EPA-R2-72-014
DECEMBER 1972 Environmental Protection Technology Series
Rehabilitation of a
Brine-Polluted Aquifer
Office of Research and Monitoring
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
EPA-B2-72-CM
December 1972
REHABILITATION OF A BRINE-POLLUTED AQUIFER
By
John S. Fryberger
Project 1U020 DLN
Project Officer
Leslie G. McMillion
Robert S. Kerr Water Research Center - EPA
P. 0. Box H98
Ada, Oklahoma JU820
Prepared for
OFFICE OF RESEARCH AND MONITORING
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, B.C. 20^60
For sale by the Superintendent at Documents, U.S. Government Printing Office, Washington, D.C. 20W2 - Prico $1.25
-------
EPA REVIEW NOTICE
This report has been reviewed ;by the Environmental Protection
Agency and approved for publication. Approval does not signi-
fy that the contents necessarily reflect the views and poli-
cies of the Environmental Protection Agency, nor does mention
of trade names or commercial products constitute endorsement
or recommendation for use.
-------
ABSTRACT
A detailed investigation was made of one (among several
noted) incident where a fresh-water aquifer has been pol-
luted by accepted disposal of oil-field brine through an
"evaporation" pit (an unlined earthen pit) and later a
faulty disposal well. The present extent of the brine
pollution is one square mile; however, it will spread to
affect 4 1/2 square miles and will remain for over 250
years before being flushed naturally into the Red River.
Detailed chemical analyses show changes in relative concen-
trations of constituents as the brine moves through the
aquifer.
Several rehabilitation methods are evaluated in detail,
including controlled pumping to the Red River and deep-we 11
disposal. None of the methods that are both technically
feasible and permissible show a positive public benefit-cost
ratio.
Although real economic damage both present and future results
from this brine pollution, rehabilitation is not now econom-
ically justified. The report emphasizes that greater effort
is needed to prevent such pollution, which not only affects
ground-water resources but also affects water quality in
interstate streams.
This report was submitted in fulfillment of Grant No. 14020
DLM under the partial sponsorship of the Office of Research
and Monitoring, Environmental Protection Agency.
11
-------
CONTENTS
Sectfon Page
I Conclusions 1
II Recommendations 3
III Introduction 5
General 5
Location 5
Objective 5
Scope 8
Project History 8
IV Geology and Hydrology 11
General 11
Extent of Ground Water Pollution in Miller
County 12
V Value of Water Polluted 19
VI Delineation of Pollution 21
History of Pollution in Project Area 21
Test Dril.ling and Sampling 22
Physical Delineation 26
Chemical Delineation 32
VII Rehabilitation Methods 39
General 39
Containment 39
Bentonite Wall 39
Accelerated Discharge 39
WaterDrive 40
Pumping to Red River 40
Use 45
Secondary Recovery 45
Blend ing for Irrigation 45
Desalinization 46
Deep Wei 1 Disposal 46
General 46
Nacatoch Disposal Well 49
Smackover Disposal Well 50
VIII Benefit-Cost Ratios 55
IX Acknowledgments 59
X References 61
-------
FIGURES
PAGE
1 LOCATION 6
2 GEOLOGICAL COLUMN 13
3 DURATION CURVE OF DAILY FLOW INTO RED RIVER 14
4 RELATIONSHIP BETWEEN RIVER FLOW AND CHLORIDE
CONTENT IN RED RIVER 15
5 LOCATION OF TEST WELLS IN PROJECT AREA 23
6 CONTOURS OF CHLORIDE CONCENTRATION AT BOTTOM
OF ALLUVIUM 27
7 CONTOURS ON TOP OF SHALE 28
8 WATER LEVEL CONTOURS JUNE 22, 1970 30
9 WATER LEVEL CONTOURS JANUARY 12-13, 1972 31
10 NORTH-SOUTH SECTION SHOWING BRINE DISTRIBUTION 33
11 EAST-WEST SECTION SHOWING BRINE DISTRIBUTION 34
12 RELATIONSHIPS BETWEEN DISTANCE FROM PIT AND
CONCENTRATION OF SELECTED CHEMICAL PARAMETERS 37
13 MEAN MONTHLY DISCHARGE OF RED RIVER .. 41
14 CALCULATED WATER TABLE CONTOURS AROUND FOUR
PROPOSED PRODUCTION WELLS 43
15 NOMOGRAPH: TO DETERMINE FLOW REQUIRED IN RED RIVER
THAT WOULD LIMIT THE CHLORIDE INCREASE
TO 10 MG/L AT ANY PUMPED WATER
CONCENTRATION 44
16 DISTANCE-DRAWDOWN CURVES 48
-------
TABLES
No. Page
1 Chloride Content and Temperature of
Samples from Test Wells 24
2 Summary of Water Level Elevations 29
3 Chemical Analysis of Samples from Selected Sources 36
4 Summary and Benefit-Cost Ratios of Rehabilitation
Methods 57
vii
-------
SECTION I
CONCLUSIONS
1. Improper oil-field brine disposal, first from an
"evaporation" pit (an unlined earthen pit) and later
from a faulty disposal well, caused pollution of one
square mile in a shallow alluvial aquifer.
2. The polluted area will spread downstream to contaminate
4 1/2 square miles of ground water before discharging
naturally into the Red River and will affect water
quality in the area for over 250 years.
3. Two other polluted areas have been outlined by test
drilling, and four more areas of high chlorides have
been found by testing private wells in Miller County.
4. Dilution of metals and other chemical parameters in the
brine was observed to be not always in proportion to
the dilution of chlorides as the brine moves through the
aqui fer.
5. Of the numerous rehabilitation methods examined, pumping
into the Red River and deep-well disposal are the most
feasible solutions of those that are technically practi-
cal. However, none of the methods are economically
justified at this time.
6. Because of the extremely long-term effect of ground-water
pollution and its eventual discharge into interstate
waters, considerably more effort is justified to insure
that such pollution is completely stopped.
7. State agencies responsible for controlling pollution
caused by brine disposal should enact and enforce stringent
pollution control regulations.
-------
SECTION II
RECOMMENDATIONS
Although the original objective of this project was to
demonstrate the feasibility of rehabilitating a brine-
polluted aquifer, it now appears that none of the rehabil-
itation methods that are technically sound and permissible
are economically justified at this time. However, the long-
term economic damage caused by such ground-water pollution,
its effect on interstate waters, and the widespread occur-
rence of brine pollution are justification for greater
participation by the responsible state and federal agencies.
Therefore, it is recommended that EPA formulate brine handling
and disposal guidelines in cooperation with the oil producing
states. For example, the standards should outlaw brine
disposal into unlined "evaporation" pits, except in the rare
cases where it's positively established that no water pollu-
tion will result under guidelines approved by EPA. Further-
more, the standards should require the use of injection
tubing and a monitored fluid-filled annulus for disposal wells.
Such regulations should apply to all brine handling systems,
not just to future construction, because of the relatively
high incidence of pollution caused by older installations.
Furthermore, because of the cost of policing and effectively
enforcing such regulations, some form of federal assistance
to the responsible state agencies may be justified. The
overall enforcement costs could be partly offset by a system
whereby noncompliance with the regulations would result in
a substantial fine to the offending party.
It is further recommended that additional observation wells
be constructed as required, and a continuing monitoring
program be established for this project area. The objectives
of such a program would be: (1) to observe the continuing
distribution of the brine in space, time, and concentration
as it spreads in the aquifer; (2) to warn downstream irriga-
tors and potential irrigators of the impending brine encroach-
ment; and (3) to observe the actual lag in brine movement
compared with natural ground-water flow as an aid in evaluating
the natural flushing process. Such a continuing program
based on this wel1-documented incident will benefit not only
the immediate ground-water users but will also provide a
sound basis for evaluating the long-term effects of all similar
ground-water pollution.
In addition, it is recommended that all other areas where
similar ground-water pollution exists be sufficiently outlined
by test drilling and sampling to describe their extent and
chloride concentration. The objectives of this program would
-------
be: (1) to warn ground-water users and potential users near
the polluted areas of the impending danger; (2) to locate
pollution incidents where early rehabilitation would be
technically feasible and economically justified before
further spreading of the brine occurs; and (3) to form an
inventory of first-choice water sources for oil-field water
flood operations.
-------
SECTION III
INTRODUCTION
General
Activities related to alleviating pollution of our ground-
water resources may be divided into two categories- (1)
those activities designed to stop pollution now taking
place, and (2) those activities to rehabilitate ground-water
reservoirs which have already become polluted. Although
this project deals primarily with rehabilitation, it is
hoped that the costs of the remedial measures presented and
the real economic damage caused by such pollution will
stimulate considerably greater efforts by state and federal
agencies in the preventive category.
This project deals with the pollution of a valuable shallow
ground-water aquifer by the disposal of oil-field brine
through first an unlined "evaporation" pit and later through
a faulty disposal well. Although the use of unlined
evaporation" pits (which should be called seepage pits) is
now outlawed in some states and some rules have been adopted
regarding salt-water disposal wells, still considerable
pollution is taking place because of the lack of sufficient
surveillance and enforcement. This report examines in
detail a singular occurrence of such pollution and the costs
involved in rehabilitation. If rehabilitation steps are
not undertaken, the polluted ground water will spread and
eventually discharge into the Red River, an interstate stream
Location
The project is in Miller County in the southwest corner of
Arkansas, see figure 1. The sources of the brine pollution
are a disposal pit and a disposal well located in the SW 1/4
of the SE 1/4 of Section 14, Township 16S, Range 26W, which
is about 2 1/2 miles southwest of the town of Garland City
and 2 1/2 miles west of the Red River.
This particular polluted area occupies about one square mile
and affects the west half of that part of the alluvial
floodplain on the west side of the Red River. The floodplain
is flat, productive farmland, which lies 222 feet above sea
level at the project area.
