. For example, chloride content in
the Green River at Mundfordville, Kentucky, increased from pre-oil-devetopment levels of 10,600 tons in
1957 to post-oil-development levels of 305,000 tons in 1959, which is equal to a 3,000 percent increase
(Krieger and Hendrickson, 1960). Indiscriminant disposal of brines onto the land surface has caused
"vegetative kill" areas in many old oil and gas fields in the past, and may still be a cause of serious
degradation of surface and ground water as a result of leaching. In an effort to reduce water pollution,
evaporation pits were introduced as a way of brine disposal under the premise that the volume of brine
could be reduced by evaporation. As a consequence, during 1961 atone, within the Ogallala outcrop of
the Texas High Plains, more than 66 million barrels of brine were disposed of on the surface, primarily in
117
-------
unlined surface pits (Burnitt and others, 1963). This represented approximately 55 percent of that
year's total brine production in the area. From a single oil field in Winkler County, Texas, more than
500,000 acre-ft of brine were disposed of between 1937 and 1957 into unlined surface pits or into a
communal disposal lake (Garza and Wesselman, 1959), adding millions of tons of salt to soils and causing
contamination of local water wells. In Colorado, 27 million barrels of brine were disposed of in unlined pits
in 1969 (Hold, 1971). After some years of pit use, it was eventually realized that most of the disposed of
brine did not evaporate but infiltrated instead into the ground through the often porous pit bottom,
creating a multitude of point-contamination sources of brine. For example, in the vicinity of Cardington,
Ohio, an area of 13 mi2 was affected by brine pollution only three years after the first successful oil well
had been drilled. Disposal of brine into surface pits and indiscriminant surface dumping of salt water by
contract truckers was the major cause of this contamination (Lehr, 1969). As a consequence of this
pollution, an entire well field had to be abandoned. Similar uncontrolled surface discharges were reported
by Baker and Brendecke (1983) in Utah, where water haulers may dispose of brine into unlined trenches,
surface depressions on undeveloped land, or into roadside ditches.
Release of salt from soil underlying former disposal pits can affect ground water for long times and in
repeated plumes. Cyclicity of salt release from the soil is caused with every precipitation period that
flushes salt from the vadose zone into the saturated zone. This causes extremely high salt concentrations
under some pits even many years after their abandonment, as measured by Pettyjohn (1982) among
others under a pit in Ohio (Cl of 36,000 mg/L after 8 years) and by Richter and others (1990) under a pit in
Texas (Cl of 20,750 mg/L after 20 years). Even longer time periods will be needed to see salt
concentrations return to background levels. Contamination of ground water caused by an unlined surface-
disposal pit in southwest Arkansas covers an area of approximately one square mile and is estimated to last
for another 250 years (Fryberger, 1972). The Oklahoma Water Resources Board (1975) estimated that it
will take more than 100 years to flush all the salt from a 9 mi2 area contaminated by pits in the Crescent
area, Oklahoma. Disposal pits that have been used for only relatively short periods also can contaminate
soil and ground water, especially if large amounts of brine have been disposed of. For example, Lehr
(1969) reported contamination of ground water from two pits at Delaware, Ohio, which had been used for
only 15 months, during which they had received more than 225,000 barrels of brine. As a result of this
disposal, chloride concentrations in ground water increased from background levels of less than 10 mg/L
to more than 35,000 mg/L. A single pit can cause ground-water deterioration on a wide scale. Rold (1971)
estimated that one pit caused a 27 ppm per year salinity increase in the Severance ground-water basin of
northeast Colorado. This pit had received only 200 barrels of brine per day but had been excavated into
7 ft of gravel that directly overlay the local fresh-water aquifer (Rold, 1971). The amount of fresh water lost
to this kind of salinization can be substantial, such as in the case of Miller County, Arkansas, where
seepage from three disposal areas contaminated approximately 60 million gallons of fresh water (Ludwig,
1972).
118
-------
Often, highly saline fluids are also used in the oil and gas industry during the drilling of boreholes.
These fluids are temporarily stored in pits and can cause fresh-water contamination if not disposed of
properly after drilling has been completed. Murphy and Kehew (1984) estimated that approximately
360 million ft3 of brine have been buried in shallow pits in North Dakota during pit closure, causing
variable degrees of soil-water and ground-water contaminations locally.
3.4. ib. Injection wells
Injection of salt water is done either for enhanced recovery or for brine disposal. Enhanced recovery
occurs in the producing formation, which typically is not a fresh-water aquifer or an aquifer containing
potentially usable ground water (TDS <10,000 ppm). In some instances, especially in shallow oil fields,
however, a hydraulic connection may exist between the oil pool and usable water updip. These areas
(Rg. 45) are especially prone to salt-water contamination. Where this hydraulic connection does not exist,
it may be created if injection pressures exceed lithostatic pressures (bottomhole pressure greater than
1 psi/ft of overburden).
Disposal of salt water may occur in any formation that is capable of accepting large amounts of fluids,
with regulations concerning depths of disposal and water quality in the disposal zone varying from state to
state. Where injection operations are faulty, ground-water contamination may occur where injection wells
penetrate fresh-water units and salt-water units. Some of the possible mechanisms of mixing between
brine and fresh water are direct injection into a fresh-water or potable-water unit, causing failure of the
injection well adjacent to a potable-water zone or to a zone in hydraulic connection with potable water,
intrusion of salt water into potable water as a result of increased pressure and natural leaky conditions
between the units, hydraulic fracturing, or upward migration of brine along the outside of the casing (Miller,
1980). Contamination may also be associated with insufficiently plugged, abandoned boreholes that
penetrate the injection zone, as discussed next.
3.4. ic. Plugged and abandoned boreholes
Exploration for oil and gas has created millions of boreholes that penetrate shallow, fresh-water
aquifers and deep, saline-water aquifers. Many of these holes may have been plugged and abandoned in
a condition that may allow communication between the different water types at some time (Rg. 46). Flow of
brine into fresh water along abandoned boreholes can occur where the brine unit or the fresh-water unit
are not sealed and where the hydraulic head of the brine unit is higher than the hydraulic head of the
fresh-water unit. Higher hydraulic heads in brine units may be natural or man-induced, the latter either by
injection into the brine unit or by the lowering of heads in the fresh-water unit. Most critical are those wells
that were abandoned without sufficient plugs and casing in the early stages of oil and gas exploration and
many relatively shallow (up to several hundred feet deep) seismic shotholes. But even boreholes that
119
-------
EXPLANATION
Shallow oil field
R County Regular shallow oil field,
exact locations not determined
QAI7237
Figure 45. Map of shallow oil fields in the United States (from Ball Associates, Ltd., 1965).
-------
ro
Brine layer from
Saline water leaking wells
stream
Water table
Oil producing reservoir
Producing
oil well
Plugged oil
well
Cement plug
Water
injection
or
waterflood
well
- Cement,
j
Unplugged
or
improperiy
'ugg
lie
Direction of brine movement
QA17238c
Figure 46. Schematic diagram illustrating possible communication scenarios between deep saline aquifers
and shallow fresh-water units through boreholes (from Bumitt, 1963).
-------
have been plugged and abandoned in a condition sufficient to protect potable water at the time of
abandonment may present potential contamination hazards in the future, as the life of steel casings,
depending on the quality of the cement job, may only last for 5,10, or 20 years (Rold, 1971).
Inadequately plugged and abandoned boreholes may provide pathways for upward migration of
brines where natural potentials of upward flow exist. As such, they act similar to faults and fractures that
allow vertical migration of saline waters. In the absence of a flow barrier, hydraulic heads in deep brine
formations higher than the water table or potentiometric surface of an overlying fresh-water aquifer will
cause mixing of brine with fresh ground water. This mixing is chemically similar to the mixing of oil-field
brine with fresh ground water. The history of salinization, geologic conditions, and the local or regional
character of the occurrence of saline ground water are often important parameters to distinguish between
this natural mixing of saline formation water with fresh ground water and the mixing of disposed of oil- or
gas-field brine with fresh water.
3.4.2. Oil-Reid Brine Chemistry
Overall salinity, concentration ranges of individual constituents, and the types of. chemical
constituents vary much more in oil and gas brines than in previously discussed brines, that is, halite-
solution brine (chapter 3.2) and sea-water intrusion (chapter 3.3). As seen in most natural brines, a strong
correlation exists between sodium and chloride concentrations (Fig. 47). The Na/CI weight ratio is typically
less than approximately 0.60, often reflecting high concentrations of other cations, especially calcium
(Fig. 47). The relatively large scatter on plots of major cations and anions versus chloride (Fig. 47),
especially of SO, versus Cl, suggests significant differences between oil-field brines from different areas
or reservoirs. These differences can possibly be used to identify the source of contamination in areas
where production is from more than one reservoir.
The most common constituents dissolved in oil and gas brines are the major cations sodium, calcium,
magnesium, and potassium, and the major anions chloride, sulfate, and bicarbonate. Other constituents
fall into the general ranges of approximately 100 ppm for strontium, 1 to 100 ppm for Al, B, Ba, Fe, and Li,
a few parts per billion in most waters for Cr, Cu, Mn, Ni, Sn, Ti, and Zr, and a few parts per billion in some
waters for Be, Co, Pb, V, W, and Zn (Rittenhouse and others, 1969). With respect to sea water, Cr, Li, Mn,
SI, and Sr are commonly more than twice as abundant, whereas Cu, K, Ni, and Sn are commonly less than
half as abundant. In some brines, certain elements may be at concentrations high enough to make
extraction economically feasible.
3.4.3. Examples of Geochemical Studies of Oil- and Gas-Reid Brine Pollution
As Is the case with most salinization sources, identification of oil- and gas-field brine contamination is
easy as long as it is the only suspected or possible source of salt wafer. Background chloride
122
-------
100000
o>
«
z
10000 -
1000 -
100
100
1000
10000
Cl (mg/L)
100000 1000000
100000 i
10000 -,
o
E 1000-3
100 -t
10
100
1000
r i. ..... | »
10000
Cl (mg/L)
100000 1000000
100000
• 10000 '
*J 1000 -j
s 1001
10 i
100
1000 10000 100000
CI(mg/L)
1000000
100000
10000
1000 •
100
10
100 1000 10000 100000
CI(mg/L)
1000000
10000 -5
1000 -
100 -.
1 •
100 1000 10000
CI(m8/L)
S
.01
.001
.0001-
.00001
.000001
100000 1000000
100 1000 10000 100000
C1(mQ/L)
1000000
QA17239C
Figure 47. Bivariate plots of major ions and Br/CI ratios versus chloride for oil-field brines from California
(data from Qullikson and others, 1961), Texas (data from Kreitler and others, 1988), West Virginia (data
from Hoskins, 1947), and Kentucky (data from McGrain and Thomas, 1951). Large scatter indicates a high
variability in chemical composition.
123
-------
concentrations in most ground waters are in the milligrams per liter to a few hundred milligrams per liter
range, which is in big contrast to chloride concentrations of several tens of thousands of milligrams per liter
found in most produced oil and gas waters. For example, a chloride increase from less than 100 mg/L to
contamination levels of greater than 1,000 mg/L indicated contamination caused by an unlined blow-down
pit used during drilling of a gas well in northeast Ohio (Knuth and others, 1990). Kalkhoff (1986) used a
cut-off value of 50 mg/L in a study of ground-water contamination in parts of Mississippi, above which brine
contamination by oil and gas activities was indicated. Contamination by inadequately designed disposal
wells has caused increases of chloride concentrations in ground water and springs in the Upper Big
Pitman Basin of Kentucky from baseline values of 4 to 43 mg/L to contamination levels greater than
50,000 mg/L (Hopkins, 1963). In the Upper Green River Basin, also in Kentucky, contamination by oil-field
brine was indicated by an increase from background levels of less than 10 mg/L to levels exceeding
1,000 mg/L (Krieger and Hendrickson, 1960). Monitoring of chloride levels in observation wells at oil
fields can easily detect sudden contamination caused by salt-water spills and water quality improvement
subsequent to spills or disposal-pit closures (Fig. 48). Richter and others (1990) determined soil-chloride
concentrations and dissolved chloride in water underlying former brine-disposal pits for identification of
contamination potentials associated with abandoned pits. Comparison of soil-chloride concentrations from
areas outside the pit (background levels) with those under the pit, and knowledge of the amount and
salinity of disposed brine, allowed determination of the salt percentage that had been flushed out and the
salt percentage that remained to be flushed out. In one example, soil-chloride concentrations were up to
three orders of magnitude higher under the pit (5.8 mg/cm3) than outside of the pit (0.007mg/cm3) even
though the pit had been abandoned more than 20 years ago (Fig. 49). Considering the amount of brine
disposed of into the pit system (100,000 barrels), Richter and others (1990) estimated that approximately
four percent of the original brine remained in the soil. Although this percentage appears to be relatively
low, ground-water chtorinity was still very high under the pit (20,750 mg/L). Ground-water contamination
spread from this pit at least 0.5 mile downgradient, as evidenced by a chloride concentration of
12,190 mg/L in a shallow testhole sample. Besides chloride concentrations, other constituents are often
used to identify oil- and gas-field brine contamination. In the previous example, bivariate plots of major
cations versus chloride identified disposal pits and eliminated local, saline ground water as the source of
salinization (Rg. 50), which is indicated by a mixing trend between pit water and testhole waters that does
not coincide with the local ground-water trend.
Krieger and Hendrickson (1960) used Piper plots to demonstrate mixing between oil-field brine and
fresh water, which was suggested from high chloride concentrations. On the Piper diagram, the
contamination is indicated by a straight-line mixing trend from a Ca-Mg-HCO3 type fresh water to the Na-CI
type oil-field brine. Williams and Bayne (1946) used bar graphs to display differences in mixtures of fresh
water and saline formation water from mixtures of fresh water with oil-field brine in Kansas. Mixtures
between oil-field brines and fresh water were reflected in relatively tow percentages of magnesium and
124
-------
o
Salt-water
line leak
Discontinued use of
disposal pits
Septembers, 1963
Discontinued use of disposal pits
• September 6, 1963
Salt-water.
line leak
• • -. 196263 64 65 66 67- 68 69 70 71
. . Year ,
QA17234C
Figure 48. Monitoring of chloride concentrations in ground water as an aid in identifying salt-water
contamination (from Miller and others, 1977).
125
-------
0 -i
-5 -
•10 -
-15
I
£1
- D - 76
234
Cl (mg/cm3)
QA11659C
Figure 49. Relationship between soil chloride and depth inside (testhole No. 76) and outside (testhole
No. 75) an abandoned brine-disposal pit. High chloride concentrations in soil underneath the pit indicate
that residua) salt had not been flushed out since abandonment (from Rfchter and others, 1990).
..•»*•
126
-------
(a)
(b)
60 •
45 •
O 3° -I
15 -
+ + •
78.**+ . 75C
% 75b
77
200 4oo
; Cl (meq/L)
76
600
50
40
'2 2° '
75a»
o '
• 77
75c
75b
200 400
'Cl (meq/L) '
76
600
(c)
(d)
600
400 -
0)
ra
200 -
^"7
77 -
75c
75b
200 400
Cl (meq/L)
76
600
20'
10
•f +•
++* 75a
75c
77
200 400
' Cl (meq/L)
76
600
QA1165BC
Figure 50. Bivariate plots of major ions versus chloride from water-supply wells (crosses) and test holes
(dots) in the vicinity of an abandoned brine-disposal pit (testhole No. 76). Test hole samples plot
intermediate between the pit sample and local ground water, indicating that the salinity is derived from the
pit and not from local saline ground water that affects water-supply wells (from Richter and others, 1990).
127
-------
sutfate, whereas mixtures of fresh water with saline formation water contained higher magnesium and
sulfate percentages. Bumitt and others (1963) used a combination of bar graphs, data posting, and Stiff
diagrams to graphically depict changes in ground water quality in Ogallala outcrop areas of Texas (Rg. 51).
The changes, resulting from oil-field contamination, were (1) an increase in Na, Cl, Ca, and Mg
concentrations, (2) base exchange between Na in the contaminating brine and Ca and Mg in soils and
caliche, causing a decrease in the Na/(Ca+Mg) ratio, and (3) low ratios of HCOjyCI and SO4/CI. Salinity
diagrams (Fig. 17) were used by Nativ (1988) to differentiate between this salinization caused by oil-field
brine from salinization caused by evaporation from a shallow water table. The former is typically associated
with Na-Ci facies, whereas the latter is typically associated with Na-SC>4 facies. Stiff diagrams and
geographic mapping of major chemical parameters were also used by Leyings (1984) in a study of oil-field
pollution in the East Poplar field, Montana. Elevated TDS, Na, and Cl concentrations reflected the extent
of ground-water movement away from the pollution site on isocontour maps, whereas Stiff diagrams
illustrated the change from low-TDS, Na-HCO3 waters (background levels) to high-TDS, Na-CI waters
(produced oil-field waters).
Leonard and Ward (1962) were among the first to use the Na/CI ratio to distinguish oil-field brine from
halite-solution brine in Oklahoma. One type of brine, derived from salt springs in western Oklahoma,
typically shows a Na/CI weight ratio in the range of 0.63 to 0.65, which suggests that nearly pure halite
(Na/CI weight ratio of 0.648) is the source of sodium and chloride in those brines. Another type of brine in
the same area consistently has Na/CI weight ratios less than 0.60, and the ratio decreases with the
increase in chlorinity (Fig. 25). This type of brine was derived from oil-field operations. Based on this
difference in ratios between the two potential sources, Leonard and Ward (1962) determined that halite
solution accounted for the salinity in the Cimarron River samples (#5 and #6, Fig. 25), whereas oil-field
brines accounted for all or most of the saynity in samples from the Little River (#1 through #4) and the
Arkansas River (#7 and #8). Surface-water degradation of streams in Kansas was studied by Gogel (1981)
using the same technique. Samples from the Ninnescah River alluvium ranged from 0.65 to 0.67 in Na/CI
weight ratios, indicating that halite solution in the underlying Wellington aquifer is the source of salinity. In
constrast, very low Na/CI ratios, ranging from 0.54 to 0.28, along Salt Creek, suggest oil-field
contamination. The Na/CI ratio was also used by the Oklahoma Water Resources Board (1975) to
distinguish halite-solution brine from oil-field contamination in parts of Oklahoma. Ratios of Na/CI (0.38)
identified oil-field brine as the source and chloride mapping (isochlors) identified the location of point
sources as well as the extent of contaminant plumes (Fig. 52). Kalkhoff (1986) used the minor
constituents bromide and strontium and the Na/CI ratio to support his conclusions of oil-field pollution in
the example noted above. In this case, a Na/CI ratio less than 0.6 was considered indicative of oil-field
pollution.
Whittemore (1984) pointed out that oil-field brines in Kansas usually have higher Ca/CI and Mg/CI
ratios and tower SO4/CI ratios than halite-solution brines. The same relationship was found for halite-
128
-------
3500
3000-
c 2500 -
.Q
oj 2000-
Q.
.52
•^ 1500 -
^1000-
500-
Permian brine
r.iOOOepm
119,186 ppmCl
Contaminated ground water
Native ground water
r-Sepm
HCO.
0 ,
4000 ft
0 1000 m
Contour interval 10 ft
© Irrigation well ; , . •
•o- Domestic, wall
cj) Abandoned well
t> Contaminated well
n Brine disposal pit .
-3570— Approximate elevation of
' static water table (ft)
CI278 Chloride concentration (ppm)
SO
QA17240C
Figure 51. Identification of salinization sources using bar graphs, Stiff diagrams, and contouring of
constituent concentrations onto maps (from Bumitt and others, 1963).
129
-------
EXPLANATION
Chloride (mg/L)
.......... 7 .:::.::::;a ::::::::: :R4w;::::::: :ioj::::::::: ;n t •••••• • ..• --
is j 17 ; is i....:;;.. is
I
<
35 ::::::.:: 36
iijiniiiH
I mi
QAI724I
Figure 52. Mapping of point source of salinity and the resulting salt-water plume through contouring of
chloride concentrations from water wells (Oklahoma Water Resources Board, 1975).
130
-------
dissolution brines and deep-basin brines in Texas by Richter and Kreitler (1986a,b). However, these ratios
appear to work as tracers best when little chemical reactions occur after mixing of the respective salt-water
source and fresh water. To avoid this change of chemical constituent ratios by mechanisms other than
mixing or dilution, Whittemore and Pollock (1979) suggested the use of minor chemical constituents,
such as bromide, iodide, and lithium, that are relatively conservative in solution. Of those, bromide
concentrations and Br/CI weight ratios are the most widely used tracers because bromide is similarly
conservative as is chloride, and because a significant difference in Br/CI ratios between most oil-field
waters and halite-solution brines could be established (for example, Whittemore and Pollock, 1979;
Whittemore, 1984; Richter and Kreitler, 1986a,b). In Kansas, oil-field brines typically have Br/CI ratios
greater than 10 x 10~4, whereas halite-solution brines typically have Br/CI ratios less than 10 x 10~4
(Whittemore, 1984). This difference can be used to calculate mixing curves between fresh water
composition and the range of values for either endmember (Fig. 26). By superimposing values from
testholes in the Smoky Hill River area, Whittemore (1984) was able to show that halite solution is the
dominant mechanism of salinization in that area (Fig. 26a). In contrast, in the Blood Orchard area south of
Wichita all of the observation-well samples indicate mixing of fresh water with oil-field brines (Fig. 26b).
Note in figure 26b that the sample obtained from the Arkansas River suggests halite solution; this
chemical signature was derived from areas upstream from the Blood Orchard area. Also, some testhole
samples plot between the two mixing fields, suggesting the mixing of fresh water with both halite-solution
brine and oil-field brine. The Br/CI ratio can also be used to identify mixtures derived from different oil-field
brines as long as ratio ranges do not overlap in respective brine endmembers.
In addition to the ratios discussed above, Whittemore and Pollock (1979) pointed out the usefulness
of I/CI ratios in differentiating between oil-field brines and halite-solution brines. Halite-solution brines
typically have I/CI weight ratios less than 1 x 10~5, whereas oil-field brines in Kansas have ratios greater
than 2 x 10~5 (Fig. 31). A similar relationship in I/CI ratios between halite-solution brine and deep-basin
brine was found by Richter and Kreitler (1986a,b) for samples from salt springs and shallow test holes in
north-central Texas (Fig. 28). In the Anadarko Basin, brines in Pennsylvanian and Mississippian strata are
characterized by some of the highest iodide concentrations ever recorded in natural brines, with iodide
concentrations of up to 1,400 mg/L (Collins, 1969). These high contents of iodide may be due to
concentration by shallow-water fauna and flora and additional diagenetic concentration. In brines being
more or less equal in iodide concentration, boron appears to be a good tracer to distinguish
Pennsylvanian brines with tow boron contents from Mississippian brines with high boron contents (Collins,
1969).
Novak and Eckstein (1988) used discriminant analysis and modified graphical techniques in
combination with selected ionic ratios in a study of ground-water quality deterioration in northeastern Ohio
to determine the relative importance of oil and gas brines versus road salt. By applying the ratios of Ca/Mg,
Na/Ca, Na/Mg, Na/CI, K/CI, K/Na, Mg/K, Ca/K, CI/Mg, Cl/Ca, and (Ca+Mg)/(Na+K), not only was it possible
131
-------
to Identify brine contamination but it was also possible to identify the stratigraphic origin of the brine
source. Other chemical and isotopic tracers of brine sources, such as bromide, strontium, oxygen-18,
deuterium, carbon-13, and sulfur-34, were not considered by Novak and Eckstein (1988) because
emphasis was put on easy and inexpensive analytical techniques. Depending on the number of chemical
analyses available, Novak and Eckstein (1988) suggested two methods of investigation. When only a few
samples were available, graphic representation of ratios on modified Piper, Schoeller, or Stiff diagrams
appeared appropriate, whereas discriminant analysis proved valuable when a large amount of analyses
had to be handled.
Morton (1986) identified brine contamination of fresh ground water by oil and gas operations in east-
central Oklahoma using the following major criteria: (1) Cl greater or equal to 400 mg/L, (2) Br greater or
equal to 2 mg/L, and (3) (Na+CI)/TDS greater or equal to 0.64. Supporting evidence for brine
contamination was provided by (4) Li/Br less or equal to 0.01 for Cl greater or equal 400 mg/L, (5) Na/CI
approximately 0.46, (6) Na/Br approximately 92, and (7) Br/CI approximately 4.8 x 10~3. These criteria
were established by Morton (1986) through a series of constituent and constituent-ratio plots. On plots of
Na/Br versus Br (Fig. 53), of Na/CI versus Br, and of Br/CI versus Cl, a bromide concentration of greater
than 2 mg/L appears to indicate less scatter and ion ratios that are similar to or are typical for brines. A
bromide value of 2 mg/L was measured in samples with chloride ranging from 250 to 500 mg/L. A value of
400 mg/L chloride was selected by Morton (1986) as the brine Index because the mean Br/CI ratio of 4.8 x
10~3 intersects the bromide line of 2 mg/L at that concentration. This value of 400 mg/L chloride also
appears to define the cutoff between large and small scatters on a Na/CI versus Cl plot (Fig. 53). Accepting
the brine index of 400 mg/L chloride, a (Na+CI)/TDS index of S0.64 is suggested. The (Na+CI)/TDS index
and the 400-mg/L-chIoride index can be used to evaluate available data that don't include dissolved
bromide as an analyzed parameter, as is most often the case.
Mast (1982) suggested the use of mixing diagrams (Rg. 32) for differentiating between mixing of
fresh water with naturally saline ground water and mixing of fresh water with oil-field brine in Kansas. This
technique is discussed in chapter 3.2 on halite solution.
3.4.4. Significant Parameters
Oil-field brines have some of the highest Br/CI ratios found in natural salt waters. Ratios typically are
greater than approximately 10 x 10"4 In oil-field brines and less than 10 x 10"4 in halite-solution brines
(Whittemore and Pollock, 1979; Whittemore, 1984; Richter and Kreitler, I986a,b). Ratio differences
between these two potential endmembers of mixing with fresh water are generally big enough to allow
differentiation of the respective source in brackish water down to chloride concentrations of a few
hundreds of milligrams per liter, although identification is best at high concentrations.
132
-------
•
.
^ i.ooo-
.2 :
£ -
.c
2>
'3
J "
*•••'
ffi
18
100 —
-
Limit of sddium/bromid
_ Data from brine analyses by
various oil companies
e ratios
./when bromide concentration < 2 mg/L • Ground-water sample
*•.,
• I
* •
* 4
• /*
! • •'•;
: «-.:\
• • .1
1. .• . '. A
, i !'..v:-v-.
i • .*,-»
! ! i«J.ro>
I* • • • 9 ' '
. * •; *. •. ;
..-•*-. •-._ •;•.•". ;
.? Questionable validity
Petroleum-associated-brine index:
^ bromide concentration > 2mg/L
• ?
* '
••-..'•:.• .•-..'• • - • - : •
*. Sea water •
' • * • • . ••-•«•• •"•- '
••>' .••"• fJ.. * ".. . • . *
• * •*. .• Median sodium/bromide
* • ^ ratio= 92 when bromide
• • * " - concentration > 2 mg/L
J • i • •
O o
i i iii iii j -i iiiiiiii . i~ i. .1,111111. i iiiiiii
0.1 1.0 , 10 100 1,0
Br(mg/L)
Na / Cl (weight ratio]
b o
•iiiil i i iiiiiii
.'.1.
• • ** •
\. « ' . ' * * *
'^JVv*:'1-
* *Ss.^«s * *
• • »^^L" •** * • * *
• * • o
*
•? .?
0 . -10 v, •'-. 100
_ Data from brine analyses by
.various oil companies
• Ground-water sample
.? Questionable validity
/Petrtj'eum-associated-brine index:
bromide concentration > 400 mg/L
Median sodium/chloride
ratio = 0.46 when chloride
* , concentration > 400 mg/L
• . / . • • Sea water*
»"• •••%f •" ..•-••. •"•
1,000 , • 10,000 •-" •.•' 100,000 r.ocx
CI(mg/L)
QA17242C
Figure 53. Determination of geochemfcal constituent criteria that indicate brine contamination of fresh
ground water in parts of Oklahoma. Through a series of constituent plots, background values and mixing
values can be established, based on theoretical mixing relationships between fresh water and brines (from
Morton, 1986).
133
-------
The ratio of Na/CI works well to distinguish oil-field brine from halite-solution brine at high chloride
concentrations. Sodium and chloride occur in halite at equal molar concentrations (Na/CI molar ratio = 1;
Na/CI weight ratio » 0.648). Brines that originate from solution of halite within a shallow ground-water flow
system will exhibit a similar ratio as long as concentrations are high enough so that the Na/CI ratio is not
affected appreciably by ion exchange reactions. In most oil-field brines molar Na/CI ratios are much less
than one.
3.4.5. State-by-State Summary of Oil-and Gas-Reid Contamination
Production of oil and gas once was or is now occurring in parts of 25 states (Rg. 4). Associated with
this production are different degrees of ground-water problems, as summarized on a state-by-state basis
in this section.
Alabama: Ground-water contamination from brine has occurred in the Pollard, Gilbertown,
Citronelle, and South Cariton oil fields. Principal sources of contamination in oil fields are (a) unlined
disposal pits, (b) leaks from pipelines, and/or (c) spills (Miller and others, 1977).
Arkansas: The disposal of brine and brackish water produced with oil and gas has caused
deterioration of several ground-water basins in Arkansas. The Arkansas Department of Pollution Control
and Ecology (1984; Morris, 1988) estimated that approximately 20,000 acres are affected by petroleum-
Industry waste and saH water.
Ground-water salinization has been reported in Independence, White, Woodruff, Chicot, Ashley,
and Union Counties (Atkinson and others, 1986). In Miller County, brine-polluted ground water
(Cl>40,000 mg/L) can be found in river alluvium along the Red River west of Garland City. A faulty disposal
well and disposal of brine into a surface pit accounted for a brine plume covering one square mile. It is
estimated that this plume will extend within the next 250 years to 4.5 mi2 before all salt is flushed into the
river. Two other polluted areas and four possibly polluted areas have been found in this county
(Fryberger, 1972) where more than 60 million gallons of fresh water have been contaminated by brine
disposed into surface pits (Ludwig, 1972). Disposal of oil-field brines also contaminated many streams and
associated ground waters. Streams that have been most affected are in the southern part of the state and
include Bodcaw, Comie, Smackover, and Lapile Creek, Bayous Dorcheat, and the southern reaches of
the Ouachita River (Scatf and others, 1973).
Oil-field brines in southern Arkansas are characterized by relatively high calcium (30,000-
50,000 mg/L) and bromine (1,500 to greater than 3,000 mg/L) concentrations (Collins, 1974).
Colorado: Disposal of brine and inadequate abandonment of exploration and production wells has
led to numerous occurrences of ground-and surface-water pollution. Uncased exploratory oil wells in the
White River area caused an increase in the river's salt toad from 9.9 million Ib/year to as high as
52.9 million Ib/year (Feast, 1984). For example, a 1,837 ft deep oil test drilled in 1915 was allowed to
134
-------
discharge approximately 52,000 metric tons of TDS per year (TDS of 19,200 mcj/L at a discharge rate of
1,350 gpm) over a 53-year period in unrestricted flow to the White River in western Colorado. Subsequent
plugging of this well caused other nonflowing wells in the area to flow and created saline seeps in the
vicinity of these wells (van der Leeden and others, 1975). During 1969, approximately 27,000,000 billion
barrels of salt water were disposed of in surface pits, affecting entire ground-water basins. For example,
the South Platte River Basin exhibits annual increases in TDS, with total concentrations currently ranging
from 500 to 4,000 ppm. One single disposal pit in the Severence Basin at Weld County contributes
approximately 27 ppm TDS to the entire ground-water basin (Rold, 1971).
Georgia: An abandoned oil-test well in Glynn County allowed upward flow of saline water from
saline zones below 2,000 ft into several fresh-water zones between 610 and 920 ft. Chloride
concentrations vary from 40,000 ppm in sands at 4,150 ft depth to as little as 18 ppm at 1,500 ft, resulting
in a discharge-water concentration of 7,780 ppm at land surface (Wait and McCollum, 1963).
Indiana: An abandoned, unplugged exploration borehole at Terre Haute polluted three high-
capacity water wells 2,000 ft away with saline water containing 8,600 ppm of chloride. Static water level in
the deep hole was 28 ft above the static water level of the shallow aquifer (Gass and others, 1977).
Kansas: Many areas in the state have been affected by oil and gas.activities. Past disposal of oil-
field brine still yields salt water to surface water in some oil-field areas (Krieger and others, 1957).
Oil-field brine is a major source of saline water in the Walnut River Basin, contributing perhaps
40 percent of the TDS load downstream from the confluence of the Walnut and Whitewater Rivers
(Leonard, 1964). Shallow ground water in part of the Little Arkansas River Basin in south-central Kansas
has been contaminated by oil-field brines and municipal wastes. Oil-field brines enter shallow ground
water through corroded or improperly cased wells, by upwelling of injected brine, and by leakage from
disposal ponds (Leonard and Kleinschmidt, 1976). Oil-field brine also mixes with fresh ground water and
with saline formation water along the eastern side of the Elm Creek Valley in Barber County (Williams and
Bayne, 1946) and around Salina (Atkinson and others, 1986).
Kentucky: Salt-water problems in Kentucky are generally associated with the oil and gas industry
and not with overpumping and inflow of natural saline ground water, although most of the state is
underlain by shallow saline ground water (Atkinson and others, 1986).
Oil-field brines have contaminated shallow ground water in the Upper Big Pitman Creek Basin.
Contamination started with the practice of brine disposal into ponds, sinks, and local drainage ways and
continued with brine injection into the shallow subsurface (175 ft below land surface) due to the presence
of inadequately abandoned test wells. Chloride concentration increased from 60 ppm prior to oil
production to as high as 51,000 ppm after oil production. Secondary-recovery operations may increase
the extent of contamination if abandoned wells are not plugged (Hopkins, 1963).
Brine disposal at oil fields have contaminated water wells and springs in the upper Green River Basin,
especially in Green and Taylor Counties. Before oil production (August 1958), river water was of the
135
-------
Ca-Mg-HCO3 type with an average chloride concentration of less than 10 ppm. With the development
of the oil field, the water changed to a Na-CI type with chloride concentrations frequently exceeding
1,000 ppm. Chloride discharge of the Green River at Munfordville increased from 10,600 tons in 1957 to
305,000 tons in 1959. An increased chloride load is reported as far as 100 miles downstream from the
areas of heaviest brine production (Krieger and Hendrickson, 1960).
Upward flow of oil-field brine from disposal zones in the Louisville Limestone through abandoned oil
and gas wells has contaminated shallow ground water in the state; chloride concentration increased locally
from 60 mg/L prior to oil production to 51,000 mg/L after oil production (Gass and others, 1977).
Louisiana: Oil and gas activities in salt-dome areas in the southern and northern parts of the state
are potential sources of saUnization. (Atkinson and others, 1986). Oil-and gas-field operations contribute
large amounts of brine to surface-water bodies along the Louisiana coast, as state regulations allow
disposal of brine into naturally saline or otherwise unusable water. Since 1938, millions of barrels of brine
have been disposed of into lakes, canals, bayous, and marshes in Lafourche Parish, contaminating
sediments and surface waters and destroying vegetation (Hague, 1989). In 1986, approximately
2.6 million barrels of oil-field brine, which is equivalent to 70 percent of the state's total brine production,
were being discharged into surface water at 698 stations along the coast (Van Sickle and Groat, 1990).
Lately this practice has raised new concern because of occasionally high concentrations of naturally-
occurring radioactive materials in oil-field brines.
Disposal of drilling and production wastes has contaminated surface and shallow ground water in
Vermilion Parish, where these wastes had been disposed of by injection, unlined surface impoundments,
land application, landfill, burial, and marsh reclamation (Subra, 1990).
Michigan: Hundreds of millions of gallons of highly mineralized water have been leaking through
abandoned boreholes during the past 80 to 100 years, creating widespread salinization throughout the
state (Gass and others, 1977). The Michigan Department of Natural Resources (1982) reported
contamination cases caused by oil-and gas-field activities in 19 counties during 1979.
Mississippi: The first successful oil well was drilled in Yazoo County in 1939. Since then, several
thousand wells have been drilled throughout the state, making brine disposal one of the major sources of
salt-water contamination in the state. According to an inventory by Miller and others (1977), in 13 of
25 counties in which brine-disposal wells are located, contamination of ground water due to oil-field
brines has been reported. Saline springs and seeps resulting from the disposal of brine into unlined
surface pits have been encountered in Wayne, WiBdnson, and Yazoo Counties. Contamination of surface
and shallow ground water by oil-field brine has also been reported in the Brookhaven, Baxterville, Pistol
Ridge, Little Creek, and Tinstey oil fields as well as at numerous areas in Adams County (Kalkhoff, 1986),
where chloride background levels in uncontaminated fresh water are less than 20 mg/L. Affected surface
waters include Tallahala Creek, Leaf River, Chickasawhay River, Eucutta Creek, Yellow Creek, and Mortons
Mill Creek (Shows and others, 1966; Baughman and McCarty, 1974).
136
-------
Montana: Locally, deterioration of ground-water quality has occurred from leaky wells and brine
disposal. Major areas of oil production are located in the Powder River Basin, the Williston Basin, and the
Sweetgrass Arch. During the period of 1950-1970, ten seismic companies drilled more than 300,000
seismic holes in Montana. There is concern that holes drilled during oil and gas activities may allow
crossformational flow between multiple aquifers as a result of head differentials in inadequately plugged
wells and test holes (van der Leeden and others, 1975).
Disposal of oil-field brine in surface pits and wells or leaky wells and pipelines have caused
contamination of surface water and shallow ground water in the East Poplar oil field, Roosevelt County.
Chloride concentrations as high as 45,000 mg/L were measured in testholes 45 ft deep. The Poplar River
exhibits a chloride increase from 20 mg/L to 880 mg/L in the area, associated with a change in water fades
from a Na-HCQs type to a Na-CI type (Levings, 1984).
Nevada: The only oil-producing area is Eagle Springs field in Railroad Valley of east-central Nevada
(Van Denburgh and Rush, 1974). Disposal of brine in that field occurred through ponds, from which the
water infiltrated the valley-fill alluvium. Since oil production began in 1954, about 500 acre-ft (through
1971) of brine containing 25,000 to 30,000 mg/L TDS were disposed of by this method. This amount is
equal to 17,000 to 20,000 tons of salt. There are indications that shallow ground water is affected by the
percolating brine, as suggested by an increase in chloride concentrations from background levels smaller
than 30 mg/L to contamination levels of 66 mg/L in a 79-ft domestic well (Van Denburgh and Rush, 1974).
