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

-------