Objective
The original objective of this project was to develop selec-
tive pumping techniques whereby a fresh-water aquifer, which
had become contaminated by brine from oil-field practices,
-------
\
FIGURE I
•«"•''"••'
LEG END
OIL a GAS FIELD
BRINE POLLUTED
GROUND WATER
. CHLOBIOC COITOIT
°° I SMALUW MOUMXWTEII
I "I/I )
EOSE Of RIVER
"•« ». ALLUVIUM
FAULT, u- UPTMDWM
0- DOWN THROWN
20/ POTENTKJ*«TRtC 5URWCE
" IN RIVER ALLUVIUM
-------
LOCATION MAP
REHABILITATION OF A BRINE POLLUTED AQUIFER
EPA PROJECT I4O20 OLN
ARKANSAS DIVISION OF SOIL a WATER RESOURCES
6CNCMAL WQMVUY MAP
MILLER COUNTY
ARKANSAS
MKAMSA3 STATE NtGMWfcr
L 0 U I
-------
could be rehabilitated. Because of problems arising from
disposal of the salt water to be pumped from the aquifer,
actual rehabilitation is not now considered feasible.
The present objectives of this report are the comprehensive
technical evaluation of the brine contamination, the
technical and the economic evaluation of various disposal
methods, and the relationship of this singular incident to
the general pollution problem. In addition, the severity
of ground-water pollution as herein presented will hopefully
lead to greater state and federal action to prevent such
pol1u tion.
The following areas are herein examined:
1. A history of the pollution.
2. A technical description of the polluted aquifer.
3. A discussion of the value of the water that was polluted
4. Technical evaluation of several possible methods of
rehabilitating-the aquifer and attendant costs.
5. Private and public benefit-cost ratio analysis of the
rehabilitation methods presented.
Project History
In 1967 a farmer brought to the attention of the state
agencies that his irrigation well had turned salty. This
1,000 gpm (gallons per minute) well was located in the NW
corner of Section 24, about 2,500 feet southeast of the
subject disposal pit. Analyses of water samples from this
well showed an increase in chlorides from 900 to 1,100 mg/1
(milligram per liter) over the two weeks before the well was
shut down.
During the summer of 1967 the Arkansas Soil and Water Conser-
vation Commission, along with the Arkansas Geological
Commission, the Pollution Control Commission, and the Oil
and Gas Commission conducted an investigation to determine
the source of the pollution. This investigation consisted
of auguring holes through the alluvium and sampling the
water-sand mix as it was brought to the surface. These test
hole samples suggested that the disposal pit was the source
of pollution to that farmer's well. In addition, under a
reconnaissance study being conducted simultaneously by the
U. S. Geological Survey, samples were obtained from existing
domestic and irrigation wells and other test holes over a
-------
20-square mile area. This more general survey delineated
two other polluted areas where chlorides exceeded 500 ppm.
All three of these areas are at or just down gradient
(south) of producing oil fields and are shown as "brine
polluted ground water" on figure 1.
In May 1968 efforts were initiated to obtain a federal
grant to rehabilitate the aquifer at the project area, and
in June 1969 the demonstration grant was awarded to the
Arkansas Soil and Water Conservation Commission. The grant
was divided into two phases. The object of Phase I was to
delineate the problem and establish in detail the best solu-
tion. Upon approval of the proposed solution, Phase II, the
construction of rehabilitation facilities, would then be
authorized. Field work was started by personnel of that
agency in November 1969.
During the latter part of 1969 and the first half of 1970,
28 ground-water sampling sites were established in the
project area to delineate the extent of the pollution. Upon
analysis of this data a Phase I Report, entitled "Rehabilita-
tion of a Brine-Polluted Aquifer," was submitted to EPA in
December 1970. This report stressed the feasibility of
disposing of the polluting salt water into the Red River;
however, this solution was not acceptable to EPA, and Phase II
was not authorized.
In January 1972, EPA authorized the Arkansas Division of Soil
and Uater Resources (previously the Arkansas Soil & Water
Conservation Commission) to finalize the project by submitting
a more detailed summary of the problem and work performed.
This report constitutes that summary and was prepared for the
Division of Soil & Water Resources by Engineering Enterprises,
a consulting firm specializing in ground water, assisted by
personnel of the Division of Soil & Water Resources.
-------
SECTION IV
GEOLOGY & HYDROLOGY
General
The floodplain of the Red River is about nine miles wide
in this area and is characterized by oxbow lakes, cut-off
meander scars, and poorly drained bayous, typical of a
mature meandering agrading river. Clean, highly permeable
sand was deposited by the river during much of its early
depositional stage. Later deposition has consisted pre-
dominately of silts and clays which form a blanket of
variable thickness overlying the alluvial sand bodies.
In the project area, which is- only about one mile from the
west side of the floodplain, the alluvium extends to about
40 feet; however, depths up to 90 feet are known according
to Ludwig. With water levels only 5 to 20 feet below land
surface, these readily rechargeable alluvial sands form an
important fresh-water aquifer. The extent of the alluvium
i s shown on figure 1 .
Much of the upland area is also covered by unconsolidated
sand and silt as terraces deposited by the Red River in
earlier times. These terrace deposits also constitute an
important source of fresh water to private wells and, as
in the case of the alluvium, are also easily polluted.
The Sparta sand of Tertiary age underlies the terrace
deposits and shallow alluvium in the general vicinity, but
it has been eroded away beneath the project area. Ludwig
describes this formation as a fine-to-medium sand, brown
and gray sandy clay, and lignite. This extensive formation
is also an important fresh-water aquifer in southwestern
Arkansas and northeastern Louisiana. Water levels and
chemical analyses of samples from the Sparta sand suggest
that the salt-contaminated alluvial water will not contam-
inate the Sparta aquifer.
The Cane River formation underlies the Sparta sand and is
the primary source of domestic water where the shallower
alluvium and terrace deposits are absent or polluted.
Yields from the Cane River are too small for irrigation
however. The Carrizo sand lies below the Cane River and is
the lowest known fresh water aquifer in the central part of
Miller County. The town of Fouke, about eight miles south-
west of the project, obtains its supply of nearly 0.23 mgd
from a 600-foot well completed in the Carrizo.
11
-------
Figure 2, "Geologic Column," shows the sequence of forma-
tions and approximate depths at the project area. These
formations dip southward at about 30 feet per mile. The
north edge of a northeast trending fault zone and a splinter
fault lie directly under the project area, as shown on
figure 1. The main fault plane dips southward 45 degrees
in the shallower sediments and steepens to about 60 degrees
with depth. Because of continued vertical movement along
the fault through geologic time, the deep formations are
displaced over 500 feet; the Annona chalk is displaced about
300 feet; but relatively little displacement can be observed
in the formations of late Eocene time.
Precipitation in southwest Arkansas averages 48 inches per
year. Although rainfall distribution is fairly uniform
throughout the average year (August and September having the
least precipitation), nevertheless during the dry springs
and summers (of years of below average rainfall) many farmers
must irrigate in order to produce a good crop. The average
annual pan evaporation is 57 inches at Hope, Arkansas, which
is about 25 miles from the project area.
The average flow of the Red River at Fulton, Arkansas, is
17,600 cfs (cubic feet per second). The flow duration curves
for this stretch of the Red River are shown on figure 3.
Figure 4 shows the chloride concentration at various flows.
In general the water in the Red River is high in hardness,
chlorides and sulfates, and without extensive treatment is
unsuited for many uses, including irrigation of some crops
according to Ludwig.
According to Ludwig about 80 percent of the total water use
in southwest Arkansas is derived from ground water and only
one municipality, Texarkana, uses surface water. In 1965,
6.89 mgd (million gallons per day) was used for irrigation,
nearly all of which was derived from alluvium in the Red
River Valley. Boisier City, which is 57 miles south of the
project area in Louisiana, is the closest downstream user of
Red River water.
Extent of Ground-Water Pollutionin Miller County
In order to obtain information on the natural (unpolluted)
chloride content of the water in the alluvial and terrace
sands, 25 additional water samples were obtained throughout
the county in January 1972. Locations of these water sample;
and the chloride content are shown on figure 1. These data
show that the unpolluted water contains chlorides ranging
from 7 to about 50 mg/1 .
12
-------
JLfiJL
RECENT
FOR NATION
u
o
O
U
<»
ARKADtLPHIA MARL
NACATOCH SAND
SARATOGA CHALK
MARLBROOK MARL
ANNONA CHALK
OZAN FM
BROWNSTONE MARL
TOKIO FM
TUSCALOOSA
FREDERIC K3BURG GRP
PALU X Y SAND
TRINITY GRP
kk^Li,
1
^:~:
P
I
-My
,s f
w
M
%
$
"^X^X"
^SjS*
XxO'*
5^
<^
^^
DEPTH
-0 FEET
REMARKS
—1000
—2OOO
-3000
COTTON VALLEY QRP
SMACKOVER
L 1ME STONE
(ABSENT UNDER PROJECT AREA)
DEEPEST FRESH-WATER AQUIFER
SALT WATER DISPOSAL ZONE
NOW USED BY MCKINNEY
BAYOU OIL FIELD
OIL PRODUCING ZONE
OIL PRODUCING ZONE
POSSIBLE DISPOSAL ZONE
FIGURE 2
GEOLOGIC COLUMN
AT PROJECT AREA
13
-------
99 99 99 9 99 8
99 98 95 90 80 10 60 50 «0 30 30
I 0.5 0,2 01 004 O.Qi
100,000
o
u
u.
u
D.OOO
0 01 O.Oi O.i 02 OS !
1000
-------
m
u_
o
>
K
RELATIONSHIP BETWEEN RIVER FLOW
AND CHLORIDE CONTENT IN REDRIVER
AT FULTON, ARKANSAS 1994-1961
Tf li-
_L
_L
100
200 300 400
CHLORIDES mg/ 1
*oo
ADAPTED FROM
STRAMEL
15
-------
Notable exceptions to the natural chloride concentrations
are as fol1ows :
1. Sec 35, T 15S, R 26W; chlorides = 230 mg/1. The
source of the chlorides is probably 1 1/2 miles
upstream at the Lenz oil field which has been
abandoned but reportedly produced considerable
salt water which was disposed of through pits.
2. Sec 7, T 16S, R 25W; chlorides = 5,650 mg/1. The
chlorides in this sample could be related to pits
in either the New Garland City oil field to the
west, or the Mayton oil and gas field to the north.
3. Sec 32 and 29, T 16S, R 25W; chlorides = 150 and
160 mg/1. The two irrigation wells from which these
samples were obtained were used to irrigate rice.
However, after losing one-half of the rice crop in
1971 due to high chlorides, they are now abandoned.
The source of the chlorides is probably from old
pits in the Cypress Lake oil and gas field, which is
located one-half mile to the north. The chloride
content was reportedly much higher when the wells
were in operation than at the time these samples
were taken .