New Mexico: Oil- and gas-field operations have caused salinity increases in the San Juan River
valley-fill aquifer and in the Pecos River valley-fill aquifer (Ong, 1988). Instances of ground water
salinization in the southern part of the state are suspected to have been caused by disposal of oil-field
brines (U.S. Geological Survey, 1984), such as in Lea County, where a leaky brine pit contaminated fresh
ground water (McMillion, 1970).
New York: Oil and gas seeps have been known to occur in western New York since historic time, as
documented by accounts about Indians collecting oil from the Seneca Oil spring near Cuba. Other sites of
oil springs are at Freedom, Allegany County, and around Canandaigua Lake. This discharge of gas or oil at
land surface is probably too small to have contaminated water supplies; an exception to this might be Oil
Creek near Cuba. Oil production is concentrated in Allegany and Cattaraugus Counties and to some
extend in Steuben and Chautauqua Counties. These counties have the greatest potential of oil-well
pollution. Among the areas of oil-field pollution are the valleys of Chipmunk and Knight Creeks and
Qenesee River, where surface-water chlorinities of up to 4,000 mg/L have been recorded (Grain, 1969). In
1973, there were approximately 5,400 operating oil and gas wells in the state (Miller and others, 1974).
Some domestic wells in Chemung County appear to have been contaminated by mineralized water that
entered shallow aquifer units through old gas wells, as indicated by an increase in chloride concentrations
from background values of less than 10 mg/L to several hundreds of mg/L (Miller and others, 1974).
Randall (1972; Miller and others, 1974) suggested similar conditions for the Susquehanna River Basin. In
137
-------
the Jamestown area, abandoned oil and gas wells, which had been drilled through beds of halite, are
suspected to allow vertical migration of salt water into shallow fresh-water aquifers (Grain, 1966; in Miller
and others, 1974).
North Dakota: Oil- and gas-field activities, which started in the early 1950's, contribute to the
salinization problem in the state. Between 1951 and 1984, an estimated 9,000 wells were drilled in the
state. Associated with drilling is the use of highly saline drilling fluid, which is commonly buried in pits and
trenches after drilling has ceased. Murphy and Kehew (1984) estimated that approximately 360 million
cubic feet of drilling have been buried that way in the state.
Ohio: In mid-1979, roughly 35,000 active oil and gas wells produced approximately 40,000 barrels
of brine per day (Templeton and Associates, 1980). These brines were produced nearly exclusively from
the Clinton, Trempealean, and Berea Formations. Disposal of these brines has caused major problems in
the state (Atkinson and others, 1986).
Water pollution associated with oil-field brines has been documented in Morrow, Delaware, and
Medina townships (Pettyjohn, 1971). Approximately 13 mi2 of land have been affected by oil-field brine in
Morrow County after only 3 years of oil production (Lehr, 1969). Contaminated ground water underlying
the affected areas appears to move like a slug and does not mix extensively with fresh water. Major causes
of contamination are disposal of brine into surface pits and the indiscriminate dumping of salt water by
contract truckers. Two disposal pits at Delaware, Delaware County, have contaminated ground water,
affecting approximately 20 acres of land along the Olentangy River (Lehr, 1969).
Ground water has been and still is being contaminated by abandoned brine-disposal pits along the
Olentangy River in central Ohio. Leaching of salt from soil underneath the pits occurs with each
precipitation event, creating repeated plumes of salt water in the area. Chloride concentrations in the
plume often exceed the concentration of the original brine disposed of in the pits (CI-35,000 mg/L) due
to brine evaporation prior to infiltration. The process of leaching out of soil salts appears to continue for
many years after ponds have been closed and reclaimed (Pettyjohn, 1982).
Leaky brine pits, insufficient surface casing, and poor cementing have caused fresh-water
contamination in Perry Township, Lake County (Novak and Eckstein, 1988).
Oil brine is spread onto roads as an anti-dust agent in 77 townships of Ohio, resulting in elevated
concentrations of heavy metals in soils (Kalka and others, 1989). Breen and others (1985) estimated that
of the 38,000 barrels of brine produced every day in Ohio, approximately 75 percent are disposed of by
road spreading, evaporation ponding, and illicit dumping. Contamination of surface and ground water from
brine spreading deviates from contamination caused by road salting (rock salt or salt-solution brine),
although most of the oil-field brine constituents appear to be absorbed in the ground within a few tens of ft
from the roads (Bair and Digel, 1990). Periodic application during winter and summer months can lead to
high chloride concentrations in ground water adjacent to highways.
138
-------
Oklahoma: Oil-field brines constitute problems in the Cimarron Terrace from Woods County
southeast to Logan County (Atkinson and others, 1986).
Many streams have been contaminated by salt water from the disposal of oil-field brines, leading to
pollution of shallow ground water recharged by these streams. The Arkansas, Canadian, Cimarron, and
Red Rivers have been affected by salt-water. Oil-field brine pollution has been reported from Garvin,
Pohtotoc, Seminole, Oklahoma, Pottawatomie, Lincoln, Okfuskie, and Creek Counties. In 1970,
approximately 15,000 salt-water injection wells existed in Oklahoma (Scalf and others, 1973).
Salt springs and seeps issuing from underlying salt beds increase the salinity of the Cimarron River
near Mocane. The salinity of the river water is increased further in the lower reaches by oil-field brines. Oil-
field brine also causes salinization of the North Canadian River downstream from Oklahoma City,
amounting to TDS concentrations at times exceeding 15,000 ppm (Krieger and others, 1957).
Salt-spring brines in western Oklahoma can be distinguished from oil-field brines using the ratio of
sodium over chloride. Spring brines typically exhibit Na/CI ratios greater than 0.60 (weight ratios),
suggesting halite dissolution to be the source of the salt water. Oil-field brines, in contrast, exhibit Na/CI
ratios smaller than 0.60, and the ratio decreases with an increase in chloride concentration (Leonard and
Ward, 1962). Brine contamination of fresh ground water in parts of south-central Oklahoma is indicated by
chloride concentrations of greater than or equal to 400 mg/L, bromide concentrations greater than or
equal to 2 mg/L, the ratio of Na+CI over TDS is greater than or equal to 0.64, the ratio of U to Br is less than
or equal to 0.01 at Cl concentrations greater than or equal to 400 mg/L, a Na/CI ratio of approximately 0.46,
a Na/Br ratio of approximately 0.92, and a Br/CI ratio of approximately 0.0048. Chloride and bromide are
the most reliable brine indicators. Water-quality degradation in the area may be caused mainly by oil-and
gas-field activities (Morton, 1986).
A large area near Crescent was contaminated by oil-field brine which seeped into terrace sands
through evaporation pits (Oklahoma Water Resources Board, 1975). It is estimated that possibly more
than 100 years will be required to naturally flush out the saline water.
Shallow soil and ground water have been contaminated by seepage from a brine-disposal pit near
Bums Flat, Washita County (Shirazi and others, 1976). Inadequately plugged wells polluted ground water
near the community of Sasakwa (Smith, 1989).
Pennsylvania: In their report, Miller and others (1974) stated that salt water produced in the state
is usually disposed of in unlined surface ponds or is discharged into the ground where it can infiltrate into
shallow fresh water aquifers. Several wells and springs in Venango County have been contaminated by
this kind of brine discharge.
Oil and gas activities have led to ground-water pollution occurrences that range in aerial extent from
1 acre to more than 50 square miles (U.S. Geological Survey, 1984). Because of the geomorphology of
the Appalachian Plateau characterized by gently dipping strata incised by major streams, any
139
-------
contamination occurring in upland areas can cause regional ground-water contamination along the flow
paths toward discharge areas (that is, in valleys) (Miller and others, 1974).
Leakage of acid mine drainage through old oil and gas wells and other open holes has caused
extensive ground-water pollution in the coal fields of the state (Gass and others, 1977).
South Dakota: In 1950 it was estimated that approximately 12,000 to 15,000 artesian wells within
the state leak water into overlying aquifers above them. Inadequately plugged test holes drilled for oil, gas,
and uranium may permit upward migration of saline water even in areas where no production is occurring.
Salt and gypsum beds in the Spearfish Formation have to be "salted up" during drilling, introducing the
potential of salt-water contamination from unlined drill pits. In 1980, there were approximately 27 brine-
disposal pits in operation. In 1985, only 50 percent of the produced brine was injected into deep
formations (Meyer, 1986).
Leakage from an abandoned well caused salinization of a municipal well in Avon (Jorgensen, 1968,
Gass and others, 1977).
Texas: More than 1.5 million wells have been drilled in the state in search for oil and gas (Texas
Water Commission, 1989). An inventory of brine production and disposal on the more than 67,000 oil and
gas leases throughout the state revealed that a total of 2,237,000,000 barrels of salt water were produced
during 1961. Of this total, about 68.7 percent (1,537,000,000 bbls) were re-injected into the subsurface,
about 20.6 percent (461,000,000 bbls) were disposed of into open surface pits, 10.1 percent
(225,000,000 bbls) were disposed of into surface water, and 0.6 percent (14,000,000 bbls) were
disposed of by miscellaneous methods such as spraying on leases and highways (Miller, 1980; Texas
Water Commission, 1989). As of 1962, 20,000 to 30,000 injection wells were in operation. In the early
years of oil and gas operations, casing programs in hundreds of oil and gas fields were insufficient to
protect fresh ground-water resources. Serious regional contamination problems may have occurred in
areas where highly pressured brine aquifers occur, such as the Rustler Formation of southwest Texas, the
Coleman Junction Limestone in west-central Texas, and the deep Miocene brine aquifers of the Gulf
Coast. Plugging of boreholes often was done by simply putting wood, mud, or rocks into the hole and dry
holes were often left uncased (McMilBon, 1965).
In the early days of exploration, oil-field brines were often diverted into surface streams, as reported
from fields in Orange and Hardin Counties. Brines were collected at central points and then pumped
through pipelines or allowed to flow through open ditches to estuaries of the Gulf of Mexico (Schmidt and
Devine, 1929). In those days, brines were often disposed of onto the land surface by purposely
spreading, adding large amounts of salt to soil in oil and gas fields and causing "vegetative kill" areas in
many instances. Leaching of salt from such areas may continue to contribute salinity to surface and
ground water.
Hundreds of instances of contamination of fresh-water wells as a result of oil-field-brine disposal into
surface ponds on the surface of the Ogallala Formation have been reported. Bumitt and others (1963)
140
-------
specify some counties (and instances), such as Andrews County (1 occurrence), Cochran County (6),
Ector County (4), Gaines County (11), Qarza County (9), Glasscock County (2), Hockley County (9),
Hutchinson County (1), Lamb County (1), Lubbock County (3), Lynn County (2), Moore County (1), Terry
County (5), and Yoakum County (6). In Winkier County water wells were contaminated by brine disposal
into unlined pits and into a communal disposal lake (Garza and Wesselman, 1959). in addition, waste
disposal from industrial plants has contaminated wells in Carson, Hockley, Ector, Howard, Moore, Terry,
and Yoakum Counties (Burnitt and others, 1963). Disposal of oil-field brine in surface pits has also caused
contamination of shallow water wells in Clerhville, Matagorda County. Chloride concentrations in affected
wells range from 940 ppm to 4,000 ppm (Shamburger, 1958).
Many streams and adjacent ground-water units have been contaminated by salt water as a result of
oil-field brine disposal, especially the Brazos and Pecos Rivers. Contamination from brine pits has been
reported in Baylor, Cochran, Colorado, Comanche, Cooke, Dawson, Ector, Gaines, Glasscock, Harris,
Karnes, Knox, Montague, Pecos, Matagorda, Runnels, Rusk, Victoria, Wilbarger, Wilson, and Winkier
Counties, just to name a few. Contamination problems associated with brine-disposal wells are known from
Coleman, Karnes, Shackelford, Victoria, and Wilbarger Counties. Improper plugging of test holes may
allow downward movement of saline water into the Trinity Sands near the town of Sherman. Unplugged
wells in areas underlain by artesian brine aquifers have also resulted in flowing salt water in Knox, Hopkins,
and Young Counties, and allowed intrusion of salt water into fresh-water zones in Crockett, Duval, Fisher,
Glasscock, Runnels, and Scurry Counties. Contamination by natural gas through faulty well construction
has been reported in Caidwell, Bastrop, Comanche, and Wharton Counties, whereas an abandoned well
in the Trinity Bay area of Chambers County allowed sea water to contaminate a fresh-water zone (Scalf and
others, 1973). In the area of Harrow, brine contamination may have been caused by corroded casing in
brine-injection wells, upward flow of injected brine through unplugged, abandoned wells, or disposal of
brine into unlined surface pits (Fink, 1965).
Salt-water spills have damaged exceptionally large areas of soil in the following counties (affected
acreage in parenthesis): Clay (35,000), Crane (31,250), Ector (15,000), Howard (82,700), Pecos
(83,510), Ward (21,500), Winkier (21,940), and Young (18,800) (Texas State Soil and Water Conservation
Board, 1984).
Utah: Oil-field brine injection has locally contaminated ground water in the Dakota, Entrada, and
Navajo Sandstones of the Montezuma Canyon area (Kimball, 1987). Approximately 97 percent of all brine
disposed of into unlined surface impoundments in the Greater AKamont-Bluebell field is lost to seepage
into the shallow aquifer system causing local contamination of shallow ground water (Baker arid
Brendecke, 1983).
West Virginia: Historic records report on salt and oil springs and shallow brine occurrences at
various localities in West Virginia. Among those are Campbells Creek, Kanawha County, and Bulltown,
Braxton County. Exploration and production of oil and gas has contributed to the salinization of shallow
141
-------
ground water and of surface water. As much as 80 bbts of brine are produced with every barrel of oil; the
brine is reinjected into the Injun Sand and the shallower Salt Sands (Bain, 1970).
In the past few decades, salt-water migration toward the land surface has been caused by vertical
leakage along hundreds of unplugged wells and test holes. These wells had been drilled during
exploration for brine, oil, gas, and coal, and commonly were abandoned uncased or improperly plugged.
In Fayette County, chloride concentration of ground water increased from 53 mg/L to greater than
1,900 mg/L within 51/2 years due to fresh-water pumpage and inflow of salt water from abandoned
boreholes. Secondary recovery of oil in an area of Kanawha County caused accelerated migration of brine
upward into fresh-water resources. Oil-field activities also caused an increase in chloride concentration
from 100 mg/L to 2,950 mg/L in an aquifer near Wallace, Kanawha County. Hydraulic connection between
a brine aquifer and an overlying fresh-water aquifer resulted in upward migration of salt water in oil field
areas of Roane County in direct response to pressure-injection disposal of produced brine, causing
chloride increases in fresh-water wells from less than 25 mg/L to more than 1,500 mg/L within several
months (Wilmoth, 1972).
Problems of oil seepage, old and leaky wells, and brine disposal in the Sisterville, Bens Run, St.
Mary's, Belmont, and Waveriy oil fields exist between Paden City and Waveriy (Cariston and Graeff, 1955).
Brine discharge to the river in the Sisterville oil field contaminated the alluvial aquifer, where salt water'is
I
reported just above bedrock at 65 to 70 ft. Brine also contaminated the alluvial aquifer through abandoned
and leaky wells. Similar conditions were reported at Cox Landing, Ceredo, and Matamoros (Cariston and
Graeff, 1955).
Wyoming: Major salt-water problems in the state are associated with oil and gas activities but not
with naturally occurring saline ground water (Atkinson and others, 1986).
Mixing of saline water and fresh water due to salt-water intrusion along abandoned oil wells has
occurred locally. Degradation of water quality has also been reported to have occurred as a result of
irrigation-return flow (Newport, 1977).
Leakage of hydrogen-sulfide gas and salt water into fresh water through oil-well holes has been
reported in the Bighorn Basin, especially in an area east and north of Worland. Other areas of oil-field
activities are located In the Laramie, Hanna, Wind River, Green River, and Powder Basins (van der Leeden
and others, 1975).
3.5 Agricultural Sources
3.5.1 Mechanism ,
Contamination of surface and ground water from agricultural activities may be associated with
irrigation, animal wastes, and commercial chemicals, such as fertilizers, pesticides, and herbicides. These
142
-------
sources may all result in increased salinity of surface and ground water. They may also occur in regions
where alternate sources of salinity are present. Because contamination of agricultural sources can result in
increased salinities, they need to be identified, especially in the contest of differentiating them from other
salinity sources. Agricultural management techniques can also lead to development of saline seep, which
will be discussed in chapter 3.6. The following section will not deal with ground-water contamination from
agricultural chemicals, such as pesticides, herbicides, and fertilizers, which were discussed in detail by
O'Hare and others (1985) and by Canter (1987).
Under natural conditions, a balance exists between the amount of salt entering the soil and the
amount of salt that is removed. This balance maintains a certain quantity of salt in the soil that is needed for
and tolerated by local vegetation. Change from natural vegetation to agricultural crops and application of
irrigation water adds salt to the system. Irrigation can deteriorate ground-water quality in two ways, by inflow
of saline water in response to heavy pumpage and by irrigation-return flow. Inflow of saline water as a result
of pumpage is included in chapter 3.1 (naturally-saline ground water). The effects of irrigation-return flow
on surface- and ground-water quality are discussed below.
Irrigation-return flow water is water that has been diverted for irrigation purposes, was not consumed
by processes such as evaporation and transpiration, and finds its way back into surface or ground-water
supplies. Irrigation without proper drainage, at insufficient or excessive amounts, the use of poor-quality
irrigation water, and solution of surface and soil salts can cause the same effect on soil- and ground-water
salinization. Many irrigation projects have caused serious problems of waterlogging and salinity because
the relationship between irrigation and drainage had not been realized (Food and Agriculture Organization
of the United Nations, 1973). Irrigation-return flow becomes concentrated in chemical constituents from a
variety of sources, such as evapotranspiration, solution of minerals, and, solution of agricultural residues,
such as animal wastes, fertilizers, herbicides, and pesticides. Major processes that occur at land surface
are evaporation, solution, and erosion. The degree to which these processes are active depends on
factors such as type of soil and rock material, topography, climate, irrigation method and rate, vegetative
cover, and the quality of the applied water (Balsters and Anderson, 1979). Infiltrating irrigation water is
subject to changes caused by transpiration, evaporation, leaching, ion exchange, and filtration.
Approximately 60 to 65 percent of the supplied irrigation water is consumed by growing crops (Law and
others, 1970). As water is consumed by plants, most of the salts stay behind. Where natural precipitation
is low or drainage is inadequate, these salts accumulate in the soil. This salt has to be removed by an
adequate drainage system to permit continued plant growth. Excess irrigation or precipitation water can
leach the salt down the soil column to the ground water where it travels to water wells or to natural
discharge areas (Fig. 54). In the USSR, five stages of ground-water salinization resulting from irrigation are
recognized: (1) an increase in salt concentration during the first years as native soil salts are dissolved;
(2) a possible reduction of salt due to higher rates of salt removal than salt dissolution; (3) increase in
salinity due to evaporation from a shallow water table which rose to (6-9 ft) below land surface;
143
-------
Irrigation
Canal
Surface runoff of
irrigation water
Salts being added to river
from ground-water discharge
Shale with soluble salts
QA17243C
Rgure 54. Transport of salt to discharge areas as a result of irrigation-return flow (from Van der Leeden and
others, 1975).
144
-------
(4) reduction through improved artificial drainage; and (5) steady-state conditions of stabilized ground-
water salinity (Framji, 1976).
Irrigation with surface water is a major cause of soil and ground-water salinization (Linger, 1977). This
problem is especially acute in Montana, Wyoming, Washington, Utah, and North Dakota, where 90 percent
of the cropland irrigation water is derived from surface water, as well as in California, Arizona, and New
Mexico, where more than 50 percent of the irrigation water is derived from surface-water bodies (Linger,
1977). Salt concentrations in return flows are from two to seven times higher than in the originally applied
irrigation water (Utah State University Foundation, 1969). In most instances, these increases are within
permissible limits. However, where salt content in the soil is high due to insufficient natural leaching, such
as in arid and semiarid areas of the western United States (Table 9), salt content in irrigation-return water
can be relatively high. Where drainage is impeded by a high water table, concentrations of soil solutions
may be 40 to 80 times higher than concentrations in the irrigation water (Doneen, 1966).
Saline-soil problems may develop in agricultural areas that are low saline and well drained under
natural conditions but inadequately drained for additional ground water from irrigation practices. In such a
case, the water table might rise within a few feet of land surface, from which water evaporates easily (see
also chapter 3.6 on saline seep). This increase in salinity within the shallow subsurface, the root zone, can
lead to salinities too high to support plant growth. Similarly, application of high-TDS irrigation water on soils
with poor internal drainage can lead to serious soil-salinity problems (U.S. Department of Agriculture,
1983).
Of the total withdrawal of ground water in the United States during 1970, approximately 35 percent
was used for irrigation purposes (Geraghty and others, 1973). This percentage is higher in agricultural-
dominated states, such as Nebraska, where 85 percent of all ground water pumped in 1970 was used for
irrigation (Engberg and Druliner, 1988). Approximately 50 million acres were irrigated nationwide in 1970,
with approximately 90 percent of that irrigated area occurring in 17 western states (Fig. 5), where
precipitation is insufficient for crop production and where evaporation is high. Irrigation-return flow is of
less concern in the eastern states, along the Gulf Coast, and in the Mississippi Valley, where precipitation
is plentiful and provides a high degree of dilution and sufficient leaching. Associated with the leaching of
soils by excess irrigation water is the flushing of dissolved nitrogen compounds (mostly nitrate) into
ground water. Sources of nitrogen are natural soil compounds, which are oxidized to nitrate and dissolved
in water preferentially under irrigated and cultivated land, animal wastes and septic tank effluent, and
fertilizers (Kreitler and Jones, 1975; Krertler, 1979).
Pollution of water by animal wastes is increasing because of the increasing number of animals being
raised and because of modem methods used in the livestock industry, which may result in higher animal
populations. Major sources of pollution are beef cattle, poultry, swine-feeding, and dairy industries (Miller,
1980). Problems associated with these industries include (a) the large supply of nutrients supplied to
145
-------
Table 9. Percentage of saline and sodic areas in seventeen western states and in Hawaii
(from Utah State University Foundation, 1969)1.
State
Arizona
California
Colorado
Hawaii
Idaho
Kansas
Montana
Nebraska
Nevada
New Mexico
North Dakota
Oklahoma
Oregon
South Dakota
Texas
Utah
Washington
Wyoming
TOTAL
Area reported
Statewide
Statewide
Statewide
7 areas
All but 3 counties
Statewide
4 areas
Statewide
Statewide
Statewide
6 areas
Statewide
Statewide
Statewide
4 areas
7 areas
23 counties and the
Columbia Basin
Statewide
Total
acreage2
1,565,000
11,500,000
2,811,532
117,418
1,880,063
421,545
1.242.7283
1,218,385
1,121,916
850,000
2.636.5003
826,650
1,490,394
1 ,697,974
2,198,950
1,390,222
2,221,484
1 261 132
36,451,893
Saline (all
Acres
398,830
3,744,951
981,828
45,550
252,945
102,330
197,671
290,000
475,600
191,000
816,630
193,750
103,361
1,196,266
275,854
512,782
266,254
279.703
10,325,305
classes)
Percent
25.5 '
32.6
34.9
38.8
13.5
24.3
15.9
23.8
42.4
22.5
31.0
23.4
6.9
70.5
12.5
36.9
12.0
22.2
28.4
1 Unpublished data from the U.S. Salinity Laboratory
2|rrigablo
3Arable
146
-------
aquatic systems and the consumption of dissolved oxygen in aquatic systems, (b) bad odor, and (c) the
addition of chemical and biological constituents to water resources (Sweeten, 1990).
The use of chemical fertilizers has doubled from 20 million tons in 1950 to 40 million tons in 1970,
with the heaviest use in the Midwest, Texas, and the Sacramento and San Joaquin Valleys of California
(Miller, 1980). In terms of tonnage applied, Illinois was the heaviest user, followed by California, Iowa, and
Texas. Commercial fertilizer contain nutrients such as nitrate, phosphate, and potassium, which are only
partly used by crops. Saffigna and Keeney (1977) reported crop recovery of less than 50 percent of the
applied fertilizer. The unused portion finds its way into surface water by runoff from excess precipitation or
irrigation and into ground water by recharge to the water table.
3.5.2. Water Chemistry
Constituents commonly analyzed in irrigation water include major cations (Ca, Mg, Na, K), major
anions (HCC>3, $04, Cl), and the minor constituents NC>3 and B. Total salinity affects the compatibility of
the water with the type of crop being irrigated, as certain crops can only tolerate certain levels of salinity.
However, in most instances, it is not the absolute salt concentration of the water that is too high, but the
accumulation of salt in soil over time, which causes injury to vegetation (Price, 1979). As water salinity
increases so, generally, does sodium. The interaction of sodium with clay minerals causes adverse soil
changes ("sodium hazard"), that is, exchangeable sodium tends to make a moist soil impermeable to air
and water. On drying, this soil is hard and difficult to till. This sodium hazard is expressed as "sodium
adsorption ratio," which represents the relative activity of sodium ions in exchange reactions with soil
(U.S. Salinity Laboratory, 1954) as determined from the ratio Na/[(Ca+Mg)/2]1/2. The relationship
between salinity (electric conductivity) and sodium adsorption ratio is commonly used to classify irrigation
water in terms of its applicability whereby the hazard increases with increasing salinity and the sodium-
adsorption ratio (Fig. 55). A decrease in calcium and magnesium concentration increases the salinity
hazard. Calcium may be lost as a result of carbonate precipitation. Magnesium carbonate is more soluble
and, therefore, is less likely to precipitate. However, magnesium enters the exchange complex of the soil,
replacing calcium (Wilcox and Durum, 1967). This reaction is enhanced by coprecipitation of calcium
carbonate. Such a loss of calcium and magnesium causes a relative increase in sodium concentration.
Frequently, boron is analyzed in irrigation water. This is done because of its toxicity to plants even at
concentrations of a few mg/L.
Comparison of water-quality measurements between flow into and out of a ground-water basin
highlights phenomena taking place within the unsaturated and saturated zones. Irrigation-return waters
are typically characterized by net decreases in Ca, HCC^, and 804 as a result of precipitation and by a net
gain of sodium, potassium, and chloride (Smith, 1966). In humid areas, the amount of irrigation and
dissolved-solids contents are generally low. The proportion of sodium and potassium salts in the drainage
147
-------
n
-c
~a
T5
o
W
o
28 -
; 209 water samples in this area
i i iMMI i iii ri
100 250 750 2250
Conductivity - micromhos/cm (EC x 106) at 25°C
Class
C1
low
C2
medium
C3
high
C4
very high
Salinity hazard
QA17244C
Rgure 55. Classification of irrigation water based on sodium-adsorption ratios and conductivity. The higher
the conductivity and the sodium-adsorption ratio, the higher the hazard involved (from Price, 1979).
148
-------
water is small when compared to iron and aluminum. In contrast, in arid areas soils contain a considerable
amount of teachable Na, K, Ca, and Mg salts (Utah State University Foundation, 1969).
In some areas, low-chloride brines from oil-and gas-field areas are used for irrigation. These brines
need to be analyzed for constituents other than just major cations and anions because of the often high
content of trace metals in oil-field waters. The high bicarbonate content in some low-chloride waters
produced with methane gas (for example, San Juan Basin, New Mexico) can increase the sodium hazard
because these waters are typically very high in sodium and very low in calcium and magnesium
concentrations.
Soil salinity is commonly expressed in conductivity terms as micromhos/cm. Conductivity ranges and
corresponding salt contents are listed in Table 10. Salinity expressed in units of rricrosiemens/cm can be
converted to ppm or mg/L by multiplying by a factor of 0.6.
3.5.3. Examples of Geochemical Studies of Agricultural Salinization
Irrigation is practiced along the Arkansas River in Colorado. As water use and re-use increases
downstream, so does the concentration of TDS (Heame and others, 1988) (Fig. 56). This increase may
change the chemical composition of individual constituents to a degree that makes it easily
distinguishable from background water quality using Stiff diagrams (Rg. 57).
Irrigated agriculture is responsible for chloride and nitrate concentrations above natural background
levels in parts of central Wisconsin (Saffigna and Keeney, 1977). Excess fertilizer is leached by excess
irrigation water because crops recover only 50 percent of the fertilizer. Absolute concentrations of
chloride and nitrate vary to a high degree in direct relation with the time of fertilization and of irrigation
application. Ratios of CI/NOs, in contrast, are relatively uniform, suggesting a common source. In fact,
applications of chloride-rich fertilizer are similar to application rates of nitrogen fertilizer. Because some of
the nitrogen is consumed by plants, whereas chloride is not, ratios of CI/NO3 are generally greater than
one. Potassium, another element enriched in the fertilizer (KCI), does not reflect this source of
contamination because of its uptake by plants (Saffigna and Keeney, 1977).
The storage and disposal of agricultural chemicals may be more critical than the application of these
chemicals on fields (Waller and Howie, 1988). In Dade County, Florida, after 50 years of agricultural activity,
fertilizer application has increased nitrate and potassium concentrations in ground water to levels that are
usually below health standards. But in areas of storage and dumping of agricultural chemicals, ammonia,
potassium, and organic nitrogen were one order of magnitude and nitrate was two orders of magnitude
above background levels. These parameters allow distinction from other salinization sources in the area,
that is, inflow of artesian saline water. Elevated concentrations of magnesium, sodium, and chloride are
associated primarily with this second source. Elevated in disposal areas were also iron and manganese,
149
-------
Table 10. Soil salinity classification in the United States (from Food
and Agriculture organization of the United Nations, 1973).
Classification Conductivity Total Salt Content
Saft free 0-4 millimhos/cm 0-1,500 mg/L
Slightly saline 4-8 1,500-3,500
Moderately saline 8-15 3,500-6,500
Strongly saline >15 >6,500
150
-------
Number of analyses
18 17 32 27 33 21 30
10,000
18
Downgradient
Percentile: percentage of analyses
equal to or less than indicated value
— 90th
— 75th
— 50th
— 25th
— 10th
Number of water-quality samples used
in box-plot analysis for the alluvial
aquifer along the indicated river reach
National drinking-water standards;
maximum recommended
contamination level (secondary)
QA1724SC
Rgure 56. Increase in IDS concentration in the Arkansas River alluvial aquifer as a result of irrigation-return
flow (from Heame and others, 1988).
151
-------
(a)
Ca
Mg
Na+ K
Fe
Cations
Anions
HC03
S04
Cl
NO2 + NO3
1.2 0.8 0.4 0 0.4 0.8 1.2
Meq/L
1.2 0.8 0.4 0 0.4 0.8 1.2
1.2 0.8 0.4 0 0.4 0.8 1.2
NO2 + NO3
.QA 17246C
Figure 57. Use of Stiff diagrams for identification of water-quality changes as a result of agricultural
activities: (a) (b) background water quality; (c) water affected by agriculture (from Denver, 1988).
152
-------
whereas other trace elements, such as arsenic, cadmium, chromium, copper, lead, mercury, nickel, and
zinc occurred below detection limits (Waller and Howie, 1988).
The impact of irrigation-return flow on surface-and ground-water quality were investigated
experimentally by Law and others (1970). Surface-return flow after three irrigation periods showed
increases in TDS of between 8.5 and 32.2 percent above irrigation-water concentrations (Fig. 58a), with
the degree of change being related to weather conditions between irrigation applications. The large
change between the June 28 and the July 18 irrigations (Fig. 58a) was caused by a dry period during
which salt accumulated in the soil. The salt was dissolved and carried away with the July 18 irrigation. A rain
event of 1.6 inches between the July 18 and the August 18 irrigations, in contrast, removed soil salts,
causing only a slight TDS increase in runoff after the third irrigation (Fig. 58a). This alternating change in
water quality is not reflected in water percolating through the soil, as indicated by a gradual TDS increase
as the irrigation season progresses (Fig. 58b).
Increases in soil salinity at this site are many times higher, up to twentyfokJ at a depth of 18 to 24
inches, than increases under nonirrigated land (Rg. 59a). These increases are associated with downward
displacement of salts, as shown by the relationship of sodium and chloride with depth (Fig. 59b). The
magnitude of difference between the nonirrigated and the irrigated profiles was caused by salts brought in
by the irrigation and by evapotranspiration. Sulfate, calcium, and magnesium concentrations don't follow
the simple depth relationship seen for sodium and chloride, but instead exhibit three significant peaks
(Fig. 59c). Natural occurrences and the solution of gypsum in the soil may control this distribution with
depth.
The chemical composition of irrigation-return water reflects local conditions and the irrigation history
of an area. For example, in the Yakima Valley of Washington, sodium and chloride concentrations show
higher net increases between applied irrigation water and outflow drainage (Najp/Naout - 4.4; Cljn/Clout =
6.5) than magnesium and sulfate concentrations (Mgjn/Mgout« 2.0; SO^n/SO^ut * 3.6) (Sylvester and
Seabloom, 1963). The overall increase reflects drainage of an area formerly waterlogged and saline
because of previous irrigation practices (Utah State University Foundation, 1969). In the Grand Valley of
Colorado, in contrast, net changes between inflow and outflow show large increases in magnesium
(Mgjn/Mgout - 11.4) and sulfate (SO^n/SO^t - 10.8), a relatively small increase in sodium (Najn/
Naout - 2.5), and even a decrease in chloride toads (CI|n/Clout - 0.6). This net change is due to
replacement of formation water in salt-rich shales and to reactions between irrigation water and soil
minerals (Skogerboe and Walker, 1973).
Irrigation and petroleum production account for ground-water deterioration in the High Plains Aquifer
of south-central Kansas. Irrigation-affected ground water is characterized by increased concentrations of
calcium, magnesium, potassium, fluoride, and nitrate, whereas ground water affected by oil-field brine is
characterized by increased concentrations of TDS, sodium, and chloride (Helgeson, 1990).
153
-------
(a)
IT
a
CO
Q
H
££UU -
2100-
2000-
1900-
1800-
1700-
1600-
1500-
1 Ann .
1 «*UU
1300-
Surface return flow
June 28 •
July 18 ,/\
August 18 /
'"'•*
\ ..»
/"" '" \ ...-»•'""
/ V""
t V
/
*
s'~""^ , ..-•
* *•*
l-:'-:^ Range in quality of
irrigation water supply
2468
Time after start of surface runoff (hr)
10
(b)
-6-
•3 -12 -
Q.
0)
•o
•5 -18 H
CO
-24-
-30-
IDS (ppm)
2.000 4,000 6,000 8,000 10,000 12,000
Irrigation supply
Surface return flow
July 18
(second irrigation)
August 7
(after 1.6 in. rainfall)
•• July 18
E3 August 18
August 18
(third irrigation)
2,000 4,000 6,000 8,000 10,000 12,000
IDS (ppm)
OA 17247c
Rgure 58. Salt content in (a) irrigation water and irrigation runoff and in (b) soil during and after three
controlled irrigation experiments (from Law and others, 1970).
154
-------
(a)
IDS (ppm) (1:1 extract)
1000 2000 3000
-12 -
-24 -
^ -36 -
d
£ -48 -
Q.
o>
TJ
"5 -60 -
-72 -
-84 -
-96 -
» Nonirrigated
profile
(b) Concentration (meq/L) (1:1 extract)
0 10 20 30
-12
-24 -
-36 -
•£ -48 -
Q.
0)
-O
15 -60 -
CO
-72 -
-84 -
-96 -
_i i i i_
Nonirrigated
profile
Cl
Na
\
•
/
<
(C)
0
Concentration (meq/L) (1:1 extract)
10 20 30 40
-12 -
-24 -
-36 -
-48 -
-60 -
-72 -
-84 -
-96 -
_i i i i_
Nonirrigated
profile
S04
Ca+ Mg
QA17248C
Rgure 59. Relationship between (a) TDS, (b) Na, Cl, and (c) Ca+Mg, 804 and depth in soil under irrigated
and nonirrigated land. Downward displacement of salts occurs under both profiles, with differences in
magnitude caused by salts brought in by Irrigation, by evapotranspiratJon, and by the natural occurrence of
gypsum in the soil (from Law and others, 1970).
155
-------
3.5.4. Significant Parameters
Degradation of ground-water quality by agricultural activities can be caused by (a) solution and
transport of chemicals, such as herbicides, pesticides, and fertilizers, (b) disposal of animal waste and
waste water from animal farms, and (c) irrigation-return flow. With respect to ground-water salinity, irrigation-
return flow is the most important source of degradation. Evapotranspiration and leaching of soil minerals
accounts for increases in most chemical components in drainage waters from irrigated areas. Typically,
chloride and sodium concentrations show the highest increases, although other constituents may be high
in some areas, reflecting local conditions. Significant parameters in irrigation-return flow may change over
time, as original soil minerals are dissolved in the initial irrigation stage of an area and minerals brought in by
Irrigation water are dissolved in subsequent irrigation phases.
A significant parameter that differentiates agriculturally induced contamination from other salinization
sources discussed in this report is nitrate. In agricultural areas, nitrate concentrations are often above
background values. Salinization associated with other sources, such as sea-water intrusion or oil-field
pollution, in contrast, is typically associated with increases in chloride, sodium, calcium, and magnesium
concentrations and with small NO/CI ratios.
3.5.5. State-by-State Summary of Agriculturally Induced Ground-Water Problems
The following section provides a state-by-state summary of some of the water-quality problems
associated with agricultural practices. This list is not complete but should serve as an overview of the
magnitude of the problem.
Arizona: Ground-water quality in deep and shallow aquifers that are hydraulicaliy connected with
surface water supplies has been deleteriously affected by irrigated agriculture, which accounts for
approximately 90 percent of all water consumption in the state (Sabol and others, 1987). In the Willcox
Basin of southeast Arizona, dissolved-soRds concentrations in the alluvial aquifer have increased as a
result of recharge of irrigation-return water containing high contents of dissolved salt. This is especially the
case where depth to ground water is less than 100 ft (Kister and others, 1966). Of the total salt load in the
Colorado River at Hoover Dam (9 million tons annually), it is estimated that approximately 37 percent are
contributed by irrigation-return water (Jonez, 1984). Recycling of ground water for irrigation purposes has
caused water-quality deterioration in the Weltton-Mohawk Irrigation and Drainage District (Effertz and
others, 1984).
Arkansas: As a general rule, approximately 75 percent of the ground water removed for rice
irrigation is consumed and 25 percent returns to aquifer systems (Holland and Ludwig, 1981). Water-
quality deterioration from irrigation-return water has occurred along the Arkansas River (Scalf and others,
1973).
