x>EPA
            United States
            Environmental Protection
            Agency
            Robert S. Kerr Environmental
            Research Laboratory
            Ada, OK 74820
EPA/600/2-89/038
July 1989
            Research and Development
Development of a
Methodology for Regional
Evaluation of Confining Bed
Integrity

-------
                                          EPA/600/2-89/038
                                          July 1989
    DEVELOPMENT OF A METHODOLOGY FOR REGIONAL
      EVALUATION OF CONFINING BED INTEGRITY
                       by
                 Gary F. Stewart
               Wayne A. Pettyjohn
           Oklahoma State University
          Stillwater, Oklahoma  74078
        Cooperative Agreement CR-814061
                Project Officer

                Jerry Thornhill
       Applications and Assistance Branch
Robert S. Kerr Environmental Research Laboratory
              Ada, Oklahoma  74820
ROBERT S. KERR ENVIRONMENTAL RESEARCH LABORATORY
       OFFICE OF RESEARCH AND DEVELOPMENT
     U. S. ENVIRONMENTAL PROTECTION AGENCY
              ADA, OKLAHOMA 74820

-------
                          DISCLAIMER
    The information in this document has been funded wholly or
in part by the United States Environmental Protection Agency
under cooperative agreement CR-814061 to Oklahoma State
University.  The report has been subjected to the Agency's peer
and administrative review, and has been approved for
publication as an EPA document.  Mention of trade names or
commercial products does not constitute endorsement or
recommendation for use.
                                 ii

-------
                          FOREWORD

     The Environmental Protection Agency was established to
coordinate administration of the major Federal programs
designed to protect the quality of our environment.

     An important part of the Agency's effort involves the
search for information about environmental problems, management
techniques and new technologies through which optimum use of
the Nation's land and water resources can be assured and the
threat pollution poses to the welfare of the American people
can be minimized.

     EPA's Office of Research and Development conducts this
search through a nationwide network of research facilities.

     As one of the facilities, the Robert S. Kerr Environmental
Research Laboratory is the Agency's center of expertise for
investigation of the soil and subsurface environment.
Personnel at the laboratory are responsible for management of
research programs to; (a) determine the fate, transport and
transformation rates of pollutants in the soil, the unsaturated
zone and the saturated zones of the subsurface environment; (b)
define the processes to be used in characterizing the soil and
subsurface environment as a receptor of pollutants; (c) develop
techniques for predicting the effect of pollutants on ground
water, soil and indigenous organisms; and (d) define and
demonstrate the applicability and limitations of using natural
processes, indigenous to the soil and subsurface environment,
for the protection of this resource.

     This report contributes to that knowledge which is
essential in order for EPA to establish and enforce pollution
control standards which are reasonable, cost effective and
provide adequate environmental protection for the American
public.                           />
                              Clinton W. Hall, Director
                              Robert S. Kerr Environmental
                                Research Laboratory
                                111

-------
                                ABSTRACT
     For safe underground injection of liquid waste, confining
formations must be thick, extensive, and have low permeability.
Recognition of faults that extend from the potential injection zone to
underground sources of drinking water is critical for evaluation of
confining-bed integrity.

     Nonproprietary geologic information from ordinary sources can be
used to map localities suspected to be injection-sensitive.  Materials
include remote-sensing imagery, aerial photographs, surface-geologic
maps, and subsurface-geologic maps; structural geologic maps, thickness
maps and initial-production maps are useful.  Persons with limited
experience can use data bases and computer mapping to generate well-
suited information, but input by experienced geologists is necessary.

     In the Southwest Enid Area and West Edmond Field, Oklahoma, rocks
at the surface reveal little evidence of subsurface faulting.  Oil
reservoirs are "tight", fractured limestones.  Cumulative-production
mapping (Southwest Enid), initial-production mapping (West Edmond),
structural contour mapping, and analysis of lineaments from stream
patterns (Southwest Enid) and Landsat imagery (West Edmond) were
combined.  Production at Southwest Enid Area seems to be correlated with
intersections of lineaments.  At West Edmond Field, lineaments suggested
penetrative fractures; an injection-sensitivity map was based on
inferences from lineaments and subsurface mapping.

     The sandstone reservoir at Burbank Field, Oklahoma, almost
certainly is jointed systematically.  Structural contour maps and
initial-production maps do not show the fracture system.  Faults are few
or are of little displacement; geometry of the channel-fill reservoir
influences production more so than natural fractures.  Fitts Pool,
Oklahoma, is in a complexly faulted graben.  Areas near bounding faults
were interpreted as injection-sensitive.  Faults in thick shales that
seal Fitts Pool are suggested by the many lineaments shown on satellite
imagery; the faults are believed to be closed.  Confining beds probably
would be effective if fluid injected did not exceed volumes withdrawn,
and injection pressures were below original formation pressure.

     Injection-sensitivity maps are for precautionary purposes.  They
are designed to protect underground drinking water, not to obstruct
responsible and conscientious production of oil and gas.

     This report was submitted in fulfillment of Cooperative Agreement
No. CR-814061 by Oklahoma State University under the sponsorship of the
U.S. Environmental Protection Agency.  This report covers a period from
July 1, 1987 to February 28, 1989.
                                       IV

-------
                           CONTENTS
Foreword	iii

Abstract	iv

Figures	vii

Tables	xi

Section 1.   Introduction 	   1
              Injection wells	   2
              Faults, joints, lineaments, fracture traces.   4
              Scope and approach of study	   6
              Methods of investigation 	   7
Section 2.   Conclusions	11
Section 3.   Recommendations. .  	  14
Section 4.   Southwest Enid Area	15
              Introduction 	  15
              Mapping of lineaments	22
              Subsurface geology 	  22
              Conclusions	31
Section 5.   West Edmond Oil Field	36
              Introduction 	  36
              Surface geology	40
              Subsurface geology 	  40
                Stratigraphy 	  40
                  Permian System 	  43
                    Hennessey Shale	43
                    Garber-Wellington Formations 	  43
                  Pennsylvanian System 	  45
                  Mississippian System 	  45
                  Devonian and Silurian Systems	45
                    Frisco and Bois d'Arc Formations ...  45
                    Haragan and Henryhouse Formations. .  .  45
                    Chimneyhill Subgroup 	  47
                  Ordovician System	47
                Abbreviated structural-geologic history.  .  47
                Interpretation of subsurface geology ...  47
                  Production-trend mapping 	  48
                  Structural contour mapping 	  50
              Mapping of lineaments	53
              Injection-sensitivity map  	  57
              Conclusions	65
Section 4.   Burbank Oil Field	67
              Introduction 	  67

-------
              Mapping of subsurface geology by computer.  .   67
                Structural geology 	   81
              Recognition of fracture trends 	   82
                Subsurface-mapping techniques	82
                  Initial-potential maps 	   82
                  Reservoir-thickness maps ........   86
                  Structural geologic maps 	   86
                  Conclusions	87
              Surface-mapping techniques 	   88
                Conclusions	88
              Injection-sensitivity map	   90
              General conclusions	92
Section 5.  Fitts Pool,  Pontotoc County	93
              Introduction 	   93
              Subsurface geology 	   93
              Mapping of fracture traces and lineaments.  .   97
              Injection-sensitivity map	  106
              Conclusions	106
Selected references	110
                                vi

-------
                            FIGURES


Number                                                     Page

    1. Locations of study areas	    9

    2. Tectonic features of Midcontinent, generalized,
       showing position of Sooner Trend	   16

    3. Location of Southwest Enid study area	   17

    4. Stream drainage patterns, Southwest Enid Area  ...   18

    5. Surface geologic map, Southwest Enid Area 	   19

    6. Locations of oil wells, gas wells and dry holes,
       Southwest Enid Area	   21

    7. Stream-lineament map, Southwest Enid Area 	   23

    8. Generalized lineament map, Southwest Enid Area. .  .   24

    9. Intersections of generalized lineaments,
       Southwest Enid Area	   25

   10. Structural geology,  top of Meramecian-Osagean  rocks,
       Southwest Enid Area	   27

   11. Structural geology,  top of Woodford Shale,
       Southwest Enid Area	   28

   12. Thickness map, top of Meramecian rocks to top
       of Woodford Shale, Southwest Enid Area	   30

   13. Contour map, oil-equivalent production, wells
       completed before 1977, Southwest Enid Area	   32

   14. Areas within which oil-equivalent production
       exceeds 100,000 and  250,000 barrels per well,
       Southwest Enid Area	   33

   15. Convergence, areas of uncommonly large
       production and areas of intersection of
       lineaments, Southwest Enid Area	   34

   16. Type log, West Edmond Field,
       rock-stratigraphic units from Woodford
       Shale to Simpson Group	   37
                                vii

-------
17.  Paleogeologic-structural geologic map,
    Oklahoma City Anticline and West Edmond Field ...   38

18.  Cross-section, central part of West
    Edmond Field	   39

19.  Pre-Pennsylvanian paleogeologic map,  central
    Oklahoma; location of West Edmond Field relative
    to Oklahoma City Anticline	   41

20.  Generalized surface geology, West Edmond
    Field and nearby areas	   42

21.  Type log, West Edmond Field; rock-stratigraphic
    units from base of surface casing to
    uppermost part, Pennsylvanian System	   44

22.  Type log, West Edmond Field; rock-stratigraphic
    units from "Oswego Lime" to Chimneyhill Subgroup
    Subgroup	   46

23.  Production-trend map, central part,
    West Edmond Field	   49

24.  Structural geology, top of Hunton Group,
    central part, West Edmond Field 	   51

25.  Structural geology, base of Hunton Group,
    central part, West Edmond Field 	   52

26.  Structural geology, "base-of-Permian" marker
    bed, central part, West Edmond Field	   54

27.  Lineaments interpreted from Landsat imagery,
    central part, West Edmond Field 	   55

28.  Lineaments interpreted from color-infrared
    imagery, central part, West Edmond Field	   56

29.  Thickness, Frisco-Bois d'Arc Formations,
    West Edmond Field	   60

30.  Thickness, total Hunton Group, West
    Edmond Field	   61

31.  Thickness, Woodford Shale, West Edmond
    Field	   62
                             viii

-------
32. Injection-sensitivity map, West Edmond
    Field	   64

33. General configuration, Burbank Field	   68

34. Type electric log, Burbank Field	   69

35. Locations of wells for data base,  Burbank
    Field	   70

36. Structural geology, Cottage Grove Sandstone,
    Burbank Field 	   71

37. Structural geology, Pink Limestone
    Burbank Field 	   72

38. Thickness of interval, Cottage Grove Sandstone
    to Pink Limestone, Burbank Field	   73

39. Thickness of Cottage Grove Sandstone,
    Burbank Field	   74

40. Thickness of net sandstone in Cottage Grove
    Sandstone, Burbank Field	   75

41. Thickness of net shale in Cottage Grove
    Sandstone, Burbank Field	   76

42. Thickness of confining unit above Cottage
    Grove Sandstone, Burbank Field	   77

43. Cumulative thickness of shale above Cottage
    Grove Sandstone, Burbank Field	   78

44. Number of shale "breaks" in stratigraphic
    section above Cottage Grove Sandstone,
    Burbank Field 	   79

45. Possible injection zones between Cottage
    Grove Sandstone and depth of about 1000 ft
    (305 m) , Burbank Field	   80

46. Initial-potential map, Burbank Sandstone,
    Burbank Field 	   83

47. Thickness of effective reservoir rocks,
    Burbank Sandstone 	   84

48. Lineaments and joint clusters, T.  26 N.,
    R. 6 E., Burbank Field	   89
                             IX

-------
49.  Injection-sensitivity map, T. 26 N.,
    R. 6 E., Burbank Field.	   91

50.  Generalized map of geologic provinces,
    Pontotoc County 	   94

51.  Structural contour map of Viola Limestone,
    Fitts Pool and nearby areas	   96

52.  Thickness of Frisco-Bois d'Arc Formations,
    Fitts Pool	   98

53.  Initial-potential production, Hunton Group,
    Fitts Pool	   99

54.  Topographic map, east-central part,
    T. 2 N., R. 7 E	100

55.  Lineament and fracture traces in
    sandstone of Simpson Group	101

56.  Lineament in upland terrain, Arbuckle
    Group	102

57.  Lineament as swale in grassland	103

58.  Lineament as poorly defined swale
    in grassland	104

59.  Satellite imagery, Franks Graben, with
    lineaments	105

60.  Injection-sensitivity map, Franks Graben	107

61.  Lineaments in Franks Graben  	  108

-------
                         TABLES
Number                                                 Page

1. Example, convergent evidence of faulting,
   West Edmond Field	59

2. Rock-stratigraphic units, eastern
   Pontotoc County	95
                             xi

-------
                          SECTION 1

                         INTRODUCTION
     The Underground Injection Control (UIC) program of the
U.S. Environmental Protection Agency (EPA) is involved with the
safe emplacement of liquid wastes in subsurface geologic
formations.  Of particular concern are hazardous toxic wastes
injected in Class I wells and the vast quantity of oil-field
brine injected in Class II wells.  EPA is required by the
Hazardous and Solid Waste Amendment of 1984 to assess the
environmental suitability of well injection.  Presently the
Agency's paramount interest and effort lie in (1) evaluation of
well construction, (2) reactions among injected waste,
formation fluids, and the geologic framework, and (3) relation
of the injection interval to the confining beds.

     A major part of environmental suitability for underground
injection is tied to the integrity of confining beds.  That is,
confining units must be of areal extent, thickness and low
permeability sufficient to prohibit the upward migration of
injected fluids or displaced formation fluids into underground
sources of drinking water.

     A major concern in confining-unit integrity is the
presence of abandoned wells, either unplugged or inadequately
plugged.  An equally important concern and one that is rarely
considered, particularly for Class II wells, is the potential
for upward migration of waste substances or formation fluids
along faults or other fractures.  Additionally, at a few places
operational experience clearly has indicated a relationship
between deep-well injection and  increasing occurrences of
earthquakes, as exemplified by the former operation of a  deep
injection well at the Rocky Mountain Arsenal near Denver,
Colorado  (Evans, 1966).  In Ohio, reactivation of stabilized
faults by deep-well injection may have been the primary cause
of one or more earthquakes in recent years.

     In a major oil field in the southwestern part of the
United States, brines have been  injected through several wells
for more than 30 years.  It was  recently discovered that
several shallow cathodic-protection wells in this field, all
located along a nearly straight  line, contain brine with total
dissolved solids that exceed  100,000 mg/1.  This explanation
seems to be warranted: The brine originates from fluids
injected at depths as great as 8,000 ft  (2400 m).  Under
pressure of injection and of differential formation pressures,
the brine migrated to shallower units — including sources of

-------
underground drinking water— along a fault or a set of faults.

     Investigations of shale-gas production in West Virginia
during the late 1970's indicated that wells drilled within 350
ft (100 m) of a fracture or linear feature mapped from aerial
photographs or satellite imagery produced substantially more
gas than did wells at greater distances.  This implies that the
linear features, which in this case probably are fractures, are
indicative of zones of substantial permeability.

     Geologists have long recognized systems of joints and
fractures, in the field and on aerial photographs.   However,
after the Earth Resources Technology Satellite 1 (ERTS-1) was
launched in July, 1972, linear features were discovered to be
far more abundant and extensive than previously considered.
Satellite imagery, which is available for the entire Earth, can
be readily used to map regional linear features.  These linears
represent both faults and joint systems.

     Presently, siting procedures for Class I hazardous-waste
wells call for an extensive examination of subsurface
conditions within 2 mi (3.2 km) of the proposed well.  Of major
concern in this case is the location of abandoned wells within
the area of influence and the description of primary hydraulic
characteristics of confining units.  However, pressure build-up
brought about by injection can extend outward for miles, even
from a single injection well.  The pressure increase could
cause formation  fluids to migrate upward through fractures to
impact underground sources of drinking  water located much
farther away than the 2-mi (3.2-km) radius from an injection
site.  Thomas (1986) provided descriptions of the long
distances across which injected fluids could migrate through
fractures.

     Regulatory controls on Class II wells are far less
stringent, with most of these wells situated at points of
convenience. Throughout decades of use, Class II wells have led
to a variety of problems; some severe problems are related to
lack of integrity of confining units.  The integrity of
confining units can be compromised by abandoned wells and open
fractures, which are potential avenues for upward migration of
formation fluids into underground sources of drinking water.

INJECTION WELLS

     In 1983, 195 hazardous-waste injection wells were in
operation in the United States  (Brazier, 1987); currently 245
such wells exist.  In addition, about 120,000 enhanced-
recovery wells are in use  (J. Lynn, Underground Injection

-------
Practices Council, personal communication, 1989).  In the
latter, water is injected into oil-producing zones to maintain
formation pressure as an aid to oil recovery. Finally, Clark
(1983) estimated that  20,000 wells were used strictly for
disposal of oil-field brine; about 38,000 wells of this type
are in operation now (J. Lynn, UIPC, personal communication).