4. Sec 21, T 17S, R 27W; chlorides = 115 mg/1. The
moderately high chloride content in this well is
probably related to operations in the South Fouke
oil and gas field.
5. Sec 1, T 15S, R 27W; chlorides = 116 mg/1. The
source of the moderately high chloride content in
this well is not apparent.
The samples obtained from wells in sections 30 and 32 in
T 16S, R 26W, were expected to be high in chlorides because
of their location in the Fouke oil and gas field. However,
the reason no pollution was observed may be because of
insufficient sampling points or because the operators in
this oil field have been more careful with their brine
disposal .
In addition to the three brine-polluted areas outlined on
figure 1, this general survey using only existing private
wells shows that there are at least four other areas in
Miller County where the shallow fresh-water aquifers have
been polluted by brine-disposal practices of oil-field
operations. Further pollution could undoubtedly be delin-
eated by a wide-ranging test drilling and sampling program.
16
-------
Although the present use of "evaporation" pits is somewhat
regulated, the practice is not specifically outlawed in
Arkansas. Furthermore, enforcement suffers due to the lack
of personnel and the lack of fines or other deterents to
viol a tors .
17
-------
a
SECTION V
VALUE OF WATER POLLUTED
The ground-water area polluted by the brine is about one
square mile. Under the natural ground-water gradient in
the alluvium of 1.4 feet per mile (figure 1) and assuming
a permeability of 1500 gpd/ft^, (Ludwig) the natural rate
of ground-water movement is about 100 feet per year. The
direction of movement is south-southeast and eventually,
after flowing about 4 1/2 miles, the salty water will be
slowly discharged into the Red River which is the natural
discharge avenue for all ground water in the area. At the
estimated rate of movement, the salty water will not reach
the river for approximately 250 years.
It could be concluded, then, that ground-water occupying
at least one square mile (but not necessarily the same
square mile) will remain polluted for at least 250 years.
It is this author's opinion that because of dispersion,
adsorption and vagarious characteristics of the aquifer,
the entire 4 1/2 mile-long path of this one-mile wide body
of salt water will remain contaminated for a much longer
period than 250 years. This view is shared by others
(Collins).
The damage that has already taken place because of this
particular pollution incident consists of the loss of a
high capacity irrigation well valued at $4,000 and the
partial loss of one year's rice crop on 120 acres valued
at $36,000, for which water from the well was a necessity.
The future monetary loss which will result because of the
pollution is of course impossible to evaluate accurately.
Nevertheless, the following figures are presented to estab-
lish an order-of-magnitude value. All of the estimates
conservatively assume that only one square mile of irrigible
land is removed from irrigation.
Ri ce - At $150/acre profit over 640 acres for 250 years =
124 ,000,000 loss in profit income. (Irrigation is manda-
tory for rice farming.)
Cotton - At $35/acre difference in profit between irri-
gated and non-irrigated cotton for 250 years over 640
acres = $5,600,000 loss.
Soybeans - At $20/acre difference in profit between
irrigated and non-irrigated soybeans for 250 years over
640 acres = $3,200,000 loss.
19
-------
Furthermore, if the pollution had affected a municipal water
supply and if that town were forced to construct a surface
supply to replace the lost ground-water source, then the
difference in water cost to the town would be about 20i£/1000
gallons. Assuming that the square mile allowed production
of 1 mgd on a sustained yield basis, then the added cost to
the town would be $73,000 per year to replace the lost water
source, and over 250 years would total $18,250,000.
It is fully recognized that value projections such as in
the above examples are not meaningful in any exact sense.
Although the real value loss to the national economy result-
ing from this singular incident is not now significant,
ground-water pollution is highly significant considered on
the large scale and in time. On the other hand, the real
value loss does have present-day significance to the property
owners affected. Furthermore, with time the total area
affected will probably Increase fourfold and the value loss
due to non-irrigibi1ity will also increase.
20
-------
SECTION VI
DELINEATION OF POLLUTION
History of Pollution in Project Area
Since 1955, when the McKinney oil field was discovered, a total
of 46 test holes and wells, 15 of which are still producing,
have been drilled primarily into the Paluxy sand. Salt water
was produced along with the oil starting in 1957 and has
increased substantially as the field approaches depletion.
Accordi ng
fol1owi ng
to records
amounts of
of the Arkansas
salt water have
Oil and Gas Commission the
been produced by the field
June
Year
1957
1958
1959
1960
1961 (June)
1961-1970
Total Barrels
Produced
22,200
49,700
78,000
114,500
1 18,500
3,828,803
Disposal
Method
Pit
Pit
Pit
Pit
Pit
Combination pit
di sposal wel 1 .
and
Between July 1961 and August 1967 the Parks #1 disposal well
adjacent to the pit was in operation; however, the water was
stored in the pit before injection into the well. It is con-
sidered likely that most of the produced brine seeped through
the pit during this period. In August 1967 after the initial
study was made of this problem, tanks were provided for storage
and the pit was no longer used. All of the water was injected
into Parks #1 disposal well under pressure.
During the course of the field work on this project, a hydro-
graph from observation well #6, about 500 feet from the disposal
well, showed a marked relationship between the water level in
the alluvial aquifer and the periods of operation of the pump
on the disposal well. An investigation of the disposal well
showed that the injection pressure was zero, whereas when the
well was first put into operation 300 to 400 psi was required
to pump the brine into the disposal formation, the Tokio formation
The overall conclusion was that the casing of the disposal well
had indeed corroded through and the brine was escaping into the
alluvial aquifer. By December 1970, a new disposal well was
completed (Parks #3), and the old well was abandoned and plugged.
It is not possible, based on the available records, to accurately
calculate the quantity of brine that seeped or was injected into
the alluvial aquifer. However, this quantity of brine is esti-
mated to be about 2,700,000 bbls based on the amount of salt in
21
-------
the aquifer. Using this figure as the amount of brine soaked
or injected into the aquifer, and knowing the approximate
quantity that soaked through the "evaporation" pit, it may be
estimated that the disposal well had been injecting into the
alluvium at highly reduced pressures for approximately 1 1/2
years before detection.
If this disposal well had been constructed using injection
tubing and a fluid-filled annulus the chances for pollution
through corroded casing would have been greatly minimized, and
detection would have been much easier. However, this safeguard
is not required by Arkansas.
Test Drilling and Samp1i ng
The nature and extent of the pollution in the project area has
been determined by the construction of 36 permanent test wells
at 28 locations most of which were drilled during 1969-70. The
locations of the test wells are shown on figure 5.
The following procedures were used in the construction and
sampling of these test wells.
1. A 3-inch hole was augured through the 9 to 15 feet of
surface soils and clay.
2. A 2 1/2-inch pipe was driven through the sand to the shale
at the bottom of the aquifer. Driving was stopped at
various depths, and samples of the water were obtained as
this pipe was installed.
3. The permanent 2-inch diameter, 2 1/2-foot long plastic
well screen was installed to the desired depth inside the
2 1/2 inch pipe, and the pipe was pulled back to expose the
screen.
i
4 At all locations the primary well point was set just above ;
the shale At some locations additional well points were
installed in adjacent holes to provide permanent sampling ;
wells at higher elevations.
5 Initial and subsequent water samples have been obtained by
pumping each test well using a vacuum pump for 5 to 10
minutes in order to obtain representative formation water
samples.
Tahlp 1 shows the elevation and depth of the well points for
e h location and lists the chloride content and temperature for
all the samples that have been obtained from each test well,
including those samples that were taken during construction and
subsequent samples.
22
-------
\\ ,
1000 2000
3OOO
IE
1
SCALE IN FEET
LEGEND
15 14
_ j_—Section Corner
22 j 23
2.7 Test Well Location
A A' Cross Section ( See
Figures IO 8 II )
FIGURE 5
LOCATIONS OF TEST WELLS
AND SECTIONS
-------
TABLE 1
CHLORIDE CONTENT AND TEMPERATURE
OF SAMPLES FROM TEST WELLS
Well
No.
1
la
2
2a
3
3a
,1h
4
5
5a
6
7
7a
8
9
10
11
12
13
14
Elev.
Sam-
pi e
206
191
185
202
192
186
205
208
196
187
182
198
206
207
192
178
192
182
197
213
193
185
212
192
187
201
194
189
193
188
206
191
179
207
195
181
191
183
194
184
212
1 92
188
Depth
16
32
38
24
30
36
20
14
26
35
40
26
17
15
30
44
30
40
27
10
30
38
10
30
35
22
30
35
30
35
15
30
42
15
27
41
30
38
30
40
9
30
34
11/69-^
mg/1
1 ,060
45,000
47,000
49,500
56,500
300
4,700
47,500
48,000
160
.13,500
47,500
13,500
39,000
680
45,500
46,500
1 ,000
1 ,200
12,700
600
625
1 9,500
39,500
1 ,300
800
575
300
250
3,775
500
525
475
29,000
2,225
2,700
14,100
^70
92
98
65
65
68
66
65
64
66
66
64
12/70
mg/1
59,000
17,500
55,000
27,500
55,000
10,000
285
54,000
49,000
7,000
55,000
17,500
750
1 ,150
41 ,000
800
7,470
800
38,000
23,000
2/71
mg/1
51 ,500
2,150
49,500
13,500
50,500
8,000
280
47,000
45,000
4,000
48,000
14,500
600
1 ,050
36,000
600
9,500
400
33,000
22,500
°F
84
83
92
90
65
65
64
65
68
66
64
65
64
64
66
64
8/71
mg/1
48,800
980
47,800
12,470
45,520
4,840
249
44,000
40,200
2,450
45,560
1 5,940
525
790
28,440
520
10,210
535
29,360
22,500
°F
84
79
84
80
66
66
66
66
65
65
67
66
66
68
65
66
66
65
65
65
1/72
mg/1
45,000
950
48,000
11 ,000
47,500
3,900
240
49,000
44,000
2,600
45.000
13,000
600
800
25,500
550
13,500
690
28,500
24,000
°F
64
66
6$
64
65
65
65
(Conti nued )
24
-------
TABLE 1 (Continued)
CHLORIDE CONTENT AND TEMPERATURE
OF SAMPLES FROM TEST WELLS
Wei 1
No,
15
16
17
17a
18
l8a
19
20
21
22
23
24
25
26
27
28
Elev.