156
-------
California: Agriculture is extensive throughout most of the Central Valley, parts of Imperial,
Riverside, and San Bernardino Counties, and many coastal and Southern California counties (Lamb and
Woodard, 1988). Ground-water contamination by pesticides is widespread throughout these areas, as
indicated by a monitoring survey of water wells from 1979 through 1984. Of 8,190 wells tested, 2,522
(31 percent) were contaminated by pesticides (Lamb and Woodard, 1988). In the San Joaquin Valley,
selenium concentrations in agricultural drainage waters are high, exceeding 1 mg/L in some places
(Deverel and Gallanthine, 1989).
Approximately 32 million gallons of fresh water are used daily for irrigation and livestock in California
(Miller; 1980). Irrigation water contributes an average of approximately 1.18 tons of salt per acre per year
(Thome and Peterson, 1967). Degradation of ground water as a result of irrigation has been reported from
the San Joaquin Valley (Fuhriman and Barton, 1971), where imported irrigation water adds nearly 2 million
tons of salt to soil and water each year (U.S. Geological Survey, 1984). Irrigation-return water also causes
IDS increases in some of the southern basins, such as in the Ventura Basin (IDS as much as 2,500 mg/L)
and the Salinas Basin (TDS as much as 1,000 mg/L) (U.S. Geological Survey, 1984).
Colorado: Irrigation-return waters have caused TDS increases in ground water in the High Plains
aquifer, the San Luis Valley aquifer system, the South Platte alluvial aquifer, and the Arkansas River alluvial
aquifer (Heame and others, 1988). Generally, shallow ground water appears to be more affected by
irrigation-return water than deep ground water. In parts of the San Luis Valley, insufficient drainage of
saline soils had led to land abandonment in the early 1900's (Siebenthal, 1910). Approximately
37 percent (700,000 to 800,000 tons per year) of the total salt toad in the Upper Colorado River Basin is
attributed to saline flow from irrigation in the Grand Valley of western Colorado. This contribution is a result
of excess irrigation water and seepage from irrigation canals dissolving salts and subsequently discharging
into the river (Skogerboe and Law, 1971; van der Leeden, 1975). The repeated use of surface water (as
much as seven times) for irrigation purposes within a 65-mi stretch from Denver to Kuner caused an
increase in mineralization of surface and ground water in the South Platte River Valley (van der Leeden
and others, 1975,),
irrigation and feedtots contribute to ground-water contamination in Weld County (RoM, 1971). High
nitrate concentrations from animal waste or fertilizer teachate have been reported along Black Squirrel
Creek, the High Plains aquifer (Heame and others, 1988), and in the San Luis Valley aquifer (Edelmann
and Buckles, 1984).
Delaware: Coastal Plain aquifers have been affected by agricultural nutrients, such as nitrate and
nitrite, especially downgradient from poultry farms (Ritter and Chirnside, 1982). Most of the known
problem areas are located in Sussex and Kent Counties (Denver, 1988). Other constituents above
background levels as a result of agricultural activities include chloride, calcium, sodium, and potassium
(Denver, 1988).
157
-------
Florida: Of all the eastern states, Florida uses by far the most water for irrigation. This is because the
state's rainfall is concentrated within a few months of the year, whereas the climate permits the growing of
crops more or less year-round (Geraghty and others, 1973). Associated with long growing seasons is the
heavy use of pesticides, which has affected more than 1,000 public and private water-supply wells (Irwirt
and Bonds, 1988). u
In Dade County, storage and dumping of agricultural chemicals contributes more severely to high
concentrations of ammonia, potassium, and organic nitrogen in ground water than does the application of
these chemicals onto fields (Waller and Howie, 1988).
Illinois: It is estimated that 97 percent of all rural-domestic water systems are supplied from shallow
aquifers (Voelker and Clarke, 1988). These shallow aquifers are especially vulnerable to contamination by
irrigation-return waters.
Kansas: Irrigation-return flows have caused increased concentrations of calcium, sodium, sulfate,
and chloride in ground water of north-central Kansas ground water (Spruill, 1985).
Montana: Agricultural practices contribute to the widespread occurrence of dryland saline seep
throughout most of the state (see also chapter 3.6). Irrigation-return flow with sometimes high
concentrations of nitrate have contaminated water wells in Rosebud County (van der Leeden and others,
1975).
Nebraska: Increased usage of irrigation water may lead to future problems in the Big Blue and Little
Blue River Basins, the western parts of the Republican River basin, and most of Box Butte and Holt
Counties (Exner and Spakfing, 1979).
Nevada: Contamination potentials from agricultural sources in Nevada are caused by mineralized
irrigation-return water, feedtot and dairy-farm effluents, and agricultural chemicals (Thomas and Hoffman,
1988). Large arsenic concentrations in ground water in parts of Churchill County have been traced to
irrigation-Induced leaching of soils (Thomas and Hoffman, 1988).
New Mexico: Ground-water problems in deep and shallow aquifers that are hydraulically
connected with surface-water supplies have been deleteriously affected by irrigated agriculture, which
accounts for approximately 90 percent of all water consumed in the state (Sabol and others, 1987). In the
Rio Grande Basin, TDS content increases progressively in a downstream direction, with the concentration
at Fort Quitman (1,770 ppm), Texas, being nearly 10 times the concentration at the Otowi Bridge
(180 ppm), just below the Colorado-New Mexico state line (Fireman and Hayward, 1955). Water-quality
deterioration from irrigation-return water has also occurred along the Pecos River (Scatf and others, 1973).
Oklahoma: Investigations of irrigation-return waters identified several instances of surface and
ground-water deterioration. Compared with the quality of water applied, TDS in irrigation-water return flow
Increased by about 20 percent and in percolating soil water by more than 500 percent (Law and others
1970; van der Leeden, 1975). Water-quality deterioration from irrigation-return water has occurred along
the Arkansas River (Scalf and others, 1973).
158
-------
Texas: In the Rio Grande Basin, TDS content increases progressively in a downstream direction as
a result of irrigation-return flows. IDS concentration at Fort Quitman, Texas, (1,770 ppm) is nearly 10 times
the concentration at the Otowi Bridge (180 ppm) just below the Colorado-New Mexico state line (Fireman
and Hayward, 1955). Water-quality deterioration from irrigation-return water has also occurred along the
Pecos River (Scalf and others, 1973) where some wells in the principal irrigation areas of Reeves County
have experienced significant increases in TDS (Ashworth, 1990).
Approximately one million acres have been affected by irrigation to the degree that they are
considered saline (EC>4 millimhos/cm). Counties with more than 10,000 acres of saline soil are (acreage
in parenthesis): Cameron (275,000), Chambers (20,000), Culberson (44,000), El Paso (57,000), Hidalgo
(350,000), Hudspeth (68,000), Jefferson (10,250), Maverick (10,000), Reeves (101000), San Patrick)
(10,000), Ward (30,000), Willacy (10,000), and Zavala (50,000) (Texas State Soil and Water Conservation
Board, 1984).
Utah: Irrigation-return water has caused increased salinity of surface water in the Uinta River Basin
(Brown, 1984). In the Price and San Rafael River Basins of east-central Utah, irrigation-return flow led to
TDS increases from background levels of 400 to 700 mg/L to contamination levels of 2,000 to 4,000 mg/L
(Johnson and Riley, 1984). The same process caused a TDS increase of 2,000 percent in the Sevier
River within a 200-mile-tong stretch (ThOrne and Peterson, 1967). Other areas where ground-water
deterioration has occurred in response to irrigation pumpage include the Pahvant Valley and the Beryl-
Enterprise area (Waddell and Maxell, 1988), as well as the Curlew Valley, where high pumpage rates from
irrigation wells resulted in upward movement of saline ground water from deep aquifers (Bolke and Price,
1969).
Washington: Irrigation-return flows with sometimes high concentrations of nitrate have
contaminated water wells in the Odessa area and in Snohomish County (van der Leeden and others,
1975).
Wyoming: Mineralized irrigation-return flows have contaminated ground water in alluvial aquifers
along the Shoshone, Bighorn, and Big Sandy Rivers (Wyoming Department of Environmental Quality,
1986).
3.6 Saline Seep
Salinization associated with saline seep is considered separately from the previous section on
agricultural sources because of its significance in the Plains Region of the United States and Canada.
3.6.1. Mechanism
, Saline seep was defined by Bahls and Miller (1975) as 'recently developed saline soils in
nonirrigated areas that are wet some or all of the time, often with white salt crusts, and where crop or grass
159
-------
production is reduced or eliminated." It differs from salinization as a result of irrigation-return flow through
the dominant salinization mechanism, that is, evaporation from a shallow water table accounts for high
salinities in saline-seep systems, whereas leaching of soil salts is the major mechanism that accounts for
High salinities in irrigation-return flows. ,' "
Intensive studies of the saline-seep phenomenon in the United States have been more or less
restricted to the state of Montana, even though the problem exists elsewhere. In many localities in the
Northern Great Plains, saline seep causes loss of productive farmland, impassable roads, flooded
basements, stock deaths, and salinization of surface and ground water (Ouster, 1979). Conditions that
favor saline-seep development are widespread across the Northern Great Plains (Fig. 6), covering an area
exceeding 200,000 mi2 throughout most of North Dakota, approximately half the area of Montana and of
South Dakota, and several Canadian provinces (Bahls and Miller, 1975).
Several conditions have to be met for saline seep to develop, such as excess percolation of
recharge water, soluble soil or aquifer minerals, a low-permeable unit at relatively shallow depths, an
internally drained flow system, and evaporation. Agricultural activities often play a significant role in
creating or intensifying some of these conditions. Excess percolation is often caused by destruction of
natural vegetation and drainage ways (for example, by terracing of land), by excessive irrigation, or by
practices such as fallow cropping or by planting of crops that use little water, Undisturbed ecosystems are
characterized by optimal environmental adaptation of plant covers that withstand a wide spectrum of
environmental changes, such as dry and wet periods or hot and cold periods. A great diversity of natural
plant cover provides a certain degree of protection to individual plants. This adaptation and protection is
mostly tost in monoculture systems. With regard to saline-seep development, the most important
characteristic of undisturbed systems is the efficient uptake of water through the variety of natural plants,
which prevents large increases in water-table elevation. Disturbed systems are often less efficient in
uptake and allow a greater portion of water to percolate through the soil and to recharge the local ground-
water flow system (Bahls and Miller, 1975). Initial stages of saline seep are often indicated by prolonged
soil-surface wetness following heavy rains. Fallow areas can undergo a water-table rise of 1 to 15 ft during
years of above average spring precipitation (Miller and others, 1980). Although some of this water will be
used up during the growing season, low stands of previous years are normally not reached, causing
steady increases in water levels over the previous years. In eastern Montana, the water table within the
glacial till has been rising an average of four to ten inches per year (Bahls and Miller, 1975). Where the
water table intersects or is dose to land surface in response to these water-level increases, water-logging
or seepage will occur (Fig. 60). A water table within three feet of land surface signals potential problems, as
water will move up from this depth to the surface by capillarity (U.S. Department of Agriculture, 1983).
Depending on the mineralogy of material encountered along the flow path, water-logged or seepage
areas will be more or less saline. Minerals that are critical in salinization by excess percolating water and by
the increased water table include pyrrte, sodium-rich clays, carbonates, gypsum, sodium and magnesium
160
-------
•ft m
•15T-4
800ft
I 1
200m
Glacial till
Shale
Fluvial gravel
Interbedded sandstone and shale
Ground-water table
Ground-water flow line , '
Salt
OA172249C
Figure 60. Diagrammatic cross section of ground-water ftow with saline seep in topographically tow'areas
and in intermediate topographic position (from Thompson and Ouster, 1976).
161
-------
sulfate, and nitrate (Thompson and Ouster, 1976; Kreitler, 1979). These minerals are normally bound to
the soil but are dissolved and transported to discharge areas by excess recharge. Evapotranspiration
increases the salinity of discharge water in seepage areas.
An impermeable layer at shallow depths increases the potential of saline seep by preventing deep
percolation and by creating a perched water table. The presence of a perched water table is indicated
during drilling when a saturated zone is penetrated before passing through powder-dry shale. In Montana,
this shale layer is generally encountered within 70 ft of land surface (Thompson and Custer, 1976). Where
the shale layer occurs within 30 ft of land surface, the potential for saline-seep development is very high.
Where the overlying glacial till is thicker than 30 ft, water-table elevation has not yet reached land surface
(Bahls and Miller, 1975). In Texas, a correlation exists between soil texture and seep occurrence.
Approximately 80 percent of all saline seeps in that state are controlled by low permeabilities of fine and
fine-loamy soils (Neffendorf, 1978).
Seep water may be locally, not regionally, derived, as indicated by the rapid rise in static water level in
response to precipitation. The size of the seep (discharge) area is directly related to the size of the upland
(recharge) area (Bahls and Miller, 1975). Recharge and discharge areas may be relatively close to each
other and small in size or encompass thousands of acres. Runoff from discharge areas can greatly impair
surface-water quality. For example, some of the largest rivers of Australia have seen a threefold increase in
TDS concentrations within the 50-year period from the 1910's through the 1960's because of saline seep
(Peck, 1978). In southern Australia, approximately 430,000 ha of previously productive farmland are
affected by saline seep as a result of clearing the Indigenous vegetation for farming purposes (Peck,
1978).
In some instances, saline seep is mistaken for oil-field pollution or for saline water emerging from
great depths along geologic structures. Characteristics that positively identify saline seep include during
drilling, wet material is encountered above a dry substrate, the static water level in saline seeps responds
rapidly to precipitation, the water table reflects topography, ground-water chemistry is typical for shallow,
local ground water rather than deep ground water, and changes in drainage or cropping practices in
recharge areas affect the size of seepage areas (Custer, 1979).
3.6.2. Water Chemistry
During a field check of seep salinities, Custer (1979) noticed that seeps with low specific
conductivities were most often associated with sandstone units or colluvium derived from nearby
sandstones, whereas seeps with high specific conductance were associated most often with shale units.
The Increased salt content in the shale units (up to 60,000 micromhos at 25°C) is derived from the
weathering of pyrite, solution of carbonates and sulfates, and cation exchange (Thompson and Custer,
1976). Oxidation and dissolution of pyrite creates acidity that becomes available for hydrolysis reactions
162
-------
and solution of carbonates (Donovan and others, 1981) and often leads to high selenium concentrations
in ground water. Breakdown of chlorite, illite, and feldspars contributes Mg, K, Na, and SiO2 to seep
waters, which makes for a strong correlation between these constituents and TDS. Sodium is also derived
from adsorbed positions in smectites through exchange reactions involving hydrogen and calcium ions
(Donovan and others, 1981). Gypsum dissolution may be one of the most dominant leaching processes in
water with good correlation between 804 and TDS (Fig. 61). These chemical reactions produce a
remarkably uniform water chemistry in Montana, from tow-TDS (1,500 to 3,000 mg/L), Ca-HCOs type
waters in recharge areas to high-TDS (4,000 to 60,000 mg/L), Na-Mg-SC>4 type water in discharge areas
(Miller and others, 1980). Sulfate concentrations of up to 33,000 mg/L have been reported from test
holes in seep areas in the Fort Benton area, Montana (BahIs and Miller, 1975). With the exception of
recharge waters, salinization leads to supersaturation with respect to calcite, dolomite, and gypsum in
many seep-water systems (Rg. 62). Chloride concentrations are normally relatively low but minor and trace
constituents, such as NOs, Al, Fe, Mn, Sr, Pb, Co, Zn, Ni, Cr, Cd, U, and Ag, are relatively high, which
distinguishes them from most other naturally saline ground waters (Bahls and Miller, 1975).
Saline-seep waters in parts of northeastern North Dakota differ from most seep waters in Montana by
a dominance in Mg-CI type as opposed to Na-SC>4 type (Sandoval and Benz, 1966). A general absence of
gypsum accounts for the low sutfate concentrations in those waters, which reach TDS concentrations of
up to 34,000 ppm. Water-table elevations in salt-affected areas range from 1.5 to 5 ft (Sandoval and Benz,
1966). Waters associated with saline seep in Australia are generally of the Na-CI type, in contrast to dryland
saline seep in North America (Peck, 1978). Oceanic salts found in rainfall may be an important source of
these solutes (Hingston and Gailitis, 1976; Peck, 1978).
White salt crusts frequently occur on the ground at seep areas. Thompson and Custer (1976)
identified the minerals Loeweite (Na-j 2^7(804) 13) and, at several locations, Thenardite (Na;2SO4).
Major chemical processes operating in saline seep systems and discussed above are summarized in
figure 63.
3.6.3. Examples of Geochemical Studies of Saline Seep
Thompson and Custer (1976) used triangular diagrams (modified Piper diagrams) for displaying
chemical characteristics of seep waters in Montana. These waters are typically of the Na-Mg-SO4 type, with
some Ca-HCO3 waters (Fig. 64). Ground-water ftow from recharge areas to discharge areas induces
chemical changes, such as ton exchange, involving Ca in the water and Na on soil particles. Ground-water
below dryland-farm sites near Rapelje, Stillwater County, is of the Na-Mg-Ca-SO4 type, with TDS
concentrations of about 9,000 ppm (Thompson and Custer, 1976). Rock weathering is the source of
dissolved constituents, including processes like leaking of formation water rich in sodium chloride,
oxidation of pyrite, solution of carbonate minerals, exchange of calcium for sodium on exchange sites, and
163
-------
o
o
o
*—
X
I
16-
12-
8-
4-
8 12 16
IDS (mg/L x.1000)
20
24
QA172500
Figure 61. Correlation between 804 and TDS concentrations in seep waters from the Colorado Group,
Montana, suggesting gypsum solution as the major contributor to saUnity (from Donovan and others,
1981).
164
-------
Calcite
Dolomite
!
0.8-
0.6-
0.4-
0.2-
n n -
-0.2-
• Calcite
± Dolomite
A
•
A ' A
t
Supersaturated
* Undersaturated
-3.0
-2.0
-1.0
IDS (g/L)
1
1
U.3-
0.2-
0.1-
-n 9 -
o Gypsum o _,—-""""*
o ^^,— "
^ -*•"***'" °
/
/ Supersaturated
/ Undersaturated
12
16 20
TDS (g/L)
24
28 32
QA17251C
Figure 62. Relationship between saturation states and salinity in well waters affected by saline seep,
north-central Montana. Seep saBnizatton typically leads to saturation with respect to calotte, dolomite, and
gypsum (from Donovan and others, 1981).
165
-------
RECHARGE
Leaching along fractures
^^•(gypsum, carbonates, ion exchange)
i Vcdose II ————_/' \
j hydrolysis , " " — A
JMg. K, Si02|
/on «cha^??|
Soft, weathered shale
DISCHARGE
Precipitation of
No-SCM, Mg-SOd salts
'(evaporite) \^^
Soline_seep
dolomite precipitation
Fresh shale (bedrock)
OAI7252
Figure 63. Summary of chemical and transport processes operating in a saline-seep system (from
Donovan and others, 1981).
166
-------
Na SO4i
Saline seep
• Discharge
Recharge
' HC03
C03
OA 17253e
Figure 64. Modified Piper diagram of chemical composition of ground water in saline seeps, Montana.
Seep water typically is of a Na-Mg-SC>4 type as a result of sulfate-salt solution and ion exchange (from
Thompson and Ouster, 1976).
167
-------
precipitation of calcium, sodium, and magnesium sulfate. Below cultivated land, high nitrate
concentrations have been measured. Deep percolation of recharge water during fallow years leaches the
nitrate into ground water. Below uncultivated land, in contrast, nitrate is absent in ground water
(Thompson and Custer, 1976).
Experimental leachates of cored material from seepage areas bear little similarity to actual saline-seep
areas (Donovan and others, 1981). Apparently, the chemistry of ground water is affected by underlying
bedrock or aquifer geology, which changes the composition to a degree that cannot be duplicated in the
laboratory. Gypsum dissolution and ion exchange are dominant processes governing the major
constituents (Na, Mg, 864) in some waters. High bicarbonate concentrations and roughly equimolar
Ca/Mg ratios indicate that carbonate dissolution is additionally an important process in other waters
(Donovan and others, 1981).
High water tables and poor drainage conditions contribute to high salinity in soils along the Red River
of North Dakota. Ground-water salinity of up to 40,000 ppm has been measured in some areas (Benz and
others, 1961). The water is of the Mg-Ca-CI-SC-4 type in the most saline areas and of the Mg-Ca-SC>4-CI
type in moderately saline areas. In contrast to seep areas in Montana, this occurrence of saline water is part
of a regional flow system instead of a local flow system; not weathering, but vertical upward flow of saline
water from the Dakota sandstone, is the source of salinity. This is indicated by the chemical similarity
between Dakota Sandstone Formation water and seep waters (Benz and others, 1961).
The condition of dryland saline seep in Texas is often blamed on oil-field pollution. Although the
visual appearances of brine- and seep-affected soils are very similar, both often being void of any
vegetation and covered with white crusts, chemical characteristics are quite different. In most instances,
total salt content as well as sodium and chloride concentrations are appreciably higher in brine-
contaminated areas than in dryland seeps. The difference in chemical composition between seep-
affected and unaffected soils in the Central Rolling Red Plains of Texas was investigated by the U.S.
Department of Agriculture (1983). All major ions increase significantly in the soil within the upper few feet
below seeps, reflecting evaporation.
A variety of salinization sources, including saline seep, natural discharge of saline formation water,
and oil-field pollution, is known to occur in parts of West Texas. In a two-county salinization study, Richter
and others (1990) developed methods to distinguish saline-seep waters from the other salinization
sources. In both counties, fresh ground water (Cl <250 mg/L) is of similar fades type, dominated by Ca,
Mg, HCC>3, and SCX* (Fig. 65). At chloride concentrations greater than 250 mg/L, however, chemical
fades differ in those two counties, suggesting that different salinization mechanisms are dominant. While
ground water in Runnels County remains a Ca-Mg to mixed-cation water, ground water in Tom Green
County exhibits a trend of increasing Na proportions with increasing chloride proportions. The trend
toward Na-dominated water may indicate mixing with a Na-CI-dominated saline water, whereas the
unchanged cationte composition may suggest a mechanism that doesn't change ionic proportions, such
168
-------
(a)
CK250 mg/L
(b)
TOM GREEN COUNTY
Cl>250 mg/L
RUNNELS COUNTY
QA 5899
Figure 65. Piper diagram of shallow ground water in two adjacent counties of West Texas. Different
salinization mechanisms are suggested from data distributions in the cation triangles. At chloride
concentrations greater than 250 mg/L, a mixing trend between fresh Ca-HCOs and saline Na-CI water
appears to cause the shift in cation percentage in Tom Green County (a) (b). In Runnels County, in
contrast, cation percentages do not change, suggesting a salinization mechanism that doesn't change
relative concentration ratios, such as evaporation (c) (d) (from Richter and others, 1990).
169
-------
as evaporation. Testhole samples from the two counties plotted on bivariate plots can be used to identify
the possible sources of salinity (Fig. 66). Among samples from Runnels County, slopes of Mg and 804
plotted against Cl on logarithmic scales are not significantly different from 1.0 (a « 0.05), whereas among
Tom Green County samples, slopes of Ca, Mg, and 804 plotted against Cl are significantly different from
1.0 (a » 0.05). The differences are interesting because ground-water evaporation without mineral
precipitation gives rise to ionic relationships with unit slope because ionic concentrations increase without
changing molar ratios of chemical constituents. Therefore, it is reasonable to suspect that Runnels County
samples are influenced by evaporation. Tom Green County samples and subsurface brines have similar
ionic relationships, suggesting that the mixing of fresh ground water and brines is an important salinization
mechanism In that county (Richter and others, 1990).
In some areas, agriculturally induced saline seep can be distinguished from other salinization
sources by consideration of contaminants that are typical for agricultural pollution, such as nitrate. Kreitler
and Jones (1975) identified extremely high nitrate concentrations averaging 250 mg/L (range from
<1 mg/L to >3,000 mg/L) in shallow ground water of Runnels County, Texas. Nitrogen isotopes of soil and
ground-water samples identified natural soil nitrogen and animal waste as the major sources of high nitrate
levels. The process that produced these high concentrations in ground water was summarized by Kreitler
and Jones (1975) as follows: "Dryland farming since 1900 has caused the oxidation of organic nitrogen in
the soil to nitrate. During the period 1900 to 1950, nitrate was leached below the root zone but not to the
water table. Extensive terracing after the drought in the early 1950's has raised the water table
approximately 20 ft and has leached the nitrate into ground water. Tritium dates indicate that the ground
water is less than 20 years old." High nitrate concentrations and the nitrogen-isotopic composition of
nitrate indicate that dryland saline seep is the dominant source of salinization in that area and not oil-field
brine, which typically is very low (less than 1 mg/L) in nitrate content.
3.6.4. Significant Parameters
Saline-seep water chemistry is governed by evaporation, resulting in an increase of all constituents
in the water. The increase is reflected on constituent plots as evaporation trends, in contrast to mixing
trends toward a saline-water source observed for the other salinization sources discussed in this report,
with the exception of irrigation-return waters. At low salinities, this increase is characterized by more or less
constant constituent ratios of major ions, such as Ca/CI, Mg/CI, or SO^CI. With increasing salinity, mineral
precipitation will change these ratios as carbonates and sulfates begin to form. Precipitation products will
vary from area to area depending on the chemical composition of soil and water. Where sources of sulfate
are abundant, dissolved sulfate concentrations may by far exceed the concentration of dissolved chloride,
which distinguishes seep water from most other saline ground water. Miscellaneous trace constituents
170
-------
(b)
o>
(C)
3 -
ra
?
o 21
v>
34 5
Log C.l (mg/L)
a co
3 4
Log Cl (mg/L)
5 6
OA11671C
Figure 66. Variation in (a) calcium, (b) magnesium, and (c) sulfate concentrations with chloride
concentration in shallow ground water from a two-county area in parts of West Texas. Samples collected
from the eastern county (solid dots, Runnels County in Figure 65) plot similar to the theoretical
evaporation Hne of these constituents (unit slope), whereas samples collected from the western county
(crosses, Tom Green County in Figure 65) trend toward the composition of subsurface brines (squares)
(from Richter and others, 1990).
171
-------
may serve as good tracers on a local basis, as these are more concentrated in evaporated ground waters
than in most mixing waters between fresh ground water and brine at similar salinities.
3.6.5. State-by-State Summary of Saline Seep Occurrences
Relatively few published reports on saline-seep occurrences in the United States have been found,
with the exception of the states of Montana, North Dakota, Oklahoma, South Dakota, and Texas. The
information found is summarized in this section.
Montana: Potential for saline-seep development exists in central, northern, and eastern Montana.
Saline seep Is caused by dryland agriculture and the crop-fallow rotation system necessary for moisture
conservation and small-grain production on the scale practiced in the state. In 1969, 51,200 acres had
been affected by saline seep (Ferreira and others, 1988). A 1971 survey revealed that approximately
80,000 acres of nonirrigated cropland had been lost to saline seep. In the following four years, an
additional 100,000 to 150,000 acres were affected by saline seep (van der Leeden and others, 1975).
Affected acreage had increased to 280,000 acres by 1983 and to 380,000 acres by 1984 (Ferreira and
others, 1988). Serious conditions have appeared in north-central and northeastern Montana where saline
seep is increasing at a rate of over 10 percent a year. Growth and areal extent of saline seep in the Nine
Mile watershed illustrate the destruction of land due to saline seep. In 1941 only 0.1 percent of the total
land area had been affected by saline seep. The affected area had grown to 0.4 percent in 1951, to 9.1
percent in 1966, and to 19.4 percent in 1971 (Fig. 67). Saline-seep development is most pronounced
where the glacial till is less than 30 ft deep. The till is underlain by a thick marine shale that is impermeable
to water. Both the till and the shale contain an abundant supply of natural salts that are picked up by
excess water and are stored within the perched water table aquifer. Where the water table intersects the
land surface, evaporation of the water leads to accumulation of salts. It is estimated, from measurements at
Fort Benton, that the water table within the glacial till is rising an average of four to ten inches a year,
bringing the water table closer to land surface at topographically low areas (Bahls and Miller, 1975).
Ground-water samples are typically high in TDS (some exceeding 50,000 mg/L), sulfate (some exceeding
30,000 mg/L), and nitrate (in the hundreds of mg/L). Community ground-water supplies at Nashua, Wiota,
and Frazer have been contaminated by saline seep, and nitrate poisoning of livestock has been reported
from the Fort Benton and Denton areas (Bahls and Miller, 1975). Concentrations of greater than 25,000
mg/L TDS (Na, Mg-SO4 type) have been reported for seep water in the Fort Benton area, one of the first
areas to be affected.
Approximately 16 percent (198,000 acres) of the state's irrigable land was considered saline or
alkaline in 1960 (Fuhriman and Barton, 1971). Geological conditions that favor development for saline
seeps exist over an area of 12,500 mi2 (van der Leeden and others, 1975). Because of the state's
172
-------
30
0> 0)
"IS'
c c
fl) =
—' W '
11510-
I!
1935
1940
1945:
1950
1955
Year
1960 1965
;1970 . ,1975
--• ' ' QA17254C
Figure 67. Rate of growth of a saline seep near Fort Berrton, Montana, from 1941 through 1971 (from
Thompson and Ouster, 1976).
173
-------
position as headwater recharge area of the Missouri River Basin, continued degradation of surface waters
by saline seep in Montana can affect many downstream users (Bahls and Miller, 1975).
North Dakota: With the exception of the eastern state-border area, all land areas of North Dakota
can be considered potential saline-seep areas (Bahls and Miller, 1975). Miller and others (1980) estimated
that saline soils have developed on 20,000 to 40,000 ha of nonirrigated farm land. Of the state's total
irrigable land surface, approximately 31 percent (817,000 acres) was considered saline or alkaline in 1960
(Fuhriman and Barton, 1971). In the Red River Valley alone, approximately 400,000 acres'of land are
affected by excessive salt concentrations (Benz and others, 1961).
Oklahoma: In the 1970's, farmers in Harper County became aware of the effects of saline seep. By
1985,1,300 ha of about 65,000 ha wheatland in the county were known to be affected. The size of the
individual saline seep areas ranged from 5 to 100 ha (Berg and others, 1987).
South Dakota: The major salt-water problem in the state is associated with the occurrence of
saline seep (Atkinson and others, 1986), which exists throughout most of the northern, central, and
northeastern parts of the state (Bahls and Miller, 1975). More than 70 percent (1,196,000 acres) of the
state's irrigable land surface was considered saline or alkaline in 1960 (Fuhriman and Barton, 1971).
Texas: The Texas State Soil and Conservation Board (1984) estimated that approximately 5 million
acres of land can be considered saline. Dryland saline seep occurs over large parts of Texas, especially
since the 1950's. Approximately 50 percent of the initial saline seep observations fall within the time
periods of 1950 to 1959 and 1970 to 1974 (Neffendorf, 1978). The largest number of seeps occurs
within the Central Rolling Red Plains, where 1,230 areas of dryland seep cover more than 90,000 acres in
39 counties (U.S. Department of Agriculture, 1983). These saline spots were not evident 75 to 100 years
ago. Saline seep areas in that area are generally separated from deeper brine aquifers by impervious soft
shale bedrock. This is in contrast to natural saline springs and salt flats in the Rolling Plains, which occur
where streambeds and valleys are incised into deep-brine aquifers. Saline spots are also created by oil-
field contamination, with a similar visual appearance of being devoid of any vegetation. Levels of salinity,
however, are usually higher in oil-field contaminated seeps than in dryland saline seeps (U.S. Department
of Agriculture, 1983). The average size of seeps in the Central Rolling Red Plains was 74 acres, with
95 percent identified as increasing in size (U.S. Department of Agriculture, 1983). West of Estelline in
Hail County, saHne seeps are caused by the removal of natural vegetation and cultivation of the soil
(Blurrtzer, 1981). These practices and the natural occurrence of an impermeable clay or shale layer has
caused the development of perched ground-water conditions, allowing evaporation and salinization at
topographically low areas. Richter and others (1990) identified saline seep as a major source of saline
ground water in the Runnels County area.
Counties with the largest acreage affected by dryland saline seep in Texas are (approximate acreage
in parentheses): Baylor (17,000), Cameron (26,000), Coleman (10,000), Collingsworth (11,000), Foard
(13,000), Hardeman (27,000), Haskell (15,000), Hutchinson (19,000), Kleberg (16,000), Knox (14,000),
174
-------
Nueces (10,000), Ward (24,000), Wilbarger (30,000), and Zavala (27,000) (Texas State Soil and Water
Conservation Board, 1984).
•«. 3.7 Road Salt
3.7.1. Mechanism
Salt has been used as an efficient road dejcing agent for a considerable amount of time, with good
results regarding its primary purpose of providing safe travel during winter months. Economic benefits of
street salting are numerous, such as improved fuel efficiency and reduction of costs associated with
accidents, but they don't come without negative side effects. Some of the environmental effects are the
contamination of surface runoff, of surface waters, such as lakes and streams, of soils, and of ground
water.
The degree of contamination potential of water resources is directly related to the number of years
that salt is applied to a given stretch of road and to the amount of salt applied to that stretch each year.
More than four million tons of sodium chloride and calcium chloride were used in the United States during
the winter of 1966-1967 (Field and others, 1973). This number increased to 10.5 million tons during
calender year 1990, representing the approximate average for the past 5 years (Salt Institute, Virginia,
personal communication, 1991). Due to regional weather conditions, approximately 95 percent of road-
salt usage occurs in 'eastern and north-central states (Field and others, 1973) (Table 11). Consequently,
most water pollution is reported from those states (Table 11), where average application rates often
exceed 20 tons of salt per lane-mile (Hutchinson, 1973). The actual amount used depends on weather
conditions, such as temperatures and frequency of snowfalls, but also on population numbers and local
policies. Increases in population, road networks, and sometimes more aggressive or wasteful application
has led to a general increase in salt usage over the years, doubling every five years since 1940 (Fig. 68). It
probably can be assumed that the use of salt as a deicing material will continue to fluctuate through the
years depending on weather conditions. In the future, however, it will more likely increase than decrease,
as more and wider roads are being built. In Massachusetts, an eightfold increase in salt usage occurred
between 1954 and 1971 (Miller and others, 1974). This increase in salt application corresponds to an
overall increase in chloride concentration in the state's ground water (Fig. 69). State-by-state application
rates reported (Fig. 7, Table 11) often mask the local nature of very high usage and very high
contamination potentials. For example, Monroe County, New York, alone used between 109,000 and
224,000 tons of salt each winter between 1965-1966 and 1972-1973, which represented approximately
30 percent of New York's total use in 1966-1967 (Diment and others, 1973).
Contamination of water resources can occur from storage piles of salt as wall as from application of
salt onto roads. Uncovered storage piles may cause some of the highest, local contamination potentials
175
-------
Table 11. Reported use* (in tons) of road salt and abrasives during 1966-1967 (data from
Reid and others, 1973), during the winters of 1981-1982 and 1982-1983 (data from
Salt Institute, undated), and associated known pollution problems listed by state
(modified from Geraghty and others, 1973, and miscellaneous sources listed in chapter
3.7.5).
Contamination
yes
yes
yes
yes
yes
yes
yes
yes
yes
Slate
Arkansas
California
Colorado
Connecticut
Delaware
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Maine
Maryland
Massachusetts
Michigan
Minnesota
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Dakota
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
1966-67
1,000
11,000
7,000
101,000
7,000
1,000
249,000
237,000
54,000
25,000
60,000
99,000
132,000
190,000
409,000
398,000
34,000
4,000
10,000
4,000
118,000
51,000
7,000
472,000
17,000
2,000
511,000
7,000
1,000
592,000
47,000
2,000
3,000
28,000
89,000
77,000
2,000
55,000
225,000
1,000
NaCI
1981-82
2,510
13,600
22,460
103,201
8,913
11,000
304,184
313,365
64,000
35,490
73,275
51,676
155,758
262,000
397,000
118,587
90,963
2,817
22,221
8,500
138,692
138,692
16,000
443,000
45,264
8,222
401,285
9,300
500,010
56,280
4,345
-
79,540
71,904
178,500
10,000
90,636
236,790
5,000
1982-83
856
-
10,896
51.934
7,053
11,000
206,000
116,650
60,400
31,630
32,960
49,202
82,499
178,500
229,000
127,957
75,111
3,245
24,899
9,831
93,813
35,700
23,000
300,000
36,573
8,719
184,341
18,770
456
231,000
29,297
3,697
-
79,720
65,647
95,000
7,500
52,709
229,803
6,340
CaCIa
1966-67
_
-
:
3,000
1,000
-
10,000
6,000
2,000
2,000
1,000
1,000
1,000
6,000
7,000
14,000
3,000
-
'
-
-
6,000
-
5,000
2,000
1,000
12,000
.
•
45,000
1,000
1,000
-
-
1,000
22,000
-
9,000
3,000
-
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
* States not included are due to unavailability of data.
176
-------
108 =
in
o
•=
CD ..
5 1'
CO
105 =
104
,00
.00000.0—
o° U.S. rock salt
o oo
Oo ° U.S. deicing salt
1940
1950
1960
Year
1970
Figure 68. Salt production and salt usage for deicing. The use of salt for road deicing increased steadily
between 1940 and 1970 at a rate that exceeds that of the increase in production of salt from brine,
evaporation, and rock salt (fromDiment and others, 1973).
177
-------
900-
800-
700-
600 -
500-
400 •
300-
200-
100-
EXPLANATION
Total chlorides applied to highways by Massachusetts Department of Public Works
Average chlorides in ground-water 5-20 mi from coast
Average chlorides in ground-water 20-56 mi from coast
Average chlorides in ground-water 56-100 mi from coast
Index: 1955 = 100
1955
1960
1965
Year
i i i
1970 .
QA172560
Figure 69. Correlation between Increase in salt usage applied to highways and chloride concentrations in
ground water, Massachusetts, 1955-1971 (from National Resources and Agriculture Committee, 1973;
Miller and others, 1974).
178
-------
because of the concentrated nature of brine originating at those piles after precipitation events. Such a
case was described by Wilmoth (1972) for Monroe County, West Virginia, where brine seeped into a
fractured carbonate aquifer. There it mixed with fresh water, and subsequently appeared in a water well
located approximately 0.3 mile away from the storage pile. The number of storage piles goes into the
thousands, with each holding several hundreds to thousands of tons of salt every fall (Miller and others,
1974). These storage sites are also often used as wash stations of salt-spreading trucks, with drainage
from salt piles and the wash areas being fed into dry wells (Miller and others, 1974). Road spreading of salt
may lead to contamination during the spreading phase or during the runoff phase. If the salt is applied as a
powder, salt particles may become airborne and transported considerable distances downwind. Similarly, if
the salt is applied as a brine solution, a fine spray or mist can be transported by wind (Jones and Hutchon,
1983). Melting of ice and snow on the road creates brine solutions which enter drainage ditches or
roadside fields. Extensive drainage-ditch networks in urban areas may catch most of the saline surface
runoff. This is the case in the city of Buffalo, where an estimated 90 percent of the applied road salt ends
up in the city's sewer system (Rumer and others, 1973). In rural areas, removal of salt by surface runoff may
be less efficient (for example, 50 percent, Bubeck and others, 1971), with the remaining salt being
flushed into soil and then into ground water. While drainage ditches can divert the salt problem away from
highways, such as in the case of Buffalo, which discharges its water into the Niagara River (Rumer and
others, 1973), runoff into unlined ditches or onto fields concentrates the problem along highways. The
pollution problem generally decreases with increasing distance from the roads, but as the salt content in
soil and ground water increases over the years the extent of contamination may increase (Hutchinson,
1973). Miller and others (1974) reported pollution several thousand ft away from sources of salt, with a
penetration to depths of almost 400 ft in a few wells. This coukj cause problems for wells typically located
near roads for easy accessibility. Wilmoth (1972) estimated that more than half of all the water wells in West
Virginia are located within 100 to 500 ft of salt-treated roads which would place them relatively close to the
area of potential contamination at present or sometime in the future. Similar conditions probably exist
throughout most of those areas where road salt is used.