     Underground injection presently is the least expensive
method of waste management.  EPA estimated that for hazardous
waste, disposal by deep-well injection costs roughly $8 per
ton,  ($8.80 per metric ton) disposal in a surface impoundment
about $28 per ton ($31 per metric ton), and disposal in a
landfill about $50 per ton  ($55 per metric ton), assuming full
compliance with federal regulations. In contrast, resource
recovery, treatment, and incineration can cost as much as $718
per ton  ($791 per metric ton)(Gordon and Bloom, 1987).

     Because underground injection is the least expensive, and
in some cases, the only practical option available for liquid-
waste disposal, there is a trend toward greater reliance on
deep-well injection.  Generators of waste probably will rely
even more heavily on deep-well injection.

     In approximately 95 percent of all Class-I wells, wastes
are injected into zones that lie below usable water resources.
Brasier  (1987) determined that the average depth of hazardous-
waste injection wells is 4,000 ft (1200 m).   Typically, 2,800
ft (850 m) of strata separate the injection zone from shallower
aquifers that contain water with dissolved-solids
concentrations of 10,000 rag/1 (10,000 ppm) or less.

     Injection wells for commercial purposes originated in the
oil and gas industry.  Beginning in the middle 1930's,
injection wells were used instead of open evaporation pits,
which had been employed to dispose of highly corrosive,
commonly chemical-laden brines and drilling fluids; injection
wells also were used to enhance oil recovery.  Injection of
toxic and hazardous chemical waste from the steel and chemical
industries began in the 1950's.  The practice of injection came
into favor following the enactment of environmental laws
designed to protect surface waters from pollution  (Gordon and
Bloom, 1987).

     The objective of deep-well injection is the disposal of
liquid wastes into a suitably porous and permeable formation,
in such a fashion that it does not impinge upon the human
health and safety or the environment.  To that end, several
geologic and engineering-design criteria must be taken into
account in planning a deep-injection well.

-------
     Engineering concerns typically center on well design and
operations. Many methods have been developed or adapted to
insure the mechanical integrity of an injection well, including
cement-bond logging and pressure testing.

     Geologic standards of concern are much less precise than
criteria of engineering design.  A typical geologic assessment
of a potential injection-well site should include (1) the
geologic-hydrologic environment, (2) structural geology, (3)
physical and chemical properties of rocks, and (4) chemical
characteristics of fluids in the subsurface.

     A potential injection formation must be evaluated for
suitability. To be viable as an injection zone, the formation
must (1) have no value as a resource, (2) have sufficient
porosity, areal extent, and thickness to accept the anticipated
volume of liquids, (3) be located in a seismically inactive
area, and  (4) contain water that is compatible chemically with
wastes to be injected. Furthermore, it should be sealed above
and below by confining beds with  strength, thickness, and low
permeability, altogether sufficient to prevent vertical
migration of injected fluids or formation fluids from the
disposal zone.

     Information concerning the general lithology,
distribution, and structural configuration of a rock unit
potentially capable of accepting wastes generally can be
obtained. However, collection of geologic information typically
depends upon rock cores, records of nearby wells, logs of
nearby open holes, and seismic surveys.  Information may be
detailed but, at best, at one locality it provides little more
than a one-dimensional sample of subsurface characteristics.
Extrapolation and assumptions about homogeneity or the lack of
it are used to infer formation qualities between data points.

FAULTS, JOINTS, LINEAMENTS AND FRACTURE TRACES

     A critical geologic factor is recognition of penetrative
faults — that is,  delineation of faults that extend from the
potential  injection zone to the surface, or to an underground
source of drinking water. Such faults may breach confining beds
and may be passageways  for migration of wastes into shallow
ground-water supplies.

     Although delineation of faults  is difficult, with adequate
subsurface data the presence and configuration of some deep
faults can be discerned. Geologic study  of the subsurface
herein called "subsurface studies")  in the detail necessary is
expensive, tedious, time-consuming,  and  commonly  inadequate to
determine  the full scope and magnitude of faulting.  In

-------
addition, to determine localities  where a fault intersects the
surface is critical; this is an objective that subsurface
studies normally cannot produce.

     Close inspection of aerial photographs, satellite imagery,
surface-geologic maps and topographic maps ordinarily reveals
numerous alignments in landforms, streams and vegetation.  The
longer of such straight features have been called "lineaments,"
a useful concept of which is encompassed in the definition set
out by Lattman (1958, p. 569): "A photogeologic lineament is a
natural linear feature consisting of topographic (including
straight stream segments) vegetation, or soil tonal alignments,
visible primarily on aerial photographs or mosaics, and
expressed continuously for at least one mile, but which may be
expressed continuously or discontinuously for many miles."

     The shorter of such aligned fractures have been called
"fracture traces," as described by Lattman  (1958, p. 569): "A
photogeologic fracture trace is a natural linear feature
consisting of topographic (including straight stream segments),
vegetation, or soil tonal alignments, visible primarily on
aerial photographs or mosaics, and expressed continuously for
less than one mile.  Only natural linear features not obviously
related to outcrop pattern of tilted beds, lineation and
foliation, and stratigraphic contacts are classified as
fracture traces.  Included in this term are joints mapped on
aerial photographs where bare rock is observed."

     Although the definitions shown above make reference only
to mapping on aerial photographs, the criteria can be applied
to mapping of lineaments and fracture traces on satellite
imagery, topographic maps, and surface-geologic maps.

     Evaluation of remotely sensed data provides an attractive
alternative to subsurface investigations. Many faults or
fracture traces can be located by study of linear features on
aerial photographs or satellite images. Fracture-trace analysis
using remote sensing has many advantages.  The technique
provides an inexpensive, simple, and speedy method of analysis
of large geographic regions.

     Unfortunately, fracture-trace analysis is decidedly
subjective.  In some instances, interpretations vary
significantly from person to person.  Some mapped features are
discovered to be results of human activity, or otherwise to be
nongeological.  Many such errors can be isolated and corrected
through geologic field work.

     Assuming that all anthropomorphic features have been
eliminated from the set of lineaments and fracture traces

-------
mapped, the interpreter must determine which of these linear
features correspond to shallow joints and fractures and which
could represent significant deep-seated faults. Shallow
fractures and joints are much more numerous at the surface of
the earth than faults. Therefore, it is logical to assume that
many lineaments and fracture traces are results of surficial
fractures or joints.

     Such discontinuities commonly do not extend to significant
depths, and therefore do not necessarily threaten  confining-
layer integrity. On the other hand, faults significantly
disrupt confining layers and unsealed faults can be conduits
for fluid migration.

     Mapping of regional lineaments by means of satellite
imagery, in conjunction with aerial-photographic
interpretation, geologic field work, and study of history of
injection and production in oil  fields, provide a means to
develop maps of lineaments, joints, and faults. The maps can be
used to delineate areas where the integrity of confining units
potentially is disrupted.  In turn, this information can be
used in regulatory procedures involved in the permitting of
Class I and Class II wells. That is, the methodology can lead
to a fundamental, scientifically based approach to estimate
whether pressure in an injection zone is sufficient to cause
migration of injected or formation fluids into underground
sources of drinking water.  Lineament maps, when combined with
geologic maps and subsurface data, can be used to develop
"sensitivity maps."  Sensitivity maps indicate those areas
where there is significant probability of poor or questionable
confining-unit integrity and, therefore, the possibility of
unacceptable contamination of an underground source of drinking
water.  State or federal regulatory personnel could use
sensitivity maps to suggest or require further subsurface
investigation to establish safety of injection, or (in clear-
cut exceptional cases) to forestall or prohibit the
installation of an injection well.

SCOPE AND APPROACH OF STUDY

     The objective of this research is to define methods to
develop deep-well injection-sensitivity maps.  The principal
working hypotheses are as follows:
      (1)  Synthesis of data about areal geology, lineaments and
fracture traces permits the mapping and isolation of localities
that would be suitable or unsuitable for Class I or Class II
injection wells.
      (2)  A body of information  that can be used reliably to
infer  likelihood of fractured and permeable confining beds
exists  in the form of remotely sensed data —  linear features

-------
manifest on earth-satellite imagery, aerial photographs,
surface-geologic maps, and topographic maps.  A component
hypothesis is that numerous linear features can be mapped in
the field and shown to be images of fractured bedrock;
therefore, the linear features mapped remotely can be dealt
with on the warranted assumption that they represent lineaments
and fracture traces.
     (3) Lineaments and fracture traces are evidence of zones
of weakness in bedrock and indicate permeability distributions
that are likely to be greater than that of the adjacent
bedrock.  Fractures presumed to underlie them are judged to
have originated in crystalline basement rocks and to have
propagated toward or to the surface.  They would be paths along
which fluids could migrate if pressure distributions are
modified by injection of wastes.

     Basic premises on which this research is based are as
follows:
     (1)  Fracture traces and lineaments are correlated with
joints, faults or directional zones of weakness in reservoirs
of selected oil and gas fields.
     (2)  Correlation of fracture traces, lineaments and
fractured reservoir rock justifies the working assumption that
discontinuities have been translated vertically, and that all
strata above the reservoir are disrupted locally, or might have
been disrupted locally.

     Substantial amounts of published information suggest that
the fundamental working hypotheses are stable.  Reliability of
the premises described above seems to be justified by the work
of Alpay  (1973), Trantham and others  (1980), Rausch and Beaver
(1964), Schridder and others (1970), Brown and Forgotson
(1980), Komar and others (1973), and Feder (1984).

     Effectively, the research cited above has shown
substantial evidence of correlation between strikes of joints,
faults, fracture traces or lineaments at the surface and
fractures in reservoirs.  The correlation is not one-to-one of
course, but relations are shown to be strong enough to justify
the assumption that fracture traces and lineaments imply
similarly oriented fractures in the subsurface — despite
geologic and geographic differences in the areas investigated.

METHODS OF INVESTIGATION

     Research proceeded along these lines:
     (1)  Selection of (a)  oil and gas fields to serve as
models and (b) reservoirs known to be fractured systematically
or strongly suspected to be fractured systematically.
     (2)  Documentation of orientations of fracture-systems and

-------
abundances of fractures, insofar as was practical.
     (3)   Documentation of fractures and fracture traces above
and around the oil and gas fields by analysis of remotely
sensed imagery, including data from earth-orbiting satellites
and standard aerial photographs.
     (4)   Testing of validity of fractures in bedrock and
fracture traces mapped as described above, by methods of
standard geologic field mapping.
     (5)   Comparison of orientations of fractures and fracture-
traces at the ground surface with trends of fractures
documented in oil and gas reservoirs described above, to the
level of documentation that data permitted.

     Oil fields selected are judged to have analogs at
numerous localities elsewhere in the United States.  Geology of
these areas generally conforms with that of other injection
regions in the United States, specifically those in so-called
"hard-rock" terrain, onshore and inland from Tertiary-
Pleistocene fields of the Gulf Coast.

     Study areas selected included the Sooner Trend, the West
Edmond Field, the Burbank Field, and the Fitts Pool  (Figure 1),
all of which are  in Oklahoma.  The Sooner Trend, West Edmond
Field, and Fitts Pool produce from carbonate rocks;  fractures
have strongly influenced the locations and rates of production
in the Sooner Trend and West Edmond Field, and they  are
suspected to have had marked effect upon reservoir performance
in the Fitts Pool.  The major reasons for specifying these four
fields were their proximity, the large data base available, and
knowledge of and experience with the sites.  Despite the fact
that all the fields are in Oklahoma, the methodology should be
transferable readily because structural geologic controls and
stratigraphy are similar to a large percentage of oil fields
elsewhere in the United States.

     The products described herein evolved from a mix of
published work  and original work  (for example, mapping of
fracture traces, geologic field mapping,  subsurface  mapping,
and analysis of commonly available data about production from
oil and gas reservoirs).  If methods and products discussed
here are practical, and  in application elsewhere are robust
 (that is, economically  and temporally feasible, and  applicable
under a large  range of  conditions) then it is because the
methods were developed  under conditions that are general.  The
data were of common sorts and they were nonproprietary.  Under
such constraints some operating assumptions necessarily are
regarded as general geologic truths, verifiable by reference to
the literature.  These  are the  premises described above by
reference to published work.
                                 8

-------
                     OKLAHOMA
Figure 1. Locations of study areas: Burbank Field
          in Osage County, Southwest Enid Area in
          Garfield and Major Counties, West
          Edmond Field in Oklahoma and Logan
          Counties, and Fitts Pool in Pontotoc
          County,  Oklahoma.

-------
     Under the overall advisement of Gary F. Stewart and Wayne
A. Pettyjohn, experienced graduate students were assigned to
evaluate the four study areas.  Evaluation  of the Fitts Pool
and the Pontotoc County mapping project were conducted by James
0. Puckette and Michael R. Thornhill, the West Edmond Field by
Lonnie G. Kennedy, the Sooner Trend by Lyle G. Bruce, and the
Burbank Field by Kevin D. Flanagan and James 0. Puckette.
Although the general approaches to evaluation were similar,
each investigator emphasized slightly different techniques, and
some differences in methods evolved as a function of
differences in local geologic conditions.  This report was
integrated by Stewart and Pettyjohn.

     Separate accounts of work in the Southwest Enid Area, the
West Edmond Field, Burbank Field, and Pontotoc County are set
out by Bruce  (1989), Kennedy  (1989), Flanagan  (1989) and
Thornhill  (1989).
                                 10

-------
                           SECTION 2

                          CONCLUSIONS
     1. In the Southwest Enid Area of the Sooner Trend, the
soft sedimentary bedrock and widespread cover of
unconsolidated materials tend to minimize direct surface-
geologic evidence of faults. The concentrated area of
exceptionally productive oil and gas wells seems to be
positively but generally correlated with areal density of
stream-lineament intersections.  This relation suggests that
fracture-enhanced permeability in the "tight" Mississippian
carbonate-rock reservoir is shown to some degree by stream-
lineaments, particularly by the intersections of lineaments.
Therefore, in terrain where surface- and subsurface-geologic
conditions are similar, working hypotheses about injection
sensitivity should be concentrated on areas where stream-
lineaments intersect.

     2. Initial-production trends in West Edmond Oil Field
indicate that in reservoir rocks of the Hunton Group
permeability is strongly fracture-influenced.  Faults in West
Edmond Field dip almost vertically. In similar reservoirs,
valid interpretation of faulting can be based on initial-
production trends, conspicuous and linear deviations in strike
or dip of mapping daturns, and repeated or missing rock-
stratigraphic units.

     3. At Burbank Oil Field, trends of lineaments mapped from
satellite imagery are correlated closely with orientations of
fractures described in published accounts of detailed study of
the reservoir.  Fractures at the surface and in the reservoir
are similar in trend.

     4. At Burbank Oil Field, initial-potential and initial-
production maps are of little use in detection of fracture-
trends in the reservoir.  The Burbank Sandstone is
multistoried, multilateral, channel-fill sandstone; abrupt
variations in porosity and permeability are governed more by
internal geometry of the sandstone than by fracturing. Large
production units are common; therefore oil-production records
combine the yields of more than one well, and individual-well
performance is obscured.  For this reason, initial-production
maps are marginally informative.

     5. In Pontotoc County, numerous faults delineated by
subsurface geologic mapping were not shown by lineaments.
Many of these faults seem to have formed before Middle
Pennsylvanian time; presumedly, in structural movements that
                                11

-------
took place after the Middle Pennsylvania!* they were not
rejuvenated enough to affect overlying strata detectably.
However, the extensive Ahloso,  Franks, and Stonewall Faults,
and a large fault in northern Pontotoc County are partly
manifest as lineaments.

     6. Aerial photographs and satellite imagery are quite
useful for mapping of lineaments.  Most lineaments mapped from
aerial photographs correspond to linear transitions in
topography or to stream-drainage patterns.  Lineaments mapped
from satellite imagery tend to be the larger in extent, and in
general seem to match better with faults mapped by standard
field-geologic methods.  Major lineaments mapped from aerial
photographs correlate well with lineaments mapped from
satellite imagery.  Aerial photographs are better for
small areas and detailed work.

     7. Judicious constructive-bias mapping allows projection
of some faults from petroleum reservoirs to shallow marker
beds.  Trends of faults in shallow beds can be evaluated in
conjunction with lineaments mapped from stream trends, aerial
photographs, Landsat imagery, color-infrared imagery and
topographic maps.  Convergence of evidence permits the
assessment of terrain for injection sensitivity.  Zones of
sensitivity accent localities within which underground
injection of fluid would be advisable only after careful study
demonstrated acceptable levels of risk.