Sam-
ple
192
187
190
178
195
185
203
206
191
183
203
192
182
209
191
181
193
188
212
192
189
196
186
193
183
183
184
194
189
199
191
Depth
30
35
30
42
30
40
20
15
30
38
20
30
40
12
30
40
30
35
10
30
33
30
40
30
40
42
39
28
33
25
33
1 1/69-4
mg/1
650
610
235
260
18,500
50,000
1 ,830
490
3,850
1 ,175
8,550
745
415
375
600
500
4,000
500
500
4,500
1 ,500
2,050
2,750
i/70
°F
64
65
65
66
66
65
12/70
mg/1
825
47,000
1 ,000
4,300
550
10,400
590
570
450
2,750
53,000
57,000
1 ,280
9,800
3,250
5,350
2/71
mg/1
1 ,050
1 ,000
45,500
700
4,000
440
11 ,500
1 ,695
530
505
405
4^000
47,000
49,500
11 ,500
5,500
°F
64
64
64
64
65
65
66
66
65
67
68
8/71
mg/l
922
327
42,000
680
1 ,930
416
1 ,900
2,543
494
478
366
2,110
42,320
45,120
11 ,380
5,300
°F
65
65
65
66
66
66
66
65
66
64
65
66
66
67
64
66
1/72
mg/1
1 ,200
300
43,500
530
3,000
450
12,000
3,700
550
480
385
2,050
45,000
47,500
12,500
4,600
°F
65
65
66
65
65
65
65
65
66
66
25
-------
Physical Delineation
Figure 6 shows the chloride concentrations at the bottom of
the alluvium as determined by samples taken from those test
wells positioned just above the clay. From this plan view
of the polluted area it is notable that the salty water has
spread generally southward in the direction of ground-water
flow. The irregular shape of the polluted zone may be attri-
buted to variations in permeability within the aquifer and to
irregularities in the top surface of the shale.
Figure 7 shows the contours on top of the shale based on the
test-well construction data. It should be noted that a shallow
valley extends to the northeast and southwest just north of
the disposal pit. The orientation of this valley corresponds
generally with the northeast-southwest orientation of the
polluted water observed on figure 6. Because the salt water
is heavier than the fresh water, it settles to the bottom of
the aquifer; hence, its movement is somewhat controlled by
the topography of the underlying shale.
Table 2 lists all of the water-table elevations as measured
in the test wells throughout the monitoring period.
Figure 8 shows the contours on the water surface as determined
by water-level measurements made on June 22, 1970, when the
faulty salt-water disposal well was still in use. A water-level
mound is evident surrounding the disposal well which was located
adjacent to the pit. Also notable from this figure is the
water-surface depression south of the disposal well.
In January 1972, water levels were again measured resulting in
the contours shown on figure 9. The mound is absent indicating
that disposal into the aquifer has been stopped. However, the
water-surface depression south of the pit is even more pronounced
than in figure 8. This depression is caused by the difference
in density between the salt water and the fresh water.
The density of the brine was determined to be 1.057 g/ml .
Assuming the density of the fresh water to be 1 g/ml, the follow-
ing may be calculated based on data at TW 4.
h$ = hf where hs = theoretical height of salt water
s hf = height of fresh water = 32 feet (Figure 10)
D = density of salt water = 1.057 g/ml
Then the theoretical difference in height of the salt water and
fresh water is 32 - 30.3 = 1.7 feet.
26
-------
IOOO ZOOO 3OOO
I . I ~l
SCALE IN FEET
LEGEND
Concentrations In mg/l
FIGURE 6
CONTOURS OF CHLORIDE CONCENTRATION
AT BOTTOM OF ALLUVIUM
ADAPTED FROM STRAMEL
27
-------
IOOO
—r
2000
I
3000
SCALE IN r E E T
LEGEND
U.S.G.S. Seo Level Datum
FIGURE 7
CONTOURS ON TOP OF SHALE
( BOTTOM OF ALLUVIAL AQUIFER )
ADAPTED FROM STRAMEL
28
-------
TABLE 2
SUMMARY OF WATER LEVEL ELEVATIONS
ELEVATIONS
(In Feet Above Mean Sea Level)
Test
Well
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
4/22/70
214.23
21 5.35
213.47
213.29
213.17
213.89
214.44
214.96
213.90
214.58
213.76
213.72
214.20
214.02
213.74
214.16
213.56
214.36
213.92
213.87
214.89
214.79
215.27
12/70
211 .70
211 .63
211 .26
21 1 .18
211 .20
211 .90
21 1 .08
212.82
211 .96
212.61
211 .81
211 .63
212.37
" ?11.70
211 .48
212.13
211 .43
212.33
211 .94
211 .82
212.83
212.47
213.28
211 .94
212.27
210.75
212.17
211 .88
2/71
212.26
212.13
211 .71
211 .63
211 .54
212.18
212.65
213.28
212.45
213.05
212.22
212.31
212.66
212.15
212.15
212.65
211 .95
212.81
212.40
212.42
213.32
213.00
213.69
312.56
211 .74
211 .27
21 2.28
212.49
8/71
211 .68
212.23
212.01
211 .75
211 .61
212.61
212.74
213.43
212.84
213.08
212.19
212.26
212.90
212.11
212.28
212.46
212. Tl
213.22
212.49
213.35
213.49
213.10
213.78
212.47
211 .77
211 .31
211 .23
212.50
1/72
212.05
212.01
211 .56
211 .35
211 .36
212.15
212.97
213.39
213.04
212.94
211 .96
212.08
212.88
212.54
212.22
212.75
212.09
212.91
212.45
212.25
213.47
213.78
213.90
213.20
211 .53
211 .35
212.22
213.14
29
-------
SCALE IN FEET
LEGEND
U.S.G.S. Seo Level Dolum
FIGURE 8
WATER LEVEL CONTOURS
JUNE 22. 1970
ADAPTED FROM STRAMEL
30
-------
1000 20 OO
3000
SCALE IN FEET
LEGEND
U.S.G, S. Sea Level Datum
FIGURE 9
WATER LEVEL CONTOURS
JANUARY 12-13. 1972
31
-------
The actual difference in water-surface elevations is 213 -
211.5 = 1.5 feet as determined by the measured water-level
elevation in TW 4 and the extrapolated 213-foot water-surface
contour line shown on figure 9. This close agreement
between the calculated and observed water levels indicates
that the depression in the water surface south of the pit is
caused by the density difference.
The distortion of the water-level contours around the body of
salt water, as evident in figure 9 one year after injection
into the aquifer was stopped, may be attributed to the
slightly higher viscosity and higher density of the polluted
water which condition would tend to make the polluted water
flow more slowly than the fresh water.
Figure 10 is a north-south cross section through the pit and
disposal well showing the vertical distribution of salty water
along the section. (See figure 5 for location of sections.)
It should be noted from this figure that the source of brine
appears to be very high in the alluvium even though the pit
was not in use at the time the data were taken. Apparently
the break in the casing of the faulty disposal well, which
was then in use, was in the top half of the sandy part of the
alluvium. Also note that the salt water sinks rapidly and
flows along the bottom part of the aquifer.
Figure 11 is an east-west cross section located 600 feet
south of the pit and disposal well. The concentrated brine
continues to sink and is spreading laterally as shown by this
section.
Two rather anomalous conditions are also evident from figure 11
The high chlorides near the top of the sandy part of the allu-
vium at the east end of the section probably result from brine
having been dumped into Moccasin Creek and seeping into the
ground. The other anomaly is that the chlorides in TW 28 (west
end) are higher than in TW 24 even though TW 24 shows a topo-
graphic low in the shale. There is an abandoned oil well in
the NE corner of section 22 about 1 ,500 feet from TW 28. It
is possible that this old well is purging salt water into the
alluvium thereby causing the apparent anomaly. However, it
may also be explained by differences in permeability. A
highly permeable sand streak at TW 28 could result in more of
the brine flowing past that test well than would flow past
TW 24. Another possible explanation is that the drainage
ditch west of TW 28 had also been used for salt-water disposal,
and residual salt is still leaching downward into the aquifer.
Chemical Delineation
All of the chloride determinations that have been made on
samples taken at different times are listed on table 1. A
partial record of temperatures is also shown on that table.
32
-------
OJ
GJ
1*0
ISO
OOO CHLORIDES IN mg/l 11/69-4/70
• SAMPLE TAKEN DURING DRILLING
PERMANENT SAMPLING POINT
FIGURE 10
-« NO RTH-SOUTH *>
SECTION SHOWING BRINE DISTRIBUTION
SCALES
HORIZONTAL' l" « SOO*
VERTICAL > l"« 10'
-------
220
2ZO
ZIO
zoo
ZOO
OOO CHLORIDES IN mg/l 11/69— 4/?0
• SAMPLE TAKEN DURING DRILLING
~*~ PERMANENT SAMPLING POINT
FIGURE II
•« WEST-EAST *•
SECTION SHOWING BRINE DISTRIBUTION
SCALES
HORIZONTAL! l"« 600'
VERTICAL ' l" • 10'
-------
It may be noted that the highest temperature, 98°F, was
recorded in November 1969 on a sample taken from TW 2 just
south of the pit and near the faulty disposal well. The
most recent available temperature from the same test well
shows that the temperature has dropped to 84 F. In general
the temperature of the warm injected brine appears to be
absorbed by the sand and fresh water rather rapidly away
from the injection point. However, once the sand is warmed,
the temperature dissipates slowly with time.
In January 1972 samples were obtained of the brine, from key
test wells progressively farther from the brine source, and
from an uncontaminated private well, located 1 1/2 miles
east of the project (see figure 1). Relatively complete
analyses were performed on these samples to determine the
changes in chemical characteristics of the brine as it flows
through the alluvium and is mixed with the fresh water. The
results of these analyses are shown on table 3.
In order to better visualize the chemical changes as the brine
becomes progressively diluted, comparative graphs were prepared
of key constituents. The graphs are shown on figure 12.
The chloride graph on figure 12 may be taken as the standard,
considering that chlorides should not precipitate nor the
concentration otherwise change except in direct ratio with
the degree of dilution. Inspection of the graphs for dissolved
solids and calcium show that these constituents are diluted
in the same general ratio as chlorides. The dilution of
several other constituents such as bromide, flouride, lead,
strontium, manganese, nickel, and aluminum, although not
exactly proportional to chlorides, may not be related to
extraneous factors considering the accuracy of quantitative
analysis at low concentrations.