3.7.2. Road-Salt Chemistry
Very few complete chemical analyses of road-salt brines or road-salt-affected water have been
published. Where endmember chemistry of salt solutions are of interest for statistical evaluation or for
mixing calculations, some researchers have used analyses of halite solutions prepared in the laboratory.
This was done, for example, by Novak and Eckstein (1988), who reported a Na-CI water with very low
concentrations of other major chemical constituents. Molar concentrations of sodium and chloride in those
analyses are equal, reflecting dissolution of nearly pure halite. Street salt is not pure halite in many
179
-------
instances, however, but instead consists of a mixture of sodium chloride and calcium chloride, with
sometimes relatively high concentrations of potassium and boron and low concentrations of sulfate.
3.7.3. Examples of Geochemical Studies of Road Salting
Water wells along highways in Maine showed increases from background levels of 3-4 ppm in
sodium and chloride to 70-76 ppm average sodium and 150-171 ppm average chloride caused by street
salting. The increase in sodium lags behind the increase in chloride because sodium ions are retained in
the soil whereas chloride ions enter the ground-water system unrestricted (Hutchinson, 1973). Chloride
concentrations were also used by Diment and others (1973) to assess surface-and ground-water
contamination in Monroe County, New York. Although two other sources of high chlorinity, such as
sewage and naturally saline formation water, were present in Monroe County, chloride concentrations
were used successfully after water-budget considerations had identified these sources as minor
contributors to the salinity problem.
In two case studies reported by Aulenbach (1980), brine seepage from salt storage piles in New York
was suspected to contaminate water wells. Due to the highly permeable nature of the affected aquifers—
fractured limestone in one case, gravel in the other case—rhodamine dye was used successfully to
positively identify the salt-storage piles as the point sources of pollution. The tracer was detected in the
affected wells after 2 weeks and 3 months, respectively.
Novak and Eckstein (1988) used discriminant analysis and constituent ratios between major cations
and chloride to distinguish water possibly contaminated by road salt from water possibly contaminated by
oil-field brine. The rationale for using only the constituents Ca, Mg, Na, K, and Cl was that this allowed
inclusion of older analyses, which did not report $04, HCC>3, or ar|y minor constituents and because
emphasis was put on constituents that are determined routinely and inexpensively. Using contaminated
ground water from Perry Township, Ohio, as test samples, oil-field brines and salt solutions as the two
potential endmembers of salinization, fresh ground water as the third endmember, and the ion ratios
Ca/Mg, Na/Ca, Na/Mg, Na/CI, K/CI, K/Na, Mg/K, Ca/K, CI/Mg, Cl/Ca, and (Ca+Mg)/(Na+K), Novak and
Eckstein (1988) were able to eliminate street salt as a possible source. Discriminant analysis grouped the
contaminated waters as more similar to oil-field brines or to fresh ground water than to salt solution. Snow
and others (1990) followed the approach of Novak and Eckstein (1988) in using modified Stiff diagrams
(with constituent ratios instead of concentrations as end points) to distinguish sea-water intrusion from
road-salt contamination in coastal wells of Maine. In addition to graphical display in Stiff diagrams, these
sources of salinity were best distinguished using bivariate constituent plots of Br versus Cl and of 864
versus Cl. Road-salt contamination was characterized by Br/CI ratios and SCtyCI ratios being substantially
lower than in samples affected by sea-water intrusion (Snow and others, 1990).
180
-------
Knuth and others, 1990, differentiated road-salt contamination in Ohio (location not specified) from
deep-formational brine disposed of in surface pits during drilling of a gas well and naturally occurring saline
water in the Meadville Formation by analyzing Br and Cl concentrations in fresh, uncontaminated water, in
gas brine, and in contaminated water wells. Construction of mixing curves on a Br/CJ versus Cl plot
identified two wells (Fig. 70), which had experienced increases in chloride concentrations of only several
mg/L to 70 mg/L and which displayed low Br/CI ratios (14 x 10"4 to 60 x 10"4), as affected by road salt
(end-member Br/CI ratio of approximtely 1 x 10"4). Br/CI ratios of approximately 100 x 10"4 helped to
identify two wells as affected by gas brine (end-member Br/CI ratio of approximately 115 x 10~4),
whereas wells affected by upward flow of naturally saline ground water displayed intermediate Br/CI ratios
3.7.4. Significant Chemical Parameters
By far the most widely used parameter in identification of street-salt contamination is the chloride ion.
Chloride is a good tracer because it is the most conservative ion dissolved in ground water, it is the most
abundant ion in street-salt solutions, and it is analyzed on a routine basis. Background chloride
concentrations are known for a vast number of water wells all over the country. Because contamination
from street salt is a seasonal phenomenon with high chloride concentrations in spring runoff and
decreasing (dilution) concentration throughout the remainder of the year, deviation of chloride
concentrations from background levels are a good measure of the degree of salt contamination in most
instances. Accumulation of salt may occur in the soil and in ground water, which means that background
levels may increase over the years. When salt-brine runoff infiltrates the vadose zone and the saturated
zone, sodium is often absorbed to soil and aquifer material. Therefore, the Na/CI ratio may be smaller in
salt-affected ground water than in salt-affected surface water.
Because of its conservative nature once dissolved in ground water, bromide can be a good tracer of
salinity. Expressed as Br/CI weight ratios, it can be used to differentiate salinity derived from road salt
(halite) as opposed to oil- and gas-field brines, deep-formation waters, and sea water, as halite solution
produces some of the lowest Br/CI ratios measured in naturally saline waters.
On a local basis, high concentrations of calcium and chloride may be indicative of road-salt
contamination where large amounts of CaCl2 are added to the salt mixture.
Under certain circumstances, dye tracer (rhodamine) may be useful for identifying point sources of
alleged street-salt contaminations.
The literature review identified very few complete chemical analyses of ground water affected by
road salt. This may indicate that the parameters discussed are sufficient for identification of road-salt
contamination in most instances.
181
-------
10*1
CO
10'=
10°.
.Zone G
10'
I I I I I II II I I II I fill
,0 ml
101
103
Cl (mg/L)
I1I I I f III
10*
105
106
Zone A Mixing zone for fresh water and salt-solution brine
Zone B Mixing zone for fresh water and Meadville Formation watar
Zone C Mixing zone for fresh water and gas-well brine
OA17257C
Figure 70. Bivariate plot of Br/CI ratios versus Cl for selected water samples from northeastern Ohio.
Mixing curves delineate salinization by road salt (zone A), by deep saline ground water (zone B), and by
gas-field brine (zone C) (from Knuth and others, 1990).
182
-------
3.7.5. State-by-State Summary of Road-Salt Issues
The amounts of road salt used by and listed for individual states below, in Table 11, and in figure 7
represent the tonnage of salt applied mostly by state highway departments, as reported during salt
surveys. These numbers should be considered conservative because they do not include salts applied
by towns and cities with their own deicing programs and because not all agencies respond to such
surveys.
Connecticut: Approximately 82,000 tons of road salt were used in the state during the winter of
1965-1966, which translated to 9 tons per single-lane mile. In the following winter (1966-1967), road-salt
application increased to 104,000 tons (Field and others, 1973), the same rate which was.applied 10 years
later during the winter of 1976-1977 (Bingham and Rolston, 1978), The large variability of salt usage is
reflected in the amounts of salt used during the winters of 1981-1982 and 1982-1983, reported at
103,201 and 51,934 tons, respectively (Salt Institute, undated). High chloride and sodium-ferrocyanide
concentrations in some water supplies have been traced to salt-storage areas in the state (Scheldt, 1967;
in Reid and others, 1973).
Illinois: The reported use of sodium and calcium chloride as road-deicing agents during the winter
of 1966-1967 amounted to 259,000 tons (Field and others, 1973). This is a 58 percent increase from the
previous year (164,000 tons), during which already more than 20 tons of salt were applied per lane mile
(Geraghty and others, 1973). During the winters of 1981-1982 and 1982-1983, salt usage varied from
304,000 to 206,000 tons, respectively (Salt Institute, undated).
Contamination by road salt is indicated by an abrupt increase in salt content during the major spring
thaws with smaller subsequent increases. Minimum salt increases are generally observed in late fall and
early winter. Ground-water contamination is less abrupt but shows a consistent increase in yearly minimum
chloride levels when compared to levels recorded prior to heavy street-salt usage (Walker, 1970). In
Peoria, contamination of several water wells was caused by leakage of brine from a salt-storage pile into an
old storm sewer and subsequent leakage from the sewer into shallow ground water (Walker, 1970).
Malno: Approximately 100,000 tons of sodium and calcium chloride were applied in Maine during
the winter of 1966-1967 (Field and others, 1973). Usage was only about half that amount during
1981-1982 and 1982-1983 (Salt Institute, undated). The average annual application rate of road salt for
the past several years (1973 date) has been 25 tons per two-lane mile (Hutchinson, 1973). This
application has led to increased salt concentrations in ground water in many areas of the state. Sodium
and chloride levels in soils next to highways exhibit a direct relationship with the number of years salt has
been applied. In areas where salt had been applied for 20 years, the sodium levels rose over a distance of
60 ft from the shoulder and to a depth of 18 inches. However, surface waters have not been markedly
affected owing to the high degree of dilution of saline solutions (Hutchinson, 1973).
183
-------
Seasonal variation of chloride concentrations in some water wells in Maine suggest contamination by
road salt. While chloride content in various aquifers in the state is normally less than 20 mg/L, the three-
year (1967-1969) average April chloride content of water from approximately 100 sampled wells was
171 mg/L; the highest concentration of sodium was 846 mg/L and that of chloride was 3,150 mg/L.
Concentrations were less in August than in April, which is the month of greatest snow melt and runoff from
highways (Miller and others, 1974).
Maryland: Approximately seven tons of road salt were applied to each single-lane mile during the
winter of 1965-1966, for a total of 45,000 tons (Miller and others, 1974). By the next winter (1966-1967),
this amount had tripled to approximately 133,000 tons of salt (Field and others, 1973), similar to the
amount used during the winter of 1981-1982 (Salt Institute, undated). •
Massachusetts: During the winter of 1965-1966, approximately 126,000 tons of sodium and
calcium chloride were applied as road-deicing agents, which is equivalent to nearly 21 tons per single-lane
mile (Miller and others, 1974). Salt use increased during the winter of 1966-1967 to 196,000 short tons,
during the winter of 1981-1982 to 262,000 tons, and during the winter of 1982-1983 to 178,500 tons
(Salt Institute, undated). Many occurrences of water-supply contamination due to road salt have been
recorded in the state. Among them are the town of Burlington, the town of Becket, Mystic Lakes, and
wells in the Weymouth, Braintree, Randolph, Holbrook, Auburn, Tyngsboro, Charlton, and Springfield
areas (Reid and others, 1973).
Michigan: During the winter of 1966-1967, approximately 416,000 short tons of sodium and
calcium chloride were applied throughout Michigan for deicing of roads. During the wnter of 1981-1982,
salt usage was only slightly lower (397,000 tons), but during the winter of 1982-1983, salt usage dropped
to 229,000 tons (Salt Institute, undated). Contamination from salt-storage piles has occurred in Manistee
County, where a well located 300 ft away from a storage pile contained chloride concentrations of up to
4,400 mg/L (Schraufnagel, 1967). An unprotected salt-storage pile also may have contaminated portions
of the Black River limestone near the village of Rock (Moore and Welch, 1977). The Michigan Department
of Natural Resources (1982) identified 33 known and 86 suspected incidents of salt-storage or road-salt
contamination cases for the 1979 calendar year. Known contamination, mostly from salt-storage areas,
occurred in 19 counties. In an additional 29 counties contamination was suspected. Of the total 203 road-
salt storage facilities, more than 50 percent had known or suspected contamination problems during that
year.
Minnesota: Approximately 412,000 tons of sodium and calcium chloride were used during the
winter of 1966-1967 (Reid and others, 1973), an amount significantly above that used during the winters
of 1981-1982 (118,587 tons) and 1982-1983 (127,957 tons) (Salt Institute, undated). The average
application rate is 15 tons per highway mile (Schraufnagel, 1967, Hutchinson, 1973).
New Hampshire: During the winter of 1965-1967, approximately 12 tons of road salt were applied
to every single-lane mile in New Hampshire. This totals more than 83,000 tons of sodium and calcium
184
-------
chloride salt (Miller and others, 1974). The following winter, approximately 118,000 tons of road salt were
used. In the early 1970's, application rates had increased to an average of 150,000 tons per year
(Morrissey and Regan, 1988). The number of wells being abandoned each year as a result of street-salt
contamination increased from four wells in 1953 to 37 wells in 1964 (Moore and Welch, 1977). By 1965,
more than 200 road-side wells had to be abandoned due to contamination by road salt; chloride
concentration exceeded 3,500 mg/L in some of those wells (Field and others, 1973). As of January 1987,
approximately 79 percent of all contaminated wells in the state were contaminated by road salt (Morrissey
and Regan, 1988). This high number of abandoned and contaminated wells may be caused by the
accumulation of salt over the past 50 years, during which more than 4 million tons of salt were used within
the relatively small area of this state.
New York: During the winter of 1965-1966, approximately 250,000 short tons of sodium chloride
and calcium chloride were used in New York. This is equivalent to 7.5 tons per single-lane mile (Miller and
others, 1974). By the winter of 1966-1967, salt usage had increased to approximately 477,000 short
tons (Reid and others, 1973). Between 1965 and 1973, the use of deicing salt in Monroe County (Lake
Ontario) alone varied between 109,000 and 224,000 metric tons per winter. This has caused (a) a fourfold
rise of chloride concentration in Irondequoit Creek and Irondequoit Bay waters, (b) maximum chloride
concentrations in creeks ranging from 260 to 46,000 mg/L during winter season, (c) a decrease in mixing
of bay waters due to density stratification, and (d) possible increases in chloride concentrations in water
wells. Approximately 50 percent of the salt that is used is removed by surface runoff. The other 50 percent
are stored in soils and ground water (Diment and others, 1973).
Aulenbach (1980) reports of the successful use of Rhodamine WT dye for tracing contamination of
water wells by nearby salt-storage areas in New York. In the two cases, the dye, which had been
distributed around the storage areas, reached the domestic wells after two weeks in one and after three
months in the other. Chloride concentrations in those wells had increased to greater than 1,700 mg/L as a
result of salt-water seepage from the uncovered salt-storage piles.
Approximately 23,000 tons of deicing salt are discharged annually through sewer systems into Lake
Erie by the City of Buffalo (Rumer and others, 1973).
North Carolina: During the winter of 1966-1967, approximately 19,000 tons of sodium chloride
and calcium chloride were used throughout the state (Field and others, 1973). About the same amount,
18,000 tons, was used during the winter of 1973-1974 (Miller and others, 1977), but usage increased
substantially during the winters of 1981-1982 (45,264 tons) and 1982-1983 (36,573 tons) (Salt Institute,
undated). In Haywood County a stockpile of salt caused contamination of a water well which was indicated
by a maximum chloride content of 1,320 mg/L in the well water (Miller and others, 1977).
Ohio: During the winter of 1966-1967, approximately 523,000 short tons of sodium and calcium
chloride were used as road salt throughout the state (Field and others, 1973). This one application of salt
has caused local salintzation of ground and surface water. Chloride concentrations increase downstream
185
-------
from storm-drain outlets in the Olentangy River near Columbus (Kuhlman, 1968). Knuth and others, 1990,
differentiated road-salt contamination in Ohio (location not specified) from deep-formational brine
disposed of in surface pits during drilling of a gas well and naturally occurring saline water in the Meadville
Formation by analyzing Br and Cl concentrations in fresh, uncontaminated water, in gas brine, and in
contaminated water wells. Construction of mixing curves on a Br/CI versus Cl plot identified two wells,
which had experienced increases in chloride concentrations of only several mg/L to 70 mg/L and which
displayed low Br/CI ratios (14 x 1Q-4 to 60 x 10""4), as affected by road salt (endmember Br/CI ratio of
approximately 1 x 10"4) (Rg. 70). Br/CI ratios of approximately 100 x 10"4 helped to identify two wells as
affected by gas brine (endmember Br/CI ratio of approximately 115 x 10"4), whereas wells affected by
upward f tow of naturally saline ground water displayed intermediate Br/CI ratios (50 x 10"4 to 100 x 1O"4).
Pennsylvania: During the winter of 1966-1967, approximately 637,000 short tons of sodium and
calcium chloride were used as road salt throughout the state (Field and others, 1973). This is a
136 percent increase compared to the previous year (270,000 tons; average of more than 20 tons per
lane-mile; Geraghty and others, 1973), translating to an average application rate greater than 50 tons per
lane-mile. Application rates were much lower, however, during the winters of 1981-1982 (500,000 tons)
and 1982-1983 (231,000 tons) (Salt Institute, undated).
South Carolina: Only about 50 tons of road salt were used throughout the state during the winter
of 1973-1974 (Miller and others, 1977). Usage increased to 3,450 tons during the winter of 1981-1982
and 3,697 tons during the winter of 1982-1983 (Salt Institute, undated).
Vermont: Deiting salts are major sources of elevated sodium and chloride concentrations in some
areas of the state; more than 30 percent of the wells that produce water from contaminated aquifers are
contaminated by salt from road application or storage (Cotton and Butterfield, 1988). During the winter of
1965-1966, approximately 84,000 short tons of road salt (sodium chloride and calcium chloride) were
applied throughout the state, averaging approximately 18 tons per single-lane mile (Miller and others,
1974). Only slightly more than that, approximately 90,000 short tons (Reid and others, 1973) were used
during the winter of 1966-1967. Most of this salt may end up in surface streams, as documented by
Kunkle (1971) for a watershed that received 63 to 100 tons of salt per year and had a calculated 83
ton/year excess of NaCI when compared to background levels.
Virginia: During the winter of 1966-1967, approximately 77,000 short tons of sodium chloride and
22,000 short tons of calcium chloride were used in Virginia (Field and others, 1973). Application rates
were less during the winter of 1973-1974, amounting to approximately 63,000 short tons (Miller and
others, 1977), but more than twice that much (178,500 tons) during the winter of 1981-1982 (Salt
Institute, undated). Contamination of water wells by salt-storage areas has been reported by Miller and
others (1977) for Goochland and Dinwiddie Counties.
West Virginia: During the winter of 1966-1967, approximately 64,000 short tons of sodium
chloride and calcium chloride were applied as road salt throughout the state (Field and others, 1973).
186
-------
Incidents of deteriorating water quality due to road-deicing salts are increasing every year. Chloride
concentrations of more than 10,000 mg/L have been observed in runoff from heavily salted areas of Grant
and Raleigh Counties. It is estimated that more than half of all the water wells in the state are located within
100 to 500 ft of roads that receive some salt treatment. Several hundred tons of road salt stored
unprotected have caused well contamination near Union, Monroe County. In the absence of any other salt
source, chloride concentrations in road-salt affected wells increased from about 25 mg/L to as high as
7,200 mg/L (Wilmoth, 1972).
Wisconsin: During the winter of 1966-1967, approximately 228,000 tons of road salt were applied
throughout the state (Field and others, 1973). Similar amounts were used during the winters of 1981-
1982 (236,790 tons) and of 1982-1983 (229,803 tons) (Salt Institute, undated). The average application
rate of road salt in the state is 15 tons per mile (Schraufnagel, 1967; Hutchinson, 1973). Chloride
concentrations in surface runoff from highways as high as 10,250 mg/L have been measured in parts of
the state. This caused seasonal changes in surface-water concentrations from background levels of 0.5 to
2 mg/L dissolved chloride to maximum levels of 45.5 mg/L (Moore and Welch, 1977). Chloride
concentrations in lakes that receive highway-salt runoff may not be uniform but stratified, as reported for
Beaver Dam Lake by Schraufnagel (1967). Chloride concentrations in this lake increased from 8 mg/L at
the top to 33 mg/L at the bottom at 15 ft.
187
-------
4. GEOCHEMICAL PARAMETERS
This chapter will provide a brief introduction to characteristics of individual parameters that have been
used in salinization studies in the past, and that may have been selected for a particular problem study
during the previous chapters. Also included are references to field and laboratory methods and
approximate analytical costs.
It is absolutely crucial in a salinization study to know which methods and parameters are the best to
use for a particular problem. Time and money may play an important role, but technology does also, as
better methods are available now than were available in the past, and still better methods will be available in
the future.
Through the years, a variety of chemical constituents and constituent ratios have been used as
possible tracers of salinity sources (Table 12). Parameters most often used include the major cations Ca,
Mg, Na, the major anions HCO^, 804, Cl, some minor elements (K, Br, I, Li), and some isotopes (180,2H,
3H, 14C). Some of these constituents are more useful than others, as discussed in the following sections
that deal with individual chemical and isotopic constituents listed in Table 12. Constituents are listed in
alphabetical order.
4.1. Discussion of Individual Parameters
Aliphatic Acldsj: High concentrations of short-chain aliphatic acids (acetate, propionate, butyrate,
valerate) in some oil-field/deep-basin brines give way to very low or zero concentrations in tow-saline
mixing waters. This is due to dilution and progressive biodegradation of the organic acids (Hanor and
Workman, 1986, Kreitler and others, 1990). Therefore, their usefulness in salinization studies is limited to
differentiation of little-diluted endmember brines.
Alkalinity: Alkalinity represents the capacity of a solution to neutralize acid. It is determined in the
field by titration of a sample aBquot (50 to 100 ml are sufficient in most instances) to an endpoint pH of
i . . -.'..'.
approximately 3.0 using a strong acid (6N ^804) (Brown and others, 1970; Wood, 1976). Alkalinity
measurements in the laboratory are typically slightly lower than actual values (Roberson and others, 1963).
At the ph range of most natural waters, alkalinity is represented mainly by the dissolved carbon dioxide
species HCO^- and CC>32~; commonly, alkalinity is reported as bicarbonate (HCC>3~) or calcium
carbonate (CaCOs). To convert alkalinity expressed as calcium carbonate to alkalinity expressed as
bicarbonate, the amount (in mg/L) of calcium carbonate is multiplied by the factor of 1.22. Alkalinity is
generated by the action of dissolved atmospheric carbon dioxide and soil carbon dioxide in the water on
carbonate rocks, such as limestone. Respiration by plants and the oxidation of organic matter in the soil
and in the unsaturated zone increase CO2 content over atmospheric concentrations (0.03 percent).
Additional sources of carbon dioxide result from biologically mediated sulfate reduction, metamorphism of
188
-------
Table 12. Geochemical parameters used for identification of salinity sources.
Chemical Parameter
Sallnlzatlon Sources
Natural saline water
versus others
Halite-solution brine
versus others
Sea-water intrusion
versus others
Page
Cl 28
Br, I, S-34,18Q, D, Br/CI, Na/CI, I/CI, I 28
Mg/CI, K/CI, Ca/CI, (Ca+Mg)/SO4, Sr . .. 28
K/Na, Br/TDS 62
(Ca+Mg)/(Na+K), Na/CI, 62
Ca/CI, Mg/CI, SCvt/CI, 66
Br/CI 66
K/CI, (Ca+Mg)/SO4, I/CI 68
18Q/D, I/CI, SO4/(Na+K), SO4/TDS, SO4/CI 68
Cl 88
Major ions (Piper) 92
14C,3H 96
l/CI, B, Sa, I 98
18O, 2H, 13C 98
Ca/Mg, CI/S04,B/CI, Ba/CI 100
Br/CI 100
Oil-field brines
versus others
Agricultural effluents
versus others
Saline seep
versus others
Road salt
versus others
Cl, Major ions
Na/CI
Ca/CI, Mg/CI, SO4/CI, Br/CI
I/CI, Major ton ratios, Cl, Br, (Na+CI)/TDS,
Li/Br, Na/Br, Na/CI, Br/CI
Cl, NO3, CI/NOs
K.TDS
SC-4
Ca/CI, Mg/CI, SO4/CI
NO3
a
Major ion ratios, Br/CI
Dye
123
128
128
131
132
149
149
163
168
170
180
180
181
189
-------
carbonate rocks, and outgassing from rocks in the Earth's mantle (Hem, 1985). In some instances,
especially in oil-and gas-field associated waters, short-chain aliphatic acids may contribute to total alkalinity
of the water sample and sometimes have been confused with inorganic alkalinity. Decarboxylation of
short-chain aliphatic acids (for example, acetic acid) in oil and gas fields may also contribute large amounts
of carbon dioxide (Carothers and Kharaka, 1978).
In most natural waters alkalinity ranges from a few tens of mg/L to <1,000 mg/L. Low concentrations
occur in soils and rocks low in calcium carbonate, whereas high concentrations are reported in water
flowing through carbonate-rich soils and aquifers; very high concentrations can be found in sandstone
aquifers as Na-HCC-3 waters. Concentrations > 1,000 mg/L have been reported, especially in waters
associated with oil and gas reservoirs. In the latter case, other constituents, such as short-chain aliphatic
acid anions (for example, acetate, valerate) may contribute to or represent the bulk component of
alkalinity. Extremely high alkalinity concentrations (as HCOs) of close to 20,000 mg/L have been reported
by Kaiser and others (1991) in deep ground water associated with coal deposits in the San Juan Basin,
New Mexico. According to these authors, the high inorganic-alkalinity concentrations are bacterially
derived through CO;>-releasing processes, such as sulfate reduction and methanogenesis. The low
concentrations of calcium and magnesium in the same waters prevent removal of dissolved HCC>3 from
solution through precipitation of carbonate minerals.
Concentrations of bicarbonate are usually much higher (several hundred mg/L) in fresh waters than
in oil-field brines (several tens of mg/L), making it a potential tracer of oil-field brine pollution. However,
care must be exercised not to compare bicarbonate in fresh water with total alkalinity in oil-field brine
because tltratlon of total alkalinity will include any short-chain aliphatic acids present in the brine.
Bicarbonate concentrations depend strongly on pH and CC>2 partial pressure and therefore, are easily
changed by a change in chemical environment along the flow path of ground water. This makes
bicarbonate a marginal tracer of salt-water sources although it has been used as constituent ratio of
CI/HCO3 in combination with other ratios of chemical constituents to identify salinization trends (for
example, Collins, 1969; Bumitt and others, 1963).
Argon: Argon is produced by radioactive decay of potassium-40. Its concentration dissolved in
ground water is dependent on the temperature of the recharging water and thus could allow
differentiation between salt waters originating in different geographic areas and differentiation between
continental saline water and sea-water intrusion (Custodio, 1987). No salinization studies using argon
have been conducted in the United States, limiting the usefulness of argon as a tracer of salinization
sources because of the lack of data and documentation of its usefulness.
Bromine: Bromine, present in water as the bromide ion Br~, is generally a very good tracer of
salinization sources in combination with chloride. Both constituents are relatively conservative, that is,
once in solution they are not easily removed by processes such as ion exchange (because of their large
size) or precipitation (because of high solubility). High concentrations occur in sea water (65 mg/L), many
190
-------
geothermal waters (several tens of mg/L), and in deep-basin brines and oil-field brines (several tens to
greater than 2,000 mg/L). Most saline waters encountered in salinizatioh studies are undersatu rated with
respect to halite, that is, the Br/CI ratio in a mixing water is not affected by mineral precipitation but reflects
contributions by the principal salinization source(s). The ratio of Br/CI (or CI/Br) is especially well suited to
distinguish halite-dissolution brine from oil-field/deep-basin brines because only small amounts of
bromide are incorporated into the crystal structure of halite during evaporation (distribution coefficient
<1.0). When this halite crystal is dissolved by fresh water, the resulting Br/CI ratio in solution will be small.
The Br/CI ratio is typically one or more order of magnitude smaller in halite-dissolution brines (Br/CI <5 x
10~4) than in oil-field/deep-basin brines. In brines that have gone through a halite-reprecipitation stage,
bromide concentrations are high because proportionally more chloride than bromide was incorporated
into the halite deposits, rendering the final solution higher in Br/CI than the initial solution. Continued
dissolution and recrystallization of halite further lowers Br/CI ratios in the halite and increases Br/CI ratios in
the solution. Therefore, a brine that originated by dissolving halite and which later went through a halite-
reprecipitation stage may show relatively high Br/CI ratios (Land, 1987), such as basinal brine, which would
then have a Br/CI ratio very different from a halite-dissolution brine that originated just by dissolving halite.
Salinization derived from mixing of fresh water with sea water or sea spray is also generally recognized by
higher Br/CI ratios (34 x 1CT4) than are halite-solution brines (typically <10 x 1Q-4).
As is the case for all tracers, the beneficial nature of Br/CI ratios as tracers of salinization sources
decreases when the ratios are similar to both endmembers under consideration. That is, oil-field brines
from different reservoirs may have Br/CI ratios too similar to allow distinction between them. Also,
differentiation of salt-water sources using Br/CI ratios works best at high concentrations of TDS. Many
investigators have made use of the Br/CI (CI/Br) ratio in their studies, including Collins, 1969, Patterson
and Kinsman, 1975, Whittemore and Pollock, 1979, Kreitler and others, 1984, 1990, Morton, 1986,
Richter and Kreitler, 1986a,b, and Richter and others, 1990. The conservative nature of bromide and
chloride ions have also been used to distinguish the mixing of fresh ground water with street runoff as
opposed to the mixing of fresh ground water with sewage effluent, as street runoff is typically relatively
high in bromide content (Behl and others, 1987).
The chemical character of bromide and the relatively tow cost of bromide determination (Table 13) in
the laboratory make this an excellent constituent to be used in salinization studies. Whittemore (1988) and
Banner and others (1989) warned, however, that many Br/CI ratios bas6d on data from the earty literature
may be in error because of past difficulties with bromide determination in the laboratory.
Calcium: Calcium is the most abundant alkaline-earth metal and is a major component of minerals in
most aquifer types. The concentration of calcium in ground water is governed by the abundance of
calcium-bearing minerals in the aquifer material, equilibrium conditions between the solid, solution, and
gas phases, and the presence of minerals with high cation-exchange capacities (Hem, 1985). Calcium
concentrations in ground water are high in aquifers, consisting of limestone, dolomite, gypsum, or
! •. ' '
191
-------
Table 13. Approximate costs of chemical and isotopical analyses
of constituents covered in this report, as reported by various
laboratories (no specific laboratories are recommended).
Constituent ($) Cost per sample
Na,K,Mg,CaIU,SiO2,B,Ba (ICP-OES) 32.00
Cl (Ion Chromatography) 10.00
SO4(do.) 19.00
Br (do.) 10.00
NO3(do.) 20,00
I (Spectrophotometry) 10.00
Br(do.) 10.00
pH 10.00
Alkalinity 10.00
TDS 5.00
TOO 20.00
Oxygen-18 and deuterium 80.00
Tritium 280.00
14C 225.00-440.00
34s 85.00
192
-------
gypsiferous shale. In those environments, calcium concentrations are typically much higher than chloride
concentrations. Addition of salt water, normally much higher in chloride concentrations than in calcium
concentrations, from one source can be detected using the change in ion ratios such as Ca/CI. However,
if more than one source of high chtorinity is suspected or if a high-Ca brine is a suspected endmember of
mixing, the ratio of Ca/CI may be of little use to differentiate between these sources (Leonard, 1964). Ion
exchange on clay minerals or albitization of plagioclase or K-feldspar can enrich waters in calcium at the
cost of sodium (Banner and others, 1989). Because calcium concentrations are highly dependent on pH,
partial pressure of CC*2, and the availability of carbonate minerals, the amount of calcium In ground water is
more often related directly to chemical processes such as precipitation,' dissolution, and ion exchange
than to simple mixing of fresh water with saline water.
Calcium analysis is performed in the laboratory at a modest cost together with other major and minor
cations (Table 13). Also, sampling is uncomplicated (see Major Ions, this chapter), which justifies
determination of calcium as a standard technique in water-quality studies!
Chloride: More than three-quarters of the total amount of chloride in the Earth's outer crust,
atmosphere, and hydrosphere is contained in solution as Cl~ ion in the oceans (Hem, 1985). This reflects
the chemically conservative nature of chloride, that is, once in solution, it is not easily removed by
processes other than precipitation at very late evaporation stages. Concentration in rain water may vary
from 1-20 mg/L at the coast to less than 1 mg/L further inland. Ocean spray and wind transport may also
cause high chloride concentrations in coastal surface and ground water. Most fresh-water sources contain
chloride in the mg/L to the tens of milligrams per liter range, the upper limit for drinking water being 250
mg/L (U.S. Environmental Protection Agency, 1977). Concentrations commonly increase with depth and
with distance from recharge areas, approaching the concentration of sea water (approximately 19,000
mg/L) or greater than that in many sedimentary basins. Deep formation waters and especially waters
associated with oil and gas often contain chloride concentrations in excess of 100,000 mg/L. Major natural
sources of chloride in ground water are dissolution of late-stage evaporites (NaCI), flushing of saline
waters retained in predominantly fine-grained sediments since deposition, and sea-water intrusion.
Anthropogenic sources of chloride in ground water include highway deicing salts, industrial, domestic,
and agricultural wastes, oil-field brines, and pumping-induced salt-water intrusion, including sea-water
intrusion. Evaporation of water, naturally and human-induced, increases chloride concentrations in
surface and ground water. As a direct result of its conservative nature and involvement in nearly all
processes of salinization of ground water, chloride is the most-often used parameter to identify
deterioration of water quality. Most studies dealing with saynization due to street-deicing salt or with sea-
water intrusion make use of chloride mapping to define the extent of salt-water movement. In general,
because of its relatively conservative nature once in solution (see also discussion of Bromine, this
chapter), chloride concentrations are excellent tracers of single salt-water sources. A wide combination of
constituent ratios of X/CI (X - Ca, Mg, Na, K, SO4, HCCg, Br, I, Li) is used to distinguish between two or
193
-------
more salt-water sources, such as sea water, oil-field brine, halite-dissolution brine, connate water, or
sewage. Chloride is determined in the laboratory by titration, using a sample aliquot from the anion sample
bottle, or by ton chromatography (Table 13).
Iodine: Iodine is being used successfully as geochemical indicator for gas and oil reservoirs due to
its close association with volatile fatty acids (Carothers and Kharaka, 1978) and argillaceous deposits
containing organic matter (Collins, 1967). Similar to bromide and chloride, iodine is very soluble (as iodide
Ion H and does not readily substitute into mineral phases because of its large ionic size (Frape and Fritz,
1987). Although not particularly abundant, it is therefore concentrated preferentially in the aqueous
phase during water-rock interactions which makes it a favorable tracer of salinity sources. Because of its
relatively narrow range, Whittemore and Pollock (1979) consider iodide a potentially useful tracer (as I/CI)
to distinguish among different types of brines. Lloyd and others (1982) used (a) the ionic ratio of l/Sr to
distinguish between different water sources in an alluvial aquifer in Peru, (b) I/CI ratios to distinguish
between old and modem saline ground waters in the Lincolnshire Chalk aquifer of England, and (c) I/CI to
distinguish saline water in a limestone aquifer from saline water in gypsiferous beds in Qatar. A very
irregular distribution of Iodine concentrations result from the application of fertilizers containing iodine
(Lfoyd and others, 1982).
The concentration of iodide in most fresh and brackish waters is less than 1 mg/L. Carbonate
horizons in soil profiles appear to act as natural barriers to iodine migration (Whitehead, 1974). Iodide
concentrations are also low in sea water (0.05 mg/L; I/CL - 2.6 x 10~6), whereas oil-field waters may
contain several tens or hundreds of milligrams per liter of iodide (I/CI >10~5). Collins (1969) reported
unusually high concentrations of iodide in excess of 1,400 mg/L in some Anadarko Basin brines. Low I/CI
ratios In sea water and high I/CI ratios in most oil-field and deep-basin brines may allow differentiation
between these two sources of salt water.
Isotopes: The stable isotopic composition of a water sample may be indicative of the source of the
water or of the source of the mineral content dissolved in the water. Carbon and sulfur isotopes reflect
water-rock interactions and thereby may mask the origin of the water. At low temperatures, oxygen and
hydrogen isotopes are much less affected by water-rock interaction than by fractionation processes
before recharge to the water table and, therefore, are more indicative of the origin of the water than of the
dissolved mineral content. Stable isotopes of oxygen and hydrogen (18O and 2H, deuterium) are used
frequently to differentiate between waters originating at different recharge areas. This is based on the
occurrence of fractionation processes between lighter and heavier isotopes, that is, (a) during
evaporation, the heavier isotope becomes abundant in the solution and the lighter isotope becomes
abundant in the vapor phase and (b) during precipitation, consecutive rainfall events become lighter in
their isotopic composition. Therefore, coastal rain is isotopically heavier than rain occurring further inland.
Also, evaporatively concentrated recharge water is isotopically enriched when compared to recharge
water not affected by evaporation prior to infiltration. Enrichment or depletion is expressed relative to the
194
-------
isotopic composition of standard mean ocean water (SMOW), arbitrarily being assigned a 5180 and 50
composition of zero. Within humid climates, the standard meteoric water line defined by Craig (1961)
represents the relationship between oxygen-18 and deuterium in most recharge water. Arid zone
recharge waters are better represented by a meteoric water line that is shifted toward isotopic enrichment
(Welch and Preissler, 1986). Oxygen-18 and deuterium can be useful to distinguish local meteoric water
that dissolved halite in the shallow subsurface from a regional oil-field brine (for example, Richter and
Kreitler, 1986a,b) or any other brine that is derived from a different recharge area. In a study of formation
waters from central Missouri, Banner and others (1989) used these ratios to trace the origin of the water all
the way to the Front Range of Colorado. However, a basic problem exists in the fact that the water
component of a brine may have a very different origin than the majority of the solute components, that is,
water chemistry and isotopes do not necessarily reflect the same source (Kreitler and others, 1984, Frape
and Fritz, 1987). At high temperatures, waters will equilibrate with oxygen-18 of the aquifer material and
therefore will not preserve the original recharge signature. This oxygen shift can be of value in
differentiating deep basinal waters from shallow meteoric waters.