     8. Confining-potential is an important counterpart to
assessment of injection sensitivity.  Documentation of
thickness and extent of storage formations and confining beds
should be included in mapping for injection sensitivity. In
most instances, formation-thickness maps would meet basic
needs for information.

     9. A computer data base allows quick generation of maps,
including structural contour maps, interval-thickness maps,
and cumulative-thickness maps of confining beds and storage
formations. Routine use of data bases would allow geologists
with limited experience in subsurface geology to generate
several kinds of maps.  Experienced geologists could
synthesize the information and construct injection-sensitivity
maps.

     10. Integration of geologic material from sources diverse
in nature and quality may produce interpretations that are
strongly inferential but nevertheless close to the truth.
Under some kinds of review, such interpretations may be
regarded as being unacceptably subjective.  Structural
geologic maps and formation-thickness maps constructed by
                                12

-------
computers — where algorithms are conventional,  have been
tested repeatedly, and are of third-party origin — are likely
to be regarded as "unbiased".  They may provide interpretations
of structural geology and thicknesses of beds that would be
acceptable to persons with vested interests and opinions that
are quite different.  Computer-based maps should provide
minimally controversial baseline interpretations, and thereby
should reduce the probability of conflict between regulating
agency and producer.

     11. Nonproprietary geologic information from ordinary
sources can be used under general methods to produce maps
showing localities that are suspected to be injection-
sensitive.  The necessary materials generally are inexpensive
and readily accessible.

     12. Deliberations based on mapped zones of injection
sensitivity should include consideration of the likelihood
that (suspected) faults are sealed.  If faults penetrative
from oil reservoir to the surface were unsealed, the trap
should be emptied of petroleum, seepage should be evident at
the surface, or seepage should have been evident (and perhaps
recorded) before the oil field was developed.

     13. Interpretation of surface and subsurface geology for
injection-sensitivity mapping, including study of remote-
sensing imagery, aerial photographs and geologic maps would be
done best by experienced geologists who are trained
empirically for pattern recognition.  Synthesis of geologic
information commonly results in significantly different
interpretations among geologists.  Mapping for injection
sensitivity inevitably will generate adversarial positions
between regulating agencies and producers.  Therefore, both
parties must bear in mind the fact that injection-sensitivity
maps are precautionary devices.  They are designed to protect
underground supplies of drinking water, and not to obstruct
the responsible and conscientious producers of oil and gas.
                                13

-------
                           SECTION 3

                        RECOMMENDATIONS
     Methods described in this report are believed to be
simple and robust enough to be applicable in many parts of the
United States, under a wide range of geologic conditions.
Nevertheless, trial elsewhere should be illuminating,
especially if carried out under operational conditions.
Certainly the probability exists that partly different
approaches would be necessary outside the Midcontinent; for
example, some of the structural geologic conditions and
circumstances of confining beds in Mesozoic, Tertiary and
Quaternary formations of the Gulf Coast are markedly different
from the Midcontinent.

     Mapping of injection-sensitive areas could benefit from
compilation of historical data.  In some areas of the United
States oil and gas seeps were numerous long ago.  Because of
nearby production and consequent pressure depletion, the large
majority no longer are active.  Such possible conduits from
subsurface to surface should be evaluated, especially if
injection of fluid under high pressure is planned.  Clearly,
oil and gas traps exist because the reservoir is sealed
completely, or is so nearly sealed that migrating petroleum can
be regarded as having been arrested — in the short-term
perspective of geologic time.  In either case, integrity of the
seal is relative to pressure in the reservoir.  Put simply,
confining beds seal a trap  if pressure in the reservoir is not
sufficient to force oil and gas across the barrier. The
conclusion follows that fluids injected into a reservoir under
pressures less than or equal to the original reservoir pressure
should not breach the confining beds.

     Records of brine intrusion into fresh-water aquifers might
permit detection of contamination that is not related to
leaking wells.  Sampling of streams during base-flow conditions
might indicate localities where water from the deep subsurface
emerges.

     In the instances where methods and maps of the kinds
described herein are applied, we hope that a spirit of cordial
and cooperative interest in protection of underground sources
of drinking water can be maintained by regulator and producer
alike.  The intent of this endeavor was to contribute
information that in the long run would tend to maximize the
benefits of all subsurface resources.
                                14

-------
                           SECTION 4



               SOUTHWEST ENID AREA, SOONER TREND

                         INTRODUCTION
     In the attempt to  correlate fractures in rocks of the
subsurface with geologic data from the surface, to study an
area  with  abundant oil and gas wells and  fractured
reservoir rocks is necessary.  The westerly dipping carbonate-
rock strata of the Mississippian Meramecian and Osagean Series
in the Sooner Trend  of  Oklahoma (Figure 2)  compose a
fractured reservoir  (Nelson,  1985).  The trend is about  20 mi
(30 km)  wide; it extends for approximately 60 mi (100 km)  on a
homocline on  the  northeastern shelf of  the  Anadarko  Basin.
The trap developed where a system of fractures penetrates a
thick section of limestones and dolomites with low matrix
porosity. According to Harris (1975),  the limits of commercial
production of oil and gas are a function of fracture-controlled
permeability.  Cumulative production  from the Sooner Trend is
more than 270 million barrels of oil and 650 billion  cubic
feet of gas from more than 4,900 wells (Petroleum Information,
1982) .

     The study area is in the central part of the Sooner Trend;
it is referred to herein as the Southwest Enid Area.  It
includes 324 sq mi (839 sq km) in parts of Major and Garfield
Counties, and a small area in Kingfisher County, Oklahoma
(Figure 3).   The area is farmland with a few  small towns; in
the northeastern part are the city of Enid and Vance Air Force
Base (Figure 4).

     The terrain is of very gentle relief; most of it is
cultivated in wheat.  Bedrock is eroded easily, and outcrops
are rather sparse.  Topography evolved by differential stream
erosion of the comparatively soft bedrock and overlying
Quaternary materials. In general, bedrock stratigraphy and
structural geology influenced the overall development of the
gentle hills only slightly. Development of stream valleys seems
to have involved the partial adjustment of streams to subtle
stratigraphic and structural variations in bedrock.

     The northeastern two-thirds of the study area is underlain
by sandstones,  siltstones  and  shales  of  the Permian Salt
Plains Formation, Bison Formation, Cedar.Hills Sandstone, and
Flowerpot Shale,  and by Quaternary terrace alluvium (Figure
5).  The southwestern one-third of the area is underlain mostly
by Quaternary terrace alluvium, alluvium, and eolian deposits;
                                15

-------
                           NORTHWEST OKLAHOMA
                                SHELF
                                           NORTHEAST
                                           OKLAHOMA
                                            PLATFORM
          MMADQR ARCH
Figure  2.  Generalized  tectonic map  of Midcontinent
           region, showing  location of Sooner  Trend,
           on northeastern  shelf of Anadarko Basin.
           (After Huffman,  1959.)
                            16

-------
                     OKLAHOMA
T. 21 N.
T. 20 N.
       R. 9 W. R. 8 W. R. 7 W.
Figure 3. Location of  9-township Southwest Enid
          Area, Garfield,  Major and Kingfisher
          Counties, Oklahoma.
                         17

-------
R. 9 W.
R. 8 W.
                           R. 7 W.
                                                     T. 22 N.
                                                     T. 21 N.
                                                     T. 20 N.
                    Miles
0      Kilom
                   Kilometers
            10
Figure 4. Stream network,  Southwest Enid area.
          Drainage pattern traced from topographic
          maps.   Southwestern part of area  is
          underlain by young, unconsolidated
          geologic materials, and drainage  network
          is  not integrated.
                         18

-------
     R. 9 W.
R. 8 W.
R. 7 W.
                                               T. 22 N.
                                               T. 21 N.
                                               T. 20 N.
              0     Miles     6
              T  t   ii   i  i  T
                  Kilometers    1,0
                   i t  i  i  t i  i i
Figure 5. Surface geologic map,  Southwest Enid
          Area.  In ascending stratigraphic order,
          bedrock is  Permian Salt Plains Formation
          (Psp), Bison  Formation (Pbi),  Cedar Hills
          Sandstone  (Pen), and Flowerpot Shale
          (Pf).  Quaternary  terrace alluvium (Qt)
          and  alluvium  (Qal)  cover Permian strata
          in extensive  areas. (After Bingham and
          Bergman, 1980.)
                         19

-------
in this locality stream drainage has not evolved to an
integrated system (Figure 5).

     Oil and gas are the most valuable mineral resources in
the area. Average density of oil and gas wells is 5.2 per sq
mi (2 per sq km); approximately 1700 wells have been drilled
(Figure 6). As of January, 1987, more than 52 million bbl of
oil  and 475 billion cu ft of gas have been produced the nine
townships of the Southwest Enid Area (Petroleum Information
oil data and Dwight's Energy Data). Almost 90 percent of the
wells were completed in Mississippian carbonate rocks.

     Wells completed after 1976 were "infill" wells, most of
which were drilled in a partially depleted reservoir. The
post-1976 wells yield production data that are not compatible
with records of wells drilled in the few years after the field
was discovered. For this reason, production data used were from
wells completed before 1977.

     As described above, carbonate rocks of the Mississippian
System in the Sooner Trend are a fractured reservoir.
Therefore, a  map of cumulative production from wells drilled
into the reservoir before its significant depletion should be
indicative  of relative  fracture  density.

     At many localities outside the study area, faults in rocks
of the deep subsurface extend through rocks at the surface.  At
other localities faults and joints of the subsurface are
expressed indirectly at the earth's surface or are manifest not
at all.  Close inspection of aerial photographs, satellite
imagery, areal geologic maps and topographic maps ordinarily
reveals numerous alignments in landforms, streams and
vegetation — the fracture traces and lineaments discussed
previously.  Although definitions of the terms "fracture trace"
and "lineament", as set out by Lattman (1958), make reference
only to mapping on aerial photographs, the criteria can be
applied to mapping of lineaments and fracture traces on
satellite imagery, topographic maps, and standard geologic maps
of the surface.

     This principal operating assumption is considered to hold
true and to be justified or justifiable by repeated
observation: In places where faulting and folding have affected
the evolution of landforms not at all or only to a small
extent, landforms are curvilinear in overall form or in general
outline.  This working assumption is useful through
contradiction.  In some regions, straight-line topographic
features contrast strongly with the predominant curvilinear
forms.  The straightness of these landforms is such an uncommon
attribute that a sense of empirical and intuitive probability
                                20

-------
         R. 9 W.
             R. 8 W.
                                                     R. 7 W.
 ^ •   • • •
       > A A
• • A »   •$>
 * •» V -a^ -
  •  * A <$>  ^
.•* *. *•+  :
 *
i
 * •
       *  <*>   •
    •   * *   *
 >   •   k   •  • '•
 r . ^  **  •  **.•
            * * ••$>
                   ..* •
                  *    *   (»„'        *
          *     *     •    |   *       •••'£'•


     .'         *    -,•<$>..'$'* *t^     *'*'»
     •_•_.*.	^JS-.it—*	!*•	L.VCJL.-A<%_5«!> ._*_^	-^
•***i
      <(^>   <^> |tt <§>  <^ 'j^>   •
*    *   ••;    *   *  .      *
             iv.»*   *
         **•*   **^  .
                                                               ^
                                                                 W

                                                               • *
                                          * •* ^
                                          •  •!•

                               * <^»  * * •  »»  *
                                      •*•
                                               • * *   *   *   «
                                                    A * A   A  <
                                             <* • A * * .  *  . • * .
                                              *
                                                                   T. 22 N.
                                                                *
                                             /x'VT'-<^-aarv ^r~i,'> ^ yX~;/~-
                                             T^*^^^^^*^^^'*- ^*-;*' ,,-^<$'.'9^>?^<»V'*>^.'^,<:$> <^»T^'*' <»>/*>,•»>„
                       •*  >*.**^.*32A^
                                             *r-»r-»T r* «fc'A-^ A-^",
                                                          «• •» A. A   •'  A^^t.*^.*,  *. .*>.!«. !
                                  '•   '        '      '
                               Miles
                     0       Kilometers      10
                     i   i  i  i  i   i  i  i  i   i  I

         Figure 6.  Locations of oil  and gas  wells drilled
                    before  1977  (simple asterisk within
                    diamond),  dry holes drilled before 1977
                    (cross  with open  circle,  within  diamond),
                    oil and gas wells drilled after  1977
                    (simple asterisk),  and dry holes drilled
                    after 1977   (simple  cross  with open
                    circle).
                                   21

-------
leads to the inference that folds, faults or joints in bedrock
have influenced evolution of the landforms.

     The terms "empirical" and "intuitive" are evidence of
subjectivity inherent in mapping of lineaments and fracture
traces.  Because areal geology is so variable from region to
region, the mapper or mappers must establish the limits of
features that are more-or-less linear but judged to be random
and therefore not the results of structural discontinuities in
bedrock.  This limit necessarily is set by experience — by
study of the local terrain and by empirical definition of two
general kinds of landforms: (a) Those that are rectilinear but
nevertheless reasonably probable to have developed without
strong influence of joints, faults or folds, and (b) those that
are so rectilinear as to be improbable in the absence of
influence or control of joints, faults or folds.

                     MAPPING OF LINEAMENTS
     A detailed map of stream drainage (Figure 4)  was made
on topographic maps. Streams were traced as far headward as
possible. Because quadrangles at the scale 1:125,000 were not
available for coverage of the entire area, quadrangles at the
scale of 1:62,500 (approximately 1 in. per mi (1.6 cm per km))
were used.

     Each stream pattern was examined; straight-line segments
and angular reaches of channels were marked, if they were
judged to be anomalous. Changes in directions of channels on
the order of 90 deg., or of an acute angle were regarded as
being particularly suggestive.  (Compare Figures 4 and 7.)
Four general orientations of lineaments are apparent (Figure
7): Northward, eastward, northwestward, and northeastward.

     The map of lineaments was generalized  (Figure 8), in the
expectation of usefulness of such a map for comparison with
maps related to subsurface geology.  Numerous short lineaments
arranged end-to-end were considered to be evidence of a larger
lineament that by ordinary inspection was not mappable in its
entirety; therefore, the short lineaments were combined. A
pattern of  high-  and  low-density  areas was revealed (Figure
8), but the lineaments were so abundant that reduction of the
data was required.  Intersections of lineaments were mapped
 (Figure 9).

                      SUBSURFACE GEOLOGY


     From public records available at Oklahoma Well Log


                                22

-------
 R. 9 W.
R. 8 W.
R. 7 W.
                                                      T. 22 N.
                                                      T. 21 N.
                                                      T. 20 N.
                     Miles
            0      Kilometers
            T i  i  i  i  i   ii
Figure 7. Stream network, Southwest Enid Area,  with
          segments marked that were judged  to show
          anomalous straightness, anomalous
          intersections with other streams,  or
          both.
                          23

-------
R. 9 W.
R. 8 W.
R. 7 W.
                                                       T. 22 N.
                                                       T. 21 N.
                                                       T. 20 N.
                      Miles
                    Kilometers      1.0
               i  i  i  i  i   i i   i  i  I
Figure 8.  Stream network, Southwest Enid Area, with
           stream-lineaments  extended and
           intersections marked by circles.
                          24

-------
R. 9 W.
R. 8 W.
                       R. 7 W.
                                                       T. 22 N.
                                                       T. 21 N.
                                                       T. 20 N.
            9   .
Miles
            9
Kilometers
            1,0
Figure 9.  Southwest Enid Area.   Circles show
           locations of intersection of generalized
           stream-lineaments.
                         25

-------
Library, Tulsa, Oklahoma,  data on each of the 1,688  oil and
gas wells in the Southwest Enid Area were studied,  including:
     Location, accuracy of 2.5 acres (1 hectare).
     Well status (i.e., oil well, gas well,  or dry hole).
     Year completed.
     Datum elevation (ordinarily the kelly-bushing elevation).
     Availability of wireline logs.
     Depth to top of Meramecian Series of Mississippian System.
     Depth to base of Mississippian System.
     Cumulative oil production to January 1, 1987.
     Cumulative gas production to January 1, 1987.
     Fracture-treatment of the well.
     Pay zone or zones.

     Wireline logs were reviewed; kelly-bushing elevations and
depths of the top of Meramecian strata and the base of the
Mississippian System were recorded. Only data from such
sources were used for construction of thickness maps and
structural geologic maps.  Scout-ticket tops were considered
not to be reliable. Of 1,688 wells in the study area, logs of
813 were available. These 813 wells were distributed in such a
manner that the entire area can be considered to have been
sampled adequately.

     Data were entered into a spread-sheet program; thickness
of the Meramecian-Osagean stratigraphic section was calculated
for each well. Cumulative production was estimated for  each
well, as an oil-equivalent quantity. Added to oil production
(in thousands of barrels) was a number calculated by conversion
of gas production to oil production. One billion cubic feet of
gas was estimated to be equal to 75 thousand barrels of oil.