The pronounced decrease in barium may be explained by its
precipitation as barium sulfate, noting the presence of sulfate
in the native ground water but the absence of sulfate close
to the injection point. Strontium also appears to have been
precipitated, probably as a sulfate, as the brine mixed with
the native water. Both strontium and barium are more soluble
in brines than in fresh water (Davis and Collins) which also
may explain their abrupt decrease in concentration. Although
boron shows a similar concentration decrease, its precipitation
would not be anticipated, and the reason for its similar
pattern is not known.
Other constituents such as iodide, iron and zinc are note-
worthy in that the concentrations of these elements at TW 3
are significantly higher than in the brine. Three explanations
may be offered to explain these anomalous concentrations.
(1) These elements may combine with anions that adhere loosely
to the sand grains and do not move far from the point of
35
-------
CHEMICAL
FROM
TABLE 3
ANALYSES OF SAMPLES
SELECTED SOURCES
Samples Taken 1/12 - 13/72
Concentrations in mg/1 unless otherwise indicated
Distance from Pit
Relationship to Flow
(Arkansas Pollution
Control Lab)
Spec Cond mmhos
Chlorides
PH
Sul phates
Nitrate NO,
Total Solids
Diss Solids
Susp Solids
Total Hardness
Ca Hardness
Calcium
Magnesium
Methyl Orange Al k
Bromide
(USGS Lab Ark)
Iodide
Fl our i de
Potass i urn
Phenol s
(EPA Lab Ada)
Iron
Barium
Stronti urn
Zinc
Lead
Manganese
Nickel
Aluminum
Boron
Brine
0
111 ,000
49,000
7.0
1
84,770
84,750
20
15,400
12,200
4,880
778
110
570
2.8
2.2
160
00
21 .0
87.0
256
0.13
5.8
5.0
1 .0
2.0
19.8
Well
#2
20 ft
Well
#3
650 ft
••— Directly in 1
111 ,000
48,000
6.5
1
rn i
79,728
78,622
1 ,106
14,800
11 ,100
4,440
889
71
560
7.4
2.2
140
58.0
76.0
248
8.4
5.8
6.0
1.1
1 .5
20.4
111 ,000
47,500
6.6
Mel 1
111
3000 ft
i ne — ••
33,300
13,500
6.6
1 200
oride Interferen
80,313 21,784
78,609
1 ,703
15,700
11 ,200
4,480
1 ,094
103
610
8.0
1 .8
150
60.0
48.0
246
4.2
5.9
7.0
1.0
1 .7
20.6
21 ,594
190
8,200
5,300
2,120
705
328
162
2.7
0.9
20
8.4
1 .0
25
1 .2
2.5
3.6
0.4
1 .3
0.3
Well
120
36600 ft
Offset
12,200
3,700
6.9
150
6,877
6,853
24
3,700
2,540
1 ,016
282
383
34
.8
0.2
8.7
4.5
1 .0
3
0.86
0.8
2.6
0.2
0.90
0.3
Private
Well
12,000 ft
Offset
333
3.5
7.0
1
1 .7
207
194
13
162
102
41
15
120
0.81
.0
0.0
0.8
1.45
0.12
0.02
0.06
0.1
0.02
0.04
0.10
0.2
36
-------
hi m - S
**»»
» *» £ t » »»
* » * £
CHLORIDE
* * *
i- i- i-
o
o
o
?"
Q
O
If)
V
g
•0
JO
—
.0
11"
o
^"
CD
—
CT>
o
(D
"""
^
OT
•n
-
CM if}
. KJ
n5
o
CD
00
*
o
CD
^_
•»
—mm.
0
OJ
1^
WW.
c?
h-
O
CM
~1°.
-, r
DISSOLVED
SOLIDS
CALCIUM
BROMIDE
001 DE
TLOURIDE
0
O V
LEAD
BARIUM
IRON
STRONTIUM
ZINC
MAGNESIUM
(—
«
— 1
•0
—1
fvj
fvj
O
O
V
o
"~
q
"1°
"No
o| y
o
(M
__
r~
_•
"!
crt —
1X1
CT5
d
-------
injection. This explanation applies particularly to iodide
which may be adsorbed by the sand. (2) The injected brine
may have contained higher concentrations of these elements
in the past than it now contains. Although no chemical
records are available, all of the brine has been derived from
the same general formations now producing; therefore, this
explanation is considered unlikely. (3) It should be noted
that the pH decreases significantly then increases again
with greater distance from the pit. The lower pH may dissolve
iron coating the sand grains in the alluvium resulting in the
increased iron concentrations. Carbonates in the alluvium
would also be dissolved by the lower pH which would explain
the observed increase in alkalinity (table 3) and the subse-
quent increase in pH. As the pH increases again toward the
outer fringes of the brine, the dissolved iron reprecipitates.
The initial decrease in pH is not explained, but it may be
related to complex changes in the bicarbonate-carbon dioxide
balance in the native water and the sodium-calcium ratio in
the injected brine. The samples were not analyzed in the
field; hence, the pH as determined in the laboratory may not
accurately reflect the field conditions. These complex
relationships may help explain the iron concentration pattern,
but it is difficult to include the similar zinc pattern
under this explanation because zinc would probably not be
found as a coating on tJie sand grains as would iron. The
adsorption theory may better explain the zinc concentrations.
It may be concluded from the comparative dilution of these
chemical parameters that few elements remain in solution in
the exact ratio of the chlorides and that considerable care
must be exercised in using ratios as a "finger printing" tool
for identifying brine sources. It is apparent that consider-
ably more research is needed in this area of geochemical
reactions of brines polluting fresh-water aquifers.
38
-------
SECTION VII
REHABILITATION METHODS
General
In reviewing possible rehabilitation methods, it is necessary
to consider the degree of rehabilitation desired. For instance,
it would be possible to contain the water in place thereby
preventing the future contamination of an additional 3 1/2
square miles of aquifer. On the other hand, by physically
removing the contaminated water, all of the aquifer could be
restored to beneficial use.
Another consideration affecting the selection of the rehabili-
tation method is the present and most probable future use of
the water and the attendant quality requirements. The water
could be used for some purposes, such as washing sand and
gravel, as is; and, even though the tail water would require
a disposal system, the income from such a beneficial use
could partially offset the cost of disposal. Another example
is treatment or blending of the water for irrigation use which
would be less critical, hence less costly, than treatment for
domestic use. Of all the possible treatment methods and
degrees of rehabilitation, only those that could possibly apply
to this project are discussed in this report.
Rehabilitation methods are herein grouped under four general
headings which are: (1) containment, (2) accelerated discharge,
(3) use, and (4) deep-well disposal. Under each of these
general categories both the technical feasibility and economics
are discussed for each of the methods which could apply to
this project. Emphasis is placed on discussion of those
methods that initially appeared most feasible even though most
of these methods are necessarily discarded for one reason or
another.
Contai nment
I3entp_ni_t>e W.aJJ_^ Rather than allowing the salty water to spread
and" move downstream thereby polluting over four times the area
now affected, it would be possible to contain it by constructing
an impermeable underground wall across the east, west and
downstream (south) sides. Such a bentonite cut-off wall would
cost on the order of $7,000,000 as estimated in the Phase I
report.
Accelerated Discharge
If the polluted water is allowed to remain, it will not only
move downstream and contaminate an additional 3 1/2 square miles
39
-------
of ground-water for an estimated 250 years but will also
eventually discharge into the Red River under natural
ground-water flow conditions. The objectives of an accel-
erated discharge rehabilitation system would be: (1) to
limit the time period that the pollution is present, and
(2) to limit the area affected. Any method of getting the
polluted water into the Red River faster than will occur
under natural conditions could be considered accelerated
discharge. One such method is to force the polluted ground
water to the river underground, and another method is to
pump the water to the river above ground. Both of these
methods are herein discussed but with emphasis on above
ground discharge.
W.aJLeL £rj_ve_: This method would employ recharge wells posi-
tioned west of the polluted zone through which imported fresh
water would be pumped. This "water drive" would cause the
polluted water to move eastward -through the aquifer and dis-
charge into the Red River in much less time than will be
required under the natural ground-water flow rate. The result
would be faster removal of the polluted water, and pollution
under less area than will occur under natural conditions.
Of course the quality of water in the Red River would deteri-
orate somewhat as the salty ground-water flows into it.
Cost of such a system was estimated to be $1,264,000 in the
Phase I report.
P.umpjjig_ t^o_R£d__Ri_ve_rj_ A system whereby the polluting brine
would be pumped from the aquifer and discharged into the
Red River is discussed in detail in the Phase I report. Although
it was recognized that discharging a pollutant into a surface
stream is against established policy, it was argued that 'the
discharge would be regulated so that the chloride content of
the Red River would never be increased by more than 10 mg/1 -
an insignificant amount to pay considering the benefit derived
from rehabilitating the aquifer. Surface discharge has been
disallowed, but the technical considerations are herein summa-
rized for comparison with the alternate methods.
Examination of figure 3, Daily Flow Duration Curves - Red River,
shows that 50 percent of the time the flow exceeds 10,000 cfs
in the Red River. Figure 4 shows that although the chloride
content of the river varies rather widely as a function of flow,
a general relationship is present as shown by the average
chloride - flow curve. Figure 13 shows the average monthly
discharge which would affect the timing of brine discharge.
These data establish the physical parameters of the Red River
which affect its ability to receive additional salty water.
Because the rate of discharge into the Red River is not as
limiting a factor as the rate of injection into a disposal well
system, the pumping rate for removing salt water from the
aquifer can be relatively high. A practical rate would be 300 gpm
40
-------
o
a
MEAN MONTHLY DISCHARGE OF RED RIVER
AT FULTOH.ABKAHtAS
AOATIEO nmi smuMEL
t±L
m
-H-f-
d_L 44-U
1 111:
H-
Irrl:
-r-H
lifa
TB±]
tttfl
^yrfl
, !
t i I
iii^
is
y-±j
OCT. MOV. OCC. JAM. FEB. MA*. AMI. MAT JU
MOUTHS
JULY AU*. KPT.
41
-------
to be pumped from each of four wells so that the discharge
rate to the river would be 1,200 gpm (2.67 cfs). At this
rate of pumping it would only take 2.8 years to pump out the
5,400 acre feet estimated amount to be removed. It is some-
what arbitrarily estimated that three times the volume of
polluted water will have to be removed to effectively flush
the aquifer. The wells would be positioned as shown in
figure 14, which also shows the water-level contours and
flow directions that would result from the pumping.