Age-dating techniques using unstable isotopes such as carbon-14, tritium (3H), or chlorine-36 may
allow differentiation between ok) and modem sea-water intrusion (Lloyd and Howard, 1979, Custodio,
1987, Gascoyne and others, 1987). Carbbn-14 and tritium are produced by cosmic rays interacting with
nitrogen in the outer atmosphere. Because of this constant creation of carbon-14, the ratio of 14C/12C in
the atmosphere is relatively constant. Studies of carbon-14 content in tree rings suggested that the
14Q/1 ZQ ratjO has varied only slightly during the past 7,000 years (Freeze and Cherry, 1979). However,
when out of contact with the atmosphere, this ratio will decrease as a result of radioactive decay of the
carbon-14 isotope, allowing determination of the time span that has elapsed since isolation from the
atmosphere. This determination is complicated, however, by the addition of "dead carbon" from
dissolution of carbonate rocks, as these rocks contribute carbon essentially devoid of carbon-14 (Freeze
and Cherry, 1979). Large-scale atmospheric testing of thermonuclear weapons in the 1950's and 1960's
created large amounts of tritium, which now enables differentiation of pre-1950's waters (0-2 tritium units)
from post-1950's waters (>2-3 TU). Tritium has a half-life of only 12.3 years, which restricts age
determinations to only a few tens of years. The half-fife of carbon-14 of 5,730 years, in contrast, allows age
determinations of organic matter or carbon dissolved in water in the order of tens of thousands of years
(maximum of approximately 50,000 years). Chlorine-36 combines the advantages of the relatively
conservative nature of the chloride ton with a half-life of approximately 300,000 years. As such, it may be
more useful than tritium in the future, when the bomb-tritium peaks will have decayed as a result of its short
half-life. Chtorine-36 is produced naturally by (1) spallatJon of heavier nuclei, such as argon, potassium, or
calcium, by energetic cosmic rays, (2) stow neutron activation of argon-36, and (3) by neutron activation of
chlorine-35 (Benttey and others, 1986). In addition, neutron activation of chlorine-35 in sea water by
weapons testing in the South Pacific between 1952 and 1958 caused enrichment of chlorine-36 (Bentley
195
-------
and others, 1986). The difficulty of analyzing small quantities of the isotope in the past have been
overcome in recent years by the use of Tandem Accelerator Mass Spectrometry. However, costs are high
(up to $750 per sample) and existing data are sparse, both of which limit the wide-scale usefulness of this
isotope in salinization studies at this time.
Strontium isotope ratios (87Sr/86Sr) may be useful to distinguish brines from different oil pools or
stratigraphic units (Chaudhuri, 1978), reflecting strontium isotopic ratios of the host rocks. This isotope
ratio \s not used routinely in salinization studies and is probably best used to test a theory that was
previously established using more conventional methods.
Sulfur Isotopes (34S/32S) may be used to Identify the source of sulfide or sulphate in ground water,
which In turn may be useful to differentiate between salinization sources. For example, sulphate may
originate from the solution of evaporite minerals, such as anhydrite or gypsum, giving the water a
characteristic Isotopic composition typical for the unit in which the minerals occur. Solution of sulfide
minerals or decomposition of organic matter may provide a different isotopic composition that is
characteristic and distinct for another water-bearing unit. In a study in Ohio, Breen and others (1985)
suggested the use of sulfur isotopes to distinguish between isotopically heavy sulfate in brines and
isotopically light sulfate in ground water. Dutton and others (1989) used this ratio to differentiate between
two potential brine sources in parts of West Texas.
Determination of isotopes requires sophisticated laboratory equipment and techniques that are not
routinely provided by just any laboratory. Certain isotopes are only determined in one or just a few
laboratories in the country and costs can easily run into the hundreds of dollars per water sample. At a
combined analysis cost of approximately $100 per sample (Table 13), determination of oxygen-18 and
deuterium is probably too costly to be done on a routine basis but may prove very helpful for individual
samples identified by other methods as being characteristic of a certain water type or water source.
Sampling for oxygen-18 and deuterium is relatively easy, requiring only 250 or 500 ml of filtered water.
Similarly straightforward is collecting water samples for tritium and sutfur-34. A one liter glass bottle is filled
completely with filtered sample water for later analysis of tritium content. Laboratory analysis of sutfur-34
requires field treatment of the filtered sample, using small amounts of HCI and Cd-acetate or Zn-acetate.
Much more elaborate is field-sample collection and preparation for carbon-14 determination (Feltz and
Hanshaw, 1963; International Atomic Energy Agency, 1981). Depending on the amount of dissolved
carbon in the water sample, conventional methods of carbon-14 analysis may require processing of
hundreds of liters of water to precipitate the typically required 4 grams of carbon, using CO2-free
ammoniacal strontium-chloride solution as precipitation agent. For example, 2,000 liters of sample water
would be needed at a bicarbonate content of 10 mg/L. However, the required amount of sample water
would only be 40 liters at a bicarbonate content of 500 mg/L. At a cost of laboratory analysis in the order of
$250 per sample, determination of carbon-14 may not be practical in most investigations. This is especially
true for the even costlier method of determining carbon-14 using the Tandem Accelerator Mass
196
-------
Spectrometry (cost at approximately $450 per sample), which otherwise has the big advantage of
requiring much smaller amounts of carbon (in the milligram range), greatly reducing the amount of water
needed.
In summary, collection and measurement of isotopes are expensive, often elaborate, and constitute
techniques that may not be conclusive as tracers of salt-water sources by themselves. More often and
probably more importantly, isotopes are used to support conclusions drawn from the evaluation of
chemical constituents.
Lithium: Once in solution, lithium is not readily removed by exchange reactions or secondary
minerals and it will accumulate in solution depending on time and availability in the host rock (Frape and
Fritz, 1987). Therefore, it may be used as a good indicator of the degree of water-rock interactions.
Whittemore and Pollack (1979) observed a wider range ofU/CI ratios than of Na/CI, Br/CI, and I/CI ratios in
Kansas oil-field brines as a whole, but much narrower ranges in LJ/CI ratios than in the others when
restricted to a particular geographic area. This would suggest that LJ/CI ratios may be good indicators for
local salinization sources. However, lithium concentrations in most fresh and brackish ground waters are in
the ug/L range and close to analytical detection limits. Therefore, lithium is probably used best at high
chloride concentrations, that is, at low degrees of dilution of the original salinization source.
Magnesium: Magnesium in ground water is derived mainly from dolomite, limestone, and
ferromagnesian minerals. In limestone terrain, an increase in magnesium concentration along the flow path
is likely because dissolution of limestone causes an increase in magnesium but subsequent precipitation
of calcium carbonate removes only little magnesium from solution (Hem, 1985). Dolomite solution and
subsequent precipitation of calcite may also contribute to high magnesium concentrations, as suggested
by Senger and others (1990) for the Glen Rose Formation of Central Texas. However, physical and
chemical processes that may control the amount of magnesium are multiple, including dispersion,
complexation, adsorption, desorption, precipitation, and solution. Concentrations of magnesium are less
than calcium concentrations in most natural ground waters. In sea water, however, magnesium content is
more than three times that of calcium (1,359 mg/L Mg vs. 410 mg/L Ca; Goldberg and others, 1971).
Therefore, the Mg/Ca ratio may allow detection of sea-water intrusion into coastal fresh-water aquifers as
long as ion exchange, dissolution, or precipitation does not affect either ton to a large degree. Because of
its nonconservative nature, magnesium concentrations or Mg/X (X - Ca, Na, K, $04, Cl) ratios are used
within a suite of other ratios but seldom constitute the sole indicator of a salt-water source. Like other major
cations, determination of magnesium is included in standard ICP (Inductively Coupled Plasma-Atomic
Emission Spectrometer) techniques at a relatively tow cost.
Major Ions: Most chemical analyses of ground water include only the major cations Ca, Mg, Na, and
possibly K, and the major anions HCCg, 80-4, and Cl. In addition, many analyses reported in the literature,
especially those being 30 years old or older, include a calculated value for Na+K, determined from the
difference in meq/L of major anions and the cations Ca+Mg. This practice prevents a quality check (ion
197
-------
balance) of the analytical work and should be avoided whenever possible. Analytical problems and costs
are the main reason for the often small number of parameters reported. In salinization studies, major
cations and anions are used mostly within concentration ratios related to chloride. Ratios are generally
preferred to absolute concentrations because dilution of salt water by fresh water affects ratios of
constituents to a smaller degree than individual concentrations. Chemical reactions, such as ion
exchange, precipitation, and dissolution, change ionic ratios and present obstacles for the use of these
ratios In determining salt-water sources at low concentrations. In many instances, chemical reactions
between water and the aquifer material mask salt-water sources at low concentrations. The conservative
ion chloride can sometimes be used to evaluate the degree of chemical reactions. For example, if the
chloride concentration indicates a salt-water to fresh-water ratio of 1:10 in the mixture (calculated from
known end-member concentrations) but calcium concentrations indicate a 1:5 mixture, the difference in
calcium concentration may be explained by water-rock interactions (see also chapter 6 for calculation of
mixing ratio). Therefore, chemical ratios of major cations and anions have to be applied carefully, keeping
in mind chemical and physical processes that may change these ratios.
Sampling for major cations and anions (for procedures see Brown and others, 1970) requires two
500 mL, polyethylene bottles. Both samples are filtered using a 0.45 u. membrane filter. In addition, the
cation sample is treated with a HCI solution to a pH below 3.0. Laboratory costs for all major cations and
anions combined amount to approximately $60 per sample (Table 13).
Minor Ions: In contrast to the major tons which exist in soluble form in many different environments
in the lithosphere and hydrosphere, minor elements and trace elements are less abundant and are often
concentrated only under certain conditions. Concentrations seldom exceed 1 mg/L, but nevertheless
these elements may reflect differences in aquifer compositions between different sites; differences that
are not reflected in concentrations or concentration ratios of major ions. Anthropogenic factors may often
lead to unnaturally high concentrations of minor constituents, creating excellent tracers of ground-water
contamination. On the other hand, minor and trace elements in concentrated brines often occur in
\
concentrations so small that significant dilution with fresher waters renders them close to or below
detection limits. Also, many of the minor and trace elements may easily be precipitated or absorbed during
flow and mixing with other waters due to a change in the chemical environment. For example, barium and
strontium can be removed as sutfates when encountering water that has dissolved gypsum or anhydrite
(Whittemore and Pollack, 1979). The occurrence of iron in one of its oxidation states, which govern its
solubility, is very dependent on pH and Eh, making Fe an unfavorable tracer. In general, precipitation as
oxides, hydroxides, carbonates, sulfates, phosphates, etcetera, removes many of the minor and trace
elements from solution. Physical and chemical processes that may control amounts of these elements in
ground water are dispersion, complexation, acid-based reactions, oxidation-reduction, precipitation-
dissolution, and adsorptJon-desorption (UNESCO, 1980).
198
-------
Nitrate: Excessive concentrations of nitrate in drinking water may cause methemoglobinemia
("blue baby" syndrome). Drinking water standard has been set at 44 mg/L nitrate, which is equivalent to 10
mg/L nitrogen. This standard is increasingly exceeded, especially in rural areas, where a variety of nitrate
sources exist, and where private well-water use is still high. Sources of nitrate include (a) natural soil
nitrogen converted to nitrate by bacteria, accelerated by cultivation of land, (b) septic tank drainages,
(c) animal waste effluents, and (d) commercial fertilizers.
Nitrogen also occurs in ground water in the form of nitrite (NC>2~) and ammonia (NH4+);
concentrations of these compounds are generally much smaller than nitrate concentrations, however.
Nitrate content has been used widely to identify pollution from organic waste, that is, from septic tanks and
animal wastes, especially when combined with a relatively high chloride content, which is also present in
those waters. In contrast, high chloride concentrations in association with low nitrate concentrations has
often been used for identification of a salinization source other than agricultural effluents or septic tanks,
as all other salinization sources (oil-field brine, halite, road salt, etcetera) are low in nitrate.
Where the source of nitrate is unknown, the isotopic composition (15N/14N) can be useful for
differentiating soil nitrate from fertilizer nitrate or from organic-effluent nitrate (Kreitler and others, 1978;
Kreitler, 1979). However, determination of nitrogen isotopes is relatively expensive (Table 13).
To prevent bacterial creation or destruction of nitrate between sampling and analysis, it is important
to treat the water sample in the field. Thompson and Ouster (1976) reported concentration increases in
the order of 100 ppm in untreated samples relative to acidified samples within three days of sample
collection. After several months, the same authors also detected small nitrate increases in acidified
samples. Therefore, Thompson and Ouster (1976) recommend that nitrate water samples should be
filtered, acidified, cooled, and analyzed quickly. R. F. Spalding (personal communication, 1991)
recommends filtering and freezing of the samples without acidification before shipment to the laboratory in.
frozen form (note that sample bottles should not be filled completely to avoid breakage).
Potassium: Potassium is added to ground water mainly from the weathering of feldspars and clay
minerals. But more easily than it is incorporated into solution it is taken out of solution by fixation into clay
minerals (Craig, 1970, Frape and Fritz, 1987). Leaching of potassium from illite is part of the salinization
process observed at saline seeps in Montana (Donovan and others, 1981), suggesting its potential use as
a salinity tracer. In Ohio, Breen and others (1985) demonstrated that the ratio of K/Na can be useful to
distinguish between brines from different sandstone units. High concentrations are often found in mine
waters, hot-spring waters, and in syMte-dissolution brines. Potassium concentrations are often lowest in
halite dissolution brines, highest in brines from carbonate aquifers, and intermediate in brines from
sandstone units (Rittenhouse and others, 1969). Despite these differences, however, potassium
concentrations or K/X ratios (X - Oa, Mg, 804, Cl, Br) are mostly used within a suite of other constituent
ratios for detection of salinization sources.
199
-------
Selenium: Selenium in ground water is a byproduct of pyrite oxidation and dissolution and is often
found in high concentrations in irrigation and saline-seep areas. Areas underlain by black shales or shale-
derived materials (Donovan and others, 1981; Deverel and Gallanthine, 1989) appear to be affected most.
Selenium minerals are also associated with some of the uranium deposits in sandstones in the Western
United States. As such, its usefulness as a salinization tracer may be restricted to certain geographic
areas. Recommended limit of selenium in drinking water is 0.01 mg/L (U.S. Environmental Protection
Agency, 1975)
Silica: With the exception of some high-temperature waters, silica concentrations in most natural
waters are commonly less than 100 mg/L. Fogg and Kreitler (1982) observed a general decrease in silica
content in ground water away from recharge areas, which could be explained by precipitation of quartz
cement or precipitation in authigenic clays. In oil-field brines, silica concentrations are highly variable
(Rittenhouse and others, 1969), depending largely on lithology, residence time, and temperature. Silica
content has been used in a study in the Netherlands to distinguish young salt water from old salt water
(Custodio, 1987), but little other work related to salt-water sources has been documented.
Sodium: Sodium in ground water is derived mainly from the decomposition of feldspars and from
sodium salts. Once in solution, it tends to stay there unless minerals having high cation-exchange
capacities and available exchange sites are present. Sodium is commonly used in salt-water studies as a
constituent ratio of Na/CI, with a weight ratio of 0.65 in brine being characteristic of halite dissolution and a
ratio of less than 0.60 in brine being characteristic of oil-field/deep-basin brine (for example, Leonard and
Ward, 1962; Oklahoma Water Resources Board, 1975; Whittemore and Pollack, 1979; Gogel, 1981;
Richter and Kreitler, 1986a,b). Cation exchange between Ca in the aquifer material and Na in the water
leads to a decrease in Na/CI ratios. Albrtization of plagioclase or K-feldspar may also lead to a decrease in
Na/CI ratios (Land and Prezbindowski, 1981). This is the case in most oil-field/deep-basin brines and
wherever Na-type water replaces Ca-type water in the presence of suitable exchange sites. An increase in
Na/CI ratios will occur when fresh water replaces marine water (Custodio, 1987). The Na/CI ratio is applied
most successfully to differentiate halite-dissolution brine from oil-field/deep-basin brine, as documented
in numerous studies in Kansas, Oklahoma, and Texas and to document mixing of fresh water with salt
water. It usually is not applied to differentiate between brines from different stratigraphic units because of a
general overlap in ratios. In combination with other constituents, however, sodium may be used
successfully to distinguish between different brines on a local basis. This was shown by Breen and others
(1985), who used the ratio of K/Na to distinguish between brines from three sandstone units in eastern
Ohio.
Sodium makes up a major portion of the cation composition in most ground waters, but often is
determined by the difference between the sum of anions and the sum of calcium plus magnesium,
expressed as meq/L This practice should be avoided whenever possible, because analytical errors that
could be detected through cation-anion balance of analyzed constituents will remain undetected.
200
-------
Sulfate: Sulfate occurs in often high concentrations in nonsaline ground water where it is derived
primarily from the decomposition of iron suifides and solution of gypsum. To a lesser degree it is derived
from volcanic sources, as a result of the combustion of fossil fuels, and from industrial and mining activities
(Craig, 1970). Sulfate is easily reduced bacterially below about 80°C and thermally at higher temperatures;
therefore, sulfate-dominated solutions are rare in the deep subsurface (Land, 1987). Because of these
low concentrations in most brines, the constituent ratios of SCXi/CI or SOLIDS have been used
occasionally for detection of brine mixing with fresh water. However, because sulfate concentration can be
readily altered by chemical and biochemical processes, the usefulness of this parameter in salt-water
studies is limited. Therefore, ratios of SO4/X (X - Cl, TDS, Ca+Mg) are used best within a suite of ionic
ratios. The recommended maximum limit of sulfate in drinking water is 250 mg/L (U.S. Environmental
Protection Agency, 1975).
Total Dissolved Solids: The concentration of TDS in ground water is an overall measure of water
quality. Certain ranges of TDS values are often used to define terms such as fresh, brackish, saline, or
brine. Many different classification ranges are being used in the literature to define the same or similar
terms; the two major ones are shown in Table 2.
The amount of TDS is a general indicator of water quality and is a useful parameter for quality
monitoring. In studies of saline water, TDS is closely related to Cl concentrations and is used primarily in
mapping to illustrate the location and extent of poor-quality ground water. TDS is of limited use in
identification of salt-water sources except in simple one-source scenarios.
The concentration of TDS in solution is determined either as the dry residue after evaporation or is
calculated from the individual concentrations of major cations and anions. For calculation of TDS, HCC>3
concentrations are converted to carbonate in the solid phase using a gravimetric factor (mg/L HCOs *
0.4917 = mg/L CO^); this assumes that half the bicarbonate is volatilized as CC-2 and H2O and that the
carbonate value obtained corresponds to conditions that would exist in dry residue (Hem, 1985).
4.2. Summary of Field Techniques
The field-sampling methods described next follow for the most part the generally accepted
procedures, and were applied successfully by the authors in numerous salt-water studies. Some of the
suggestions made here may not meet with everyone's approval. For further references on sampling
methods, the reader may refer to Brown and others (1970), Wood (1976), Lico and others (1982), and
Hem (1985).
Collection of ground-water samples within salinization studies will be either from established wells,
from testholes drilled for the particular study, or from surface waters. The purpose of ground-water
sampling is to collect a water sample representative of the aquifer of interest. Installation of monitoring
wells and sample collection induce changes, which may make it impossible to obtain a truly representative
201
-------
sample (Pennine, 1988). Fortunately, these changes are most profound on dissolved gases, organics,
and trace metals, which are of minor importance in most salt-water studies.
Sample recovery will generally be with the help of a pump or a bailer. Wells may be pumped or bailed
for some time before sampling to avoid collection of a water that has been changed through long storage
time in the well bore and is not representative of true formation water. Pumping or bailing of three bore
volumes before sampling is generally considered sufficient. Garner (1988) recommended that sample
collection may be started when in-line monitoring values of pH, Eh, temperature, electrical conductivity,
and dissolved oxygen do not vary more than 10 percent per casing volume pumped. Puls and others
(1990) suggested that turbidity be monitored in addition to these parameters, and that the pumping rate
should be close to the actual ground-water flow rate. Some constituents, such as lead and cadmium,
however, may not stabilize even if those monitored water-quality parameters have stabilized, as reported
by Pennine (1988).
Sampling for most constituents will require filtering in the field (0.45 jam filter), which is done very
easily in the case of pumped samples using in-line filters. Filters will be discarded after each sample to
avoid cross-contamination between samples, which, at a cost of approximately $15.00 per filter may
appear expensive at first but probably is the most convenient, time-saving, and safest sampling method.
Although filtering in the field is a standard technique, Snow and others (1990), who compared analytical
results of filtered and unfiltered water samples, did not find a statistical difference between these two
groups. When planning sampling of salt-water wells it may be an advantage to conduct sampling from the
location of lowest salinity to the location of highest salinity, which makes equipment cleaning between
sampling easier and decreases the probability of contaminating a tow-TDS water with a high-TDS water
through the repeated use of sampling equipment. Repeated use of sampling equipment requires
generous rinsing of equipment using detonized water. A convenient check that sufficient rinsing was
performed is provided by addition of some drops of silver nitrate into a sample of rinse water. In the case of
Insufficient rinsing, the sample will turn cloudy from the formation of silver chloride, indicating that cleaning
should be continued. Such a check is needed especially when bailing samples, because those need to
be filtered using a specially designed reusable filter chamber, into which disposable filter paper is inserted
before the sample is forced into sample bottles by nitrogen gas. The use of nitrogen gas is recommended
when contact with atmospheric gas is to be kept at a minimum. In most salinization studies, however,
atmospheric contact is of minor concern and a foot pump may suffice for forcing the samples through the
filter.
Special care has to be taken when collecting water samples from oil and gas wells. To avoid too much
contact of sample equipment with oil, oil-water mixtures are collected in a bucket with a drum tap at the
bottom. After several minutes, water and oil will separate and water can be drained from the bucket.
Draining through a glass wool filter will remove most of the residual oil in the water before transfer to the
202
-------
filter bottle for final sample collection. Afterwards, the filter bottle may require cleaning with soap and
plenty of deionized water.
In addition to samples collected for laboratory analyses, some relative unstable parameters are
generally determined in the field. These include pH and alkalinity, which are determined during and
through titration. It is general practice to perform this alkalinity titration as soon as possible after sampling to
avoid out-gassing of the sample. In most salinization studies, the alkalinity value is used for not more than
the cation-anion balance or for the plotting of ionic percentages in Piper diagrams, in which cases slight
changes are of little concern. Therefore, collection of a separate sample for alkalinity titration to be
performed at the end of a sampling day can greatly increase sampling efficiency without adverse effects
on sample quality and data evaluation. It is the authors' personal experience that alkalinity titration
performed on a completely filled sample bottle within 12 hours of sample collection has produced
satisfactory sample results, that is, cation-anion balances always satisfied a pre-determined error tolerance
of less than three percent.
It is a good practice to design field-data sheets in the office before going into the field. These data
sheets will include a form that specifies the samples to be taken and the bottles and preserving agents to
be used. This information can be included on a sample-summary sheet, to be filled out at each sample
point. A different form will be used for the alkalinity titration.
203
-------
5. DATA AVAILABILITY AND SELECTION
5.1. Sources of Data
, i .
Among the variety of available data sets, the one maintained by the U.S. Environmental Protection
Agency, STORET, is probably the largest. Other data sets are maintained by the U.S. Geological Survey,
WATSTORE, by individual state agencies (for example, Texas Natural Resources Information System
[TNRIS], in Texas), and by commercial data services (for example, Petroleum Information, Dwights). These
large data bases are available on magnetic tape, on diskette, or as printouts either for a fee or free of
charge. Oftentimes, site-specific data can be retrieved from these sources. Local water-chemistry
laboratories, Individual researchers, and published reports can be excellent sources of data not included
on some of the federal, state, or commercial data banks. However, these data are mostly available only as
paper copies. Sorting and compilation of chemical analyses can be very costly, as indicated by Hiss
(1970), who reported acquisition and data reduction costs of approximately $5.00 per usable record
during a study of saline water data from southeastern New Mexico and western Texas. These costs appear
relatively high but are much less than costs of field collection and laboratory analyses, which may run into
several hundreds of dollars per analysis.
Quality and completeness of analyses vary to a large degree between and within data sets. Some of
the problems encountered when working with existing data sets are described below, using the following
data retrievals conducted for this study. Retrieval #1 (STORET): Complete chemical analyses from water
wells and springs in the United States in which chloride concentrations are greater than or equal to the
drinking water standard of 250 mg/L This request resulted in 99,915 analyses from 18,772 stations.
Retrieval #2 (STORET): Locations and chloride concentrations of water wells in the United States. This
request resulted in 734,091 chloride analyses from 212,678 stations. Retrieval #3 (TNRIS): Locations and
chloride concentrations of all water wells in Texas. This request resulted in 71,835 chloride analyses from
33,463 stations.
The completeness of analyses varied to a high degree in Retrieval #1, that is, only approximately
50 percent of all analyses include all major cations and anions. Selection of minor constituents or
isotopes reduces the availability, and with it the use of this data set, even further, as, for example, only
2,716 bromide values, 1,466 iodide values, and 826 oxygen-18 values are included in the set.
Requesting complete analyses or a large number of parameters per analysis, in this case resulted in a
huge data file that is difficult and expensive to work with. Separation into several small data files, for
example, one of major constituents and another one of isotopes, will eliminate some of the problems of
working with large files.
204
-------
Although a relatively good coverage for almost the entire country was provided by Retrieval #2
(Fig. 71), it is apparent that there is a lack of data in some areas, such as Texas. This lack is artificial, as
indicated by Retrieval #3 (Rg. 72), which illustrates a much higher data density from the TNRIS data set
when compared to the STORET data set. This suggests that working with more than one established data
set can greatly increase data coverage and that care should be exercised during selection of a certain data
bank.
Existing data sets often contain a bias toward certain data. For example, an evaluation of ground1
water chlorinity in the state of Tennessee using the STORET data base would actually be an evaluation of
water quality from a relatively small area in Tennessee (Rg. 71). Obviously, most of the samples available
for this state were derived from a study that was confined to a certain area. Site-specific studies also may
tend to be biased toward a specific source of water, such as shallow ground water of potable quality
(biased toward fresh water), sources of ground-water contamination (biased toward saline water), ground-
water quality in formation A (biased toward fresh or saline water), or ground-water quality in a specific
county. Most large, computerized data bases of water quality (with the exception of petroleum-related data
banks) consist of analyses from observations wells and from municipal and domestic wells, which by their
own nature are for the most part of the best quality available in any area. This may be reflected in Retrieval
#2; 88 percent of the stations have chloride concentrations in water that are less than the drinking-water
standard of 250 mg/L. Whenever a poor-quality water is encountered during drilling or when a water well
has gone bad, the well is most likely abandoned, no additional analyses are obtained, and a new well
furnishing better water is installed. Indiscriminant mapping may identify this as an improvement in water
quality in the area or it may be interpreted as an area without salinization problems. The small number of
chloride analyses with chloride concentrations greater than 250 mg/L (9 percent of all stations with
chloride analysis) suggests that the STORET data set may be biased toward tow-TDS waters. Another
problem with the use of of large, existing data sets is the often unknown source of the data and the lack of
information regarding well depths or producing formations. For example, only approximately 30 percent of
all analyses in Retrieval #1 included some kind of depth specification. A depth specification is important,
however, as most of the country is underlain by saline ground water at some variable depth.
Besides data bases of chemical analyses, literature data bases can be used to assist in data
searches. For example, for this report, the Water Resources Abstract data base, which contains
approximately 170,000 records on diskette, was used for screening of keywords, such as salt water, brine,
salinization, saline seep, etcetera. A variety of computer-based data bases are available to researchers
(Table 14), as described by Atkinson and others (1986) and Canter (1987), for example.
205
-------
ro
o
a>
\ '.' ^^?fe?^l®S^|^--^X' ?' /?
, ^^^«1ftft^^^Sff^¥^^^/
QAI72S8
Figure 71. Location map of ground-water stations for which a chloride value is available at U.S. Environ-
mental Protection Agency's data base STORET (Retrieval #2; approximately 200,000 stations).
-------
m.
it •• \ «
•".' :"P%*>: v*/
'• '*'.?••'*• ?**•"<'/
• •' ^'fir SJf
* . *" •••*.• f ^T; 11
*\ . ^^,'-f -.' "
I? ' " . » t*. .* l
Figure 72. Location map of water wells in Texas for which a chloride value is stored at the Texas Natural
Resources Information System data bank (approximately 33,000 stations).
207
-------
Table 14. Description of literature data bases (from Canter, 1987).
1. Agricola. Years covered 1970-1978. Contains all the information of the National Agricultural
Library and comprehensive coverage of worldwide journals and monographic literature on
agricultural subjects.
2. Agricola. Years covered 1979-1984. See above. The two Agricola data bases contain 2 million
records.
3. Compendex. Years covered 1970-1984. Contains the machine-readable version of the
Engineering Index which includes engineering and technological literature from 3,500 journals
and selected government reports and books. Size of data base—1.4 million records.
4. Conference Papers Index. Years covered 1973-1984. Contains records of more than 100,000
scientific and technical papers presented at over 1,000 major regional, national, and international
meetings each year. Size of data base—1.0 million records.
5. CRIS/USDA. Years covered 1982-1984. Contains active and recently completed agricultural
research sponsored by the USDA or state agriculture institutions. Size of data base—31,000
records.
6. Dissertation Abstracts. Years covered 1961-1984. Contains dissertations from U.S. institutions
as well as Canada and some foreign schools. Most of the abstracts are for degrees granted after
1980. Size of data base—852,000 records.
7. NTIS. Years covered 1964-1984. Contains government sponsored research, development, and
engineering plus analyses prepared by federal agencies, their contractors and grantees.
Unclassified, publicly available reports are available. Size of data base—1.1 million records.
8. Pollution Abstracts. Years covered 1970-1984. Contains environmentally related literature on
pollution, its sources and control. Size of data base—107,000 records.
9. SSIE Current Research. Years covered 1978-1982. Contains reports of both government and
privately funded scientific research projects either in progress or recently completed in all fields of
basic and applied research in the life, physical, social, and engineering sciences. Size of data
base—439,000 records.
10. Water Resources Abstracts. Years covered 1968-1984. Contains materials collected from over
50 water research centers and institutes in the United States and focuses on water planning, the
water cycle, and water quality. Size of data base—173,000 records.
208
-------
5.2. Selection of Data Criteria
Before starting with the evaluation of a set of existing chemical data, it is worthwhile to establish
certain criteria that must be met by the data. These criteria will vary with the nature of information required.
Exact well locations (for example, latitude and longitude) will be important in a study of a contaminant
plume from a leaky salt-water well, whereas specification of a county code may be sufficient in a regional
water-quality study. In the former case, analyses of waters from unknown locations will be discarded from
the beginning or an effort will have to be made to identify well locations. Similarly, the date, method, or
purpose of sample collection may be of great importance. The date allows identification of chemical
changes over time and the establishment of precorrtamination concentrations. Mixing of Current and
historical data without dates, in contrast, may lead to wrong conclusions regarding the geographic
distribution of water-quality changes. The method of sampling will determine how reliable analyses of
certain constituents may be. For example, lab-determined alkalinity values may suffice in a study in which
alkalinity values are used for not much more than a mass balance, but may not be acceptable in a study of
equilibrium conditions within carbonate systems. The purpose of sampling may be reflected in an
underlying bias toward a certain water type; for example, monitoring of municipal wells will be biased
toward good-quality waters, whereas sampling within a salt-water contamination study will be biased toward
poor-quality water. When using large, existing data bases, some of these parameters may not be available,
especially when different sources contribute to that data base.
The problem of completeness of information includes the availability of certain chemical parameters.
Most analyses of ground water include the major cations (Ca, Mg, and Na) and the major anipns (HCCL,
: _.,-,. • , , .,,.-. . £
SO4> and Cl). Provided these constituents are all determined in the laboratory or in the field, a mass
balance error can be determined (difference of sum of cations and sum of anions over the sum of cations
and anions [in meq/L]; |ecatk>ns-eanions|/[ecations+eanions]). The number of data available, the quality of
the water, and the personal quality criteria established by the investigator will determine what kind of error
is considered acceptable. This error boundary will most likely be high when any sample discarded would
represent a substantial loss, that is, when data coverage is very sparse, and when dealing with very fresh
waters. The upper margin of error generally is located somewhere between 5 and 10 percent. In the case
that one parameter, sodium, was calculated by difference instead of analyzed, the balance error will be
zero percent. Availability of an adequate number of analyses that had not been determined in such
fashion will conclude if calculated values should or should not be used.
The availability of existing data often determines the technique to be used for evaluation of water
chemistry. For example, chloride and bromide concentrations are often used with good results to
distinguish between halite-solution brine and oil-field brine. But because bromide is not a part of standard
water analyses, only a relatively limited amount of data may be available. Also, time and money will often
determine which technique or data base to use. Availability of a free data base (computer tape) of water
209
-------
chemistry from state or federal agencies can considerably cut down on costs and time, as opposed to
having to compile data from published sources, water-quality files, or chemical laboratories. On the other
hand, collection and analysis for isotopes may be very time consuming and expensive, as costs per single
analysis may go into the hundreds of dollars (Table 13). When using commercial, outside laboratories for
sample analyses, the following observations should be kept in mind (Rice and others, 1988): (a) the
reliability of laboratory analyses should not be taken for granted, (b) analytical reliability may not be
reflected in the price charged by laboratories, and (c) quality assurance programs benefit both the
customer and the laboratory.
How to graphically display and statistically evaluate data selected from available sources or collected
In the field will be discussed briefly in chapter 6.
210
-------
6. GRAPHICAL AND STATISTICAL TECHNIQUES
Evaluation of chemical analyses often starts with graphical display and statistical manipulation of
physical and chemical data. Which technique is used depends largely on the amount of data and on the
type of information that is needed. In salt-water studies, techniques are used that maximize the separation
of chemical characteristics between potential salt-water sources and illustrate to which salt-water source a
contaminated water sample belongs.
6.1. Graphical Techniques
Graphical techniques are used to (a) illustrate the chemical character of a single analysis, (b) compare
the characteristics of several analyses, (c) assist in identifying the relationship that exists between water
samples, and/or (d) to calculate mixing ratios between fresh water and the contaminating source.
Among these techniques, the simplest ones illustrate a single parameter or analysis. Contouring or
posting of chemical parameters onto maps or cross sections (Fig. 73) is done for locating areas of
abnormal chemical composition, suggesting an erroneous value, a possible point source of mixing (for
example, a leaky well), or more than one source of water. As such, these maps are also useful in
identifying local positions of contaminant plumes or regional changes in water quality. In order to represent
several ions that make up a chemical analysis either several contours, several maps, or other methods that
combine the ions in a convenient form, need to be used. These other methods include bar graphs (Fig.
74a), pie charts (Fig. 74b), and polygonal (Stiff) diagrams (Fig. 74c).
On bar graphs and pie charts, concentrations of major cations and anions are illustrated by different
sizes of representative areas. The overall shape and size of the graph may stay the same, in which case
relative concentrations (percentages) are indicated. On bar graphs, absolute concentrations of individual
ions can be displayed by changing the sizes of representative areas. A change in overall size of the pie
chart, in contrast, can be used to express changes in TDS concentrations. Extending relative areas
according to concentration scale similar to the one used in bar graphs can be used to present
percentages as well as absolute concentrations on pie charts. Bar graphs were used by Williams and
Bayne (1946) to display differences in mixtures of fresh water and saline-formation water from mixtures of
fresh water with oil-field brine in Kansas. Relatively low percentages of magnesium and sulfate in oil-field
waters are reflected in mixtures between oil-field brines and fresh water; mixtures of fresh water with saline
formation water, in contrast, is characterized by higher magnesium and sulfate percentages. Pie charts are
often used to illustrate water quality on maps, as was done on a national scale by Feth and others (1965).
Stiff diagrams allow a quick comparison between samples primarily through a change in shape, which
is caused by different ionic compositions. This is provided by consistent plotting of cations to one side
and anions to the other side of a vertical zero line and by connecting the end points. The distance of each
211
-------
(a) r
EXPLANATION
• Control point
/ Total disiolvnd solids
(mg/L)
10 mi
WHARTON CO
Colorado R.
Beaumont
Chicot
Evangeline
-IOOO
e
(§)
—soo-
EXPLANATION
Well bottom
Profile intersection
Total dissolved
solids (mg/L)
Base of fresh water
Ca-HCO-j
•: Mixed-cation-HC03
Na-HC03
Na-CI
o
I-
10 mi
10 km
OAI7260
Rgure 73. Graphical illustration of chemical analyses by contouring of individual parameters onto (a) maps
and (b) cross sections (from Richter and others, 1990; and Dutton and Richter, 1990).
212
-------
(a)
4-
2-
12-6
15-1
[ I Na + K [:;-::\ Ca
SO4
Hardness
(b)
15-1
-Na
SO4
17-3
12-6
SO4
(c)
Na + K
Ca
Mg
Fe
10
Cations
Cl
HCO;3
SO4
C03
17-3
Anions
iiiiii^
30 25 20 15 10 5 0 5 10 15 20 25
Meq/L Meq/L
(d)
Meq/L
-100
01 5 10
50 100
Na + K
| |
Scale of radii
(total of meq/L)
Ca | [ S04
Cl [777777:] HCO3
-10
-1.0
-100
Na
-1000 Cl S04
-1000
1000
-100
FA/
-10
-10 '-
-10
HC03
- 1000
-100
-10
10
12-6
15-1
17-3 -_..__.
QA17261C
Figure 74. Presentation of major ions in form of (a) bar graphs, (b) pie charts, (c) Stiff diagrams, and
(d) Schoeller diagrams (from Hem, 1985).
213
-------
end point from the vertical zero line is proportionate to the concentrations (in meq/L) of the respective
constituent, allowing easy identification of hydrochemical fades. Stiff diagrams were used by Bumitt and
others (1963) to graphically depict changes in ground water quality in Ogallala outcrop areas of Texas
(Fig. 51). Stiff diagrams and geographic mapping of major chemical parameters were also used by
Levings (1984) in a study of oil-field pollution in the East Poplar field, Montana. Elevated IDS, Na, and Cl
concentrations reflected the extent of ground-water movement away from the pollution site on isocontour
maps, whereas Stiff diagrams illustrated the change from low-TDS, Na-HCOs waters (background levels)
to high-IDs, Na-CI waters (produced oil-field waters).
These single-analysis diagrams are of limited use when large data bases are to be considered. In
such a case, a large number of analyses needs to be combined within one graph, such as in a Schoeller
diagram (Schoeller, 1935), Piper diagram (Piper, 1944), or a bivariate plot.