     Data from the spread-sheet were  entered into the Jupiter
Mapping Program, the principal algorithm of which is
"neighborhood-based" (Watson and Philip, 1987). Data
calculated for each intersection of a generated grid are based
on a weighted well-value; data at each well-location and
distances from the grid intersection to nearby wells are taken
into account.  Each well-value also is honored, as long as
density of wells is not more than one well per grid unit.  The
grid size was defined so that each well-value was integrated in
contour mapping.

     Structural contour maps were made of the top of Meramecian
strata and the base of Mississippian rocks  (top of underlying
Woodford Shale)  (Figures 10 and  11).  The working hypothesis
tested was that faults in Mississippian rocks would be shown or
indicated at places where configurations of datums are
anomalous — where changes in dip are uncommonly abrupt, where
dip is reversed in short distances, or where the structural


                                26

-------
R. 9 W.
R. 8 W.
R. 7 W.
                                                        T. 22 N.
                                                        T. 21 N.
                                                        T. 20 N.
             0       Miles        6
             Ti   ill   i   I
                    Kilometers      1,0
               i  i  i  i  i  i  i   i  i  I
Figure  10.  Southwest Enid  Area.   Structural  contour

            map, top of Meramecian strata.  Contour

            interval 100  ft.
                          27

-------
        R. 9 W.
R. 8 W.
r
                                               R. 7 W.
                                                             T. 22 N.
                                                             T. 21 N.
                                                             T. 20 N.
                            Miles
                          1	i	L

                          Kilometers
                                i — i
       Figure 11. Southwest Enid Area.   Structural contour
                  map, base of Mississippian System (top
                  of Devonian Woodford  Shale).   Contour
                  interval 50 ft.   Observe increase in
                  detail, most of which is due to smaller
                  contour interval.
                                28

-------
fabric seems to show alignments. A  first-order trend surface
of the top of Meramecian rocks (a best-fit planar surface)
shows that regional dip is south-southwestward at less than 1
degree.

     Several  small anomalous areas of closed contour lines are
shown in Figures 10 and 11.  A fault was interpreted  by Harris
(1975) on the eastern side of the syncline that trends through
the northwestern part of T. 20 N., R. 9 W. (arrows, Figure 11).
The syncline is uncommonly large and asymmetrical; by
comparison to other folds in the study area,  its size and shape
were taken as justification for the inference of a normal fault
that strikes northeastward, downthrown to the west.

     Figure 12 shows thickness of rocks between the top of the
Meramecian Series and the base of the Mississippian System. In
the Southwest Enid Area, the post-Mississippian, pre-
Pennsylvanian unconformity is on Chesteran strata; the
Meramecian-Osagean strata probably were shallow beneath the
surface, and could have been part of a fresh-water aquifer.
Numerous areas in Figure 12 are mapped as localities where the
reservoir is thin; these circular, ovate and elliptical
patterns are suggestive of terrain where rock has been eroded
locally by dissolution.  If these features are interpreted
correctly, their distribution may have been influenced by
systematic fractures in the rock, a relation that is well
documented (for example, see Thornbury, 1956, p. 337; also
Jennings, 1985; Bogli, 1980).  Otherwise, Figures 10, 11 and
12 show no evidence that is firmly indicative of the abundance
and orientations of fractures in the reservoir rock; the maps
show no conclusive evidence about fracturing of the reservoir.

     In the Southwest Enid Area oil and gas have been produced
from rocks as old as Ordovician and as young as Pennsylvanian.
However, 88 percent of the wells were completed in Meramecian-
Osagean rocks, and about 90 percent of the petroleum has been
produced from these Mississippian strata.

     The statement that the Mississippian reservoir of the
study area is permeable mostly because of systematic fractures
(Nelson, 1985; Harris, 1975) seems to provide a stable premise.
Accordingly, this inference would seem to follow: If
permeability in the reservoir is correlated directly with
abundance of fractures, with lengths, heights and openness of
fractures, then relative density of fracturing in the reservoir
should be manifest in a general way  by volumes of oil and gas
produced from place to place. (The qualifying phrases
"relative density" and "general way" take into consideration
lithic variation in the reservoir, perforated intervals of
different lengths, fracture-treatments of different types,


                                29

-------
                     R. 8 W.
                   R. 7 W.
                                                     T. 22 N.
                                                     T. 21 N.
                                                     T. 20 N.
                                                   J
                      Miles
              0      Kilomi
              i  i  i  i i   i
Kilometers      10
     i  i  i  i  I
Figure 12. Southwest  Enid Area.  Thickness of
           strata  between top of Meramecian Series
           and base of Mississippian System.
           Contour interval 50 ft.  Hachured closed
           lines show areas where the stratigraphic
           sequence is thin.
                          30

-------
operators with different practices, and other variables that
are exceedingly difficult to document in systematic fashion.)

     Figure 13 shows cumulative oil-equivalent production from
wells completed before 1977.  (Production from wells drilled
thereafter probably was reduced by overall depletion of the
reservoir.)  If production of oil and gas is related directly
to systematic, closely spaced fractures in the reservoir, then
patterns of contour lines suggest that fractures are
concentrated in T. 21 and 22 N.,  R. 7 and 8 W.  If indeed such
a relation exists, then fractures seem to trend eastward,
northeastward, and northwestward (see arrows, Figure 13).

     At several places in T. 20 N., R. 8 and 9 W., and T. 21
N., R. 9 W., large cumulative production per well is owing to
commingling of petroleum from Mississippian rocks and strata of
the deeper Silurian-Devonian Hunton Group (Figure 13).   In
Figure 14 these localities are not shown, and terrain where
exceptionally large amounts of production from Mississippian
rocks is outlined.  Figure 15 shows areas of large cumulative
production per well, and the outlines of areas where fractures
are judged to be uncommonly numerous.

     In general, the patterns seem to be .positively and
generally correlated.  Two areas in which the patterns seem
definitely not to match are (a)  the northeastern part of the
study area, where the city of Enid and Vance Air Force Base
diminished the effect of analysis of stream patterns,  and (b)
the northwestern part of T. 21 N.,  R. 8 W.  If only the region
where cumulative production of more than 250,000 bbl per well
is considered, correlation of cumulative'production and
abundance of  (inferred) fractures seems to be significantly
greater (Figure 15).

                          CONCLUSIONS
     In the Southwest Enid study area, tracts of exceptionally
productive oil and gas wells seem to be positively but only
generally correlated with areal density of intersecting
lineaments.  This general observation could be the basis for a
deductive argument that where cumulative production of wells is
uncommonly large and no fundamental change in matrix porosity
and permeability of the reservoir is evident, the reservoir is
more fractured than elsewhere. Accordingly, if confining beds
are cut by throughgoing fractures, then the penetration is more
probable in terrain where lineament intersections are abundant.

     In terrain geologically similar to this part of the Sooner
Trend,  where structural geology is simple and bedrock at the
                                31

-------
 R. 9 W.
R. 8 W.
                      R. 7 W.
       COMMINGLED PRODUCTION,

       MISSISSIPPIAN AND SILURIAN-

       DEVONIAN ROCK
                                                         T. 22 N.
                                                         T. 21 N.
                                  T. 20 N.
                      Miles
             9
Kilometers
           10
Figure  13.  Southwest  Enid Area.   Contour lines  show
            cumulative production  from Mississippian
            rocks, and locally from Mississippian
            and Silurian-Devonian  rocks.  Contour
            interval,  50,000 barrels of oil-
            equivalent production.
                          32

-------
  R. 9 W.
   R. 8 W.
                                         R. 7 W.
                                                       T. 22 N.
                                                        T. 21 N.
                                                        T. 20 N.
                      Miles
                 j	i	i	t	L
              1
   Kilometers      10
i	i—i—i—i—i—i—i—I
Figure 14. Southwest Enid Area.  Contour lines  show
           cumulative production from Mississippian
           rocks.   Contour lines scaled as  100,000
           and  250,000 barrels of oil-equivalent

           production.
                           33

-------
 R. 9 W.
 R. 8 W.
R. 7 W.
                                                      T. 22 N.
                                                      T. 21 N.
                                                      T. 20 N.
                     Miles
                i    i   i	1	L
Kilometers
                                 10
Figure 15. Southwest Enid area.  Stippling shows
           where  intersections of stream-lineaments
           are numerous.   Oil-equivalent production
           of more  than  250,000 bbl per well is
           near up-dip limit of area of abundant
           lineament  intersections.
                          34

-------
surface is weakly resistant to erosion and almost undeformed,
maps designed to show areas of possible injection-sensitivity
should show localities where intersections of lineaments are
closely spaced.  In combination with computer-contoured
structural-geologic and unit-thickness maps, the injection-
sensitivity maps could indicate areas in which cautious
assessment of local geology should precede disposal of fluids
by underground injection — especially if injection pressures
are designed to exceed original formation pressures.

     Conventional contouring  of structural geologic maps with
unit-thickness data and lineaments taken into account can
produce an  integrated  interpretation of subsurface structural
geology.  An experienced and creative mapper may present
interpretations that in fact are closer to the truth than
mapping with somewhat less license. However, under some
circumstances  the interpretation probably would be regarded as
being unacceptably more subjective than the computer-contoured
maps set out here.  Computer-contoured maps seem to be regarded
as being "unbiased"; for this reason they may provide
interpretations of structural geology that are rather likely to
be acceptable to persons with central interests and derivative
opinions that are quite different. In regions with geologic
similarities to the Southwest Enid Area (and indeed to the
Sooner Trend), computer-generated structural geologic maps and
thickness maps, and conventionally prepared lineament-
intersection maps should provide a basis from which regulatory
personnel and petroleum companies could begin to decide whether
underground injection of liquid waste is prudent.
                                35

-------
                           SECTION 5

                     WEST EDMOND OIL FIELD

                          INTRODUCTION
     West Edmond Oil Field includes about 160 sq mi (420 sq
km),  located mainly in Oklahoma County,  Oklahoma  (Figure 1).
The field was developed in the early 1940's.   The principal
reservoir is the "Hunton Lime" — strata of the Silurian-
Devonian Hunton Group (Figure 16).   The field is on the
northern part of the plunging Oklahoma City Anticline.   The
trap in West Edmond Field is developed chiefly where a large
salient of the upper part of the Hunton Group extends eastward
past the general erosional limit and up-dip onto the flank of
the Oklahoma City Anticline (Figures 17 and 18). About 1,400
wells were completed in the Hunton, at depths averaging about
6,800 ft (2100 m).  Hundreds of wells have been drilled near
the periphery but outside  West Edmond Field; collectively,
wells in the field and in the surrounding area provide
excellent basic data about the subsurface geology.

     Excellent accounts of the early history of West Edmond
Field were provided by McGee and Jenkins (1946) and Swesnik
(1948). Four months after the discovery well was drilled only
10 or so drilling rigs were in operation; by the end of 1944,
120 rigs were drilling, and by the end of 1946, development of
the Hunton reservoir was nearly  finished; more than 1300 wells
had been completed. In the period between the discovery well in
April, 1943 and the end of 1945, the field had produced more
than 32 million barrels of oil and 46 billion cubic feet of gas
(McGee and Jenkins, 1946).  In July, 1947 the field was
unitized by order of the Oklahoma Corporation Commission;  at
this time 65 million barrels of  oil had been produced. As  of
1989, total production of oil exceeds 160 million barrels.

     In the western, down-dip part of the field, many wells
penetrated only the upper part of the Hunton Group, chiefly in
the Bois d'Arc Limestone  (Figure 18).  Although the Haragan and
Henryhouse Formations and the Chimneyhill Subgroup are less
porous and permeable than the Bois d'Arc, the strata are
fractured and the formations are productive near the eastern
limit of the field.

     In the early development of the field, most wells
initially were allowed to flow at relatively high rates, in
order to "clean up" the reservoir near the borehole.  In this
process the 24-hr production rate was measured.  Most wells in
the field were drilled on 40-acre  (16-hectare) spacing. Oil-
                                36

-------
        SYSTEM GROUP FOHMATION










2
<

"Z.
o
>
LU












z
<


D
_j
















2


O
>
Q
ac
O


























D.



(3
~2L
r\
P
z
^>
3C





















a
D
O
a:
O

<
Q




n
5
O
a:
CD

^
O
8.
5
w
a
DC
0

o
o


1 £
0 <
O U
C/J **
a: £
"• o
t



' W
5 -
0 °
< 5
* rc
*< i

X




_J
I
>-
UJ


I
O


2:
5
5
>




5



O

!*
S*.H
8 t
Q 2





0
2


O
O
_J
s
%

















































































/








































J

— — ^-


/
	 	 	

V.



^> _


=::::<
	 , —.^^ — — e-.—

| wast. *"

,
*••
y ... ^

^ o, j_ 2
__ ID \ /
0 \
Q* ~\'
^
hb*

If
	 . 	 .

0





_
C^=.
t ^


°







-j Z
-^- 	 g--^^-*.'
L_ ° ' '^
- L i
- Sh ?-





o 	 ^


~L u N

./
A
_v.
	 0 —^ 	
: Uu. 5-

- . ~9i
— — fj

^
J - »^ 	
Wi
o ™"


y-
	 	 _ p^-.—
ZZZ **• ^ Z". ~ ^ —
^ ,. .. - I" \f~ —'

<: ^_
- - -
X -.
, 	 *^
_
-- - . '






- -
-- ,

~~ \" 7s" ** **"

" "•*••-.-
	 .
-- _^ • ... %
^
-r ~~ ^" * '
*"" -" *^ .



_ _ ~-~ *
~ v
• — •

>

' -t •• ''
'. -.

.. ^


















	
7 ^i r 	





— - " '
"^ -"-•
- . A..
i .'- • — •"
»

-- -^r;
Figure 16.
Type electric log,  West Edmond  field,
showing  rock-stratigraphic units  from
Woodford Shale to Simpson Group.
                        37

-------
           R. 5 W.
                     R. 4 W.
                                R. 3 W.
                                          R. 2 W.
                                                  T. 15 N.
Figure 17,
                                                  T. 14 N.
                                                  T. 13 N.
                                                  T. 12 N.
                                                  T. 1 1 N.
                          Miles
                      9
             Kilometers
                     \°
Paleogeologic-structural geologic map,  Oklahoma
City Anticline and West Edmond Field,  showing
paleogeology of the time after deposition of Hunton
Group and before deposition  of Woodford Shale.   The
trap was formed where  salient of upper part of
Hunton Group extends up-dip  onto flank of
northward-plunging Oklahoma  City Anticline.
Contour lines indexed  in feet below mean sea level;
contour interval 250 ft  (76  m).  Modified from
McGee and Jenkins, 1946, p.  1805.)
                               38

-------
                                                                                  A'
co
\0
                    Figure 18. Cross-section, central part, West Edmond Field,
                               showing truncation of Hunton Group and sealing of
                               trap by Woodford Shale and Pennsylvanian strata.
                               Line of cross-section shown in Figure 16.  Bois
                               d'Arc Limestone is principal reservoir in central
                               part of field.  (Modified from McGee and Jenkins,
                               1946, p. 1804.)

-------
production quotas were established in the year of discovery, in
order to maximize long-range production of oil; to forestall
decline of reservoir pressure, quotas were placed on gas as
well. In 1947, the Hunton Group in the West Edmond Field was
unitized.  Field-wide reservoir-pressure maintenance was based
on injection of water or gas.  As a consequence of this
history, only the gross field production is reported to the
State of Oklahoma; single-well production data are not
available for the period from date of unitization to the
present.

                        SURFACE GEOLOGY
     The West Edmond Field is on the northwestern flank of the
Oklahoma City Anticline (Figure 19).  The anticline is near the
southern end of the Nemaha Range, a complex of faulted
anticlines that extends from southeastern Nebraska into south-
central Oklahoma.  The eastern margin of the range is faulted
(Figure 19).

     Generalized surficial geology of the study area is shown
in Figure 20.  The Hennessey Shale,  which crops out over most
of the area,  dips southwestward at about 90 ft per mi (17 m per
km).   In the northeastern part of the area Garber Sandstone is
exposed, whereas in the southwestern part the El Reno Group
crops out.  Quaternary sand, silt, clay and gravel are
distributed along the North Canadian River (southern part,
Figure 20) and all major tributaries.  To the southeast
Quaternary dune sands and alluvial terrace deposits conceal
bedrock.