The chloride content of the pumped water is, of course, an
important factor in determining the resulting chloride
increase of the river. Generally, the chloride content would
be close to 40,000 mg/1 initially but would decrease probably
within a few months to perhaps 10,000 mg/1 and continue to
decrease even more gradually throughout the pumping period.
It would be necessary to monitor the chloride content of the
pumped water and regulate the discharge as necessary depending
on river stage in order to avoid a mixed water chloride
increase greater than some limit such as 10 mg/1.
Figure 15 is presented as a graphical aid in determining that
minimum river flow at which it would be safe to discharge
1200 gpm of water with a known (by observation) chloride
content and not increase the chloride content of the mixed
water by more than 10 mg/1. For instance, figure 15 shows
that if the observed chloride content is 40,000 mg/1 in the
water pumped from the aquifer, then the river flow would have
to equal or exceed 11,000 cfs to stay within the 10 mg/1 limit
increase. Because the river flow exceeds 10,000 cfs only
about half the time according to figure 3, the discharge of the
alluvial salt water should be started in about February, accord-
Ing to figure 13, in order to discharge the most concentrated
water during the period of maximum river flow. By monitoring
both the river flow and the chloride content of the produced
water it would be possible to stop pumping at any time that the
chloride increase would exceed 10 mg/1; however, it is doubtful
that such curtailment to pumping would be necessary.
Costs for construction and operation of the Red River discharge
system were estimated in the Phase I report. These costs are
summarized as follows:
Estimated Costs - Red RiverDisposalSystem
Wells, pumps and power $ 39,000
Pipeline from wells to river 69,000
Maintenance 10,000
Personnel, supervision and operation 6 3 ,000
$181 ,000
42
-------
SCALE IN FEET
Pumping Well And
Cone Of Depression
Flow Line
FIGURE 14
CALCULATED WATER TABLE CONTOURS
AROUND FOUR PROPOSED PRODUCTION WELLS
PUMPING 300 GPM FOR 100 DAYS
ADAPTED FROM STRAMEL
43
-------
BASED ON EQUATION
Frt F.)
WHERE: cm= mfl/l CHLORIDES IN WELL WATER
Cr » mo/1 CHLORIDES IN RIVER WATER
F. = FLOW FROM WELLS =2.67 CFS
f, * FLOW IN RIVER
m
Ml
I
BASED ON <
I. DISCHARGE-CHLORIDE RELATIONSHIP IN
RED RIVER AT FULTON , FIGURE 4
2. CONSTANT WELL DISCHARGE OF 2.67 CFS
EXAMPLE"
WHEN WELL WATER CONTAINS IO.OOO mg/l
CHLORIDES. THE RED RIVER FLOW MUST EQUAL
OR EXCEED 26OO CFS IN ORDER TO LIMIT THE
INCREASE IN RIVER CHLORIDES TO IOaig/1.
111
i
H{
nt
or WATCH mom MCHAIILITATION
FIGURE 15 «tVISED fROU ST"AMEL
TO I0.g/l AT ANY PUMPED WATER CONCENTRATION
44
-------
Although disposal into the Red River is technically sound and
less costly than other methods, it is not permissible because
of its violation, however slight, of the very principle it is
intended to enhance. The final justification for the rehabil-
itation of this brine-polluted aquifer is economic gain through
agricultural production. Similarly the production of oil, and
its attendant brine waste problem, is for economic benefit. In
neither case should further pollution be allowed.
Use
Rehabilitating the aquifer by pumping the water out and putting
it to some beneficial use would be an ideal solution, and in
some cases might be practical. Three possible uses which could
apply to this project are described, each use requiring a
different level of treatment.
Se£onda/y_ R.ecpv_e.ry^ Oil fields are often repressurized by
water to increase oil production. This water flooding operation
sometimes requires fairly large quantities of makeup water in
addition to the water produced with the oil. If the polluted
water in the aquifer could be used for this purpose, no treat-
ment would be necessary. Inquiries were made to all the nearby
oil field operators; however, none expressed a need for addi-
tional water. If such a need existed, the cost for a pumping
system to deliver the polluted alluvial water would be about
$80,000 plus the cost of the pipeline to the point of use and
power costs which could be paid for by the user.
The rehabilitation approach has been successful in other areas,
according to McMillion, and has threefold benefits in that the
aquifer is reclaimed, the contaminated water is used benefi-
cially, and fresh water that may have been used for water
flooding is available for other purposes. This ideal rehabili-
tation method should be actively pursued whenever possible.
djjHj f£r_I_r_r.iga.tj_on : An intermediate degree of treatment
is dilutTon of the higK chloride water with unpolluted fresh
ground water which could enable the blend to be used for irriga
tion. The clayey soil in the river bottom where the irrigation
would take place is a limiting factor, however. A high sodium
content in irrigation water tends to cause clayey soils to
become very tight and difficult to farm. A blended water
containing only 1,000 mg/1 of chlorides would have a calculated
SAR (sodium adsorption ratio) value of 4.4 and a conductivity
of about 5,000 mmhos. This combination establishes a medium
salinity hazard which, for the clayey soils involved, would be
the maximum recommended limit.
The ratio of fresh water to salty water in order to limit the
chlorides to 1,000 mg/1 in the resulting blend is 40.2 to 1
when the salty water contains 40,000 mg/1 chlorides, and 9.3
45
-------
to 1 when the salty water contains 10,000 mg/1 chlorides.
For example, if 1,200 gpm (capacity of well that was destroyed)
were being used for irrigation, then only 30 gpm could be
pumped from the salty part of the aquifer when the chlorides
were in the 40,000 mg/1 range.
In order for a blending-use system to be effective in prevent-
ing the further spread of the polluted water, 323 acre-feet per
year (average 200 gpm) of salty water would have to be pumped
out. At the 10,000 mg/1 chloride range, this would require
3,000 acre-feet of fresh water for dilution resulting in
3,323 acre-feet of blended water per year for irrigation. If
rice could be grown with this water, then about 1,100 acres
could be irrigated. However, if more salt-tolerant crops such
as bermuda grass or cotton were grown, then at least 1,700
acres under irrigation would be needed. In practice this
would require exchanging both good and bad quality water
between several farms. Farmers owning land overlying good
quality water could object to using the blended water with the
resultant increase in costs caused by the salt build-up and
other harmful results to their soil and crops.
Furthermore, such a s-ystem in effect is pollution of the
blended fresh water to some degree by the addition of the salty
ground-water and is analogous to pumping the water to the
Red River. In fact, the results would be even less desirable
and the system much more difficult to administer than pumping
directly to the river. The cost of a well, pumping system,
power, and supervision is estimated to be about $200,000 plus
pipeline costs of about $100,000 to various irrigation users
for a total of $300,000.
._i rn z,atj CHIJ_ The final degree of treatment of the polluted
water to permit its direct use would be removal of all the
contaminants. If there were a sufficiently dire need for the
water, desalinization would be considered. It was estimated
in the Phase I report that construction and operation of a
desalinization plant would cost about $2,000,000.
Deep Well Disposal
_: An apparently attractive solution is to dispose of
the polluting brine into deep formations already containing
salt water. In examining the technical considerations involved
in pumping the polluted water out of the fresh-water aquifer
and disposing of it through disposal wells, there are two limit-
ing and opposing factors. The rate of pumping out of the
aquifer should be maximized in order to effectively capture
all of the polluted water and not allow any of the water to
bypass the pumping well and to do this in the shortest time
possible. On the other hand, the rate of injection of the
water into the disposal zone should be minimized in order to
46
-------
operate under reasonable injection pressures. This problem
is resolved by establishing the smallest quantity that can
be pumped from the aquifer and be effective, and then design-
ing the disposal system to accommodate that quantity.
In order to establish the smallest effective quantity that
can be pumped, it is necessary to examine the hydraulic
properties of the aquifer. Figure 16 shows the shape and
extent of the cones of depression around a pumping well at
different rates of pumping under the hydrologic conditions
assumed for this aquifer. Under natural conditions the aquifer
is confined by the shallow clay layer; however, as the aquifer
is dewatered by pumping, the hydraulic characteristics change
from a confined to a water-table condition. That part of the
distance-drawdown curve below the confining layer will reflect
water-table characteristics (large values for the coefficient
of storage, S), whereas that part above the confining clay
will reflect confined (very low S value) characteristics. The
proportion of the curve below the clay is primarily a function
of the pumping rate and secondarily a function of time. The
theoretical distance-drawdown curves, shown on figure 16, take
into consideration this estimated resulting change in the
value of the coefficient of storage. The curves are intended
to show the most probable drawdown conditions under the known
and assumed aquifer parameters. Based on an examination of
these curves, the pumping rate of 200 gpm is (somewhat arbi-
trarily) selected as the minimum rate that will be effective
in pumping out all of the salt water. This pumping rate
causes about two feet of drawdown at a radius of 2,000 feet
which should result in effective movement of the water toward
the well from the perimeter of the polluted area.
Based on the hydraulic gradient and permeability, the natural
flow through a cross sectional area one mile long is about
30 gpm. Therefore, if a system of closely spaced wells were
installed all across the downstream side of the polluted area,
the absolute minimum pumping rate would be 30 gpm to capture
the polluted water. Such a system would have to be operated
for over 100 years, however, in order to drain all of the salt
water. Because fewer wells and higher pumping rates are more
practical, the production system chosen for deep-we 11 disposal
employs one pumping well for which the minimum pumping rate is
200 gpm.
The position of the production well relative to the direction
of movement of the main body of salt water and relative to the
configuration of the underlying shale surface is critical in
the successful removal of the polluted water. The well should
be located just downstream (south) from the main body of
polluted water, and the well should be located in a topographic
pwiiuucu n u u c. i 9 aiiu u 11 c rrcii OMVUIU L/C iwi-aucu in a u u |j u y i u jj 11 i
low of the shale surface in order to allow the high-density
salt water to flow toward the well. Furthermore, the producti
on
47
-------
00
10 100 1000
KADIUS FROM PUMPED WELL 1M FEET
—II II *"
v*-
MMM
•X4"
*x"^ *
*
S !