On a Schoeller diagram, similarities and dissimilarities of water types are displayed in a similar fashion
as in the Stiff diagram by comparing the overall shape created by all major anions and cations making up an
»*
analysis (Rg. 74d). This plotting technique is very useful for representing changes in the relationship
between ions, as nonchanging ions plot at identical points whereas changing ions plot at different points
along their respective axes.
Piper diagrams combine major cations and major anions in separate triangles, reducing the
compositions to single points that represent meq/L percentages of each individual cation and anion
(Fig. 75a). Each analysis is presented as a single point in the diamond-shaped diagram by projecting the
anion and cation positions into that field. This technique allows convenient determination of
hydrochemical fades (Fig. 75b), as well as assistance in interpreting relationships that may exist among the
water samples, such as (a) mixing of different water types (see also Sea-Water Intrusion, chapter 3.3),
(b) cation exchange (see also Sea-Water Intrusion, chapter 3.3), (c) precipitation and dissolution
reactions, and (d) sulfate reduction (Custodio, 1987). For example, Krieger and Hendrickson (1960) used
Piper plots to graphically depict the mixing between brine and fresh water, which was suggested from high
chloride concentrations (Fig. 75). On the Piper diagram, the contamination is indicated by a straight-line
mixing trend from a Ca-Mg-HCOs type fresh water to a Na-CI type salt water.
Probably the most frequently used form of graphical presentation of chemical data is the bivariate
plot (scattergram). By plotting a physical (for example, well depth, distance) or chemical (for example, Ca,
Na/CI) variable against another, the correlation between these parameters can be identified. On these
plots, large scatter signifies little correlation between the plotted parameters, whereas any trend indicates
either mixing of two waters or evolution of one water, such as, evaporation, precipitation, or solution. More
than one large scatter may suggest two or more distinct groups of water whereas one or a few points
outside a suspected trend may suggest erroneous data points or inclusion of nonrepresentative water
samples resulting from, for example, a different origin (aquifer, well depth) or possibly point-source
contamination. Any combination of parameters may be used, but it is preferred to plot parameters for
214
-------
(a)
QA 1 72620
Ca
Cl
Figure 75. Presentation of chemical constituents on trilinear (Piper) diagram (a) (c) and classification
scheme of hydrochemical fades (b), based on major-ton percentages (from Krieger and Hendrickson,
1960, and Freeze and Cherry, 1979).
215
-------
which a correlation can be explained by chemical or physical processes. For example, a good correlation
between calcium and sulfate or calcium and alkalinity strongly suggests chemical control of these
parameters by gypsum and calcium carbonate, respectively. Another parameter pair that is used
frequently, especially In salt-water studies, is that of sodium and chloride. Solution of halite (NaCI) by fresh
ground water results in a saline water in which the molar concentrations of sodium and chloride are equal
(mNa/mCI » 1) as long as the water is not too diluted (Na/CI ratio is typically greater than unity in fresh
water). Bivariate plots of sodium over chloride reflect this process through a slope of 0.65 on a plot of
weight ratios (Rg. 76) or a slope of unity on a plot of molar concentrations. Oil-field brines also often show
a good correlation between sodium and chloride, but the slope is considerably lower in most instances
(Rg. 76). This difference in slope of Na/CI ratios between halite-solution brines and oil-field/deep-basin
brines has been used extensively for salt-water studies in Oklahoma, Kansas, and Texas (Leonard and
Ward, 1962; Gogel, 1981; Richter and Kreitler, 1986a,b). Because of its conservative nature once in
solution, chloride is the most often used parameter in this kind of bivariate plot. Regarding salinization of
fresh ground water, representation of the composition of one or more potential salinization sources and a
possible mixing trend between fresh ground water and one of these sources is of special interest. This
technique was used successfully by Richter and others (1990), for example, who were able to separate
one formation brine from another (Rg. 77) and disposal brine from naturally saline ground water (Fig. 49) in
a salt-water study in West Texas.
When dealing with salinization of ground water which, in many instances, is a mixing between fresh
ground water and saline ground water or brine, absolute concentrations vary widely but constituent ratios
only vary to a small degree. In addition, because all salinization sources are high in dissolved solids,
differences in constituent ratios are often better tracers of certain sources than are absolute
concentrations. This led Novak and Eckstein (1988) to propose the use of ratios instead of concentrations
in some of the traditional graphical techniques (Fig. 78).
When the end-member composition of fresh water and the contaminating salt water are known, the
percentage of each endmember in a mixing water can be determined mathematically or graphically
(Custodio, 1987). Using the most conservative dissolved ion in ground water, chloride (see also
chapter 4), the percentage is calculated by the following equation or read from the percentage scale of
figure 79, based on the assumption that mixing is the only dominant effect that caused the chloride
increase in the mixing water. Chloride concentrations of the endmembers are used in the equation and in
the graph, where they are end point of a mixing trend. After the mixing ratio has has been established,
theoretical mixing values for Na, Ca, Mg, etcetera, can be calculated (see example below). Deviations of
these theoretical mixing values from the true values measured in the mixing water indicate changes other
than mixing, such as ion exchange. Other conservative constituents that can be used instead of or in
combination with chloride are bromide (Fig. 79) and oxygen-18 (Arad and others, 1975).
216
-------
150,000
100,000-
O)
OJ
50,000-
CI (mg/L)
Rgure 76. Bivariate plots of Na versus Cl for halite-solution (solid dots) and deep-basin brines (open dots).
217
-------
60
o
X
a
m
20-
Pennsylvanian
oil-field brines
Log Cl (mg/L)
D
Q
Permian
oil-field
brines
5
QA11668C
Rgure 77. Use of bivariate plots for identification of mixing trends between fresh ground water and
potential salirtzation sources in parts of West Texas (from Richter and others, 1990). With increasing
chloride concentrations, testhote samples (crosses) approach the composition of Permian oil-field brines
(solid squares), as opposed to Pennsylvanian oil-field brines (open squares), suggesting that Permian oil-
field brines contribute to salinity in water wells of the area.
218
-------
10,000-,
1,000 -
100 -
.2
ni
CC
Na/Ca Na/Mg Mg/K Ca/Mg (Na + K)/ Na/K
(Ca + Mg)
QA 1726SC
Figure 78. Modified Schoeller and Piper diagrams using concentration ratios as endpoints (from Novak
and Eckstein, 1988).
219
-------
Salt water
end member
Fresh water
end member
100
so
75
100
Percent mixture (fresh water)
75
50
25
Percent mixture (sea water)
Cl (ppm)
20.000-
18.000-
16.000-
•g 14,000-
£ 12.000-
g 10.000-
*
i
re
CO
8000-
6000-
4000-
2000-
0
100
Percent sea water
Well number
-2
7-8.5
5.5-6.5
9.5 .
13.5-14.0
30-32
32.5-34.5
30
29-31
25
50
75
Percent mixture (fresh water)
75
i
50
I
25
Percent mixture (sea water)
Br (ppm)
-100
-90
-80
-70
-60
-40
-30
-20
MO
100
05
TJ
-50 ®
CT
TO
W
CD
QA17266C
Figure 79. Calculation of mixing percentages between fresh water and salt water using mixing graphs of
chloride and bromide (from Arad and others, 1975, and Custodio, 1987).
220
-------
CM =. x.cF + (i-x).cs===>x
where: C.. - Constituent concentration in mixing water,
Cp » Constituent concentration in fresh water,
Cg = Constituent concentration in salt water,
X = Fraction of fresh water in the mixture, and
1-X * Fraction of salt water in the mixture.
Example: Step 1 : Cl^ - 1 ,500mg/L; Clp = 50mg/L; Cls = 35,OOOmg/L
=3=> X = ,(35,000-1 ,500)7(35,000-50) = .96
•• . The fresh-water source is represented by 96 percent and the salt-water
source is represented by 4 percent.
Step 2: Naw = 600mg/L; Nap - 1 55mg/L; Nas - 22,OOOmg/L
with 0.96 - (22,000-Y)/(22,000-1 55)
==> Y = 22,000 - 0.96*(22,000-155) = 1,029
The theoretical sodium concentration in the mixing water is 1 ,029 mg/L; the actual concentration is
only 600 mg/L, however. Ion exchange may account for the loss of sodium, as long as another cation
shows a comparable gain to its theoretical value.
6.2. Statistical Techniques
The application of statistics depends to a high degree on the number of observations in the data
base and the nature of the required information. For the purpose of this report, only a general discussion
of readily available statistical techniques was attempted. The exception is made with Stepwise Discriminant
Analysis, which is presented as one possible technique to identify useful parameters for identification of
salinity sources;
Hem (1985) pointed out that the literature abounds with questionable applications of statistical
procedures, the major reason for this may be that a strong background in statistics as well as in water
chemistry is needed to successfully apply some of the more sophisticated statistical techniques during
water-quality studies. Generally, for investigation of water chemistry, Hem (1985) suggested the use of
statistics as a means of testing and verifying theories instead of simply creating theories from statistical
data.
Statistical techniques are the most useful and appropriate when a large data base of observations is
available. Simple averaging or determination of frequency distributions are widely used in water-analysis
221
-------
Interpretation. Both are good techniques in establishing the background water-quality data often needed
for identification of a salinization process. Together with an average value, a maximum background value is
often given, which is then used to identify anomalously high concentrations caused by mixing with a
salinization source. Frequency distributions are used on large data bases to identify any outliers or to
determine the number of data populations that make up the total data set. Outliers may indicate
contamination, faulty analyses, or data points not representative of the rest of the data. Analyses derived
from wails that produce from different aquifers or from wells located in different geographic areas may
result in more than one grouping of the data. This is important when sources of chemical changes are
investigated, because these data groups may be unrelated to each other.
During mixing between a fresh water and a salt water, absolute concentrations vary to a high degree,
whereas constituent ratios may vary relatively little. With only one potential salinization source present,
absolute concentrations will most often provide enough information for identification of the salinization
process, as concentrations exceed normal background levels. If more than one potential salinization
source is present and a significant difference in constituent concentrations exists between them,
absolute concentrations may still suffice to identify the actual source of mixing, as the concentration of a
particular parameter may increase significantly and stand out. More often, however, absolute
concentrations will not allow positive identification of the one true source of two or more potential sources
because of overlapping concentrations, especially in the case of highly diluted mixing waters. In such
cases, concentration ratios may have to be used for separation of endmembers. Na/CI and Br/CI are two
ratios that are known to work well in a number of salinization scenarios (see also chapters 3.2 and 3.4), but
which ratio works best will depend on the composition of the individual endmembers involved.
When examining scattergrams of major ionic constituents versus chloride for various potential
endmembers of salinization (for example, sea water, halite-dissolution brines, and oil-field brines), it
becomes apparent that some sources are more variable than others. Samples from halite solution (Fig. 22)
or from sea-water intrusion (Rg. 35) show relatively little scatter, indicating little variation in end-member
chemistry. When plotting oil-field brines together from different areas (Rg. 47), in contrast, much more
scatter can be observed, reflecting the different origins and mechanisms of concentration of these brines.
From this it is apparent that differentiation between sea water and oil-field brines or halite dissolution and
oil-field brines cannot be done using a single, universal chemical constituent or constituent ratio. Instead,
constituents to be used for differentiating between these brines will vary from one location to another in
the same manner as the chemistry of the oil-field brine endmembers changes. One way to determine
which constituents work best for distinguishing between salinization sources in any given case would be
plotting or calculating ratios by trial and error until a good separation has been found. A more efficient way
may be provided by Stepwise Discriminant Analysis (SDA), which is a statistical technique that identifies
variables that distinguish between two or more predetermined groups of cases (Dixon and others, 1981).
This technique was used successfully by Hitchon (1984) to group formation waters in the Western Canada
222
-------
Sedimentary Basin, by Hawkins and Motyka (1984) to identify the origin of mineral springs in parts of the
Copper River Basin, Alaska, and by Novak and Eckstein (1988) to differentiate salt water derived from road
salt from brine samples derived from oil fields in Ohio. In terms of water chemistry, SDA can be used to
determine those chemical constituents or constituent ratios that are most useful for distinguishing
between given groups of waters, such as oil-field waters and sea water. When dealing with contamination
of fresh water by salt water, the main mechanism changing the chemistry of the fresh water is mixing, which
can also be considered as dilution of the salt water. In the case of dilution, absolute concentrations vary to
a high degree, whereas concentration ratios change relatively little. Therefore, as long as the solution
doesn't get too diluted and doesn't take on the ratio characteristics of the uncontaminated water,
concentration ratios are generally better tracers of salt-water sources than absolute concentrations and
were used in the following discussion of SDA.
In SDA, each predetermined group of samples consists of a number of chemical analyses made
using a variety of parameters, for example, major ions. Any parameter or, in this case, parameter ratio (for
example, Ca/CI, CaJSO^ Mg/CI, etc.) that is specified will be used during a SDA run to calculate mean
values representative of each group. The difference between the groups of interest is then expressed in
a linear function using the differences between the group means of each parameter. The ratio (for
example, Ca/CI) that best separates the groups, that is, for which mean values are most distinct between
the two groups, is incorporated into this equation first. During this step a certain percentage of the
individual analyses in each group will be assigned correctly to their preassigned groups, that is, parameter
ratios are closer to the respective group mean than to the mean of the opposing group. However, in most
cases, some of the analyses will be assigned to the opposite group, depending on the degree of
difference between the given groups. In the next step, SDA will attempt to correctly assign the remaining
analyses to their preassigned groups by incorporating another ratio in combination with the previously
determined variable into the equation. The addition of ratios (variables) will continue in successive steps
until a maximum number of analyses was assigned correctly to its preassigned group and no further
improvement in sample assignment can be achieved by incorporating additional ratios into the equation.
Depending on the similarity between the groups, any amount of ratios will be determined, that is, two
groups that are relatively similar may necessitate inclusion of many ratios for separation of their group
members into the assigned groups, whereas two very different groups may be distinguishable by just one
or two ratios. The variables determined by SDA can then be used for further study.
To illustrate the usefulness of SDA for identification of ratios that distinguish between given groups,
several data sets of water chemistry were compiled from the published literature. Data sets include oil-field
brines, halite-solution brines, sea-water intrusion samples, and ground water. Chemical analyses were
grouped according to brine type and location and then used in a variety of combinations to determine
ratios that best separated these groups. Ratios used were those identified as useful in salinization studies
(Table 12) and for which data were available (Ca, Mg, Na, K, HCO3, SO4, Cl, Br, and I): scenarios tested
223
-------
were (a) oil-field brine versus oil-field brine, (b) oil-field brine versus sea-water intrusion, (c) oil-field brine
versus halite-solution brine, and (d) a known case of ground-water contamination by oil-field brine. The
SOA software used in the following example is part of the BMDP software package, which is commercially
available from the license holder to be run on mainframe computers or PC's. .... •
SDA was performed on all' 15 possible combinations between any two of six oil-field brine groups,
representing brines from Texas, Louisiana, California, Oklahoma, Ohio, and Canada. Among the first three
ratios selected during each run, Ca/CI, Mg/CI, Na/CI, Br/CI, and I/CI were the ones most frequently
Identified (Table 15). It is probably reasonable to assume that combinations of these ratios will also provide
good separation power between brines of the same type (oil-field or deep-basin .brines) in other areas and
that the other ratios may be useful only on a site-specific basis.
Solution of halite (chapter 3.2) and sea-water intrusion (chapter 3.3) each produce brine and saline
ground water of relatively uniform chemical character (Rgs. 22 and 35, respectively). Therefore, where
local samples from sea-water intrusion or halite solution are not readily available, samples from other areas
can be used as hypothetical endmembers with reasonable accuracy. This was done for identification of
ratios that separate oil-field brine from sea-water intrusion brine in California, Texas, and Louisiana, and
from halite-solution brine in Texas, Pennsylvania, and West Virginia. In the cases of sea-water intrusion
versus oil-field brine, combinations of 7 (out of a possible 12) different ratios were determined as the best
four ratios in the three test runs (Fig. 80), whereas in the case of halite solution versus oil-field brine,
10 ratios were identified (Rg. 81). Some ratios appear again as more useful than others, such as the Br/CI
ratio in the case of halite solution versus oil-field brine, or the HCOg/CI ratio in the case of sea-water
intrusion versus oil-field brine. It should be emphasized here that SDA determines ratios from a strictly
statistical point of view and not from a geochemical point of view. As pointed out in chapter 4, bicarbonate
is not a conservative constituent in ground water, as its concentration is affected more by interactions
between aquifer material and CO2 with water than by mixing. Therefore, although the HGOjj/CI ratio allows
good separation between some oil-field brines and sea-water intrusion samples, the ratio may be of little
help in identifying the source of salinity in water contaminated by any of these two sources. The same
consideration should be given to any variable determined through statistical methods. In the case of Br/CI
ratios, geochemical considerations support the statistical evaluation, as discussed in chapters 3.4, and 4.
Plotting of step ratios 3 and 4 in bivariate plots, as done in figures 80 and 81, does not follow the
SDA logic, as these ratios are determined after and in combination with ratios determined in step 1 and in
step 2. Nevertheless, combining these ratios in bivariate plots can support group separations. Once the
ratios are known that best separate potential endmembers of salinization, these ratios can possibly then
be used to identify which endmember is the true source of salinity in a contaminated ground water. This
can be done graphically or through the SDA feature of checking individual analyses for their similarity to
endmember compositions (group means). First, SDA checks individual samples within each endmember
group for their representativeness of that group, thus enabling the researcher to check the initial
224
-------
Table 15. Listing of constituent ratios that separate best brines from Texas, Louisiana, Oklahoma,
California, Ohio, and Canada, as determined through Stepwise Discriminant Analysis. Ratios were
determined by individual runs of any combination between two brine groups, totaling 15
combinations between the six areas.
Sequence
Ratio Sequence of Selection
Ca/CI 1,1,1,1,1,2,2*
Mg/CI 1,1,1,3,3
Na/CI 1,1,2 .
Br/CI , 1,1,2
I/CI 2,3,3,3,3,3
HCOjj/CI 1,2,2
Ca/Mg 1,2 ,
K/Br 1,3
Na/K 2,3,3
SO4/(Na+K) 2,2
Ca/Br 2,3
(Br/CI)/(Ca/Mg) 2,3
SO4/CI 2
Na/Mg 2
Ca/K 3
Ca/SO4 3
(Ca+Mg)/SO4 3
Explanation:
1 Selected as the best ratio during any one run (step ratio #1), providing the single-most
separation between two groups.
2 Selected as the second ratio after step ratio #1 during any run, providing improved
separation between two groups in combination with step ratio #1.
3 Selected as the third ratio after step ratios #1 and #2 during arty run, providing further
improvement of separation in combination with step ratios #1 and #2-
* Of the 15 combinations between the six data sets, Ca/Cl was selected five times as the first
step ratio and two times as the second step ratio.
225
-------
(a)
s
fi
.01-5
•oooi
a.
Si
B
1 •01"!
y 5 .001 •
2.£
4r ^
* ™ .0001 i
I1 1
tt .00001 :
a.
H -
a>
nivwn -
' 't^.-.
A A ^iB^t
A
A A A
A Ai A
A A
A A A
.0001
.001
.01
.00001 .0001 .001 .01
10
STEP RATIO #1:Mg/CI (weight ratio)
STEP RATIO #3: HCO3/CI (weight ratio)
(b) ,
"3
a
3
sr ,01 T
§• —
§ o
a 1 .001 T
** "^
8| ,ooon
S ~" .00001 -
Q.
}" ,000001 T
(A
*•
A. AA A ^V
A£A At^^£& V V0
Aj^ jffjOfffBP^^^ ~^
ij^ AA
A A
A «.
A
,00001 .0001 .001 .01 .1 1 10 100
STEP RATIO #1: HCO^CI (weight ratio)
(c)
iCa/CI (weight ratio]
-sis
.J ,,,::J . .L,,J ......
$
O 4>t "
£?
2
0. .01 1
JU
w 001-
..:.-'•' A
AAA^P
^fTTi ^M
A. A A>dllft& V A
A A-a^yw |
*
03/CI (weight ratio)
M
• •
2
O
§
a.
in
.00001 ,0001 .001 .01 .1 1 10 100 1000 W
^^ 1000 i
^2
CO
2
^ 100 *
*o
^,
^
3 10 •
3
o
i 1-
cc
a.
£
-
• A A A
• A*A A
^ »^nAi Tl
*•* "* TO^\AAA
0 ^9HA.A ^0* ^ ^
A ^jk^A
jp A"AIHA;
AA^rA A
A
1 10 100 1000 10000
STEP RATIO #3: Na/Mg (weight ratio)
1 n >H
1 T
•11
.on
.001 1
.0001 1
.00001 i
*.-. f
1^
A
.000001 .00001 .0001 .001 .01 .1 1 10
STEP RATIO #1: Mg/CI (weight ratio)
STEP RATIO #3: SO4/CI (weight ratio)
.-.'•. QA17447C
Figure 80. Blvariate plots of ratios determined by appHcation of Stepwise Discriminant Analysis as the
statistically best ratios to distinguish sea-water intrusion (sold dots) from oil-field brines (open triangles).
Ratios change according to the composition of oil-field brines, derived from (a) California (data from
Gullikson and others, 1961), (b) Texas (data from Kreitler and others, 1988), and (c) Louisiana (data from
Dickey and others, 1972).
226
-------
(a)
10
_o
1
£
o>
*.
2"
a
Sf
si
CC
gj .0001
.01
(b)
"> .01-
in .001
.0001 :
(e)
-« 1000 -5
I
ii 100 -
I
I 10-
8
I
cc
CL
p
to .1
'-
A
LA
........ . ,i...-.., . .A.ftfl . .......|
.1 1 10 100
STEP RATIO #1: Mg/K (weight ratio)
A •
.1 1 10
STEP RATIO #1: Na/CI (weight ratio)
.000001 .00001 .0001 .001 .01 .1
STEP RATIO #1: Br/CI (weight ratio)
a
I
O)
I
1
3
o
cc
CL
-------
grouping. If initial grouping was done well and/or if the two groups are distinct, each individual group
member will be identified by SDA as being representative of that group. If, however, an overlap of the two
groups exists because of poor grouping and/or of chemical similarity, some samples within any
endmember may actually turn out to be more similar to the mean composition of the other group than to
the mean composition of the originally assigned group.
This feature of SDA can also be used to test individual group members of a third group (for example,
apparently contaminated ground water} regarding their chemical similarity to potential endmembers of
contamination (the two groups for which separating variables had been determined). This is illustrated with
a test case of known oil-field pollution in Illinois, where salt-water disposal into pits and indiscriminant
dumping of brine has caused local ground-water contamination (Lehr, 1969). Considering oil-field brines
from Illinois as one endmember of mixing and Illinois' ground water with chloride greater than 250 mg/L
(retrieved from STORET) as the other endmember of mixing, SDA determined that the ratios Na/CI and
Ca/CI provide the most separation power between the two groups, with Ca, Mg, Na, 804, and Cl as the
only available parameters. Rgure 82a illustrates this separation, with oil-field brines generally plotting at
lower ratios than most of the ground-water samples. Water samples from the contaminated area having Cl
>200 mg/L (data from Van Biersel, 1985, and Stafford, 1987) were identified by SDA as being more similar
to the oil-field brine endmember than to the ground-water endmember and plot within the general area of
oil-field brines (Fig. 82a). Water samples from the contaminated area having Cl <200 mg/L, in contrast,
were identified as being similar to ground water. At the same time, approximately 20 percent of all ground-
water samples were classified as being more similar to the oil-field endmember than to the mean of the
ground-water endmember (Rg. 82b). Because of this overlap of possibly oil-field related samples in the
ground-water group and oil-field brines in the second endmember, the graphical separation of the two
endmembers is not optimal and should be retried after careful examination of data within the ground-water
group, that is, data retrieved from STORET and classified for the purpose of this test case as ground water
should be scrutinized for possible inclusion of oil-field samples or of other contaminated samples. On the
other hand, these 20 percent of the samples may contain some oil-field contamination and these samples
should be reevaluated to determine whether they do show evidence of contamination. The fact that the
known contaminated samples were positively identified so that this significant overlap existed supports
the conclusion that SDA can be an effective tool for identifying sources of salinity in ground water.
Using SDA and water samples from known potential endmembers of salinization, chemical
parameters that could allow differentiation of these sources in a contaminated fresh water can be
determined prior to field investigations, possibly eliminating expensive analyses of parameters that are of
little use.
228
-------
(a)
10
meq
5
ce
Q.
Ill
-fc
O
+A
+
.01
(b)
.1 ,. . , ,; . , ..... ,. , 1 . .
STEP RATIO #1: Na/CI (meq ratio)
10
10 -
.1 -s
cc
O.
LU
.01-d
.001-
.01
10
STEP RATIO #1: Na/Ct (meq ratio)
OA17445C
Rgure 82. Bivariate plot of Na/CI ratios versus Ca/CI ratios for Illinois ground water (crosses; from STORET;
Cl >250 mg/L), Illinois oil-field brines (open scares; data from Meents and others, 1952), and ground
water contaminated by oil-field brine (open triangles; data from Stafford, 1987, and Van Biersel, 1985).
Ratios determined through the use of Stepwise Discriminant Analysis (SDA) effectively separate ground
water from brines (a). Statistical evaluation of individual analysis using SDA identified all contaminated
water samples and 20 percent of the ground-water samples (open dots) as more similar to brines than to
ground water (b).
229
-------
7. REFERENCES
AlWn, D. R., and Breummer, L. B., 1988, Minnesota ground-water quality, in Moody, D. W., Carr, Jerry,
Chase, E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrologic events and
groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 313-320.
Alderton, D. H. M., and Sheppard, S. M. F., 1977. Chemistry and origin of thermal waters from southwest
England: Institute Mining and Metallurgy, Transactions, v. (B)86, p. 191-194.
Ailing, H. L., 1928, The geology and origin of the Silurian salt of New York State: New York State Museum
Bulletin 275,139 p.
Anderson, M. P., and Berkebile, C. A., 1976, Evidence of salt-water intrusion in southeastern Long
Island: Ground Water, v. 14, no. 5, p. 315-319.
Arad, Arnon, Kafri, Uri, and Fleisher, Ezra, 1975, The Na'aman springs, northern Israel—Salination
mechanism of an irregular freshwater-seawater interface: Journal of Hydrology, v. 25, p. 81-104.
Arkansas Department of Pollution Control and Ecology, 1984, Arkansas water quality inventory report
1984: Little Rock, 495 p.
Ashworth, J. B., 1990, Evaluation of ground-water resources in El Paso County, Texas: Texas Water
Development Board Report 324, 25 p.
Atkinson, S. F., Miller, G. D., Curry, D. S., and Lee, S. B., 1986, Salt Water Intrusion: Lewis Publishers,
390 p.
Aulenbach, D. B., 1980, Litigation in the contamination of individual water supplies by storage of highway
de-icing salt: Proceedings, 5th National Ground-Water Quality Symposium, Robert S. Kerr
Environmental Research Laboratory.
Ayers, M. A., and Pustay, E. A., 1988, New Jersey ground-water quality, in Moody, D. W., Carr, Jerry,
Chase, E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrologic events and
groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 369-376.
Back, William, 1966, Hydrochemical fades and ground-water flow patterns in northern part of the Atlantic
Coastal Plain: U.S. Geological Survey Professional Paper 498-A, 42 p.
Back, William, Hanshaw, B. B., Pyle, T. E., Plummer, L. N., Weidie, A. E., 1979, Geochemical significance
of groundwater discharge and carbonate solution to the formation of Caleta Xel Ha, Quintana Roo,
Mexico: Water Resources Research, v. 15, no. 6, p. 1521-1535.
Bahls, L. L., and Milter, M. R., 1975, Ground-water seepage and its effects on saline soils: Bozeman,
Montana State University Joint Water Resources Research Center Report No. 66,39 p.
Bain, G. L., 1970, Salty ground water in West Virginia with a discussion of the Pocatalico River Basin above
Sissonville: U.S. Geological Survey Circular 11,31 p.
230
-------
Bair, E. S., and Digel, R. K., 1990, Subsurface transport of inorganic and organic solutes from
experimental road spreading of oil-field brine: Ground Water Monitoring Review, Summer 1990,
p. 94-105.
Baker, F. G., and Brendecke, C. M., 1983, Seepage from oil-field brine disposal ponds in Utah: Ground
Water, v. 21, no. 3, p. 317-324.
Ball Associates, Ltd, 1965, Surface and shallow oil-impregnated rocks and shallow oil fields in the U.S.:
U.S. Department of the Interior, Bureau of Mines, Monograph 12,375 p.
Balsters, R. G. and Anderson, C., 1979, Water quality effects associated with irrigation in Kansas: Kansas
Water News, v. 22, nos. 1 and 2, Winter 1979, p. 14-22.
Banks, H. O., and Richter, R. C., 1953, Sea-water intrusion into ground-water basins bordering the
California coast and inland bays: American Geophysical Union, Transactions, v. 34, no. 4, p. 575-
582.
Banner, J. L, Wasserburg, G. J., Dobson, P. P., Carpenter, A. B., and Moore, C. H., 1989, Isotopic and
trace element constraints on the origin and evolution of saline groundwaters from central Missouri:
Geochimica et Cosmochimica Acta, v. 53, p. 383-398.
Barksdale, H. C., 1940, The contamination of ground-water by salt water near Parlin, New Jersey:
American Geophysical Union, Transactions, v. 21, p. 471-474.
Bassett, R. L., and Bentley, M. E., 1983, Deep brine aquifers in the Palo Duro Basin—regional flow and
geochemical constraints: The University of Texas at Austin, Bureau of Economic Geology Report of
Investigations No. 130, 59 p.
Baughman, W. T., and McCarty, J. E., 1974, Water resources of Wayne County: Mississippi Geol. Econ.
Topogr. Survey Bull., no. 117, p. 241-293.
Bednar, G. A., 1988, Mississippi ground-water quality, in Moody, D. W., Carr, Jerry, Chase, E. B., and
Paulson, R. W., compilers, National Water Summary 1986—Hydrologic events and groundwater
quality: U.S. Geological Survey Water-Supply Paper 2325, p. 321-328.
Behl, Elizabeth, Davis, S. N., and Gokjwitz, Joshua, 1987, CI/Br ratios as an environmental tracer of
anthropogenically altered waters: Geological Society of America, Abstract with Programs, p. 585.
Bentley, H. W., Phillips, F. M., and Davis, S. N., 1986, Chlorine-36 in the terrestrial environment, in Fritz,
P., and Fontes, J. C., eds., Handbook of environmental isotope geochemistry, v. 2, The terrestrial
environment: Elsevier, p. 427-480.
Benz, L. C., Mickelson, R. H., Sandoval, F. M., and Carlson, C. W., 1961, Ground-water investigations in a
saline area of the Red River Valley, North Dakota: Journal of Geophysical Research, v. 66, no. 8,
p. 2435-2443.
Berg, W. A., Cail, C. R., Hungerford, D. M., Naney. J. W., and Sample, G. A., 1987, Saline seep on
wheatland in northwest Oklahoma, in Fairchikj, D. M., ed., Ground water quality and agricultural
practices: Lewis Publishers, Chelsea, Michigan, p. 265-272.
231
-------
Bingham, J. W., and Rolston, J. L, 1978, Road salt storage and road network in Connecticut: U.S.
Geological Survey Miscellaneous Reid Studies Map MF-981 -A.
Bluntzer, R. L., 1981, Investigation of soil-salinity problem at four sites west of Estelline, Hall County,
Texas: Texas Department of Water Resources, Planning and Development Division Open-File
Report 81-2, 23 p.
Bluntzer, R. L, 1982, Investigation of ground-water salinity and soil-salinity problems in the Brownfield
Lakes—Red Onion Flats depression of Terry and Lynn Counties, Texas: Texas Department of Water
Resources, Planning and Development Division Open-Rle Report 82-1,96 p.
Boggess, D. H., 1970, The magnitude and the extent of salt-water contamination in the Caloosahatchee
River between La Belle and Olga, Florida: U.S. Geological Survey, Florida Bureau of Geology
Information Circular No. 62, part 2, p. 17-39.
Boggess, D. H., 1973, The effects of plugging a deep artesian well on the concentration of chloride in
water in the water-table aquifer at Highland Estates, Lee County, Florida: U.S. Geological Survey
Open-File Report No. 73003.
Bolke, E. L., and Price, Don, 1969, Hydrotogic Reconnaissance of Curlew Valley, Utah and Idaho: Utah
Department of Natural Resources Technical Publication No. 25,40 p.
Boyd, F. M., and Kreitler, C. W., 1986, Hydrogeology of a gypsum playa, northern salt basin, Texas: The
University of Texas at Austin, Bureau of Economic Geology, Report of Investigations No. 158,37 p.
Breen, K. J., Angeto, C. G., Masters, R. W., and Sedam, A. C., 1985, Chemical and isotopic characteristics
of brines from three oil- and gas-producing sandstones in eastern Ohio, with applications to the
geochemical tracing of brine sources: U.S. Geological Survey Water-Resources Investigations
Report 84-4314, 58 p.
Brennan, Robert, 1956, Sea-water intrusion in California—Preliminary chemical-quality study in the
Manhattan Beach area, California: California Department of Water Resources, Bulletin No. 63,
Appendix E, 34 p.
Brown, H. T., 1984, A case study of a salinity control program, Uirrta Basin, Utah, in French, R. H., ed.,
Salinity In watercourses and reservoirs, Proceedings of the 1983 International Symposium on State-
of-the-Art Control of Salinity, Salt Lake City, Utah: Butterworth Publishers, Boston, p. 275-284.
Brown, E. B., Skougstad, M. W., and Fishman, M. J., 1970, Methods for collection and analysis of water
samples for dissolved minerals and gases: Techniques of water-resources investigations of the U.S.
Geological Survey, Book 5, Chapter A1,130 p.
Bruington, A. E., 1972, Salt-water intrusion into aquifers: Water Resources Bulletin, v. 8, no. 1, p. 150-
160.
Bubeck, R. C., Diment, W. H., Deck, B. L., Baldwin, A. L., and Lipton, S. D., 1971, Runoff of deicing salt-
effect on Irondequoit Bay, Rochester, New York: Science, v. 175, p. 1128-1132.
232
-------
Buetner, E. C., and Charles, E. G., 1982, Large volume toss during cleavage formation, Hamburg
sequence, Pennsylvania: Geol. 13, p, 803-805
Burnitt, S. C., 1963, Reconnaissance of soil damage and ground-water quality, Fisher County, Texas:
Texas Water Commission, Memorandum Report No. 63-02,45 p.
Burnitt, S. C., Adams, J. B., Rucker, O. C., and Port erf ield, H. C., 1963, Effects of surface-disposal of oil-
field brine on the quality and development of ground water in the Ogallala Formation, High Plains of
Texas, and a hydrologic study of brine disposal into surface pits, High Plains of Texas: Texas Water
Commission, 115 p.
Cagle, J. W., 1969, Availability of grourtdwater in Wayne County, Iowa: Iowa Geological Survey Water Atlas
No. 3,33 p.
Canter, L. W., 1987, Nitrates and pesticides in ground water: an analysis of a computer-based literature
search, in Fairchild, D. M., ed., Ground water quality and agricultural practices, Lewis Publishers,
Chelsea, Michigan, p. 153-174.
Carlston, C. W., and Graeff, G. D., Jr., 1955, Ground-water resources of the Ohio River Valley in West
Virginia: West Virginia Geological and Economic Survey, v. XXII, part III.
Carothers, W. W., and Kharaka, Y. K., 1978, Aliphatic acid anions in oil-field waters—implications for origin
of natural gas: American Association of Petroleum Geologists Bulletin, v. 62, no. 12, p. 2441-2453.
Carpenter, A. B., and Darr, J. M., 1978, Relationship between groundwater resources and energy
production in southwestern Missouri: Rolla, Missouri Water Resources Research Center,
Completion Report, 113 p.
Chandler, R. V., Moore, J. D., and Gillett, B., 1985, Ground-Water Chemistry and salt-water encroachment,
Southern Baldwin County, Alabama: Geological Survey of Alabama, Water Resources Division,
Bulletin 126.
Chaudhuri, S., 1978, Strontium isotopic composition of several oil-field brines from Kansas and Colorado:
Geochimica et Cosmochimica Acta, v. 42, p. 329-331.
Childs, K. E., 1970, History of the salt, brine and paper industries and their probable effect on the ground
water quality in the Manistee Lake area of Michigan: Hydrotogical Survey Division, Water Resources
Commission, Bureau of Water Management, Michigan Department of Natural Resources, 69 p.
Clarke, J. S., and McConnell, J. B., 1988, Georgia ground-water quality, in Moody, D. W., Carr, Jerry,
Chase, E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrologic events and
groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 215-222.
Clayton, R. N., Friedman, I., Graf, D. L, Mayeda, T. K., Meents, W. F., and Shimp, N. F.,1966, The origin of
saline formation waters: I. Isotopic composition, Journal of Geophysical Research 71, p. 869-3889
Collins, A. G., 1967, Geochemistry of some Tertiary and Cretaceous age oil-bearing formation waters:
Environmental Science and Technology, v. 1, no. 9, p. 725-730.
233
-------
Collins, A. G., 1969, Chemistry of some Anadarko Basin brines containing high concentrations of iodide:
Chemical Geology, v. 4, no. 1-2, p. 169-187.
Collins, A. G., 1974, Saline ground waters produced with oil and gas: U.S. Environmental Protection
Agency, Office of Research and Development, Technology Series EPA-660/2-74-010, 68 p.
Cooper, H. H., Jr., Kohout, F. A., Henry, H. R., and Gtorer, R. E., 1964, Sea water in coastal aquifers,
relation of salt water to fresh water: U.S. Geological Survey, Water-Supply Paper 1613C, 84 p.
Core Laboratories, Inc., 1972, Chemical analyses of saline water: Texas Wafer Development Board,
Report 157, V. 2, 378 p.
Cotecchia, V., Tazioll, G. S., Magri, G. 1974, Isotropic measurements in research on seawater ingression in
the carbonate aquifer of the Salentine Peninsula, Southern Italy, Isotope Techniques in
Groundwater Hydrology 1974: International Atomic Energy Agency, Vienna, p. 445-463.
Cotton, J. E., and ButterfiekJ, D., 1988, Vermont ground-water quality, in Moody, D. W., Carr, Jerry,
Chase, E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrotogic events and
groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 501-508.
Craig, Harmon, 1961, Isotopic variations in meteoric waters: Science, v. 133, no. 3465, p. 1702-1703.
Craig, J. R., 1970, Saline Waters—Genesis and relationship to sediments and host rocks, in Mattox, R. B.,
ed., Ground-water salinity: Symposium, 46th Annual Meeting of the Southwestern and Rocky
Mountain Division of the American Association of the Advancement of Science, April 23-24, Las
Vegas, New Mexico, p. 3-29.