                       SUBSURFACE GEOLOGY
STRATIGRAPHY

     In Oklahoma, stratigraphic nomenclature of sedimentary
rock units exposed at the surface ordinarily differs
considerably from nomenclature of strata concealed in the
subsurface.  In this report nomenclature used in exploration of
the subsurface takes precedence.  Additionally, several
stratigraphically successive rock units are lithically so
similar that they cannot be discriminated on wireline logs that
date from the 1940's and 1950's.  Therefore such formations
were combined into units that have distinctive characteristics
on wireline logs, and that can be correlated from well to well.
The following terminology was used:
                                40

-------
                  M
                     \°
               Miles
Kilometers
Figure 19. Pre-Pennsylvanian paleogeologic map of central
           Oklahoma, showing location of West Edmond Field
           relative to Oklahoma City Anticline.  (Modified
           from Jordan, 1962.)  Symbols: OC - Cambrian-
           Ordovician Arbuckle Group.  Oss - Ordovician
           Simpson Group.  Ov - Ordovician Viola Group. Os -
           Sylvan Shale. Osv - Viola Group and Sylvan Shale.
           Dsh - Silurian-Devonian Hunton Group. Dw - Devonian
           Woodford Shale. DSws - Hunton Group and Woodford
           Shale. M - Mississippian rocks, undifferentiated.
           Mn - Mississippi Lime.  Me - Mississippian Chester
           Group.
                              41

-------
Figure 20. Generalized surface geology, West Edmond Field and
           nearby areas.  Symbols: Pgw - Garber-Wellington
           Formations.  Ph - Hennessey Group. Pe - El Reno
           Group. Qt - Pleistocene terrace alluvium.  Qal -
           Quaternary alluvium. (After Bingham and Moore,
           1975.)
                              42

-------
     Formations that compose the Hennessey Group are referred
     to as the "Hennessey Shale".

     The Garber Sandstone and Wellington Formation are called
     the "Garber-Wellington Formations".

     The Frisco and Bois d'Arc Limestones of the Hunton Group
     are called the "Frisco-Bois d'Arc Limestones".

     The Haragan and Henryhouse Formations of the Hunton Group
     are called "Haragan-Henryhouse  Formations."

     The Chimneyhill Subgroup of the Hunton, composed of the
     Clarita, Cochrane, and Keel Formations, is dealt with as
     one rock-stratigraphic unit.

     Stratigraphy of rocks ranging in age from Permian through
Cenozoic, including the Garber and Wellington Formations, is
derived from Christenson and Parkhurst  (1987) and Wood and
Burton (1968). The Desmoinesian Series of the Pennsylvanian
System was described by Benoit (1957).  Much of the
stratigraphy of Ordovician and Silurian strata was taken from
Swesnik (1948).  Nomenclature of the Hunton Group was described
by Amsden and Rowland  (1967).

Permian System

     In the study area, the boundary between the Permian and
Pennsylvanian Systems is open to question.  A  set of thin,
extensive limestone marker beds at about 1,500 ft  (460 m) below
the surface is near the base of the Permian; the provisional
base-of-Permian was mapped close beneath the base of the
lowermost bed in the set (Figure 21).

Hennessey Shale—
     Thickness of the Hennessey Shale is 600 to 650 ft (180 to
200 m). The Hennessey is dominantly reddish brown shale that
includes siltstone and fine grained sandstone.

Garber-Wellington Formations—
     In the subsurface the boundary of the Garber Sandstone and
the underlying Wellington Formation  is not detectable at most
places (Carr and Havens, 1976).  The two are undifferentiated
in this report (Figure 21).  Aggregate thickness of the Garber
and Wellington ranges from 800 to 1000  ft  (250 to 300 m).  The
sequence is composed of red to maroon,  lenticular strata of
fine grained cross-bedded sandstone, interbedded irregularly
with sandy, silty shale.


                                43

-------

CO
q
i-
<
s
cc
O
u_
2
O
h-
o
2
_J
UJ
^
i
cc
UJ
CD
or
<
O

w
O-
3
O
cc
O
LU
CC
s
<
z
<
UJ
>
O
cc
O
_1
O
2
3
Q
O
C/3
<
X
O






























<

-



,
- *'
5,
;>
<;
CT"
i
^
"- i
^
•'?
f
" :.
* >
^ >
""/
- - *
• f:
I
">
I'

r\
N-
>
















- ;
r.i

-"
:




i
•*
Ul
O
0)
O
O
*»*
O
CD
0
(O
O
O
O
Ivi
O

O
O

O
0
en
0 .

rg=
•
0=
?


-r
«*!
J


^
I
/
>
V
i
ft
\
F
*
>
-


{•

;
-
i
-

[;

f
1
. *

1=
t
i--
• <~-
\ ' ~
f H



-: •

-" -
.- ':




'-
: -_. :.
i: .
-. .
•..
: - ~
"~:.



---.
Figure 21. Type electric log, West Edmond Field,
           showing rock-stratigraphic units from
           base of surface casing to uppermost part
           of Pennsylvanian System. Pennsylvanian
           rocks are below wavy line at 1760 ft.
                          44

-------
Chase, Council Grove, and Admire Groups—
     Approximately 750 ft  (230 m) of red to maroon, fine
grained, cross-bedded sandstone and shale compose the Chase,
Council Grove and Admire Groups; the strata are similar to the
Garber and Wellington Formations.

Pennsylvanian System

     The Pennsylvanian System is composed predominantly of gray
shale interbedded with sandstone and thin limestones.  The
Pennsylvanian unconformably overlies older Paleozoic rocks,
ranging from the Mississippian System in the western part of
the study area to Ordovician and Silurian-Devonian rocks in the
eastern part (Figures 19 and 22).

Mississippian System

     In central Oklahoma, Mississippian rocks compose the
"Mississippi Lime" (Figure 22), a section of gray to brown
limestone and cherty limestone.  In the northwestern part of
study area Mississippian rocks are as thick as 270 ft (80 m);
in the eastern part of the study area the Mississippian is
absent because of post-Mississippian, pre-Middle Pennsylvanian
erosion on the Oklahoma City Anticline (Figure 19).

Devonian and Silurian Systems

     The Woodford Shale, Frisco, Bois d'Arc, and Haragan
Formations of the Hunton Group  (the principal reservoir in the
West Edmond Field) are classified as Devonian (Figure 18).  The
lower rock-stratigraphic units of the Hunton, the Henryhouse
and Chimneyhill, are Silurian (Amsden and Rowland, 1967).  (In
Figure 22 the Henryhouse is shown as being Devonian, because on
ordinary wireline logs the Henryhouse and Haragan cannot be
differentiated sufficiently for consistent well-to-well
correlation.)

Frisco and Bois d'Arc Formations—
     The Friso Limestone (Figure 18) is maximally 40 ft (12 m)
thick.  The underlying Bois d'Arc is made up of four lithologic
units, all limestone.  The boundary between the Frisco and Bois
d'Arc and the internal  carbonate-rock stratigraphy of the two
formations is discernible in bit cuttings and cores, but is not
consistently mappable on ordinary wireline logs.  Therefore the
Frisco and Bois d'Arc were treated as one mapping-unit.

Haragan and Henryhouse Formations—
     The Haragan-Henryhouse stratigraphic sequence  (Figure 22)
is approximately 180 ft (55 m) thick.  The Haragan overlies the
Henryhouse unconformably, but in study of subsurface geology,
                                45

-------
        SYSTEM GROUP FORMATION
PENNSYLVANIAN
/
DEVONIAN
SILURIAN


"CHEROKEE"

HUNTON GROUP


ao
o
UJ
CO

SSISSIPPI
LIME"
,5
ATOODFORC
u- b
HARAGAN-
HENRYHOUSE
-J
CHIMNEYH

























--






._

4.










-



—




^=
^



















s
^


f










—
/





k


\

\
A


/
\
)
v















.




en
o
0
O)
O)
o
o
«,










-



	

"I


^J

.-=






-
~f

~<

r

j».




t.-
i-
-=



T


^




_t£± —
^U



-
•rt*~
•»
—1
fj



^
r£-

^
!=
•^



|-
^
- -


^r
r-,
^T
H=
d-
1







U_



—


— T-
&-
— r
-




— '
i-

-
-"













B



























=

_«
UNCONFORMITY
—



















—









—









-1




s




-







O5
OO
0
o
-








-F-.

>
» |l

r1




^













^







































—
**
	



	



UNCONFORMITY

3






















fj


<.


























'


f>

/




(








i





1


-|-


f
\ •
*Lr


-f
/^

-












































— 'j
o>
CO
0
0
-4
0
0
O

O
O


-i.

I
\

A


	 :













"*t

~*








>^

A
^











^^

/
^~

X


•

,^

^




-»





^r



— .













_._


•^! ""





__,




1
^


^t











p




^



•»
"




~




'* ' ~I~T 	 1
















—
^:




•^ \ i i _L
Tr I"
/ !









^S






=r








^^






Figure 22. Type electric log, West Edmond Field,
           showing rock-stratigraphic units from
           "Oswego Lime" (Middle Pennsylvanian) to
           Chimneyhill Subgroup (Silurian).
                        46

-------
the distinction is exceedingly difficult without careful
examination of cores.  The sequence is composed of limestones
and dolomitic limestones.

Chimneyhill Subgroup—
     The Chimneyhill Subgroup (Figure 22) is about 40 to 50 ft
(12 to 15 m) thick and is composed of limestone.  Formations
that compose the group are poorly documented in the West Edmond
Field.  The Chimneyhill conformably underlies the Henryhouse
Formation and conformably overlies the Sylvan Shale.

Ordovician System

     The Sylvan Shale is the uppermost rock-stratigraphic unit
of the set of Ordovician rocks (Figure 18).   The greenish gray
shale is approximately 100 ft (30 m) thick throughout the study
area.  The underlying Viola "Lime" is compact, clean limestone,
about 100 ft  (30 m) thick.  The Viola overlies the Simpson
Group, which is composed of interbedded dolomitic clean
quartzitic sandstones, dolomite, limestone and shale.

ABBREVIATED STRUCTURAL-GEOLOGIC HISTORY

     During the early part of the Paleozoic Era, Oklahoma was
within a broad sedimentary basin.  After deposition of the
Hunton Group regional uplift was accompanied by widespread
erosion, especially in the more extensively folded and elevated
regions.  Subsidence during the Mississippian Period and
covering of the post-Devonian regional unconformity by
carbonate rocks was followed during the early part of the
Pennsylvanian Period by general exposure of terrain, except in
the deeper basinal areas, and by extensive removal of
Mississippian and older rocks from the crests of uplifts.

     The Oklahoma City Anticline is located near the southern
end of the Nemaha Range  (Figure 2), an extensive tectonic
element that is bounded on the east by a down-to-the-east
normal fault.  Entrapment of petroleum in the West Edmond Field
is associated with up-dip wedge-out of the Hunton Group, on the
northwestern flank of the Oklahoma City Anticline.  Here the
porous, permeable, fractured carbonate rocks of the Hunton are
sealed by overlying Pennsylvanian strata, chiefly shale.

INTERPRETATION OF SUBSURFACE GEOLOGY

     Subsurface geology was mapped using data from wireline
geophysical logs, scout cards, and Forms 1002A of the Oklahoma
Corporation Commission.  The Forms 1002A document the methods
of completion of wells and the rates of initial production.


                                47

-------
     Wireline logs are available for most wells;  electric logs
are the only logs generally available for wells drilled during
the 1940's.  Most of the wells drilled in the 1950's and 1960's
are documented by electric logs and microresistivity logs.
Induction-electric logs with combinations of microresistivity
logs, gamma-ray logs, compensated density-neutron logs and
sonic logs commonly are available for wells drilled after the
1960's.

     Production data were compiled for each well in the field;
data were drawn from scout cards and Forms 1002A.  Data
compiled included name of operator, date of completion,
completion-zones, methods of completion, initial oil
production, initial water production, and duration of initial-
production tests.  Lease boundaries and cumulative production
were recorded.  From this information-base wells that produced
exclusively from the Hunton Group could be identified.  Records
from these wells were used to make a production-trend map
(Figure 23).

Production-trend Mapping

     Initial-potential oil production was reported in several
manners.  The most reliable data were obtained when wells were
allowed to produce for a full 24-hr period.  Wells tested in
this fashion are coded by the letter "a" in Figure 23, an
isopotential map.  Some wells were tested over a period of a
few to several hours, typically in the range of 6 to 10 hr;
production was recorded hour-by-hour.  Data from these wells
were extrapolated to estimates of 24-hr-equivalent production.
In Figure 23, wells with initial potential estimated in this
fashion are coded by the letter "b".  Production from other
wells was reported as increments of a few hours; for example,
200 bbl produced during the first 4 hr, 160 bbl during the next
6 hr, and so on.  Duration of the total test generally was less
than 24 hr.  Such data were extended to estimates of 24-hr
production.  Data of this sort are coded as "c" in Figure 23.
In some instances a single production-value was reported for an
initial test of less than 24 hr.  The average hourly production
rate was determined and multiplied by 24 to estimate the daily
production rate.  Estimates of this sort are shown in Figure 23
by the code "d". This method produced 24-hr rates that probably
were somewhat larger than the true rates.  To minimize the
likelihood of serious error in interpretation, no data of this
type were used if the duration of the test was less than 18 hr.

     Contour interval of Figure 23 is 500 bbl per day, which
tends to smooth the error inherent in estimates of initial 24-
hr production rates.  The isopotential map shows somewhat
linear patterns of large or small production.  The basic


                                48

-------
                            R. 4 W.
                                                          T. 14 N.
Figure 23. Production-trend map,  central part,  West Edraond
           Field, showing initial-production potential in
           barrels of oil per day.   Contour interval 500 bbl
           per day.  Unbroken wide straight lines show trends
           of areas of large production.  Dashed straight
           lines show trends of areas of low production.
           Subscripts: a - Full 24-hr production test, b -
           Hour-by-hour tests extrapolated to 24-hr equivalent
           production, c - Serial tests of a few hours'
           duration, extrapolated to 24-hr equivalent
           production, d - Single-stage test of less than 24
           hr.  Average hourly production extrapolated to 24-
           hr equivalent production.
                               49

-------
pattern shown in Figure 23 extends throughout the field.
Lineations of production trends are oriented generally in
northeast- and northwest-trending sets.   Average orientation of
large initial-production lineations is N. 48 deg. E.  and N. 36
deg. W.  Basic configuration of the production-trend map
(Figure 23) suggests initial rates of production.  Systematic
fracturing of the Hunton Group in West Edmond Field has been
described previously (see Swesnik (1948), and Elkins (1969),
for example).

Structural Contour Mapping

     Figures 24 and 25 show structural configuration of the top
and base of the Hunton Group, respectively.  The upper surface
of the Hunton was eroded during the regional post-Devonian,
pre-Mississippian hiatus, and during the post-Mississippian,
pre-Pennsylvanian hiatus.  Therefore some of the variation in
shape of the upper surface -of the Hunton can be attributed to
gentle erosional relief.  Consequently,  structure of the base
of the Hunton is the more reliable estimator of general
deformation of the reservoir, especially in the eastern, up-dip
part of the field, where wells that penetrated the entire
reservoir are more abundant.

     Final interpretation of faults in West Edmond Field was
based on three general lines of evidence:  (1) Repeated or
omitted rock-stratigraphic units were bases for interpretation
of reverse and normal faulting respectively.  (Evidence of this
kind was sparse.)  (2) Abrupt differences in elevation between
or among close-by wells were regarded as being indicative of
anomalous changes in strike or dip.  Faulting seemed to be the
more probable interpretation at localities where contour lines
indicated large changes in strike in short distances, where dip
changed in an extraordinary amount, or both.  (3) The
previously mentioned linear trends in initial production were
considered to be suggestive of faulting.

     In the course of normal geologic field mapping where faults
are exposed locally in outcrops and concealed in the reaches
between, mapping of faults commonly is strongly inferential;
interpretations are made on bits of convergent evidence, some
of strongly differing kinds.  Mapping of faults in the
subsurface commonly is more inferential and consequently less
definite.  The number of faults shown in Figure 25 could have
been reduced by simple rearrangement of contour lines;
elevations posted to the map could have been explained with
fewer faults.  But if the object of mapping is to delineate
areas in which faults might extend into the sealing beds above
oil reservoirs, the mapping could be described as depiction of
working hypotheses — the illustration of localities that might
                                50

-------
                           R. 4 W.
                                                         T. 14 N.
                             Km.  1
Figure 24. Structural geology, top of Hunton Group, central
           part, West Edmond Field.  Contour interval 25 ft
           (7.6 m). Elevations relative to mean sea level.
           Broad "straight" lines: Inferred faults.  Broad
           dashed lines: Inferred faults with less supportive
           evidence.
                              51

-------
                          R. 4 W.
                                                         T. 14 N.
                         0
                         I
Mi.
                           0  Km.  1
Figure 25. Structural geology, base of Hunton Group, central
           part, West Edmond Field.  Contour interval 25 ft
           (7.6 m). Elevations relative to mean sea level.
           Subscript "e" signifies estimated elevation.  Broad
           "straight" lines: Inferred faults.  Broad dashed
           lines: Inferred faults with less supportive
           evidence.
                               52

-------
be tested for penetrative fractures, if and when the means of
testing becomes available.