/
. ^
,x
;. — 0
•M
-«*
^
R
•<•>
X
^
"
-
»
x
AW
-"
!•-"
—
^
^
X
k__
DOWN
^SUMPTIONS
PE OF PUMPIN0 -
1
*•* **
f ,,
1
i
, f.--
i
!
-
_
CUR VE S
VARIABLE
3RAOE COEFFICIENT- VARIABLE
kNSMISSIVITY • 80,000 0PD '•
IE OF PUMPIN0 •
1 t
i
2 YEAR*
1
10,000
-------
well should be constructed using about five feet of hydrauli-
cally efficient well screen set as low as possible in perme-
able material just above the shale in order to concentrate
on the dense salt water. Additional test-hole drilling would
be recommended to locate the best position for the production
well if the rehabilitation should be carried out.
N^ac_at_och_Di_S£Os^al_ We ]_!_:_ The Nacatoch sand in this area is
non-oiT-bearing, contains salt water, and is sufficiently perme-
able to be used as a disposal zone. In fact, the salt-water
disposal well now in operation in this field is completed in
the Nacatoch at a depth of 1,466 to 1,496 feet. This well
disposes daily of about 1,500 barrels of brine in about 17 hours
of operation, which is an average pumping rate of 82 gpm. The
pressure developed by the pump ranges from 200 to 400 psi.
When the pressure gets high, a detergent is added to the brine
causing the required pressure to decrease. Because of mutual
pressure interference between wells and constant injection, it
is estimated that 50 gpm per well would be the maximum practi-
cal injection rate for a multi-well disposal system.
It has been estimated that the aquifer contains 14 million
barrels (1,800 acre-feet) of salty water. However, because of
lag time and the nonhomogeneity of the aquifer, it is estimated
that three times the original volume would have to be pumped
to effectively flush the salt out of the aquifer. This would
require nearly 17 years at the 200 gpm pumping rate.
Although only one production well would be required to produce
200 gpm, four disposal wells would be necessary in order to
inject the water within reasonable pressure limits. A system
could be designed using one production well in the center with
pipelines in four directions leading to the four disposal wells
1,000 feet away from the production well. Such a system would
space the disposal wells about 1,400 feet apart which, within
practical limits, would minimize mutual pressure interference.
In order to minimize costs, the system could be operated by a
single pump -- the production pump would develop the pressure
required for injection into the disposal wells.
Estimated costs for constructing and operating a rehabilitation
system using disposal wells completed in the Nacatoch formation
are summarized as follows:
Estimated Costs - Nacatoch Pisposal System
Four disposal wells, 1,500 ft deep $160,000
One production well 6,000
Pump and motor 6,000
Pipeline from production well to disposal wells. 16,000
Controls, valves, meters, etc 12,000
Contingencies 30 .000
Total Hardware $230,000
(Estimated Costs Continued)
49
-------
Estimated Costs (Continued)
Total Hardware (Previous Page) ........ $230,000
Design and construction supervision ............. 15,000
Operation and maintenance for 17 years .......... 102,000
Power costs for 17 years ........................ 1 03 .000
Total Estimated Cost .................. $450,000
Disposal into the Nacatoch, however, has disadvantages that
render this alternative undesirable. One disadvantage is the
presence of faults in the immediate area. These faults could
form relatively impermeable barriers to the lateral flow of
the injected brine thereby causing a gradual but highly
significant increase in the pressure required to inject the
required 200 gpm. If, on the other hand, the faults were
more permeable than the undisturbed formations, then the
fault planes could provide a conduit for the upward migration
of water injected into the pressurized Nacatoch formation
thereby creating another pollution hazard.
Another significant disadvantage is the presence of so many
oil wells and test wells drilled within 1 1/2 mile radius.
Of the 46 holes drilled, none set surface casing below the
Carrizo sand - the lt>west fresh water. Furthermore, accord-
ing to Oil and Gas Commission records, of the 31 wells that
have been plugged, only in 8 wells was casing left in the
hole extending below the Carrizo. These old holes could
allow brine to migrate from the Nacatoch into any or all of
the fresh water-bearing formations, if the pressure in the
Nacatoch were raised high enough and if some of the old holes
should form conduits. The Arkansas Oil and Gas Commission
has witnessed the plugging of all wells since 1961. Many of
these wells and test holes were plugged prior to that time,
however, and the adequacy of the plugs is questionable.
- £i.!LP£.siLl_J''e.l]_: The Smackover formation is predom-
inately a limestone of Jurrasic age (Vestal). This formation
underlies most of southern Arkansas and not only produces
considerable oil but also produces brine from which bromide
is extracted. In the northern part of the project area the
depth to the top of the Smackover is estimated to be between
8,200 to 8,700 feet. One mile south of the project area a
Smackover test well was drilled in the fall of 1971. This
well reached the top of the Smackover at 9,570 feet; however,
that location is both down-dip and on the downthrown side of
the fault. Although no oil was encountered in the well, the
drilling reports did indicate that there are high permeabil-
ities in the top 100 feet of the Smackover. The bottom-hole
pressure was not measured in this test hole, but based on
other data in the area, the static fluid level is expected to
be between 500 and 1,000 feet below land surface. This exist-
ing well could be re-entered and completed as a disposal well.
50
-------
The Ethyl Corporation at their bromide plant near Magnolia,
Arkansas (about 30 miles east of the project) produces brine
from the Smackover through 22 production wells and injects
about 195,000 bbl/day back into the formation through seven
disposal wells - an average of 815 gpm per well. In the
Magnolia area the static fluid level is about 3,000 feet
below ground level. Based on information developed from the
Ethyl Corporation records and the permeability estimated from
electric logs in the Smackover test well one mile south of the
project, it is estimated that the injection specific capacity
should be on the order of 0.2 gpm per foot of head increase.
This suggests, then, that a Smackover disposal well should
take 200 gpm under vacuum (no surface pressure required) if
the static fluid level is 1,000 feet below ground level. On
the other hand, 500 feet (217 psi) of surface pressure would
be required to inject 200 gpm if the static fluid level is
only 500 feet below land surface.
The experience of operators in the area indicates that there
should not be an excessive problem of plugging of the formation
due to mixing incompatible waters provided that a buffer zone
is established and that the injected water is not exposed to
air.
Two possible systems, one using the existing test well and the
other drilling a new well, could be used to dispose of the
polluting brine into the Smackover formation. Estimated costs
for each system considering injection pressure requirements
are presented for comparison.
Estimated Costs - Smackover Disposal Well System
Using Existing Uel1 Located One Mile South of Project Site
Disposal well (casing, tubing and other
conversion costs) $130,000
Production well 6,000
Pipeline 34,000
Pump (combined production and injection) 5,000
Controls and fittings 10,000
Contingencies 20,000
Total Hardware Costs $205,000
Design and supervision 15,000
Operation and maintenance 102,000
Power costs for 17 years 83 ,000
Total (assuming injection pressure required)... $405,000
The above cost estimate assumes that the static fluid level is
500 feet below the surface and that 217 psi injection pressure
will be required to dispose of the 200 gpm. If the static
fluid level is 1,000 feet below land surface and no injection
pressure is required, then power costs can be reduced substan-
tially and the pipeline cost can reflect less pressure require-
ments, thereby reducing the total cost to $337,000.
51
-------
Estimated Costs - Smackover Disposal Well System
Drilling a New Well Near the Production Well
Disposal well $180,000
Production well 5 QOO
Syphon system !...'.'.'!! 4,'oOO
Controls and fittings 3^000
Contingencies '..'.'. 17 [oOO
Total Hardware Costs [ $210 ]000
Design and supervision 15,'000
Operation and maintenance 65,'oOO
Total (assuming no injection pressure required) $290*000
This design is based on the assumption that the static fluid
level is on the order of 1,000 feet below ground level and
that the 200 gpm injection rate would not require surface
pressure. Under these circumstances, a syphon system could be
used to transfer the water from the production well into the
disposal well, thereby eliminating power costs.
If this system were to be constructed, the disposal well would
be constructed first. By conducting injection tests, the
rate at which the we 1.1 would receive water under vacuum could
be determined. The production well system would then be
designed to furnish water at the rate of acceptance of the
disposal well. If this rate should be less than 200 gpm (but
greater than 100 gpm), then more than one production well
would be necessary in order to spread the pumping over a wider
area so that none of the polluted water would flow past the
production system.
If the rate of injection under gravity should be less than
100 gpm, pressure injection would be required in order to
dispose of the water in a reasonable time and to insure against
salty water passing the collection system. Under these circum-
stances requiring a pump and power consumption, the total cost
of a new Smackover well disposal system is estimated to be
$375,000.
The well design used for all disposal-wel1 cost estimates
incorporates important safeguards against further pollution.
First, surface casing should be set and cemented below the
lowest fresh water. Then, the production casing should be set
to the disposal zone and cemented. Next, the injection tubing
should be set with a packer just above the disposal zone. The
annulus between the tubing and production casing should be
filled with a non-corrosive fluid, and the salt water would be
pumped through the tubing. If a leak should develop in the
tubing or packer, it would be detected immediately by the change
in pressure in the annular fluid, and injection would be stopped
until the leak was repaired. The safeguard supplied by using
tubing and a fluid-filled annulus is not now required by the
State of Arkansas.
52
-------
There is no oil or gas production from the Smackover formation
closer than four miles, and none of the nearby existing wells
penetrated to that depth. Therefore, there are no technical
disadvantages or dangers connected with using the Smackover as
a disposal zone. The only disadvantage is its great depth and
the resulting high cost of a disposal well.
53
-------
SECTION VIII
BENEFIT-COST RATIOS
In the preceding sections the technical details of this
pollution incident and the possible rehabilitation methods
have been discussed. Finally, it is necessary to decide
which rehabilitation method is most feasible and whether or
not that method is justified. The most direct means of
evaluating the feasibility of this type of project is an
economic evaluation based on the benefit-cost ratio. The
determination of costs is fairly straightforward, but the
determination of the money value of benefits is more difficult
and subject to varying approaches. The following discussion
is hopefully explained in sufficient detail to allow the
reader to follow the calculations step by step and form his
own opinion as to their appropriateness.
As stated in Section V, the long-term value of water contam-
inated ranges from $3,200,000 to $18,250,000 based on present-day
crop values and potential use. In order to grow rice, for
instance, water is mandatory, whereas for cotton and other crops
water just adds to the quantity that can be harvested per acre.