Crain, L. J., 1966, Ground-water resources of the Jamestown Area, New York: State of New York
Conservation Department, Water Resources Commission Bulletin No. 58.
Crain, L. J., 1969, Ground-water pollution from natural gas and oil production in New York: U.S. Geological
Survey Report of Investigation No. 5,15 p.
Custer, S. G., 1979, Regional identification of saline-seep recharge areas, Stillwater and Yellowstone
Counties, Montana: Montana Water Resources Research Center Report No. 102,58 p.
Custodio, E., 1987, Hydrogeochemistry and tracers, in Custodio, E., Ground-water problems in coastal
areas: Studies and reports in hydrology no. 45, UNESCO, p. 213-269.
Davis M. E., and Gordon, J. D., 1970, Records of water levels and chemical analysis from selected wells in
parts of the Trans-Pecos region, Texas, 1965-68: Texas Water Development Board Report 114,
49 p.
Denver, J. M., 1988, Delaware ground-water quality, in Moody, D. W., Carr, Jerry, Chase, E. B., and
Paulson, R. W., compilers, National Water Summary 1986—Hydrologic events and groundwater
quality: U.S. Geological Survey Water-Supply Paper 2325, p. 199-204.
DeSitter, L. U., 1947, Diagenesis of oil-field brines: American Association of Petroleum Geologists
Bulletin, v. 31, p. 2030-2040.
234
-------
Deverel, S. J., and Gallanthine, S. K., 1989, Relation of salinity and selenium in shallow ground water to
hydrologic and geochemical processes, western San Joaquin Valley, California: Journal of
Hydrology, v. 109, p. 125-149.
Dickey, P. A., Collins, A. Q., and Fajardo, Ivan, 1972, Chemical composition of deep formation waters in
Southwestern Louisiana; American Association of Petroleum Geologists Bulletin, v. 56, no. 8,
p. 1530-1533.
Diment, W. H., Bubeck, R. C., and Deck, B. L, 1973, Effects of deicing salt on the waters of the
Irondequoit Bay drainage basin, Monroe County, New York, in Coogan, A. H., ed., Fourth
Symposium on Salt: Northern Ohio Geological Society, v. 2., p. 391-405.
Dion, N. P., and Sumioka, S. S., 1984, Sea-water intrusion into coastal aquifer in Washington, 1978: U.S.
Geological Survey, Water Resources Division, Water-Supply Bulletin, v. 56,13 p.
Dixon, W. J., Brown, M. B., Engelman, L, Fme, J. W., Hill, M. A., Jennrich, R. I., and Toporek, J. D., 1981,
BMDP statistical software: Berkeley, University of California Press, 725 p.
Doneen, L. D., 1966, Factors contributing to the quality of agricultural waste waters in California, in
Doneen, L. D., ed., Symposium on Agricultural Waste Waters: University of California, Davis, Water
Resources Center Report 10, p. 6-9.
Donovan, J. J., Sonderegger, J. L., and Miller, M. R., 1981, Investigations of soluble salt loads, controlling
mineralogy, and factors affecting the rates and amounts of leached salts: Montana Bureau of Mines,
Water Resources Research Center Report No. 120,61 p.
Drever, J. I., 1982, The geochemistry of natural waters: Prentice-Hall, Inc., Englewood Cliffs, New Jersey,
388 p.
Duce, R. A., Winchester, J. W., and Van Nahl, T. W., 1965, Iodine, bromine, and chlorine in the Hawaiian
marine atmosphere: Journal of Geophysical Research, v. 70, no. 8, p. 1775-1799.
Dunrud, C. R., and Nevins, B. B., 1981, Solution mining and subsidence in evaporite rocks in the United
States: U.S. Geological Survey Miscellaneous Investigations Series Map 1298.
Dutton, A. R., Fisher, B. C., Richter, B. C., and Smith, D. A., 1985, Hydrologic testing in the salt-
dissolution zone of the Palo Duro Basin, Texas Panhandle: The University of Texas at Austin,
Bureau of Economic Geology Open-File Report OF-WTWI-1985-3,15 p.
Dutton, A. R., Richter, B. C., and Kreitler, C. W., 1989, Brine discharge and salinization, Concho River
watershed, West Texas: Ground Water, v. 27, no. 3, p 375-383.
Dutton, A. R., and Richter, B. C., 1990, Regional geohydrology of the Gulf Coast Aquifer in Matagorda
and Wharton Counties, Texas: Development of a numerical model to estimate the impact of water-
management strategies: The University of Texas at Austin, Bureau of Economic Geology, Report
prepared for Lower Colorado River Authority under Contract Number IAC (88-89)0910,116 p.
235
-------
Edelmann, Patrick, and Buckles, D. R. 1984, Quality of ground water in agricultural areas of the San Luis
Valley, south central Colorado: U.S Geological Survey Water-Resources Investigations Report 83-
4281, 37 p.
Edmunds, W. M., Andrews, J. N., Burgess, W. G., Kay, R. L. F., and Lee, D. J., 1984, The evolution of
saline and thermal groundwaters in the Cammenellis granite: Mineralogical Magazine, v. 48, p. 407-
424.
Edmunds, W. M., Kay, R. L. P., and McCartney, R. A., 1985, Origin of saline groundwaters in the
Cammenellis granite—natural processes and reaction during Hot Dry Rock reservoir circulation:
Chemical Geology, v. 49, p. 287-301.
Edmunds, W. M., Kay, R. L. P., Miles, D. L., and Cook, J. M., 1987, The origin of saline ground waters in
the Cammenellis Granite, Cornwall (U.K.)—further evidence from minor and trace elements, in Fritz,
P., and Frape, S. K., eds., Saline water and gases in crystalline rocks: Geological Association of
Canada Special Paper 33, p. 127-143.
Effertz, R. J., Sidebottom, W. K., and Turiey, M. D., 1984, Measures for reducing return flows from the
Wellton-Mohawk irrigation and drainage district, in French, R. H., ed., Salinity in watercourses and
reservoirs, Proceedings of the 1983 International Symposium on State-of-the-Art Control of Salinity,
Salt Lake City, Utah: Butterworth Publishers, Boston, p. 305-314.
Engberg, R. A., and Drub'ner A. D., 1988, Nebraska ground-water quality, in Moody, 0. W., Carr, Jerry,
Chase, E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrotogic events and
groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 347-354.
Exner, M. E. and Spalding, R. P., 1979, Evaluation of contaminated groundwater in Holt County,
Nebraska: Water Resources Research, v. 15, no. 1, p. 137-147.
FairchikJ, H. L., 1935, Genesee Valley hydrography and drainage: Proceedings of the Rochester
Academy of Science, v. 7, p. 157-188.
Farrar, C. D., and Bertokf, G. L., 1988, Region 4, Central Valley and Pacific Coast Ranges, in Back, William,
Rosendheim, J. S., and Seaber, P. R., eds., Hydrogeology: Geological Society of America, The
Geology of North America, v. O-2, p. 59-67.
Feast, C. P., 1984, Meeker dome salinity investigation, in French R. H., ed., Salinity in watercourses and
reservoirs: Butterworth Publishers, Boston, p. 359-365.
Feltz, H. R., and Hanshaw, B. B., 1963, Preparation of water sample for carbon-14 dating: U.S. Geological
Survey, Geological Survey Circular 480, p. 1-3.
Ferrelra, R. F., Cannon, M. R., Davis, R. E., Shewman, F., and Arrigo, J. L., 1988, Montana ground-water
quality, in Moody, D. W., Carr, Jerry, Chase, E. B., and Paulson, R. W., compilers, National Water
Summary 1986—Hydrologic events and groundwater quality: U.S. Geological Survey Water-Supply
Paper 2325, p. 165-172.
236
-------
Feth, J. H., and others, 1965, Preliminary map of the conterminous United States showing depth to and
quality of shallowest ground water containing more than 1,000 parts per million dissolved solids: U.S.
Geological Survey Hydrotogic Atlas HA-199, 2 sheets.
Field, Richard, Struzeski, E. J., Jr., Masters, H. E., and Tafuri, A. N., 1973, Water pollution and associated
effects from street salting: U.S. Environmental Protection Agency, National Environmental Research
Center Report EPA-R2-73-257, 48 p.
Fink, B. E., 1965, Investigation of Ground-and Surface-Water Contamination Near Harrold, Willbarger
County, Texas, Texas Water Commission Report LD-0365, 23 p.
Fireman, M., and Hayward, H. E., 1955, Irrigation water and saline and alkali soils, in U.S. Department of
Agriculture, Water, The Yearbook of Agriculture, 1955, p. 321.
Fogg, G. E., 1980, Salinity of Formation Waters, in Kreitler, C. W., and others, Geology and Geohydrology
of the East Texas Basin: The University of Texas at Austin, Bureau of Economic Geology Geological
Circular 80-12, p. 68-72.
Fogg, G. E., Kreitler, C. W., and Dutton, S. P., 1980, Hydrotogic Stability of Oakwood Dome, in Kreitler,
C. W., and others, Geology and Geohydrology of the East Texas Basin: The University of Texas at
Austin, Bureau of Economic Geology Geological .Circular 80-12, p. 30-31.
Fogg, G. E., and Kreitler, C. W., 1980, Impacts of Salt-Brining on Palestine Salt Dome—Geology and
Geohydrolgy of the East Texas Basin: The University of Texas at Austin, Bureau of Economic
Geology Geological Circular 80-12, p. 46-54.
Fogg, G. E., and Kreitler, C. W., 1982, Ground-water hydraulics and hydrochemical fades in Eocene
aquifers of the East Texas Basin: The University of Texas at Austin, Bureau of Economic Geology
Report of Investigations 127, 75 p.
Food and Agriculture Organization of the United Nations, 1973, Irrigation, drainage and salinity:
Hutchinson & Co (publishers) LTD, London, 510 p.
Framji, K. K., 1976, Irrigation and salinity—a world-wide survey: International Commission on irrigation and
drainage, New Delhi, India, 106 p.
Frape, S. K., and Fritz, P., 1982, The chemistry and isotopic composition of saline groundwaters from the
Sudbury Basin, Ontario: Canadian Journal of Earth Sciences, v. 19, p. 645-661.
Frape, S. K., and Fritz, P., 1987, Geochemical trends for ground waters from the Canadian Shield, in Fritz,
P., and Frape, S. K., eds., Saline water and gases in crystalline rocks: Geological Association of
Canada Special Paper 33, p. 19-38. - ,
Frape, S. K., Fritz, P., and McNutt, R. H., 1984, The role of water-rock interaction in the chemical
evolution of groundwaters from the Canadian Shield: Geochimica et Cosmochimica Acta, v. 48,
p. 1617-1627.
Freeze, R. A., and Cherry, J. A., 1979, Groundwater: Prentice-Hall, Inc., Englewood Cliffs, New Jersey,
604 p.
237
-------
Fritz, P., and Frape, S. K., 1982, Saline groundwaters in the Canadian Shield—a first overview: Chemical
Geology, v. 36, p. 179-190.
Fryberger, J. S., 1972, Rehabilitation of a brine-polluted aquifer: U.S. Environmental Protection Agency,
Office of Research and Monitoring, Technology Series Report EPA-R2-72-014, 61 p.
Fuhriman, D. K., and Barton, J. R., 1971, Ground-water pollution in Arizona, California, Nevada, and Utah:
U.S. Environmental Protection Agency, Office of Research and Monitoring, Water Pollution Control
Research Series Report No. 16060 ERU 12/71, 249 p.
Gamer, Stuart, 1988, Making the most of field-measurable ground-water quality parameters: Ground Water
Monitoring Review, v. 8, no. 3, p. 60-66.
Garreis, R. M., and MacKenzie, F. T., 1967, Origin of the chemical compositions of some springs and
lakes, in Gould, R. F., ed., Equilibrium concepts in natural water systems: American Chemical
Society, Advances in Chemistry Series 67, p. 222-242.
Garreis, R. M., and MacKenzie, F. T., 1971, Evolution of Sedimentary Rocks: Norton, New York, p. 272
Garza, Sergio, and Wesselman, J. B., 1959, Geology and ground-water resources of Winkler County,
Texas: Texas Board of Water Engineers Bulletin 5916,200 p.
Gascoyne, M., Davison, C. C., Ross, J. D., and Pearson, R., 1987, Saline ground waters and brines in
plutons in the Canadian Shield, in Fritz, P., and Frape, S. K., eds., Saline water and gases in
crystalline rocks: Geological Association of Canada Special Paper 33, p. 53-68.
Gass, T. E., Lehr, J. H., and Heiss, H. W., Jr., 1977, Impact of abandoned wells on ground water: U.S.
Environmental Protection Agency, Office of Research and Development, Robert S. Kerr
Environmental Research Laboratory, Ecological Research Series EPA-600/3-77-095, 52 p.
Geraghty, Miller, van der Leeden, and Troise, 1973, Water Atlas of the United States: Water Information
Center, Port Washington, New York.
Gillospie, J. B., and Hargadine, G. D., 1981, Saline ground-water discharge to the Smoky Hill River
between Salina and Abilene, Central Kansas: U.S. Geological Survey Water-Resources
Investigations No. 81-43,72 p.
Gogel, Tony, 1981, Discharge of salt water from Permian rocks to major stream-aquifer systems in Central
Kansas: Kansas Geological Survey, The University of Kansas at Lawrence, Chemical Quality
Series 9, 60 p.
Goldberg, E. D., Broecker, W. S., Gross, M. G., and Turekian, K. K., 1971, Marine chemistry, in
Radioactivity in the marine environment: Washington, D.C., National Academy of Sciences, p. 137-
146.D
Grigsby, C. O., Tester, J. W., Trujilto, P. E., Counce, D. A., Abbott, J., Holley, C. E., and Blatz, L A., 1983,
Rock-water interactions in hot dry rock geothermal systems: field investigations of in situ
geochemical behavior: Journal of Voteanotogy and Geothermal Research, v. 15, p. 101-136.
238
-------
Gullikson, D. M., Caraway, W. H., and Gates, G. L., 1961, Chemical analysis and electrical resistivity of
selected California oil-field waters: U.S. Department of the Interior, Bureau of Mines Report of
Investigations No. 5736, 21 p.
Gustavson, T. C., 1979, Salt Dissolution, in Dutton, S. P., and others, eds., Geology and geohydrotogy of
the Palo Duro Basin, Texas Panhandle, A report on the progress of nuclear waste isolation feasibility
studies (1978): The University of Texas at Austin, Bureau of Economic Geology, Geological Circular
79-1, p. 87-99.
Gustavson, T. C., Finley, R. J,, and McGillis, K. A., 1980, Regional dissolution of Permian salt in the
Anadarko, Dalhart, and Palo Duro Basins of the Texas Panhandle; The University of Texas at Austin,
Bureau of Economic Geology Report of Investigations No. 106,40 p.
Gustavson, T. C., Holliday, V. T, and Schultz, G. E., 1990, Introduction, in Gustavson, T. C., ed., Tertiary
and Quaternary stratigraphy and vertebrate paleontology of parts of northwestern Texas and eastern
New Mexico: The University of Texas at Austin, Bureau of Economic Geology Guidebook 24, p. 1-
21.
Hale, Allan, 1973, Salt water vapor discharges and the environment, in Coogan, A. H., ed., Fourth
Symposium on Salt: Northern Ohio Geological Society, v. 2., p. 387-389.
Hamilton, P. A., and Denver, J. M., 1990, Effects of land use and ground-water flow on shallow ground-
water quality, Delmarva Peninsula, Delaware, Maryland, and Virginia: Ground Water, v. 28, no. 5,
p. 789.
Hamlin, H. S., Smith, D. A., and Akhter, M. S., 1988, Hydrogeology of Barbers Hill salt dome, Texas
Coastal Plain: The University of Texas at Austin, Bureau of Economic Geology Report of
Investigations No. 176, 41 p.
Hanor, J. S., 1983, Fifty years of development of thought on the origin and evolution of subsurface
sedimentary brines, in Boardman, S. J., ed., Revolution in the earth sciences—Advances in the past
half-century: Dubuque, Kendall/Hunt, p. 99-111.
Hanor, J. S., and Workman, A. L., 1986, Distribution of dissolved volatile fatty acids in some Louisiana oil
field brines, in F. E. Ingerson Festschrift; Part 1, Applied Geochemistry, v. 1, no. 1, p. 37-46.
Hanshaw, B. B., Back, William, and Deike, R. G., 1971, A geochemical hypothesis for dolomitization by
ground water: Economic Geology, v. 66, p. 710-724.
Hanshaw, B. B., Back, William, Rubin, M., and Wait, R., 1965, Relation of Carbon-14 Concentrations to
Saline Water Contamination of Coastal Aquifers: Water Resources Research, v. 1, p. 109-114.
Haque, S. M., 1989, Environmental effects of surface brine disposal in Golden Meadow Oil and Gas field
. i '
area, Lafourche Parish, Louisiana [abs]: Geological Society of America, 1989 Annual Meeting,
, Abstracts with Programs, p. A187.
Hardie, L. A., and Eugster, H. P., 1970, The evolution of closed basin brines: Mineralogical Society of
America Special Publication, Nq. 3, p. 273-290,
239
-------
Hargadine, G. D., Balsters, R. G., and Luehring, Joann, 1979, Mineral intrusion in Kansas surface waters—
a technical report: Kansas Water Office, Topeka, Kansas, 211 p.
Harrel, R. C., 1975, Water quality and salt-water intrusion in the Lower Neches River: Texas Journal of
Science, v. 26, nos. 1-2, p. 107-117.
Havens, J. S., and Wilkins, D. W., 1979, Experimental saUnity alleviation at Malaga Bend of the Pecos
River, Eddy County, New Mexico: U.S. Geological Survey Water-Resources Investigations 80-4,
65 p.
Hawkins, D. B., and Motyka, R. J., 1984, A multivariate statistical analysis and chemical mass balance
analysis of waters of the Copper River Basin, Alaska, in Hitchon, Brian, and Wallick, E. I., eds.,
Proceedings, First Canadian/American Conference on Hydrogeology—Practical applications of
ground-water geochemistry: National Water Well Association, Worthington, Ohio, p. 238-245.
Hearne, G. A., Undner-Lunsford, J., Cain, D., Watts, K. R., Robson, S. G., Tobin, R. L, Teller, R. W.,
Schneider, P. A., and Gearrtart, M. J., 1988, Colorado ground-water quality, in Moody, D. W., Carr,
Jerry, Chase, E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrologic
events and groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 181-190.
Helgeson, H. C., 1972, Chemical interaction of feldspars and aqueous solutions: in Feldspars,
MacKenzie, W. S., and Zussman, J., Procedings, NATO Advanced Study Institute, Manchester
University Press, p. 187-217.
Helgeson, J. O., 1990, Effects of agricultural and petroleum-production land use on water quality in the
High Plains Aquifer, south-central Kansas: Ground Water, v. 28, no. 5, p. 795.
Hem, J. D., 1985, Study and interpretation of the chemical characteristics of natural water: U.S. Geological
Survey Water-Supply Paper 2254, 263 p.
Hingston, F. J. and Gailitis, V., 1976: The Geographic Variation of Salt Precipitated over Western Australia:
Australian Journal of Soil Resources, v. 14, p. 319-335.
Hiss, W. L., 1970, Acquisition and machine processing of saline water data from Southeastern New
Mexico and Western Texas: U.S. Geological Survey, Water Resources Research, v. 6, no. 5,
p. 1471-1477.
Hitchon, Brian, Levinson, A. A., and Reeder, S. W., 1969, Regional variations of river water composition
resulting from halite solution, Mackenzie River Drainage Basin, Canada: Water Resources Research,
v. 5, no. 6, p. 1395-1403.
Hitchon, Brian, 1984, Graphical and statistical treatment of standard formation water analyses, in Hitchon,
Brian, and Wallick, E. I., eds., Proceedings, First Canadian/American Conference on
Hydrogeology—Practical applications of ground-water geochemistry: National Water Well
Association, Worthington, Ohio, p. 238-245.
Holland, T. W., and Ludwig, A. H., 1981, Use of water in Arkansas, 1980: Arkansas Geological
Commission Water Resources Summary No. 14,30 p.
240
-------
Hopkins, H. I., 1963, The effect of oil-field brines on the potable ground water in the Upper Big Pitman
Creek Basin, Kentucky: Kentucky Geological Survey, Series X, Report of Investigations No. 4,36 p.
Hoskins, H. A., 1947, Analyses of West Virginia brines: West Virginia Geological and Economic Survey,
Report of Investigations No. 1, 22 p.
Howard, K. W. P., and Lloyd, J. W., 1983, Major ion characterization of coastal saline ground waters:
Ground Water, v. 21, no. 4, p. 429-437.
Hubbert, M. K., 1940, The theory of ground-water motion, Journal of Geology, v. 48, p. 785-944.
Hunt, C. B., 1960, The Death Valley salt pan, a study of evaporites: U.S. Geological Survey Professional
Paper 400-B, p. B456-B457; '
Hunt, C. B., Robinson, T. W., Bowles, W. A., and Washbum, A. L, 1966, Hydrologic basin, Death Valley,
CaUfomia: U.S. Geological Survey Professional Paper 494-B, 138 p.
Hutchinson, F. E., 1973, Effect of deicing salts applied to highways on the contiguous environment, in
Coogan, A. H., ed., Fourth Symposium on Salt: Northern Ohio Geological Society, v. 1., p. 427-
434.
Ikeda, Kiyoji, 1967, Study on salt water intrusion into ground water—part 2. Geochemical research of salt
in groundwater: Geological Survey of Japan, v. 18, no. 6, p. 15(393)-33(411).
International Atomic Energy Agency, 1981, Sampling of water for 14C analysis: Isotope Hydrology
Laboratory, Vienna.
Irwin, G. A., Bonds* J. L., 1988, Florida ground-water quality, in Moody, D. W., Carr, Jerry, Chase, E. B.,
and Paulson, R. W., compilers, National Water Summary 1986—Hydrologic events and groundwater
quality: U.S. Geological Survey Water-Supply Paper 2325, p. 205-214.
Jacks, G., 1973, Chemistry of some ground waters in igneous rocks: Nordic Hydrology, v. 4, no. 4,
p. 207-236.
Jensen, E. G., 1978, The Paradox Valley unit—The problem and the proposed solution: EOS, American
Geophysical Union Transactions, v. 59, no. 12, p. 1064.
Johnson, B. A., and Riley, J. A., 1984, Price/San Rafael River Basin salinity investigation, in French, R. H.,
ed., Salinity in watercourses and reservoirs, Proceedings of the 1983 International Symposium on
State-of-the-Art Control of Salinity, Salt Lake City, Utah: Butterworth Publishers, Boston, p. 407-
416.
Johnson, K. S., Brokaw, A. L, Gilbert, J. F,., Saberian, A., Snow, R. H., and Walters, R. F., 1977, Summary
report on salt dissolution review meeting, March 29-30,1977: Union Carbide Corporation, Nuclear
Division, Office of Waste Isolation, Report Y/OWI/TM-31,10 p.
Johnston, R. H., and Miller, J. A., 1988, Region 24, Southeastern United States, in Back, William,
Rosendheim, J. S., and Seaber, P. R., eds., Hydrogeology: Geological Society of America, The
Geology of North America, v. O-2, p. 229-236.
241
-------
Jones, B. P., 1965, Geochemlcal evolution of closed basin water in the Western Great Basin, in Rau, J. L,
ed., Second symposium on Salt: Northern Ohio Geological Society, v. 1, p. 181-200.
Jones, B. P., Vandenburgh, A. S., Truesdell, A. H., and Rettig, S. L., 1969, Interstitial brines in playa
sediments: Chemical Geology, v. 4., p. 253-262.
Jones, P. H., and Hutchon, H., 1983, Road salt in the environment, in Schreiber, B. C., and Harner, H. L.,
eds., Sixth international symposium on salt, v. 2: Salt Institute, Alexandria, Virginia, p. 615-619.
Jonez, A. R., 1984, Controlling salinity in the Colorado River Basin, the arid West, in French, R. H., ed.,
Salinity in watercourses and reservoirs, Proceedings of the 1983 International Symposium on State-
of-the-Art"Control of Salinity, Salt Lake City, Utah: Butterworth Publishers, Boston, p. 337-347.
Jorgensen, D. G., 1968, An aquifer test used to investigate a quality of water anomaly: Ground Water, v. 6,
no. 6, p. 18-20.
Jorgensen, D. G., 1977, Salt-water encroachment in aquifers near the Houston Ship Channel, Texas:
U.S. Geological Survey Open-Pile Report 76-781,45 p.
Jorgensen, D. G., 1981, Geohydrologic models of the Houston District, Texas: Ground Water, v. 19,
no. 4, p. 418-428.
Kafri, U., and Arad, A., 1979, Current subsurface intrusion of Mediterranean seawater—a possible source
of groundwater salinity in the rift-valley system, Israel: Journal of Hydrology, v. 44, p. 267-287.
Kaiser, W. R., Swartz, T. E., and Hawkins, G. J., 1991, Hydrology of the Fruitland Formation, San Juan
Basin, in Ayers, W. B., and others, eds., Geologic and hydrotogic controls on the occurrence and
produdbility of coalbed methane, Fruitland Formation, San Juan Basin: The University of Texas at
Austin, Bureau of Economic Geology, topical report prepared for the Gas Research Institute under
Contract No. 5087-214-1544, (GRI-91/0072), p. 195-241.
Kalka, M. S., Eckstein, Yoram, and Dahl, P. S., 1989, Investigation of the fate of heavy metals of oil field
brines used as an anti-dust agent on roads in northeastern Ohio [abs.]: Geological Society of
America, Abstracts with Programs, p. A243.
Kalkhoff, S. J., 1986, Brine contamination of fresh-water aquifers and streams in petroleum producing
areas in Mississippi: U.S. Geological Survey Water-Resources Investigations Report 85-4117,106 p.
Kantrowitz, I. H., 1970, Ground-water resources in the Eastern Oswego River Basin, New York: State of
New York Conservation Department, Water Resources Commission Basin Planning Report ORB-2.
Kelly, T. E., 1974, Reconnaissance investigation of ground water in the Rio Grande drainage basin—with
special emphasis on saline ground-water resources: Hydrologic Investigation Atlas HA-510,
4 sheets.
Kimball, B. A., 1987, Geochemical indicators used to determine source of saline water in Mesozoic
aquifers, Montezuma Canyon area. Utah: U.S. Geological Survey Water-Supply Paper 2340,
242
-------
Kimmel, G. E, 1963, Contamination of ground water by sea-water intrusion along Puget Sound,
Washington, an area having abundant precipitation: U.S. Geological Survey Professional Paper 475-
B, p. B182-8185.
Kister, L R., Brown, S. G., Schumann, H. H., and Johnson, P. W., 1966, Maps showing fluoride content
and salinity of ground water in the Wilcox Basin, Graham and Cochise Counties, Arizona: U.S.
Geological Survey Hydrotogic Investigations Atlas HA-214, scale 1:250,000, 2 sheets.
Knuth, Martin, Jackson, J. L., and Whittemore, D. O., 1990, An integrated approach to identifying the
salinity source contaminating a ground-water supply: Groundwater, v. 28, no. 2, p. 207-214.
Kohout, F. A., 1960, Cyclic flow of salt water in the Biscayne Aquifer of Southeastern Florida:'Journal oY"
Geophysical Research, v. 65, no. 7, p. 2133-2141.
Kreitler, C. W., 1979, Nitrogen-isotope ratio studies of soils and groundwater nitrate from alluvial fan
aquifers in Texas: Journal of Hydrology, v. 42, p. 147-170.
Kreitler, C. W., 1989, Hydrogeotogy of sedimentary basins: Journal of Hydrology, v. 106, p. 29-53.
Kreitler, C. W., and Jones, D. C., 1975, Natural soil nitrate—The cause of the nitrate contamination of
ground water in Runnels County, Texas: Ground Water, v. 13, no. 1, p. 53-61.
Kreitler, C. W., Ragone, S. E, and Katz, G. B., 1978, N /N Ratios of ground-water nitrate, Long Island,
New York: Ground Water, v. 16, no. 6, p. 404-409.
Kreitler, C. W., Chapman, J. B., and Knauth, L. P., 1984, Chemical and isotopical composition of waters
from the Salina Ometepec, Baja California: The University of Texas at Austin, Bureau of Economic
Geology Open-File Report OF-WTWI-1984-41,13 p.
Kreitler, C. W., Akhter, M. S., Donnelly, A. C. A., and Wood, W. T., 1988, Hydrogeology of formations
used for deep-well injection, Texas Gulf Coast: The University of Texas at Austin, Bureau of
Economic Geology, report prepared for the U.S. Environmental Protection Agency under
Cooperative Agreement ID No. CR812786-01-0, 204 p.
Kreitler, C. W., Akhter, M. S., and Donnelly, A. C. A, 1990, Hydrotogic-hydrochemical characterization of
Texas Frio Formation that is used for deep-well injection of chemical wastes: Journal of
Environmental Geology and Water Sciences, v. 16, no. 2, p. 107-120.
Krieger, R. A., Hatchett, J. L, and Poote, J. L., 1957, Preliminary survey of the saline-water resources of
the United States: U.S. Geological Survey Water-Supply Paper 1374,172 p.
Krieger, R. A., and Hendrickson, G. E., 1960, Effects of Greensburg oil-field brines on the streams, wells,
and springs of the Upper Green River: Kentucky Geological Survey, Series X, Report of
Investigations No. 2,36 p.
Kuhlmann, Mark, 1968, A study of chloride contamination in the Olentangy River in the vicinity of
Columbus, Ohio: Unpublished Senior Thesis, Ohio State University, 26 p.
Kunkle, S. H. 1971, Effect of road salt on a Vermont Stream: Proceedings: Street Salting—Urban Water
Quality Workshop, S.U.N.Y.: College of Forestry, Syracuse University, p. 48-61
243
-------
Lamb, C. E., and Woodward, Ft., 1988, California ground-water quality, in Moody, 0. W., Carr, Jerry,
Chase, E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrotogic events and
groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 173-180.
Land, L. S., 1987, The major ion chemistry of saline brines in sedimentary basins, in Banavar, J. R., Koplik,
J., and Winkler, K. W., eds., Physics and Chemistry of Porous Media II: Ridgefield, Connecticut,
American Institute of Physics Conference Proceedings 154, p. 160-179.
Land, L. S., and Prezbindowski, D. R., 1981, The origin and evolution of saline formation water, Lower
Cretaceous Carbonates, South-Central Texas: Journal of Hydrology, v. 54, p. 51-74.
Land, L. S., and Prezbindowski, D. R., 1985, Chemical constraints and origins of four groups of Gulf Coast
reservoir fluids: American Association of Petroleum Geologists Bulletin, v. 69, p. 119-126.
Larson, J. D., 1981, Distribution of salt water in the coastal plain aquifers of Virginia: U.S. Geological
Survey Open-File Report 81-1013, 25 p.
Law, J. P., Jr., Davidson, J. M., and Reed, L. W., 1970, Degradation of water quality in irrigation return
flows: Oklahoma Agricultural Experiment Station, Bulletin 684, p. 3-26.
Lehr, J. H., 1969, A study of ground-water contamination due to saline water disposal in the Morrow
County oil field: Research Project Completion Report Project No. A-004-OHIO, submitted to Office
of Water Resources Research, U.S. Department of the Interior, 81 p.
Leonard, R. B., 1964, A method for evaluating oil-field-brine pollution of the Walnut River in Kansas: U.S.
Geological Survey Professional Paper 501 -B, p. B173-8176.
Leonard, A. R., and Ward, P. E., 1962, Use of Na/CI ratios to distinguish oil-field from salt-spring brides in
Western Oklahoma: U.S. Geological Survey Professional Paper 450-B, p. B126-B127.
Leonard, R. B., and Kleinschmidt, M. K., 1976, Saline water in the Little Arkansas River Basin area, South-
Central Kansas: The University of Kansas at Lawrence, Kansas Geological Survey Chemical Quality
Series No. 3,24 p.
Levings, G. W., 1984, Reconnaissance evaluation of contamination in the alluvial aquifer in the East
Poplar oil field, Roosevelt County, Montana: U.S. Geological Survey Water-Resources Investigations
Report 84-4174, 29 p.
LJco, M. S., Kharaka, Y. K., Carothers, W. W., and Wright, V. A., 1982, Methods for collection and analysis
of geopressured geothermal and oil field waters: United States Geological Survey, Geological
Survey Water-Supply Paper 2194, 21 p.
Lloyd, J. W., and Heathcote, J. A., 1985, Natural inorganic hydrochemistry in relation to ground water:
Clarendon Press, Oxford, 294 p.
Lloyd, J. W., and Howard, K. W. P., 1979, Environmental isotope studies related to ground-water flow and
saline encroachment In the Chalk Aquifer of Lincolnshire, England, in Isotope Hydrology 1978,
International Symposium on Isotope Hydrology: International Atomic Energy Agency, Vienna,
p. 311-325.
244
-------
Lloyd, J. W., Howard, K. W. F., Pacey, N. R., and Tellam, J. H., 1982, The value of iodide as a parameter in
the chemical characterization of ground waters: Journal of Hydrology, v. 57, p. 247-265.
Love, S. K., 1944, Cation-Exchange in Ground Water Contaminated with Sea Water near Miami, Florida:
American Geophysical Union, Transactions, v. 25, p. 951-955.
Ludwig, A. H., 1972, Water resources of Hempstead, Lafayette, Little River, Miller, and Nevada Counties,
Arkansas: U.S. Geological Survey Water Supply Paper 1998.
Lundegard, P. D., and Land, L. S., 1986, Carbon Dioxide and organic acids—Their role in porosity
enhancement and cementation, Paleogene of the Texas Gulf Coast, in Gautier, D. L., ed., Roles of
organic matter in sediment diagenesis: Society of Economic Paleontologists and Mineralogists
Special Publication 38, p. 129-146.
Lusczynski, N. J., and Swarzenski, W. V., 1966, Salt-Water Encroachment in Southern Nassau and
Southeastern Queens Counties, Long Island, New York: U.S. Geological Survey Water-Supply
Paper 1613-F, 76 p.
Macumber, P. G., 1984, Hydrochemical processes in the regional ground-water discharge zones of the
Murray Basin, Southeastern Australia, in Hitchon, Brian, and Wallick, E. I., eds., Proceedings First
Canadian/American Conference on Hydrogeology—practical applications of ground-water
geochemistry: National Water Well Association, Worthington, Ohio, p, 47-63.
Magaritz, Mordeckai, Nadler, Arie, Koyumdjisky, Hanna, and Dan, Joel, 1981, The use of Na/CI ratios to
trace solute sources in a semiarid zone: Water Resources Research, v. 17, no. 3, p. 602-608.
Magaritz, Mordeckai, and Luzier, J. E., 1985, Water-rock interactions and seawater-freshwater mixing
effects in the coastal dune aquifer, Coos Bay, Oregon: Geochimica et Cosmochimica Acta, v. 49,
p. 2515-2525.
Manheim, F. T., and Paull, C. K., 1981, Patterns of ground-water salinity changes in a deep continental-
oceanic transect off the southeastern Atlantic Coast of the U.S.A., in Back, William, ed., Symposium
on geochemistry of ground water: Journal of Hydrology, v. 54, nos. 1-3, p. 95-105.
Manheim, F. T., 1990, Potential saltwater intrusion into freshwater aquifers during exploration for and
mining of phosphorite off the Southeastern United States: U.S. Geological Survey Circular 1033,
p. 161-162.
Martin, Peter, 1982, Ground-water monitoring at Santa Barbara, California; Phase 2—effects of pumping
on water levels and water quality in the Santa Barbara ground-water basin: U.S. Geological Survey
Open-File Report No. 82-366,53 p.
Mast, V. A., 1982, The use of ionic mixing curves in differentiating oil-field brine in a fresh-water aquifer:
presented at the American Water Resources Association, The University of Texas at Austin,
December 3,1982,16 p.
McGrain, Preston, and Thomas, G. R., 1951, Preliminary report on the natural brines of eastern Kentucky:
Kentucky Geological Survey, Series IX, Report of Investigations No. 3,22 p.
245
-------
McGuire, M. J., Krasner, S. W., and Gramith, J. T., 1989, Comments on bromide levels in state project
water and Impacts on control of disinfection by-products: Metropolitan Water District of Southern
California, 17 p.
McMillion, L. G., 1965, Hydrologic aspects of disposal of oil-field brines in Texas: Ground Water, v. 3,
no. 4, p. 36-42.
McMillion, L. G., 1970, Ground-water reclamation by selective pumping: American Society of Mining
Engineers Reprint No. 70-AG-55.
Meents, W. F., Bell, A. H., Rees, O. W. and Tilbury, W. G., 1952, Illinois oil-field brines—their geologic
occurrence and chemical composition: Illinois State Geological Survey, Illinois Petroleum No. 66,
38 p.
Meisler, Harold, 1989, The occurrence and geochemistry of salty ground water in the Northern Atlantic
Coastal Plain: U.S. Geological Survey Professional Paper 1404-D, 51 p.
Meisler, Harold, Leahy, P. P., and Knobel, L. L., 1985, Effect of eustatic sea-level changes on saltwater-
freshwater relations in the Northern Atlantic Coastal Plain: U.S. Geological Survey Water-Supply
Paper 2255, 28 p.
Meyer, F. W., 1971, Saline artesian water as a supplement: Journal of the American Works Association,
v. 63, no. 2, p. 65-71.
Meyer, Michael, 1986, A summary of ground-water pollution problems in South Dakota: South Dakota
Department of Water and Natural Resources, Office of Water Quality, 18 p.
Michie & Associates, 1988, Oil and gas industry water injection well corrosion: report prepared for the
American Petroleum Institute, 25 p.
Michigan Department of Natural Resources, 1982, Assessment of groundwater contamination—inventory
of sites: Lansing, 242 p.
Miller, D. W., ed., 1980, Waste Disposal Effects on Ground Water: Premier Press, 512 p.
Miller, D. W., DeLuca, F. A., and Tessier, T. L., 1974, Ground-water contamination in the northeast states:
U.S. Environmental Protection Agency, Office of Research and Development, Robert S. Kerr
Environmental Research Laboratory, Report EPA-660/2-74-056, 300 p.
Miller, J. C., Hackenberry, P. S., and DeLuca, F. A., 1977, Ground-water pollution problems in the
Southeastern United States: U.S. Environmental Protection Agency, Office of Research and
Development, Robert S. Kerr Environmental Research Laboratory, Ecological Research Series
EPA:600/3-77-012, 329 p.
Miller, M. R., and others, 1980, Regional assessment of the saline-seep problem and a water-quality
inventory of the Montana plain: Montana Bureau of Mines and Geology, Open-Rle Report 42,414 p.