     In the case at hand, the hypothesis that some faults might
penetrate confining beds to or near the level of underground
drinking water was tested by mapping the structural geology of
shallow marker beds.  The mapping datum was the provisional
base-of-Permian marker bed (Figure 26), which is about 1500 ft
(450 m) below the surface.  The general pattern of folding at
this datum is similar to that of the base of the Hunton Group,
but appreciably gentler  (compare Figures 25 and 26).
Throughout the West Edmond Field, faults interpreted at the
level of the "base-of-Permian" marker were about one-half as
abundant as those interpreted in the Hunton Group  (Kennedy,
1989) .

                     MAPPING OF LINEAMENTS
     Formations that crop out in the study area are composed
mostly of soft sandstones and shales and surficial deposits.
In general they are weakly resistant to erosion and uplands
evolve to terrain made up of low hills where sandstone is
bedrock and gently sloping open lands where shale is bedrock.
Because extensive, resistant tabular beds are few, distinctive
and easily correlated strata are not abundant, and the bedrock
tends not to erode in a strongly differential fashion, mapping
of faults at the surface by ordinary field-geologic methods is
quite difficult and time-consuming. The working hypothesis that
faults penetrate rocks at the surface, and that they are
manifest on Landsat imagery and color-infrared imagery was
tested by mapping of lineaments.

     Landsat imagery at the scale of 1:12,000 and high-altitude,
near-infrared photographs at the scale of 1:60,000 were
examined.  Care was given to exclude man-made features.
Figures 27 and 28 show lineaments interpreted from Landsat
imagery and color-infrared imagery, respectively.  Most of the
lineaments mapped are associated with uncommonly straight
segments of streams, but a few simply are linear parts of
stream-valley walls or linear tonal anomalies in upland terrain
(see Figure 27, lineament in extreme southeastern part of
Section 28, T. 14 N., R. 4 E., and eastward-trending lineament
in Section 20, T. 14 N., R. 4 E., respectively). The
interpretations from Landsat imagery and from infrared imagery
are different in some localities (for example, compare Sections
15, 16, 20, and 28, T. 14 N., R. 4 W.) and similar in others
(for example, compare Sections 2, 10, 35, and 28, T. 14 N., R.
4 W.) .


                                53

-------
                           R. 4 W.
                                                         T. 14 N.
                         9
Mi.
                              Km.  1
Figure 26. Structural geology, "base-of-Permian" marker bed,
           central part, West Edmond Field.  Elevations
           relative to mean sea level.  Contour interval  10
           ft (3 m).  Broad "straight" lines: Inferred faults.
                              54

-------
                            R. 4 W.
                                                   T. 14 N.
                              Mi.  1
                             OKm.1
Figure 27. Lineaments interpreted from Landsat imagery,
           central part, West Edmond Field.  Broad, dark,
           straight lines: Lineaments in approximate positions
           of faults mapped by Kennedy (1989).  Narrow,  dark,
           straight lines: Lineaments not in approximate
           positions of faults mapped by Kennedy  (1989).
                              55

-------
                           R. 4 W.
                                                         T. 14 N.
                              Mi.
                           0  Km.  1
Figure 28. Lineaments interpreted from color-infrared  imagery,
           central part, West Edmond Field.  Broad, dark,
           straight lines: Lineaments in approximate positions
           of faults mapped by Kennedy (1989).  Narrow, dark,
           straight lines: Lineaments not in approximate
           positions of faults mapped by Kennedy  (1989).
                               56

-------
     If, for the sake of logical argument,  one assumes that
structural configuration of the base-of-Permian is virtually
correct as mapped, and that all faults identified at the base-
of-Permian marker penetrate to the surface, then approximately
50 percent of the lineaments interpreted from Landsat imagery
seem to correlate to faults at the base-of-Permian marker
(Kennedy, 1989).  If fewer than all the faults mapped at the
base-of-Permian marker extend to the surface, then detection of
them by mapping lineaments on Landsat imagery is somewhat more
efficient than 50 percent. The reader will  be quick to
recognize that reliability of such  statements is based on
compound probability: (Probability of correct interpretation of
configuration of the base-of-Permian marker)  X (Probability of
correct interpretation of lineaments from Landsat imagery) =
Probability of correct correlation of lineaments to through-
going faults in the shallow subsurface.  In virtually any such
interpretive geologic exercise, the probability of correct
correlation of lineaments to through-going  faults in the
subsurface would be less than 100 percent — and in most such
exercises the probability would be much less than 100 percent.

     According to Kennedy (1989) about one-half the faults
mapped at the level of the Hunton Group are indicated at the
level of the base-of-Permian marker.  In keeping with the
qualitative probability statement above, if faults at the
levels of the Hunton and base-of-Permian were mapped correctly
and if Landsat imagery were interpreted correctly, then about
one in four faults in the Hunton Group could be detectable by
Landsat imagery.  Because data in subsurface geologic mapping
can be interpreted in at least a few different ways, and
because no worker is likely to make the indisputably correct
detailed interpretation, Kennedy's  (1989) observations lead to
the conclusion that of faults in the Hunton Group, less (and
perhaps much less) than one in four should be detectable on
Landsat imagery.

                   INJECTION-SENSITIVITY MAPS


     In the exercise under discussion, the object of endeavor
was to isolate parts of the terrain within which upward
migration of injected wastes through faulted confining beds is
a likelihood that seems to merit close attention — if in fact
such localities were to exist. In subsurface geologic mapping,
one source of information — one kind of document — rarely is
definitive.  Stable conclusions, or conclusions that are
regarded as likely to be true, commonly are based on convergent
evidence.  In the ideal, conceptual case, independent sources
of  evidence are evaluated after the fact of their having taken
form.  This procedure is meritorious and it tends to minimize


                                57

-------
the drift toward a favored interpretation.   In such a manner of
operation the former source of information tends not to bias
the shape taken by the latter.  (But the interpretation of the
latter source may be biased, and constructively so (for a brief
example of constructive geologic bias see Low (1957,  p. 219-
220))).  In the practical case, an unpremeditated carry-forward
of information that seems to be valid may produce a final
interpretation that is quite useful for operational purposes.

     In the circumstance at hand,  initial-production trends
seemed to be in rectilinear patterns, and to be suggestive of
fracturing in the reservoir. Fracturing of the Hunton at West
Edmond Field and dissolution of reservoir rock have been
discussed extensively (McGee and Jenkins, 1946; Swesnik, 1948;
Elkins, 1969).  The proposition that the fabric of dissolution
is strongly influenced by patterns of fractures is supported by
rather common geomorphic evidence.  The evidence chiefly is
joint-controlled differential erosion of carbonate rocks and
evaporites, and information compiled from the mapping of caves
(for example, see Thornbury (1956, p. 337).  Moreover, the
initial-production trends are oriented more-or-less in keeping
with regional trends of fractures in rocks at the surface
(Melton, 1929; Shelton and others, 1985, p. 39).  Thus to
regard the trends of large production as being reasonable cause
for the hypothesis of many small-scale faults seemed
reasonable.  On structural contour maps, where trends of large
production were located at or near places of abrupt change in
elevation of the datum, or in strike, or dip, the mapping of a
fault seemed to be a respectably probable interpretation.
Thus, maps were used as guides from former to latter, with
trends of initial production having been regarded as a rather
strong initial basis for indication of fracturing in the
reservoir.

     The interpretation of Landsat and color-infrared imagery
(Figures 27 and 28) did not involve transfer of information
from maps of the subsurface.  Table 1 shows examples of grading
of evidence and convergence of evidence. Configuration of
structural contour maps of the base and top of the Hunton Group
(Figures 24 and 25) was influenced but not controlled by the
map of initial-production potential  (Figure 23). Thickness maps
of the Hunton Group  (Figures 29 and 30) were influenced by
structural contour maps of the Hunton.  The base-of-Permian
structural contour map  (Figure 26) was made with consideration
of the structural geologic maps of the Hunton Group.   Of the
nine sources of information, infrared imagery, Landsat imagery,
thickness of Woodford Shale  (Figure 31) and initial-production
potential effectively are independent.  The Woodford Shale
(Figure 31) was mapped  independently, to test the proposition
that faults mapped in Sections 10 and 16 and in Sections 11,
                                58

-------
                           TABLE  1.  EXAMPLE,  CONVERGENT  EVIDENCE OF FAULTING, SELECTED  LOCALITIES,

                                    T.  14  N.,  R.  4  W., WEST EDMOND FIELD
Locations
Inferred
Faults
SEC. 10 &
16
SEC. 11,
15, 21, 29
SEC. 12,
14, 23, 27

SEC. 35


Trend
NE


NE
NE


NE, NW

GENERAL QUALITY OF EVIDENCE, AS INDICATOR OF FAULTED TERRAIN
I.R.
IMAGERY
(FIG. 28)
GOOD

FAIR

GOOD


GOOD

LANDSAT
IMAGERY
(FIG. 27)
ABSENT

POOR

GOOD
(12,14,
23)
GOOD

STRUCTURE
BASE
PERMIAN
(FIG. 26)
FAIR

FAIR TO
GOOD
FAIR


POOR

THICKNESS,
WOODFORD
(FIG. 31)
FAIR

FAIR TO
GOOD
NOT
APPLICABLE

NOT
APPLICABLE
STRUCTURE,
FRISCO-BOIS
D'ARC
(FIG. 24)
FAIR

FAIR TO
GOOD
FAIR


NOT
APPLICABLE
THICKNESS,
FRISCO-BOIS
D'ARC
(FIG. 29)
GOOD

FAIR

POOR TO
NOT
APPLICABLE
NOT
APPLICABLE
STRUCTURE,
BASE
HUNTON
(FIG. 25)
FAIR

FAIR

GOOD


FAIR TO
GOOD
THICKNESS,
HUNTON
(FIG. 30)
POOR

POOR

POOR


UNCERTAIN

INITIAL-
PRODUCTION
POTENTIAL
(FIG. 23)
GOOD

GOOD

NOT
APPLICABLE

NOT
APPLICABLE
JUDGMENT OF
CONVERGENCE
FAIR

FAIR

FAIR


FAIR TO
GOOD
Ol
10

-------
                         R. 4 W.
                                                        T. 14 N.
                           0  Km.  1
                           I	i
Figure 29. Thickness of Frisco-Bois d'Arc Formations, central
           part, West Edmond field.  Contour interval 10 ft.
           (3 m) . At some localities, map contoured as if
           thickness varies in direct relation to faulting.
                              60

-------
                          R. 4 W.
                                                         T.  14 N.
                         9
Mi.
                              Km.  1
Figure 30. Thickness of total Hunton Group, central part, West
           Edmond Field.  Contour interval 20 ft  (6.1 m).  At
           some localities, map contoured as if thickness
           varies in direct relation to faulting.
                              61

-------
                    R. 4 W.
                       Miles
                    0 Kilometers 1
                     I	i
Figure 31. Thickness of Woodford Shale,  central
           part,  West Edmond Field.  Contour
           interval 10 ft (3 m).
                                                     T. 14 N.
                          62

-------
15, 21 and 29, T. 14 N., R. 4 W. would be indicated by
variation in thickness.  Figure 31 tends to support the
interpretations of faulting.  In general, evidence from color-
infrared imagery and Landsat imagery is supportive (Table 1).

     Kennedy  (1989) did not presume that the mapping of faults
was to be maximized, nor did he stress the data to project
faults to the base-of-Permian marker and to argue thereby for
penetration of confining beds.  The constructive-bias method
was used with strong geologic inference, with the underlying
intention of showing localities where evidence raised the
suspicion that confining beds might have been breached.  The
working hypothesis of faulting would serve as  a guide for
description of areas of suspected injection-sensitivity.

     Figure 32 is an injection-sensitivity map of much of T. 14
N., R. 4 W.  The purpose of the map is to show a method of
illustrating areas in which faults might be suspected to
penetrate high into the stratigraphic section — areas where
injected fluids might migrate to fault planes and upward to
shallow formations.  Convergent evidence and mapping of the
kinds described in reference to Table 1 were the basis for
inference.  Faults in West Edmond Field seem to be almost
vertical (Kennedy, 1989); Figure 32 is based partly on the
assumption that faults would be within 6 deg. of vertical.
At the outer border of the Zone of Sensitivity a fault inclined
from vertical by as much as 6 deg.  would be encountered at
about 5000 ft (1500 m).  If the injection zone were at about
3000 ft (900 m), fluid in such a well would be injected about
200 ft (60 m) from the fault zone.

     Within the localities shown as Zones of Sensitivity, plans
for disposal  of fluids by injection should include very close
study of surface-geologic evidence and subsurface-geologic
evidence from surface casing downward, in order to estimate
better the probability of faulting in confining beds.  Within a
Zone of Sensitivity, disposal by injection could follow
assembly of new information and persuasive demonstration that
the likelihood of faulting is acceptably low.  New information
might include seismic or other geophysical surveys, production
histories of oil and gas wells, histories of injection wells
already in place, and so on.   (The reader will recall that by
intent, the work reported here involved interpretation of data
taken from ordinary sources, the mapping was planned to be of
standard kinds, and the purpose was to accent localities in
which underground injection could be unwise.)

     At the outer border of the Zone of Caution (Figure 32),
injection at about 3000 (900 m) would be at a point about 1000
ft (300 m)  from a fault plane inclined from the vertical by 6
                                63

-------
                            R. 4 W.
                                                           T. 14 N.
Figure 32. Central part of West Edmond Field: Example of
           injection-sensitivity map. Broad, gently curved
           lines show positions of inferred faults.  Bounding
           patterns show Zone of Sensitivity (nonstippled) and
           Zone of Caution (stippled).  Here underground
           injection would be advisable if closer study
           generated evidence to show that  (a)  faults are
           absent, or (b) convincingly improbable, or (c)
           fault planes are sealed, or (d) some combination of
           b and c. Example,  use of inset figure: At outer
           boundary of Zone of Sensitivity, a well about 5000
           ft (1500 m) deep would penetrate a fault plane
           dipping  6 deg.  At outer boundary of Zone of
           Caution, injection at about 3000 ft (900 m) would
           be about 1000 ft (300 m) from fault plane dipping
           at 6 deg.

                               64

-------
deg. toward the injection well (see inset, Figure 32).
Assuming that disposal would be at about this depth or
shallower and that the working hypothesis of faulting is not
rejected, then special new evidence may lead to the consensus
that underground disposal could take place with defined
precautions (for example, certain volumes, pressures, and
methods of monitoring).

     Throughout deliberations about Zones of Sensitivity and
Caution, full consideration should be given to the proposition
that the (suspected) faults are sealed.  If unsealed faults
were penetrative from the reservoir rocks to the surface,
petroleum should have migrated from the trap long ago.   Or, if
the trap were young and migration were in progress, unsealed
faults should be detectable by oil or gas seeps.  Injection in
the Zones of Sensitivity and Caution may be harmless if
penetrative faults are sealed, if injection is into the
petroleum reservoir, if pressures do not exceed original
formation pressure, and if volume of fluid injected does not
exceed the volume of petroleum and salt water withdrawn in the
past.

                          CONCLUSIONS
     Initial-production trends in West Edmond Field indicate
that permeability in the Hunton Group is strongly fracture-
influenced.  Initial-production trends, conspicuous deviations
in strike or dip along linear trends and repeated or missing
stratigraphic sections combine to permit valid interpretation
of faulting in the Hunton Group.

     Prudently used constructive-bias methods of mapping allow
projection of some of the faults mapped in the Hunton Group to
a provisional base-of-Permian marker.  Some lineaments mapped
from Landsat imagery and color-infrared imagery seem to
correlate in position and trend to faults mapped in the Hunton
Group.  Convergence of information from surface and subsurface
geologic mapping permits the assessment of terrain for
injection-sensitivity.  Zones of Sensitivity and Zones of
Caution accent localities within which underground injection
would be recommended only after conscientious study, which
perhaps would require special sources of information.