In order to establish reasonable benefit-cost ratios for this
report, the difference in profit between irrigated and nonirri-
gated cotton is chosen as a realistic use. It is further
stated in Section V that the pollution will eventually affect
4 1/2 square miles; however, to be conservative it is assumed
that only two square miles will be polluted at any one time
throughout the 250-year natural flushing period.
Using the above assumptions, it is possible to arrive at the
annual difference in profit (not total income) that may be
gained if the aquifer is rehabilitated and used to irrigate
cotton. At $35 per acre difference in profit over 1,280 acres
(2 square miles) the difference in profit is $44,800.
In order to compare benefits spread over 250 years and costs,
part of which are spread over 17 years, it is necessary to
reduce all values to present worth. This is accomplished by
using the following uniform series present worth factor equation
from Taylor.
P = R
Where P = Present worth
R = Annual payment (or income)
i = Interest = 6 percent for all calculations
n = Number of years
55
-------
The present worth (?) is defined as that amount of money
deposited now at (i) interest rate in order to withdraw
(R) dollars annually as payment for benefits or costs for
(n) years. For instance, the present worth of $44,800 per
year for 250 years (same for infinite time) is $746,680; or
in other words, if $746,680 were deposited at six percent
compounded interest, it would be possible to withdraw $44,800
per year for 250 years.
The $746,680 is the present worth of added profit that would
accrue to the few landowners involved and is used to establish
the private benefit-cost ratio. In order to relate this to
public benefit, it is assumed that 25 percent of this added
profit would revert to the public as taxes. Therefore, the
public benefit is $186,670, which is used for establishing the
public benefit-cost ratio. These values ignore the future
benefit resulting from preventing discharge of the salty water
into the Red River and possible future attendant costs. The
public benefit further assumes that irrigation of the full
two square miles would actually be done by the private land-
owners and that cotton would be grown. Rice crops would
increase the annual profit considerably over dry-land cotton,
hence would increase both the private and public benefit values.
For calculating the worth of rehabilitation costs, it is
assumed that the operating and power portions will be spread
uniformly over the life of the rehabilitation project. These
cost factors are reduced to present worth and added to the
initial investment to arrive at the total present worth cost.
Table 4 summarizes the rehabilitation methods discussed and
shows the private and public benefit-cost ratios based on
present worth for each method4 Examination of this table
shows that disposal into the Smackover formation is the least
expensive method that is both technically feasible and permis-
sible under established policy. Construction of the new well
in the project area, as opposed to using the existing test hole,
offers the most advantageous range of benefit-cost ratios.
This method would be recommended if the project were to continue.
In considering the negative public benefit-cost ratio, however,
and the assumed higher priority for funds for preventive measures
rehabilitation of the aquifer does not appear economically
justified at this time.
56
-------
TABLE 4
SUMMARY AND BENEFIT-COST RATIOS OF REHABILITATION METHODS
Method
Years of Total Present-Worth
Operation Cost Cost
Benefit-Cost Ratio
Private Public
Remarks
1 . Containment
Must be
A. Bentonite Wall maintained $7,000,000 $7,000,000
forever
2. Accel crated Discharge
A"'." Water Drive 10 years $1,264,000 $1,194,306
B. Pumping to
Red River
3. Use
3 years $ 181,000 $ 17ff,749
A. Secondary Recovery 10 years $ 80,000 $ 80,000
B. Blending for Irrig. 17 years $ 300,000 $ 258,528
C. Desalinization
17 years $2,000,000 $1,692,410
4. Deep Wei 1 Disposal
ST Nacatoch Formation 17 years $ 450,000 $ 371,247
B. Smackover Formation
(1) Existing well
(using pressure)
(2) Existing well
(gravity flow, no
pressure)
(3) New well
(using pressure)
(4) New well
(gravity flow, no
pressure)
17 years $ 405,000 $ 333,727
17 years $ 337,000 $ 288,634
17 years $ 375,000 $ 320,854
17 years $ 290,000 $ 265,022
0.1;1 0.03:1 Only restricts pollu-
tion and too costly.
0.6:1 0.15:1 Too costly, benefits
too low.
4.2:1 1.05:1 Contrary to policy.
9.3:1 2.3:1 No market for water use.
2.9:1 0.7:1 Probably not acceptable
to users.
0.4:1 0.1:1 Too costly, possible
uses don't justify.
2:1 0.5:1 Not recommended because
of danger of further
pollutlon.
2.2:1 0.55:1 Technically feasible
and acceptable, not
2.6:1 0.65:1 as economical as new
well .
2.3:1 0.57:1 Technically feasible
and acceptable, hlgh-
2.8:1 0.7:1 est benefit-cost ratio
of feasible methods.
-------
SECTION IX
ACKNOWLEDGMENTS
The support and continuing interest of Mr. Leslie G. McMillion,
EPA Grant Project Officer, is especially noteworthy. Mr.
McMillion has been instrumental throughout the project in
guiding the work to a conclusion.
This report was prepared by John S. Fryberger, Ground-Water
Geologist, with the assistance of John H. Marsh, Civil and
Sanitary Engineer, partners in the firm of Engineering Enter-
prises, Norman, Oklahoma. The work was performed under contract
to the Arkansas Division of Soil and Water Resources as autho-
rized by EPA.
The Phase I report, from which considerable information was
used in this final report, was prepared by G. J. Stramel ,
Project Director for this project and Chief Engineer of the
Arkansas Division of Soil and Water Resources. Other personnel
of this Division contributed as follows: Keith Jackson had
overall responsibility for the project; Roy Smith supervised
the initial test drilling, and A. J. Bryniarskf, J. R. Young,
Al Nyitrai, and Larry White performed the field work and water
sampling. Dr. Leslie Mack, former Governor's Advisor on Water
Resources, assisted in the early planning and in efforts to
obtain the EPA Grant.
Other Arkansas state agencies have also contributed significant-
ly. The Oil and Gas Commission took part in the initial inves-
tigation and provided records on the oil fields and brine
disposal. The Pollution Control Commission has analyzed most
of the water samples. And the Geological Commission has pro-
vided technical assistance. In addition, the U. S. Geological
Survey office in Arkansas provided considerable geologic and
hydrologic data including the results of their 1967 reconnais-
sance, and performed some of the chemical analyses for this
report.
Mr. Ernst P. Hall, Program Element Manager, and Dr. James
Shackelford, Project Manager, for the Office of Research and
Monitoring, EPA, were instrumental in approving the project
and revising the project's objectives. Mr. Jack Keeley, Chief,
National Ground Water Research Program of EPA, assisted in
review, and Mr. Wm. DePrater, Chemist, Robert S. Kerr Research
Center,_ EPA, and A. Gene Collins, Chemist, Bartlesville Petro-
leum Research Center, U. S. Bureau of Mines, provided valuable
"assistance in analyzing and interpreting the complex water
chemi stry.
59
-------
The cooperation of the landowners in the project area is
especially appreciated. Considerable assistance was rendered
by Mssrs. Linn Lowe, Glen Price, and Harold Tullos of Garland
City, Arkansas.
60
-------
SECTION X
REFERENCES
1. Collins, Gene A., "Oil and Gas Wells - Potential Polluters
of the Environment?" Journal Water Pollution Control
Federation, pp 2383-2393 (1971).
2. Davis, James W., and Collins, Gene A., "Solubility of
Barium and Strontium Sulfates in Strong Electrolyte
Solutions," Environmental Science and Technology, pp 1039-
1043 (1971).
3. Ludwig, A.M., "Water Resources of Hempstead, Lafayette,
Little River, Miller, and Nevada Counties, Arkansas,"
Geological Survey Water Supply Paper 1998 (in press).
4. McMillion, Leslie, G., "Ground-Water Reclamation by
Selective Pumping," Society of MiningEngineers, AIME,
transactions - Vol 250,pp 11-15 (1971).
5. Stramel, G.J., "Rehabilitation of a Brine-Polluted Aquifer.
Phase I Report," Arkansas Soil and Water Conservation
Commi ssion ,(T970).
6. Taylor, George A., "Managerial and Engineering Economy,"
D. Van Nostrand Co. , Inc. , (1964).
7. Vestal, Jack H., "Petroleum Geology of the Smackover
Formation of Southern Arkansas," Inform at ion C i re u1 a r 14,
Arkansas Geological Commission, 19 pp (1950).
' f. S. GOVERNMENT PRINTING OFFICE : t 972 —', 1 ti- ! 50 (127)
61
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
w
REHABILITATION OF A BRINE-POLLUTED AQUIFER,
Fryberger, John S.
Engineering Enterprises
under contract to
Arkansas Division of Soil & Water Resources
5,
5, Perforptitie Organization ,"
14020 DLN
}3. type o { Report and
Period Covered
, Spoking otf .motion
Protection Agency
Environmental Protection Agency report
number EPA-R2-72-COA, December 1972.
A detailed investigation was made of one (among several noted) incident
where a fresh-water aquifer has been polluted by accepted disposal of oil-
field brine through an "evaporation" pit (an unlined earthen pit) and
later a faulty disposal well. The present extent of the brine pollution
is one square mile, however it will spread to affect 4 1/2 square miles
and will remain for over 250 years before being flushed naturally into the
Red River. Detailed chemical analyses show changes in relative concentra-
tions of constituents as the brine moves through the aquifer.
Several rehabilitation methods are evaluated in detail, including controll
pumping to the Red River and deep-well disposal. None of the methods that
are both technically feasible and permissible show a positive public
benefit-cost ratio.
Although real economic damage both present and future results from this
brine pollution, rehabilitation is not now economically justified. The
report emphasizes that greater effort is needed to prevent such pollution,
which not only affects ground-water resources but also affects water
quality in interstate streams. (Fryberger-Engineering Enterprises)
ed
j-a De^nprors *Ground-water , *Water pollution, *Pollution abatement, *Brine
disposal, Water pollution sources, Water pollution control, Water pollu-
tion effects, Path of pollutants, Aquifers, Saline water — freshwater
interfaces, Arkansas hydrology, Water chemistry, Water conservation,
Waste water disposal .
*Aquifer rehabilitation, Red River, Disposal wells, Disposal pits
05B
Send To:
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
US DEPARTMENT OF THE INTERIOR
WASHINGTON. D C. ZO24O
John S. Fryberger
Engineering Enterprises
------- |