Mink, J. F., 1960, Some geochemical aspects of sea-water intrusion in a island aquifer: General Assembly
of Helsinki, International Association of Scientific Hydrology, Publication 52, p. 424-439.
246
-------
Monterey County Flood Control & Water Conservation District, 1989, Sources of saline intrusion in the
400-foot aquifer, Castroville area, California: Report prepared by D. K. Todd Consulting Engineers,
Inc., Berkeley, California, 43 p.
Mook, W. G., 1970, Stable carbon and oxygen isotopes of natural waters in the Netherlands, in Isotope
Hydrology: International Atomic Energy Agency, Vienna, p. 163-190.
Moore, J. W., and Welch, R. C., 1977, Environmental aspects, brine usage for highway purposes: final
report conducted for the Arkansas State Highway Department, Highway Research Project 44,133 p.
Morris, E. E., 1988, Arkansas ground-water quality, in Moody, D. W., Carr, Jerry, Chase, E. B., and
Paulson, R. W., compilers, National Water Summary 1986—Hydrologic events and groundwater
quality: U.S. Geological Survey Water-Supply Paper 2325, p. 165-172.
Morris, E. E. and Bush, W. V., 1986, Extent and source of saltwater intrusion into the alluvial aquifer near
Brinkley, Arkansas, 1984: Arkansas Geological Commission, Water Resources Circular 15,35 p.
Morrissey, D. J., and Regan, J. M., 1988, New Hampshire ground-water quality, in Moody, D. W., Carr,
Jerry, Chase, E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrologic
events and groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 363-368.
Morton, R. A., and Land, L. S., 1987, Regional variations in formation water chemistry, Frio Formation
(Oligocene), Texas Gulf Coast: American Association of Petroleum Geologists Bulletin, v. 71, no. 2,
p. 191-206.
Morton, R. B., 1986, Effects of brine on the chemical quality of water in parts of Creek, Lincoln, Okfuskee,
Payne, Pottawatomie, and Seminole Counties, Oklahoma: Oklahoma Geological Survey Circular 89,
37 p.
Muiler, A. B., 1974, Interdisciplinary modeling in the analysis of the salinity problems of the Safford Valley:
Water Resources Bulletin, v. 10, no. 2, p. 245-255.
Mullican, W. F., 1988, Subsidence and collapse at Texas salt domes: The University of Texas at Austin,
Bureau of Economic Geology, Geological Circular 88-2,35 p.
Murphy, E. C., and Kehew, A. E., 1984. The effect of oil and gas well drilling fluids on shallow groundwater
in western North Dakota. Report of Investigations No. 82, North Dakota Geological Survey, 156 p.
Nadler, A., Magaritz, M., and Mazar, E., 1981, Chemical reactions of seawater with rocks and freshwater-
experimental and field observations on brackish waters in Israel: Geochimica et Cosmochimica Acta,
V. 44, p. 879-886.
National Resources and Agriculture Committee, 1973, Interim report of the special commission on salt
contamination of water supplies and related matters: Massachusetts Senate Document 1485,
January 1973.
Nativ, Ronit, 1988, Hydrogeology and hydrochemistry of the Ogallala aquifer, Southern High Plains,
Texas Panhandle and Eastern New Mexico: The University of Texas at Austin, Bureau of Economic
Geology Report of Investigations 177, 64 p.
247
-------
Neffendotf, D. W., 1978, Statewide saline seep survey of Texas: Master's Thesis,; Texas A&M University,
67 p.
Newport, B. 0., 1977, Salt-water intrusion in the United States: U.S. Environmental Protection Agency,
Office of Research and Development, Robert S. Kerr Environmental Research Laboratory Report
EPA-600/8-77-011, 30 p.
Nordstrom, D. K., 1983, Preliminary data on the geochemica! characteristics of groundwater at Stripa:
Proceedings NEA Workshop on Geological Disposal of Radioactive Waste—in situ experiments in
granite, Nuclear Energy Agency, p. 143-153.
Novak, S. A., and Eckstein, Yoram, 1988, Hydrochemical characterization of brines and identification of
brine contamination in aquifers: Ground Water, v. 26, no. 3, p. 317-324.
Nyman, D. J., 1978, The occurrence of saline ground water in the coastal area of southwestern Louisiana:
EOS, American Geophysical Union, Transactions, v. 59, no. 12, p. 1065.
O'Hare, Margaret, Curry, D. S., Atkinson, S. F., Lee, S. B., and Canter, L. W., 1985, Contamination of
ground water in the contiguous United States from usage of agricultural chemicals (pesticides and
fertilizers): Environmental and Ground Water Institute, Norman, University of Oklahoma, variously
paginated. ,
Oklahoma Water Resources Board, 1975, Salt water detection in the Cimarron Terrace, Oklahoma: U.S.
Environmental Protection Agency, Office of Research and Development, Robert S. Kerr
Environmental Research Laboratory, Report EPA-660/3-74-033,100 p.
Ong, Kim, 1988, New Mexico ground-water quality, in Moody, D. W., Carr, Jerry, Chase, E. B., and
Paulson, R. W., compilers, National Water Summary 1986—Hydrologic events and groundwater
quality: U.S. Geological Survey Water-Supply Paper 2325, p. 377-384.
Orr, B. R., and Myers, R. G., 1986, Water resources in basin-fill deposits in the Tularosa Basin, New
Mexico: U.S. Geological Survey, Water-Resources Investigations Report 85-4219,94 p.
Parker, J. W., 1969, Water history of Cretaceous aquifers, East Texas Basin: Chemical Geology, v. 4,
nos. 1-2, p. 111-133.
Pariiman, D. J., 1988, Idaho ground-water quality, in Moody, D. W., Carr, Jerry, Chase, E. B., and Paulson,
R. W., compilers, National Water Summary 1986—Hydrologic events and groundwater quality: U.S.
Geological Survey Water-Supply Paper 2325, p. 229-236.
Patterson, R. J., and Kinsman, D. J. J., 1975, Marine and continental ground-water sources in a Persian
Gulf coastal sabkha, in Frost, S. H., Weiss, M. P., and Saunders, J. B., Reefs and related
carbonates—ecology and sedimentology: American Association of Petroleum Geologists, Studies
In Geology No. 4, p. 381-397.
Peck, A. J., 1978, Sainizatton of non-irrigated soils and associated streams—a review: Australian Journal
of Soil Research, v. 16, p. 157-168.
248
-------
Pennine, J. D., 1988, There's no such thing as a representative ground water sample: Ground Water
Monitoring Review, v. 8, no. 3, p. 4-9.
PennWell Publishing Co., 1982, Oil and gas fields in the United States: PennWell Publishing Company,
Tulsa, Oklahoma, map scale 1:3,530,000.
Pettyjohn, W. A., 1971, Water Pollution by Oil-Field Brines and Related Industrial Wastes in Ohio: Ohio
Journal of Science, v. 71, no. 5, p. 257-269.
Pettyjohn, W. A., 1982, Cause and effect of cyclic changes in ground-water quality: Ground Water
Monitoring Review, Winter 1982, p. 43-49.
Pierce, W. G., and Rich, E. I., 1962, Summary of rock salt deposits in the United States as possible storage
sites for radioactive waste materials: Geological Survey Bulletin 1148, prepared on behalf of the U.S.
Atomic Energy Commission, 91 p.
Piper, A. M., 1944, A graphic procedure in the geochemical interpretation of water analyses: American
Geophysical Union, Transactions, v. 25, p. 914-923.
Piper, A. M., Garrett, A. A., and others, 1953, Native and contaminated ground waters in the Long Beach-
Santa Ana area, California: U. S. Geological Survey, Water-Supply Paper 1136,320 p.
Poland, J. P., Garrett, A. A., and Sinnott, A., 1959, Geology, hydrology, and chemical character of ground
waters in the Torrance-Santa Monica area, California: U. S. Geological Survey, Water-Supply Paper
1461, 425 p.
Pomper, A. B., 1981, Hydrochemical observations in the subsoil of the western part of the Netherlands:
Proceedings, Seventh salt water intrusion meeting, Uppsala, Sweden, p. 101-111.
Poth, C. W., 1962, The occurrence of brine in western Pennsylvania: Pennsylvania Geological Survey,
Fourth Series, Bulletin M 47,53 p.
Price, R. D., 1979, Occurrence, quality, and quantity of ground water in Wilbarger County, Texas: Texas
Department of Water Resources Report 240,229 p.
Price, R. M., 1988, Geochemical investigation of salt-water intrusion along the coast of Mallorca, Spain:
University of Virginia, Chariottesville, Master's thesis, 186 p.
Puls, R. W., Eychaner, J. H., and Powell, R. M., 1990, Colloidal-facilitated transport of inorganic
contaminants in ground water, Part I. Sampling considerations: U.S. Environmental Protection
Agency, Robert S. Kerr Environmental Research Laboratory, EPA/600/M-90/023,12 p.
Randall, A.D., 1972, Records of Wells and Test Borings in Susquehanna River Basin, New York: New York
State Department of Environmental Conservation Bulletin No 69.
Rau, J. L, 1973, Effects of brining and salt by-products operations on the! surface and ground-water
resources of the Muskingum Basin, Ohio, in Coogan, A. H. ed., Fourth Symposium on Salt: Northern
Ohio Geological Society, v. 1, p. 369-386.
Reeves, C. C., Jr., 1970, Saline lake basins of the Southern High Plains, in Mattox, R. B., Ground-water
salinity: Symposium, 46th Annual Meeting of the Southwestern and Rocky Mountain Division of the
249
-------
American Association of the Advancement of Science, April 23-24, Las Vegas, New Mexico, p. 64-
69.
Relchenbaugh, R. C., 1972, Sea-water intrusion in the upper part of the Floridan aquifer in coastal Pasco
County, Florida, 1969: U.S. Geological Survey, Florida Bureau of Geology Map Series no. 47.
Rice, G. F., Multer, D. L, and Brinkman, J. E., 1988, Reliability of chemical analyses of water samples—the
experience of the UMTRA project: Ground Water Monitoring Review, v. 8, no. 3, p. 71-75.
Richter, B. C., and Kreitler, C. W., 1986a, Geochemistry of salt water beneath the Rolling Plains, North-
Central Texas: Ground Water, v. 24, no. 6, p. 735-742.
Richter, B. C., and Kreitler, C. W., 1986b, Geochemistry of salt-spring and shallow subsurface brines in the
Rolling Plains of Texas and southwestern Oklahoma: The University of Texas at Austin, Bureau of
Economic Geology Report of Investigations No. 155,47 p.
Richter, B. C., Dutton, A. R., and Kreitler, C. W., 1990, Identification of sources and mechanisms of salt-
water pollution affecting ground-water quality—a case study, West Texas: The University of Texas at
Austin, Bureau of Economic Geology Report of Investigations No. 191,43 p.
Ritter, W. F., and Chimside, A. E. M., 1982, Ground-water quality in selected areas of Kent and Sussex
Counties, Delaware: Newark, University of Delaware Agricultural Engineering Department, 229 p.
Rittenhouse, Gordon, 1967, Bromine in oil-field waters and its use in determining possibilities of origin of
these waters: The American Association of Petroleum Geologists Bulletin, p. 2430-2440.
Rittenhouse, Gordon, Fulton, R. B., Ill, Grabowski, R. J., and Bernard, J. L, 1969, Minor elements in oil-
field waters: Chemical Geology, v. 4, p. 189-209.
Roberson, C. E., Feth, J. H., Seaber, P. R., and Anderson, Peter, 1963, Differences between field and
laboratory determinations of pH, alkalinity, and specific conductance of natural water, in Short papers
in geology and hydrology: U.S. Geological Survey Professional Paper 475-C, p. C212-C217.
Robinove, C. J., Langford, R. H., and Brookhart, J. W., 1958, Saline-water resources of North Dakota:
U.S. Geological Survey Water-Supply Paper 1428,72 p.
Roedder, E., 1984, Fluid inclusions: Mineratogical Society of America, Rev. in Min. 12, p. 644.
Rold, J. W., 1971, Pollution problem in the "Oil Patch": American Association of Petroleum Geologists
Bulletin v. 55, no. 6, p. 807-809.
Rumer, R. R., Jr., Apmann, R. P., and Chien, C. C., 1973, Runoff of deicing salt in Buffalo, New York, in
Coogan, A. H. ed., Fourth Symposium on Salt: Northern Ohio Geological Society, v. 1., p. 407-411.
Sabol, G. V., Bouwer, H., Wierenga, P. J., 1987, Irrigation effects in Arizona and New Mexico: American
Society of Civil Engineers, Journal of Irrigation and Drainage Engineering (ASCE), v. 113, no. 1,
p. 30-48.
Safflgna, P. G., and Keeney, D. R., 1977, Nitrate and chloride in ground water under irrigated agriculture in
central Wisconsin: Ground Water, v. 15, no. 2, p. 170-177..
250
-------
Salt Institute, undated, Survey of salt, calcium chloride, and abrasive use in the United States and Canada
for 81/82 and 82/83: Salt Institute, Alexandria, Virginia, 4 p., 9 tables.
Sandoval, F. M. and Benz, L. C., 1966, Effect of bare fallow, barley, and grass on salinity of a soil over a
saline water table: Soil Science Society of America Proceedings, v. 30, p. 392-396.
Sayre, A. N., 1937, Geology and ground-water resources of Duval County, Texas: U.S. Geological Survey
Water Supply Paper 776,116 p.
Scalf, M. R., Keely, M. R., and LaFevers, C. J., 1973, Ground-water pollution in the south-central states:
U.S. Environmental Protection Agency, Office of Research and Development, National
Environmental Research Center, Environmental Protection Technology Series EPA-R2-73-268,
135 p.
Schaefer, F. L., 1983, Distribution of chloride concentrations in the principal aquifers of the New Jersey
Coastal Plain, 1977-81: U.S. Geological Survey Water-Resources Investigations Report 83-4061,
56 p.
Schaefer, F. L., and Walker, R. L, 1981, Salt-water intrusion into the Old Bridge Aquifer in the Keyport-
Union Beach area of Monmouth County, New Jersey: U.S. Geological Survey Water-Supply Paper
2184, 21 p.
Scheldt, M. E., 1967, Environmental effects of highways: Journal of the Sanitary Engineering Division,
Proceedings of the American Society of Civil Engineers, v. 93, no. SA5, paper no. 5509.
Schmidt, Ludwig, and Devine, J. M., 1929, The disposal of oil-field brines: U.S. Department of the Interior,
Bureau of Mines, Report of Investigations 2945,17 p.
Schmidt, V., and McDonald, D. A., 1979, The role of secondary porosity in the course of sandstone
diagenesis, Aspects of Diagenesis, in Scholle, P. A., and Schluger, R., Society of Economic
Paleontologists and Mineralogists Special Publication 26, p. 175-208.
Schoeller, Henri, 1935, Utilite1 de la notion des exchanges de bases pour la comparison des eaux
souterraines: France, Soci6t6 Geotogie Comptes rendus Sommaire et Bulletin S6rie 5, v. 5, p. 651-
657.
Schofield, J. C., 1956, Methods of distinguishing sea-ground-water from hydrothermal water: New
Zealand Journal of Science and Technology, General Research Section, v. 37, no. 5, p. 597-602.
Schraufnagel, F. H., 1967, Pollution Aspect associated with chemical de-icing: Highway Research Record
Report 193, p. 22-23.
Schwartz, F. W., 1974, The origin of chemical variations in ground waters from a small watershed in
southwestern Ontario: Canadian Journal Earth Science, v. 11, no. 7, p. 893-904.
Senger, R. K., Collins, E. W., and Kreitler, C. W., 1990, Hydrogeology of the northern segment of the
Edwards aquifer, Austin region: The University of Texas at Austin, Bureau of Economic Geology,
Report of Investigations No. 192,58 p.
251
-------
Shamburger, V. M., Jr., 1958, Alleged well contamination in relation to brine disposal in Clemville,
Matagorda County: Texas Board of Water Engineers Contamination Report, 10 p.
Shedlock, R. J., 1978, Saline water in the glacial outwash aquifer near Vincennes, Indiana: EOS,
American Geophysical Union, Transactions, v. 59, no. 12, p. 1065.
Shirazi, G. A., Alien, D. L, and Madden, M. P., 1976, An investigation of brine contamination of soils and
ground water from a slush pit near bums flat, Oklahoma: Oklahoma Water Resources Board, 91 p.
Shows, T. A., Broussard, W. L, and Humphreys, C. P., Jr., 1966, Water for industrial development in
Forrest, Green, Jones, Perry and Wayne Counties, Mississippi: Mississippi Research and
Development Center.
Sidenvall, Jan, 1981, Fossil ground water of marine origin in the Uppsala area, Sweden: Proceedings,
Seventh Salt Water Intrusion meeting, Uppsala, Sweden, p. 40-44.
Stebenthal, C. E., 1910, Geology and water resources of the San Luis Valley, Colorado: U.S Geological
Survey Water-Supply Paper 240,128 p.
Siple, G. E., 1969, Salt-water encroachment of tertiary limestones along coastal South Carolina: South
Carolina Development Board, Division of Geology, Geologic Notes 13, v. 2, p. 51-65.
Skibitzke, H. E., Bennett, R. R., DaCosta, J. A., Lewis, D. D., and Maddock, T., Jr., 1961, Symposium on
history of development of water supply in an arid area in southwestern United States, Salt River
Valley, Arizona: International Association of Scientific Hydrology, Publication 57, p. 706-742.
Skogerboe, G. V., and Law, J. P., 1971, Research needs for irrigation return flow quality control: U.S.
Environmental Protection Agency, Water Pollution Control Research Series 13030.
Skogerboe, G. V., and Walker, W. R., 1973, Salt pickup from agricultural lands in the Grand Valley of
Colorado: Journal of Environmental Quality, v. 2, no. 3, p. 377-382.
Slack, L. J., and Pfanert, Michael, 1988, Alabama ground-water quality, in, Moody, D. W., Carr, Jerry,
Chase, E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrotogic events and
groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 143-148.
Smith, J. F., 1966, Imperial Valley salt balance: Public Information Office, Imperial Irrigation District, El
Centro, California.
Smith, Z. A., 1989, Groundwater In the West: Academic Press, San Diego, California, 308 p.
Snow, M. S., Kahl, J. S., Norton, S. A., and Olson, Christine, 1990, Geochemical determination of salinity
sources in ground water wells in Maine: Proceedings of the 1990 FOCUS conference on eastern
regional ground water issues, Ground Water Management Book 3, National Water Well Association,
p. 313-327.
Snyder, R. H., Buehrer, L. C., and Bell, F. O., 1972, A survey of the subsurface saline water of Texas, v. 1:
Texas Water Development Board Report 157,113 p.
Sorensen, H. O., and Segall, R. T., 1973, Natural brines of the Detroit River Group, Michigan Basin, in
Coogan, A. H. ed., Fourth Symposium on Salt: Northern Ohio Geological Society, v. 1, p. 91-99.
252
-------
Sowayan, A- M., and Allayla, Rashid, 1989, Origin of the saline ground water in Wad! Ar-Rumah, Saudi
Arabia: Ground Water, v. 27, no. 4, p. 481-490.
Speiran, G; K., OkJham, R. W., Duncan, D., and Knox, R. L, 1988, South Carolina ground-water quality, in
Moody, D. W., Carr, Jerry, Chase, E. B., and Paulson, R. W., compilers, National Water Summary
1986—Hydrologic events and groundwater quality: U.S. Geological Survey Water-Supply Paper
2325, p. 449-456.
Sproul, C. R., Boggess, D. H., and Woodard, H. J., 1972, Saline-water intrusion from deep artesian
sources in the McGregor Isles area of Lee County, Florida: State of Florida, Department of Natural
Resources, Division of Interior Resources, Bureau of Geology Information Circular No. 75,30 p.
Spruill, T. B., 1985, Statistical evaluation of effects of irrigation on chemical quality of ground water and
base flow in three river valleys in north-central Kansas: U.S. Geological Survey Water-Resources
Investigations Report 85-4158, 64 p.
Stafford, M. R., 1987, Hydrogeology, ground-water chemistry, and resistivity of a contaminated shallow
aquifer system in southern Bond County, Illinois: Southern Illinois University, Master's thesis.
Steinkampf, W. C,, 1982, Origins and distribution of saline ground waters in the Floridan aquifer in coastal
southwest Florida: U.S. Geological Survey Water-Resources Investigations Report 82-4052,34 p.
Stephens, J. C., 1974, Hydrologic Reconnaissance of the northern Great Salt Lake Desert and summary
hydrologic reconnaissance of northwestern Utah: Utah Department of Natural Resources Technical
Publication No. 42,55 p.
Stringfield, V. T., and LeGrand, H. E., 1969, Relation of sea water to fresh water in carbonate rocks in
coastal areas, with special reference to Florida, U.S.A., and Cephalonia (Kephallinia), Greece:
Journal of Hydrology 9, p. 387-404.
Subra, W. A., 1990, Unsuccessful oil field waste disposal techniques in Vermilion Parish, Louisiana, in
Proceedings of the First International Symposium on Oil and Gas Exploration and Production Waste
Management Practices, New Orleans, Louisiana: U.S. Environmental Protection Agency, p. 995-
999.
Sundstrom, R. W., Hastings, W. W., and Broadhurts, W. L., 1948, Public water supply in eastern Texas:
U.S. Geological Survey Water-Supply Paper 1047, 285 p.
Swarzenski, W. V., 1963. Hydrogeology of Northwestern Nassau and Northeastern Queens Counties:
Long Island, New York, U.S. Geological Survey Water-Supply Paper 1657.
Sweeten, J. M., 1990, Cattle feedtot waste management practices for water and air pollution control:
Texas Agricultural Extension Service, B-1671, 20 p.
Sweeten, J. M., Safley, L. M., and Melvin, S. W., 1981, Sludge removal from lagoons and holding
ponds—case studies, in Livestock waste—a renewable resource: Proceedings of the Fourth
International Symposium on Livestock Wastes, American Society of Agricultural Engineers, St.
Joseph, Michigan, p. 204-210.
253
-------
Sylvester, R. O., and Seabloom, R. W., 1963, Quality and significance of irrigation return flow: Journal of
the Irrigation and Drainage Division, American Society of Civil Engineers, v. 89, no. IR3, p. 1-27.
1 '
Taylor, O. J., 1983, Missouri Basin Region, in Todd, D. K., ed., Ground-water resources of the United
States: Premier Press, Berkeley, California, p. 297-343.
Templeton, E. E., and Associates, 1980, Environmentally acceptable disposal of salt brine produced with
oil and gas: Ohio Water Development Authority, 52 p.
Texas State Soil and Water Conservation Board, 1984, Preliminary Report-Assessment of soil salinity in
Texas: Texas Department of Water Resources, Contract No. 14-40019 1 AC(84-85) 1038, Statewide
Inventory of Salinity Problems.
Texas Water Commission, 1989, Ground-Water Quality of Texas—an overview of natural and man-affected
conditions: Report 89-01,197 p.
Thomas, J. M., and Hoffman, R. J., 1988, Nevada ground-water quality, in Moody, D. W., Carr, Jerry,
Chase, E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrotogic events and
groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 355-362.
Thompson, R. G., and Custer, S. G., 1976, Shallow ground-water salinization in dryland-farm areas of
Montana: Montana Universities Joint Water Resources Research Center, MUJWRRC Report No. 79,
214 p.
Thome, W., and Peterson, H. B., 1967, Salinity in United States waters, in Brady, N. C., ed., Agriculture
and the quality of our environment: American Association for the Advancement of Science,
Washington, D.C., Publication 85, p. 221-240.
Tremblay, J. J., 1973, Salt Water Intrusion in the Summerside Area, P.E.I.: Ground Water, v. 11, No. 2,
p. 21-27.
Truesdell, A. H., 1976, Summary of section III, geochemical techniques in exploration, in Proceedings,
Second United Nations Symposium on the Development and Use of Geothermal Resources, San
Francisco, California, 20-29 May, 1975 Berkeley, California, Lawrence Berkeley Laboratory, v. 1.
Tsunogal, S., 1975, Sea salt particles transported to the land: Tellus, v. 27, no. 1, p. 51-58.
Turner, S. P., and Foster, M. D., 1934, A Study of Salt-Water Encroachment in the Galveston Area, Texas,
American Geophysical Union Transactions, v. 15, p. 432-435.
Twerrter, F. R., and Cummings, T. R., 1985, Quality of ground water in Monitor and Williams Townships,
Bay County, Michigan: U.S. Geological Survey Water-Resources Investigations Report 85-4110,
41 p.
UNESCO, 1980, Aquifer contamination and protection: Project 8.3 of the International Hydrologic
Programme, prepared by the Project Working Group, Jackson, R. E., chairman and general editor,
119 p.
Unger, S. G., 1977, Environmental implications of trends in agriculture and silviculture: U.S.
Environmental Protection Agency, Athens, Georgia, EPA-600/3-77-121,188 p.
254
-------
Upson, J. E., 1966, Relationships of fresh and salty ground water in the northern Atlantic Coastal Plain of
the United States: U.S. Geological Survey Research, U.S Geological Survey Professional Paper
550-C, p. C235-C243.
Urish, 0. W., and Ozbilgin, M. M., 1985, Mean sea level—an elusive boundary, in The second annual
eastern regional ground water conference, July 16-18,1985, Portland, Maine, p. 348-358.
Utah State University Foundation, 1969, Characteristics and pollution problems of irrigation return flow:
Report prepared for U.S. Department of the Interior, Federal Water Pollution Control Administration,
Robert S. Kerr Water Research Center, Ada, Oklahoma, under Contract Number 14-12-408,236 p.
U.S. Department of Agriculture, 1975, Erosion, sediment, and related salt problems and treatment
opportunities: Soil Conservation Service, Special Projects Division, Golden, Colorado.
U.S. Department of Agriculture, 1983, Dryland salinity study—Texas-Cooperative River Basin Survey,
prepared by United States Department of Agriculture Soil Conservation Service Economic Research
Service in cooperation with Texas Department of Water Resources-Texas State Soil and Water
Conservation Board.
U.S. Environmental Protection Agency, 1973, Identification and control of pollution from salt water
intrusion: EPA Office of Air and Water Programs, Water Quality and Non-Point Source Control
Division, Report EPA-430/9-73-013, 94 p.
U.S. Environmental Protection Agency, 1975, Water programs—national interim primary drinking water
regulations: Federal Register, v. 40, no. 248.
U.S. Environmental Protection Agency, 1977, National secondary drinking water regulations: Federal
Register, v. 42, no. 62, pt. 1, p. 17143-17147.
U.S. Environmental Protection Agency, 1989, Proposed requirements for 38 inorganic and organic
drinking-water contaminants: Office of Drinking Water, Washington, D.C. 20460,10 p.
U.S. Geological Survey, 1984, National water summary 1983—Hydrotogic Events and Issues, United
States Geological Survey Water-Supply Paper 2250: U.S. Geological Survey, 243 p.
U.S. Saynity Laboratory, 1954, Diagnosis and improvement of saline and alkaline soils: U.S. Department of
Agriculture Handbook 60,160 p.
U.S. Water News, 1990a, Bromide doesn't really help California's delta headaches; v. 6, no. 7.
U.S, Water News, 19905, Salt is invading Florida Gulf Coast: v. 7, no. 3, p. 12.
Van Biersel, T. P., 1985, Hydrogeotogy and chemistry of an oil-field brine plume within a shallow aquifer
system in southern Bond County, Illinois: Southern IllinoisUniversity, Master's thesis.
Van Denburgh, A. S., and Rush, F. E., 1974, Water-resources appraisal of Railroad and Pehoyer Valleys,
east-central Nevada: Nevada Division of Water Resources, Water Resources Reconnaissance
Series, Report no. 60, 61 p.
van der Leeden, Frits, Cerrillo, L. A., and Miller, D. W., 1975, Ground-water pollution problems in the
Northwestern United States: U.S. Environmental Protection Agency, Office of Research and
255
-------
Development, Robert S. Kerr Environmental Research Laboratory, Report EPA-660/3-75-018,
341 p.
van Everdingen, R. O., 1971, Surface-water composition in southern Manitoba reflecting discharge of
saline subsurface waters and subsurface solution of evaporites, in, Turnock, A. C., ed., Geoscience
Studies in Manitoba: Geological Association of Canada, Special Paper No. 9, p. 343-352.
Van Sickle, Virginia, and Groat, C. G., 1990, Oil field brines—another problem for Louisiana's coastal
wetlands, in, Proceedings of the First International Symposium on Oil and Gas Exploration and
Production Waste Management Practices, New Orleans, Louisiana: U.S. Environmental Protection
Agency, p. 659-675.
Voelker, D. C., and Clarke, R. P., 1988, Illinois ground-water quality, in Moody, D. W., Cart, Jerry, Chase,
E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrologic events and
groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 237-244.
Vovk, I. P., 1981, Radiolytic model of formation of brine compositions in the crystalline basement of
Shields: Geochemistry International, p. 80-93.
Waddell, K. M., and Maxell, M. H., 1988, Utah ground-water quality, in Moody, D. W., Carr, Jerry, Chase,
E. B., and Paulson, R. W., compilers, National Water Summary 1986—Hydrologic events and
groundwater quality: U.S. Geological Survey Water-Supply Paper 2325, p. 493-500.
Wait, R. L., and Gregg, D. O., 1973, Hydrology and chloride contamination of the principal artesian aquifer
in Gtynn County, Georgia: Water Resources Survey of Georgia, Earth and Water Division, Hydrologic
Report 1.
Wait, R. L., and McCollum, M. J., 1963, Contamination of fresh-water aquifers through an unplugged oil-
test well In Gtynn County, Georgia: Georgia Geological Survey Mineral Newsletter, v. 16, nos. 3-4,
p. 74-80.
Walker, W. H., 1970, Salt piBng—a source of water supply pollution: Pollution Engineer, v. 1, no. 4, p. 30-
33.
Waller, B. G., and Howie, Barbara, 1988, Determining nonpoint-source contamination by agricultural
chemicals in an unconfined aquifer, Dade County, Florida: Procedures and preliminary results, in
Ground-water contamination: Reid Methods, ASTM STP 963, American Society for Testing and
Materials, Philadelphia, p. 459-467.
Wallick, E. I., Krouse, H. R., and Shakur, Asif, 1984, Environmental isotopes—principles and applications
In ground water geochemical studies in Alberta, Canada, in Hitchon, Brian, and Wallick, E. I., eds.,
Proceedings, First Canadian/American Conference on Hydrogeology—Practical applications of
ground-water geochemistry: National Water Well Association, Worthington, Ohio, p. 249-266.
Ward, P. E., 1961, Geology and groundwater features of salt springs, seeps, and plains in the Arkansas
and Red River Basin of western Oklahoma and adjacent to Kansas and Texas: U.S. Geological
Survey, Open-File Report No. 63-132,94 p.
256
-------
Welch, A. H., and Preissler, A. M., 1986, Aqueous geochemistry of the Bradys Hot Springs geothermal
area, Churchill County, Nevada: U.S. Geological Survey Water-Supply Paper 2290, p. 17-36.
Wheeler J. C., and Maclin, L. B., 1988, Maryland and the District of Columbia ground-water quality, in
Moody, D. W., Carr, Jerry, Chase, E. B., and Paulson, R. W., compilers, National Water Summary
1986—Hydrologic events and groundwater quality: U.S. Geological Survey Water-Supply Paper
2325, p. 287-296.
Whitehead, D. C. 1974, The influence of organic matter, chalk and sesquioxides on the solubility of
iodine, elemental iodine and todate incubated with soil: Journal Soil Science, v. 4, p. 461-470.
Whitehead, H. C., and Feth, J. H., 1961, Recent chemical analyses of waters from several closed-basin
lakes and their tributaries in the western United States: Geological Society of America Bulletin, v. 72,
p. 1421-1426.
Whiteman, C. D., Jr, 1979, Salt-water encroachment in the '600-foot' and '1,500-foot' sands of the Baton
Rouge area, Louisiana, 1966-78, including a discussion of saltwater in other sands: Louisiana
Department of Transportation and Development, Office of Public Works, Water Resources Technical
Report No. 19,49 p.
Whittemore, D. O., 1984, Geochemical identification of salinity sources, in French, R. H., Salinity in
watercourses and reservoirs: Proceedings of the 1983 International Symposium on State-of-the-Art
Control of Saynity, Salt Lake City, Utah, p. 505-514.
Whittemore, D. O., 1988, Bromide as a tracer in ground-water studies: Geochemistry and analytical
determination: National Water Well Association, Proceedings, Ground Water Geochemistry
Conference, Denver, Colorado, February 16-18,1988, p. 339-359.
Whittemore, D. O., and Pollock, L. M., 1979, Determination of salinity sources in water resources of
Kansas by minor alkali metal and halide chemistry: Kansas Water Resources Research Institute,
Consultant's report to Office of Water Research and Technology, U.S. Department of the Interior,
Washington, D.C., 37 p.
Wilcox, L. V., and Durum, W. H., 1967, Quality of irrigation water, in Hagan, R. M., Haise, H. R., and
Edminster, T. W., eds., Irrigation of agriculture lands: American Society of Agronomy, Madison,
no. 11, p. 104-122.
Wilder, H. B., Robinson, T. M., and Lindskov, K. L., 1978, Water resources of northeast North Carolina:
U.S. Geological Survey, Water Resources Investigations 77-81,113 p.
Williams, C. C. and C. K. Bayne, 1946, Ground-Water Conditions in Elm Creek Valley, Barber County,
Kansas, with Special Reference to Contamination of Ground Water by Oil Field Brine: Kansas State
Geological Survey Bulletin 64,1946 Reports of Studies, part 3, p. 77-124.
Wilmoth, B. A., 1972, Salty ground water and meteoric flushing of contaminated aquifers in West Virginia:
Ground Water, v. 10, no. 1, p. 99-105.
257
-------
Wilson, W. £., 1982, Estimated effects of projected ground-water withdrawals on movement of the
saltwater front in the Floridan aquifer, 1976-2000, West Central Florida: U.S. Geological Survey
Water-Supply Paper 2189, 24 p.
Winstow, A.G., Doyle, W. W., and Wood, L. A., 1957, Salt Water and Its Relation to Fresh Ground Water in
Harris County, Texas: U.S. Geological Survey Water-Supply Paper 1360-F, 33 p.
Woessner, W. W., Mifflin, M. D., French, R. H., Elzeftawy, Atef, and Zimmerman, D. E., 1984, Salinity
balance in the Lower Virgin River Basin, Nevada and Arizona, in French, R. H., ed., Salinity in
watercourses and reservoirs, Proceedings, 1983 International Symposium on State-of-the-Art
Control of Salinity, Salt Lake City, Utah: Butterworth Publishers, Boston, p. 145-156.
Wood, W. W., 1976, Guidelines for collection and field analysis of ground-water samples for selected
unstable constituents, 1976: Techniques of Water-Resources Investigations of the United States
Geological Survey, Book 1, Chapter 02,24 p.
Wood, W. W., and Jones, B. F., 1990, Origin of solutes in saline lakes and springs on the southern High
Plains of Texas and New Mexico, in Gustavson, T. C., ed., Geologic framework and regional
hydrology—Upper Cenozoic Blackwater Draw and Ogallala Formation, Great Plains: The University of
Texas at Austin, Bureau of Economic Geology, p. 193-208.
Woodruff, K. D., 1969, The occurrence of saline ground water in Delaware aquifers: Delaware Geological
Survey, Report of Investigations No. 13,45 p.
Wyoming Department of Environmental Quality, 1986, Statewide water quality assessment 1986—
Executive summary: Cheyenne, Wyoming, 26 p.
Yeh, H. W., 1980, D/H ratios and late-stage dehydration of shales during burial: Geochimica et
Cosmochimica Acta, v. 44, p. 341-352.
Zack, Allen, and Roberts, Ivan, 1988, The geochemical evolution of aqueous sodium in the Black Creek
Aquifer, Horry and Georgetown Counties, South Carolina: U.S. Geological Survey Water-Supply
Paper 2324,15 p.
258
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA/600/2-91/064
{3. RECIPIENT'S ACCESSION NO.
PB92-11965Q
TITLE AND SUBTITLE
IDENTIFICATION OF SOURCES OF GRDUNEMWATER SALINIZATION
USING GEOCHEMICAL TECHNIQUES
REPORT DATE
December 1991
. PERFORMING ORGANIZATION CODE
AUTHOR(S)
Bernd C. Richter and Charles W. Kreitler
|8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORGANIZATION NAME AND ADDRESS
The University of Texas at Austin
Austin, Texas 78713-7508
10. PROGRAM ELEMENT NO.
CBPC1A
11. CONTRACT/GRANT NO.
CR-815748
3. SPONSORING AGENCY NAME AND ADDRESS
Robert S. Kerr Environmental Research Lab.
U.S. Environmental Protection Agency
P.O. Box 1198
Ada, OK 74820
- Ada, OK
13. TYPE OF REPORT AND PERIOD COVERED
Final Report (Mav 1989 to
14. SPONSORING AGENCY CODE
EPA/600/15
5. SUPPLEMENTARY NOTES
Project Officer; Bert Bledsoe
FTS: 743-2324
B. ABSTRACT
This report deals with salt-water sources that commonly mix and
deteriorate fresh ground water. It reviews characteristics of salt-water
sources and geochemical techniques that can be used to identify these sources
after mixing has occurred.
The report is designed to assist investigators of salt-water problems in
a step-by-step fashion. Seven major sources of salt water are distinguished:
(I) Natural saline ground water, (2) Halite solution, (3) Sea-water intrusion,
(4) Oil-and gas-field brines, (5) Agricultural effluents, (6) Saline seep, and
(7) Road salting. The geographic distribution of these sources was mapped
individually and together, illustrating which ones are potential sources at
any given area in the United States. In separate sections, each potential
source is then discussed in detail regarding physical and chemical
characteristics, examples of known techniques for identification of mixtures
between fresh water and that source, and known occurrences by state.
Individual geochemical parameters that are used within these techniques are
presented in a separate section, followed by a discussion concerning where and
how to obtain them. Also provided is a description of basic graphical and
statistical methods that are used frequently in salt-water studies. An
extensive list of references for further study concludes this report.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATl Field, Group
18. DISTRIBUTION STATEMENT
RELEASE TO THE PUBLIC
19. SECURITY CLASS tTliiS Report!
UNCLASSIFIED
21. NO. OF PAGES
20. SECURITY CLASS (Tliispagi-:
UNCLASSIFIED
22. PRICE
272
EPA Form 2220-1 (R«v. 4-77) PREVIOUS COITION is OBSOLETE
*U.S. GOVERNMENT PRINTING OFFICE: ,n2-64S-003/40670 259
------- |