     The general techniques of subsurface-geologic mapping
described in reference to the West Edmond Field contrast
severely with methods of computer mapping described in Section
4.  Personnel who set out to make injection-sensitivity maps
can choose among the relative merits of computer mapping and
the more subjective constructive-bias mapping.  The fragmentary
                                65

-------
nature of subsurface geologic data allows multiple detailed
interpretations.  Geologists with strong empirical backgrounds
for pattern-recognition probably would choose to make
injection-sensitivity maps by constructive-bias mapping.
Computer mapping could be taken into account as a matter of
policy, to provide baseline interpretations that are likely to
minimize controversy between producers and regulating agencies.
                                66

-------
                           SECTION 4

                      BURBANK OIL FIELD

                         INTRODUCTION
     The Burbank Oil Field in Kay and Osage Counties, Oklahoma
(Figures 1 and 33), is a large stratigraphic trap. The
Pennsylvanian Desmoinesian "Burbank Sandstone" reservoir
(Figure 34) is encased in shale with oil and gas trapped
against the up-dip limit of the reservoir rock (Figure 33).
Discovered in 1920, Burbank Field has undergone several stages
of development, including secondary and tertiary enhanced-oil-
recovery projects.

     Hagen (1972) studied trends of subsurface fractures and of
joints in bedrock at the surface. The research produced strong
evidence of systematic fractures within the Burbank Sandstone
and in strata immediately above the reservoir.  When the
formation was pressured by water-flooding, these fractures were
opened to migration of fluids from well to well.  Hagen
indicated that primarily, overpressuring during secondary
recovery was the cause of opened fractures in and above the
Burbank reservoir.

          MAPPING OF SUBSURFACE GEOLOGY BY COMPUTER
     A data base was compiled from wireline logs of 934 wells
(Figure 35). Data pertained to the Cottage Grove Sandstone
(Figure 34), a potential storage formation, and confining beds
within and above the Cottage Grove. Maps used to evaluate
confining-bed integrity in the Burbank Field included: (a)
Structural geology of two rock-stratigraphic units (Figures 36
and 37), (b) thickness of the interval between the Cottage
Grove Sandstone and the Pink Limestone (Figures 34 and 38),
(c) thicknesses of rock units and percentage of shale in the
Cottage Grove (Figures 39, 40, and 41), (d) thicknesses and
extents of certain confining beds (Figures 42, 43, and 44), and
(e) number of formations above the Cottage Grove (Figure 45)
believed to be porous and permeable enough to transmit fluids
in large quantities.

     Lotus 1-2-3 spread-sheet program was combined with mapping
by the Jupiter Mapping Program and printing by multiple-pen
line plotter.
                                67

-------
          R. 5 E.
R. 6 E.
R. 7 E.
                                                   T. 27 N.
                                                   T. 26 N.
                                                    T. 25 N.
                            Miles
                           Kilometers
Figure 33. General configuration of Burbank  Field.   All but
           the northwesternmost part of the  oil  field is in
           Osage County, Oklahoma.  The Burbank  Sandstone dips
           southwestward; the arcuate eastern  boundary of the
           field is the up-dip limit of permeable sandstone.
           The reservoir is a multistoried,  multilateral,
           alluvial-deltaic channel-fill  reservoir.   In
           general, trends of the more permeable  sandstone
           "stringers" follow the configuration  of the eastern
           border of the field.
                              68

-------

COTTAGE GROVE SS.
COTTAGE GROVE SS. [BASE]

r» -n
g METERS C
IX sj
CHECKERBOARD LM.
vO OSWEbO LM.
i—
LU
LU
U_
-100
PINK LM.
BURBANK SS.
MISSISSIPPI LM.











-

































-



.-

~~i






-
-


\












~




-
-


—







f,
«-

-<





t
p»


c





-*

t,






5

i





-••
-
~»>—

^
Ji
^
"


-


•^.

«



c

~"

1

-
-



-
.

-
.
-
-"


-





:'
-h
-

-


-

:







-






-;
k
"

(
_!
1
r
_i.-

i
<

i
(
~
I _



2

i

j£r



"^ ?5
TC !_

^2
^
/

^2
/
_ S
'

f
— E
C
--,
»»i

2.

i-
'1

* c
>
;>






i

3t
















~





.1-
^
s

r

^
<


5

n
-
•w



-














D
=i




"

^


-


_











-




V
-






L















-





-

















10
























f.


<










^,

1
<
I
r
(1
(
-.

(
-
c
(
V

„
^*

<•
L
"^








V


d

-
^

a.

i


i
"A












u

-h
:•"


."-

^T



Figure 34. Type electric log, Burbank Field.  Important
           marker-beds are shown.  All strata shown above
           "Mississippi Lime" are Pennsylvanian.
                            69

-------
                             R. 6 E.
                                                T. 26 N.
                                                T. 25 N.
0
                           Kilometers
                                               10
Figure 35. Locations of wells in Burbank Field  from  logs of
           which data base was compiled.  Lack  of  control in
           some areas of the field is due to the fact that
           much of the drilling predated the application of
           wireline logs.
                              70

-------
                              R. 6 E.
               0
Miles
                            Kilometers
                                                  T.26 N
                                                  T. 25 N.
Figure 36. Structural geology, Cottage Grove Sandstone
           (Figure 34), Burbank Field.  Contour interval 20 ft
           (6.1 m) .  In Sections 30 and 31, T. 25 N., R. 6 E.,
           a large syncline is shown.  Figure 35 shows absence
           of data in this area.  The syncline is an artifact
           of software; mapping algorithm must consider
           absence of data in Sections 30 and 31, and from
           beyond boundaries of map.  In northernmost corner
           of map locations of wells and elevations of the
           datum are plotted.  At this scale, to include data
           and contour lines is not practical.  But scale can
           be changed, if appropriate for solution of problem.
                               71

-------
                             R. 6 E.
            Mi.
                                                 T. 26 N.
                                                 T. 25 N.
Km.
Figure 37. Structural geology, Pink Limestone, Burbank Field
           (Figure 34), an extensive tabular formation that
           is excellent for use as a mapping datum.  Contour
           interval 20 ft (6.1 m). At Burbank Field
           and nearby anticlinal and synclinal noses, and
           anticlines and synclines are superimposed on a
           regional, westward-dipping homocline.  Mapping  from
           data base, at scale shown, is not sufficient to
           detect faulting,  if indeed detectable faulting  is
           present.
                               72

-------
                             R. 6E.
     0
          J Mi.
                                                T. 26 N.
                                                T. 25 N.
Figure 38. Thickness of stratigraphic interval, Cottage Grove
           Sandstone to Pink Limestone, Burbank Field.
           Contour interval 20 ft (6.1 m).  The map was
           designed to test the working hypothesis that
           anomalous thickening or thinning of the interval
           would indicate faulting.   Thickening of the
           interval in a linear trend northeastward across
           south-central part ofT.  26N.,R. 6E. is
           suggestive of faulting (but not diagnostic of
           faulting). Hachured, closed contours show areas of
           thick rock.
                              73

-------
                            R. 6 E.
                                                T. 26 N.
                                                T. 25 N.
                                                Km.

Figure 39. Thickness of Cottage Grove Sandstone, Burbank
           Field.  Contour interval 10 ft (3 m).  Compare
           Figures 35 and 39 for evidence that the map is
           rather interetive where data are sparse, with
           tendency to show ovate "thicks" and  "thins " at
           such places.  In general the map is quite useful,
           It would be especially useful for general
           assessment of the extent and thickness of an
           injection zone.
                              74

-------
                             R. 6 E.
                                //   \Cc
                                                T. 26 N.
                                                T. 25 N.
                                               Km.

Figure 40.  Thickness of  net sandstone in Cottage Grove
           Sandstone,  Burbank Field.  Contour interval 10 ft
           (3m).   Map shows thickness of sandstone in Cottage
           Grove that should be of "reservoir-quality".
           Interbeds of  shale were eliminated in calculation.
           The map would be useful for general assessment of
           storage-unit  potential.  Effects of sparse data and
           map-boundary  effects are apparent by comparison
           of Figures 35 and 40.
                              75

-------
                              R. 6 E.
                                                 T. 26 N.
                                                 T. 25 N.
                             Miles
                           Kilometers
Figure 41. Thickness of net shale  in Cottage Grove Sandstone,
           Burbank Field.  Contour interval 10 ft (3 m) .   This
           map would be useful  for broad assessment of
           potential confining  beds within the Cottage Grove
           Sandstone.
                               76

-------
                             R. 6 E.
                                                T. 26 N.
                                                T. 25 N.
                           Kilometers
10
Figure 42. Thickness of confining unit directly above Cottage
           Grove Sandstone, Burbank Field.  Contour  interval
           20 ft (6.1 m).  Maps of this type would be useful
           in general evaluation of confining-bed potential.
           Merging of lines is due to small scale and small
           contour interval, both of which could be  modified
           during use of the software.
                              77

-------
                               R. 6 E.
                                                  T. 26 N.
                                                  T. 25 N.
                                                 Km.
Figure 43. Cumulative thickness of shale above Cottage Grove
             Sandstone, Burbank Field.  Contour interval 50 ft
             (15.2 m).  The map shows general distribution of
             total shale (and total confining-bed potential) in
             stratigraphic section overlying Cottage Grove.
             Boundary-effects and sparse-data effects are
             apparent (for example, northeastern corner of map
             and hachued contour in southwest-central part of T.
             26 N., R. 6 E.).
                               78

-------
                             R. 6 E.
     I	1 Mi
Mi.
                                                 T. 26 N.
                                                 T. 25 N.
                                                      Km.
Figure 44. Total number of shale "breaks" in stratigraphic
           section above Cottage Grove Sandstone/ Burbank
           Field.  Each shale break is 20 ft (6.1 m) thick or
           thicker, by definition.  In effect, the map shows
           the number of confining beds above the Cottage
           Grove.  This map would be useful for general
           evaluation of the study area. False patterns
           showing no shale breaks are shown at some places,
           owing to scarcity of data and to map-edge effects.
                              79

-------
                              R. 6 E.
                                                 T. 26 N.
                                                 T. 25 N.
               0
J Mi.
0  1
                                                Km.
Figure 45. Possible injection zones between top of Cottage
           Grove Sandstone and depth of about 1000 ft  (305  m),
           Burbank Field.  Map shows number of sandstone
           formations as thick as  about 6 ft (2m) or
           thicker.  These data also would be useful for
           estimation of the number of reservoirs into which
           upwardly mobile, ovexpressured fluids might escape.
                               80

-------
STRUCTURAL GEOLOGY

     Configuration of the Pink Lime marker bed shows that
structural geology fundamentally is homoclinal, with dip
westward at about 1/2 deg.  The homocline is interrupted
by anticlinal and synclinal noses, and by a few anticlines and
synclines (Figure 37).

     Structural geology of the Cottage Grove Sandstone is
similar.  Figure 36 shows general westerly dip with few closed
folds.  However, because no data were available in the
southwesternmost part of T. 25 N., R. 6 E. the software
interpreted a closed syncline; minimal elevation is shown to
be less than -980 ft.

     Figure 38 indicates rates of variation in thickness of the
stratigraphic section between the Cottage Grove Sandstone and
the Pink Limestone marker.

     The Cottage Grove Sandstone extends throughout the study
area; its thickness and extent (Figure 39) indicate that the
formation has good potential as a fluid-injection reservoir.
Figure 40, a net-sandstone-thickness map shows the general
amount of reservoir-quality sandstones at specific sites,
exclusive of shale interbedded with sandstone  (Figure 41).
Figures 39, 40 and 41 illustrate that the Cottage Grove is a
heterogeneous reservoir.  Nevertheless, sandstone in the
Cottage Grove is thicker than 50 ft  (15 m) at most places.

     The Cottage Grove is overlain by a thick sequence of
clayey shale (Figure 42).  At some places this shale is thicker
than 200 ft (60 m), but at other localities it is as thin as
about 20 ft (6 m).  Thicknesses of about 100 ft (30 m) are
common; indeed in about 80 percent of the wells in Burbank
Field, Cottage Grove Sandstone is overlain directly by 100 ft
(30 m) or more of clayey shale.  In only about 6 percent of the
wells is the shale thinner than 80 ft (24 m).  (Thicknesses of
zero feet shown in Figure 42 are regarded as being artifacts
of the software.)

     Figure 43 shows the extent and thickness of cumulative
shale above the Cottage Grove interval.  Throughout the study
area the Cottage Grove is covered directly or indirectly by a
substantial thickness of confining shales; altogether,
cumulative shale is as thick as about 175 ft (53 m)  to 600 ft
(180 m).

     Figure 44 delineates the areas where multiple layers of
confining shale — 20 ft (6 m) thick or thicker — overlie the
Cottage Grove.   Based on the assumption that a stratigraphic


                                81

-------
unit of shale 20 ft (6 m)  thick or thicker would also be
extensive laterally, this map should be valuable in estimating
the confining-potential of shales.

     Figure 45 simply shows areas where possible injection-
reservoirs are above the Cottage Grove Sandstone and below the
general depth of about 1000 ft (300 m).  However, few of these
formations are as thick and extensive as the Cottage Grove.
The map also could be used for a general assessment of the
number of reservoirs into which injected fluids might escape,
if fluids were to travel upward through abandoned wells or
along unsealed fractures.

                RECOGNITION OF FRACTURE TRENDS
SUBSURFACE-MAPPING TECHNIQUES

     Initial investigation of geologic data pertaining to
Burbank Field and available to the public indicated that three
subsurface-mapping techniques could be useful in detecting
fracture trends in the reservoir: (1) An  initial-potential map
of wells drilled early in development of Burbank Field  (Figure
46),  (2) an effective-reservoir thickness map of the Burbank
Sandstone  (Figure 47), and (3) a structural contour map of the
Pink Limestone marker bed, which is  close above the Burbank
(Figure 34; general example of map is in Figure 37).  Initial-
production maps were considered to be unbeneficial, because
most leasehold blocks in the Burbank Field are 160 acres  (65
hectares) or more.  Spacing of wells is 10 acres  (4 hectares).
Therefore oil produced from more than one well would have been
measured at a tank battery.  Because most fracture trends are
believed to be quite narrow, on a 160-acre  (65-hectare) tract,
the large production of wells affected by fractures could have
been offset by small production from wells not affected by
fractures.  This production-monitoring problem was aggravated
by the historical formation of secondary-recovery units within
boundaries of the field, which makes the establishment  of
cumulative-production values for single wells nearly
impossible.  Moreover, no satisfactory method was found for
discounting the effects of artificial fracturing of the
reservoir.

Initial-potential Maps

      The initial-potential map of wells producing from  the
Burbank  Sandstone  (example, Figure 46) was based on data from
wells drilled before 1936.  For an initial-potential map to be
effective  in fracture-trend study of the Burbank reservoir,
factors influential on initial potential must be considered,
                                82

-------
                               R. 5 E.
R. 6 E
oo
to
                                          0  Km.  1
                                                 •
               Figure 46
                                            map/ Burbank Sandstone, Burbank
                                  Stippling shows areas within which initial-
                          potential production of wells was more than 1000
                          bbl per day.   Index to contour lines: 250 to 500
                          bbl per day;  500 to 1000 bbl per day; 1000 to 2000

                                       day?
                                                                                — T. 27 N.

-------
                               R. 5 E.      R. 6 E.
oo
                                                                                   T. 27 N.
                                              Km.  1
                                              	I
                 Figure 47. Thickness of effective  reservoir rocks,  Burbank
                            Sandstone, Burbank  Field.  Contour interval 25 ft
                            (7.6 m).  Stippling shows  areas within which
                            Burbank Sandstone is thicker than 75 ft (23 m).

-------
including: (a) Ages of wells and effects of reservoir-pressure
depletion on initial production, (b) completion techniques
(thickness of reservoir drilled and artificial stimulation),
(c) operator-introduced bias in reporting initial-production
rates, and (d) changes in reservoir character.

     To eliminate the effects of reservoir depletion on initial
potential, only data for wells drilled in the early stages of
Burbank Field were used.

     Completion techniques used were similar for all wells.
Most wells were drilled entirely by cable-tool rigs or else
the Burbank Sandstone was drilled with cable tools.  This
basic technique resulted in relatively undamaged reservoir
faces in the borehole, and hydrostatically unbalanced
conditions that allowed oil and gas to flow freely into the
borehole.  In most wells the reservoir was fractured by
nitroglycerin, to enhance flow.  Therefore, "after-shooting"
flow rates were used where available.  Because of the ages of
the wells used in the study, no flow rates after hydraulic
fracturing were used to map initial-production rates.

     Most wells in the Burbank Field were drilled into the
Burbank Sandstone until a satisfactorily large flow of oil and
gas was developed or until the entire sandstone section had
been exposed.  This practice was not followed near the western
border of the field because operators recognized an oil-water
contact.  They drilled into the top part of the reservoir,
exposed several feet of rock, and stopped drilling above the
oil-water contact.  The amount of reservoir exposed by
drilling seems not to have had an overriding effect on
initial-production rates; very large initial flow-rates were
recorded from wells drilled close to the oil-water contact.

     Operator-introduced bias due either to inflation or
suppression of initial flow rates is believed to have been
insignificant in most areas of Burbank Field.  As was noted
earlier, leases in the field covered at least 160 acres  (65
hectares); much variation in initial potential can be
recognized within the boundaries of a 160-acre (65-hectare)
tract.  However, operator-introduced bias cannot be eliminated
entirely; this fact creates doubt about the validity of
large initial-potential trends situated along boundaries
between leases of different operating companies.

     The most significant factor in initial potential of
wells in the Burbank Sandstone was variation in reservoir
quality or thickness.  The Burbank Sandstone is composed of
channel-fill units, "stacked" by cut-and-fill, and
multilateral.  Individual channel-fill, "shoestring" sandstone
                                85

-------
units or combinations of units were porous, permeable, highly
productive reservoirs.  However, the uppermost and youngest
sandstone unit was an exceptionally good reservoir across the
entire field.  Many wells near the oil-water contact drilled
only a few feet of this uppermost unit, but they produced at
larger rates than wells where much thicker sections of the
Burbank were drilled and the uppermost unit was absent. The
highly productive uppermost sandstone has significant impact on
reliability of fracture-trend recognition in Burbank Field.

Reservoir-thickness Maps

     Owing to the stacking of channel-sandstone units,
sandstone thicker than 75 ft (23 m) is common, as is abrupt
thinning.  The initial-potential map (Figure 46) and
reservoir-thickness map  (Figure 47) indicate that large
initial-potential trends and thick reservoir trends are
correlated closely over the entire field, except in part of T.
27 N., R. 5 E.  An easterly trend of large initial potential
is along the northern boundaries of Sections 22, 23, and 24.
This trend appears to be "normal" to the northerly and
northwesterly trends of thicker Burbank sandstone reservoir-
rock; it may be evidence of fracture-enhanced initial
production.  However, in form of a counter-argument, different
operating companies drilled wells in Sections 13, 14, and 15
and wells in Sections 22, 23, and 24.  Operator-introduced
bias cannot be eliminated as a possibly significant origin of
large differences in initial potential of wells along the
boundary between these two areas.

Structural Geologic Maps

     In study of West Edmond Field, structural contour maps set
out considerable evidence to support the interpretation of
faults in the reservoir.  Structural geologic maps of parts of
Osage County are published and  copyrighted by the Osage Tribe
of Native Americans.  Structural contour maps of Burbank Field
were reviewed, but no significant evidence of  faulting was
recognized.  Electric-log surveys are  scarce over large areas
of the field, which makes the search for missing stratigraphic
section on logs quite difficult.  Such omissions of  strata
commonly  indicate normal faults of small throw, which would not
be manifest  in ordinary  structural contour mapping.
Nevertheless, the hypothesis that  structural geologic mapping
would indicate faulting  in the  reservoir was tested  by mapping
the  configuration of  the Pink Limestone, which  is a  short
distance  above the Burbank Sandstone  (Figure  34).  At  a contour
interval  of  10 ft  (3  m)  the map showed considerable  evidence of
folding by differential  compaction of  shale between  the Pink
Limestone and the Burbank  "shoestring"  sandstones,  but little
                                86

-------
of the evidence would be explained best by faulting.
Considering the available published maps,the scarcity of
evidence about faulting recoverable from them, and the paucity
of electric-logs in some areas of the field, the conclusion was
reached that for the purpose of detecting faults of small
displacement, structural contour maps yield little useful
information.

Conclusions

     Study of the Burbank reservoir by ordinary subsurface-
mapping methods and testing for recognition and delineation of
fracture-trends had these results:
     (1) Abrupt changes in thickness, and the general linear
geometry of the Burbank channel-fill sandstone reservoir make
mapping of fracture-trends by standard subsurface-geologic
methods an endeavor of small reward.
     (2) Fracture trends may be identifiable in some sandstone
reservoirs where initial-potential trends or initial-
production trends are "normal" to reservoir-thickness trends,
or otherwise are conspicuously detectable.
     (3) In the Burbank Field, initial-potential maps are
valuable for mapping trends of highly productive rock, but
initial-production maps cannot be used effectively for this
purpose.
     (4) The likelihood of operator-introduced bias in
reporting initial-production rates seems to justify reserved
judgment about validity of large initial-production trends
located along boundaries of leases owned by different
operators.
     (5) In the Burbank Field, the quality of reservoir
exposed for production had more influence on initial
production rates than total thickness of reservoir drilled.
     (6) Effects on initial-potential production by
reservoir pressure depletion and hydraulic-fracturing
techniques were eliminated by using wells drilled early in
development of the field.
     (7) Explosion-fracturing of the Burbank reservoir had
greater effect on initial productions of wells with small
natural-production rates than on wells with large natural-
production rates.
     (8) Study of the Burbank reservoir strongly indicates that
abrupt changes in thickness and geometry make most channel-fill
sandstone reservoirs generally unsuited for use in developing
methods to detect fracture trends by standard methods of
mapping. Attention should be directed toward sandstone-
reservoirs of other depositional settings.
                                87

-------
                  SURFACE-MAPPING TECHNIQUES


     Satellite imagery of the eastern part of Burbank Field was
studied to map lineaments.  Imagery consisted of a Band-7,
black-and-white image from Landsat 5, at an approximate scale
of 1:1,000,000.  A zoom-magnifier enlarged the image by a
factor of five.  Lineaments mapped on the magnified image were
grouped with topographic-map lineaments and with joint clusters
mapped by Hagen (1972).   Satellite-image lineations follow
stream-drainage patterns closely and also linear features
detectable on topographic maps.  From an empirical point of
view, a strong relationship seems to exist between orientations
of joint patterns mapped by Hagen (1972) and lineations
detectable on satellite images and topographic maps.

     Figure 48 shows the mapped orientations of joint patterns
recognized by Hagen (1972) from surface geologic mapping and
photogeologic mapping.  These joint patterns are represented by
single lines on the map, but they were shown as joint-clusters
on Hagen's map.  Principal orientations of joint clusters,
satellite lineaments, and topographic features correlate rather
closely with recognized subsurface fracture trends determined
by Hagen (1972) and by Trantham and others (1980).

     Satellite lineaments tend to be somewhat longer than joint
clusters and to be strongly related to drainage patterns.  This
suggests that the lineaments could be taken as evidence of
fractures that penetrate a thick part of the stratigraphic
column.

Conclusions

     Satellite imagery provides good evidence of large
lineaments that are oriented primarily northeastward and that
seem to be related closely to stream-drainage patterns.
Topographic maps are useful for delineating drainage-related
lineaments. Orientations of some of the major streams may be
due to superposition of drainage and adjustment to zones of
weakness in bedrock.  (For added information see Melton,
1959.)

     Figure 38 was constructed to test the working hypothesis
that penetrative faults in Burbank Field might be suggested by
abrupt, linear changes in thickness of the stratigraphic
section between the Cottage Grove Sandstone and Pink Limestone.
In the lower middle part of T. 26 N., R. 6 E., trending east-
northeast  (arrows, Figure 38), crowded contour lines show
evidence of uncommon thickening.  This zone of transition seems
loosely correspondent to a broad aggregation of joint clusters
                                88

-------
                         R. 6 E.
                                                      T. 26 N.
Figure 48. Lineaments and joint clusters, T. 26 N., R. 6 E.,
           Burbank Field. Long, broad, dark lines are
           lineaments mapped from remote-sensing imagery.
           Short, straight lines of intermediate width show
           localities and trends of joint clusters mapped by
           Hagen (1972).   Short, narrow, "broken" lines are
           topographic-map lineaments, many of which are
           related to drainage patterns.  Observe
           concentration of northeast-trending joint clusters
           and lineaments in southern part of township.
                             89

-------
and lineaments (Figure 48).

                   INJECTION-SENSITIVITY MAP

     Examination of lineaments evident on satellite imagery led
to expression of localities with possible sensitivity to fluid
injection.  This mapping was based on working assumptions
believed to be worthy in conservative judgment of injection-
potential:
     (1) Systematic joints are in the Burbank Sandstone,
and are in the rock unit next above (Hagen, 1972; Dickey, 1979).
     (2) Under overpressured injection, this network of
fractures is a conduit for fluids (Dickey, 1979).
     (3) Communication between wells exceeds 1 mi (1.6 km) at
some localities (Dickey, 1979, p. 384).
     (4) Anomalously dense and extensive concentrations of
lineaments suggest penetrative or incipiently penetrative
fractures in the sedimentary-rock column.

     Systematic fracturing within the Burbank Sandstone has
been documented well.  The effect of fracturing seems evident
in the record of water-flooding  (see summary account by Dickey,
1979, p. 384-387).  The concept of abundant joints in an
orderly arrangement within the Burbank reservoir seems to be
supported formidably by physical evidence.  Of course, joints
in the reservoir and confining beds would not be detectable by
the standard methods of subsurface mapping used here. If  faults
penetrate reservoir and confining beds and extend to the
surface, then they are not made apparent by subsurface-mapping
methods and they are not clearly evident in the surface
geology.  General trends of lineaments are in keeping with
trends of joint clusters mapped by Hagen  (1972).  Hagen and
Trantham and others  (1980) described evidence to support  the
assumption that fractures in the Burbank Sandstone and
fractures in rocks at the surface are  strongly  similar.   The
provisional operating assumption follows that lineaments  may be
manifestations of deep-seated zones of weakness  in bedrock, and
that they might reasonably be interpreted  as localities within
which fluids injected under extraordinary  pressure could
penetrate confining beds.

     Figure 49 shows areas of suspected injection sensitivity.
Because of the lack of  clear-cut evidence  to favor
correspondence of lineaments and penetrative fractures, the
areas are described as  Zones of  Caution.   For injection wells
planned within the Zone of Caution, the proposition of nearby
faults that penetrate storage zone and confining beds should
be rejected by consideration of documented evidence, or an
acceptable level of risk should be demonstrated.  Injection
should be into partly depleted reservoirs  or reservoirs that


                                 90

-------
                          R. 6 E.
                                                         T. 26 N.
                         0	Mi.   1
                             Km. 1
Figure 49. Injection-sensitivity map, T. 26 N., R. 6 E.,
           Burbank Field.  Stippling outlines Zones of
           Caution, areas within which lineaments and joint
           clusters are suggestive of faulted rocks in the
           subsurface.  In Zones of Caution geologic evidence
           should accumulate to show that risks associated
           with injection of fluid are acceptable.
                              91

-------
                      GENERAL CONCLUSIONS
     A data base and mapping by computer allow quick generation
of various maps, including structural contour maps, interval-
thickness maps, and cumulative-thickness maps of confining beds
and potential storage units.  Routine use of data bases and
mapping software would allow geologists with limited experience
in subsurface geology to collect, enter and manipulate data to
develop a large variety of maps.  This method would free
experienced geologists for generation of maps that would be
more definitive for judgment about injection-sensitivity of
specific areas.

     Initial-potential and initial-production maps are of
seriously limited usefulness in delineation of suspected
trends of fractures in the Burbank reservoir.  Large leases
made initial-production maps effectively useless, whereas
methods of reporting initial flow rates and operator-
introduced bias reduced confidence in mapping from initial-
flow potentials.

     The most significant variable in initial-potential mapping
of the Burbank Sandstone was the overall character of the
reservoir rock and abrupt changes in reservoir quality
associated with narrow, channel-fill sandstones.  Although
study of initial-potential maps led to identification of
several localities where fractures were suspected to have
influenced production, this explanation was rejected on the
grounds that the production could show strong effects of the
operators' management of leases.

     Structural contour maps are not informative for the
purpose of detecting fractures in the Burbank reservoir, even
at the contour interval of 10 ft (3m).

     Remote-sensing imagery, including satellite imagery and
topographic maps, provides information useful for conjectural
mapping of fracture trends and for construction of injection-
sensitivity maps.
                                92

-------
                           SECTION 7



                  FITTS POOL, PONTOTOC COUNTY

                         INTRODUCTION
     Pontotoc County is divisible into five geologic provinces
(Figure 50): (1) the Hunton Arch, (2) the "Northeast Province",
(3) the Lawrence Uplift, (4) Franks Graben, and (5) the
Arbuckle Uplift.  The Arbuckle Uplift, Franks Graben, and
Lawrence Uplift are bounded by extensive faults (Figure 50).
In the course of this research, general attention was given to
the geology of Pontotoc County, but of particular concern was
the Fitts Pool, in Franks Graben (Figure 50).

                      SUBSURFACE GEOLOGY
     Cropping out on the deeply eroded Arbuckle Uplift are
chiefly carbonate rocks of the Arbuckle and Simpson Groups
(Table 2). Strata that crop out in Franks Graben mostly are
Pennsylvanian shales and sandstones.  Thick shales, primarily
of the Boggy Formation (Table 2), are the seal above the Fitts
Pool (location, Figure 50).

     The Fitts Pool has produced oil and gas from the
Ordovician Simpson Group and Viola Group, the Silurian-Devonian
Hunton Group, and strata of the Pennsylvanian System.  The trap
is a large faulted anticline (Figure 51); most production from
rocks older than Pennsylvanian is related to folding of strata,
but some entrapment in Pennsylvanian rocks is stratigraphic.
The Pennsylvanian confining beds are in the range of 2000 to
3000 ft  (600 to 900 m) thick, but locally within the Franks
Graben some of the faults shown in Figure 51 penetrate
Springeran and Morrowan strata. The Franks Fault Zone, the
Stonewall Fault, and the Ahloso Fault to the north are
traceable at the surface, although the Stonewall Fault is
obscure locally, and the Ahloso Fault is not evident throughout
the length shown in Figure 50.

     In study of West Edmond Field, initial-potential
production maps were used to advantage, because on the whole,
reservoir rocks are limestones of the Hunton Group.  At Fitts
Pool, of the several reservoirs, only the Hunton Group has
generally similar lithic properties; the Hunton is suited for
use of initial-potential production as a means to estimate the
effects of fracturing in the reservoir.  At Fitts Pool the more
porous and permeable formations in the Hunton Group are the
                                93

-------
Figure 50. Generalized map of geologic provinces, Pontotoc
           County.  Major faults include large, unnamed fault
           in the subsurface that extends through northern part
           of Township 4  North.
                            94

-------
Table 2.  ROCK-STRATIGRAPHIC UNITS, EASTERN
          PONTOTOC  COUNTY (AFTER MORGAN  (1924),
          BARKER (1951),  MISER (1954), GILLERT
           (1952))
SYSTEM
Pennsylvanian
Mississippian
Devonian
Silurian
Ordovician
STAGE
Virgilian
Missourian
Desmoinesian
Atokan
Morrowan
Springeran




ROCK-STRATIGRAPHIC
UNITS
Vanoss Formation
Ada Formation
Vamoosa Formation
Belle City Limestone
Francis Formation
Seminole Formation
Holdenville Formation
Wewoka Formation
Wetumka Shale
Calvin Sandstone
Senora Formation
Stuart Shale
Thurman Sandstone
Boggy Formation
Savanna Formation
McAlester Formation
Atoka Fromation
Wapanucka Limestone
Wapanucka Shale
Union Valley Limestone
Union Valley shale
Springer Formation
Caney Shale
Welden Limestone
Woodford Shale
Hunton Group
Sylvan Shale
Viola Group
Simpson Group
Arbuckle Group
(upper part)
                      95

-------
              R. 6 E.
R. 7 E.
R. 8 E.
                                                           T. 2 N.
                                                           T. 1 N.
                           Miles
                        Kilometers
Figure 51. Structural contour map, Fitts  Pool  and nearby
           areas.  Datum: Tog of Viola Group.   Broad,  curved
           black lines are traces of faults that  cut Viola
           Group.  Contour interval 100 ft  (30 m)  north of
           Stonewall Fault and 200 ft  (60 m) south of
           Stonewall Fault.  Fitts Pool is large  oil field in
           T. 2 N., R. 7 E., on north side of  large convex-
           southward fault.  Oil and gas  are trapped in
           anticline, with closure against fault  that  extends
           from Stonewall southwestward to southeastern corner
           of T. 2 N., R. 6 E.
                               96

-------
Frisco and Bois d'Arc (see Figure 16, p. 37, for stratigraphic
position).  The upper part of the Hunton Group is eroded, and
thickness of the Frisco-Bois d'Arc section varies accordingly.
Therefore study of initial production centered on localities
where high-production trends are inconsistent with thickening
of the Frisco-Bois d'Arc sequence.   (Compare Figures 52 and 53,
with special notice given to broad dashed lines.)  Insofar as
was practical, calculation of initial-potential production
eliminated variation due to (a) methods of reporting
production, (b) completion techniques,  (c) ages of wells
relative to decline of reservoir pressure, and (d) lease-line
protection (Puckette, 1989).  The amount of error that remains
is believed to not affect placement of the 1000 bbl-per-day
contour intervals.  Because the Fitts Pool is located where
rocks generally are strongly folded and faulted,  definite and
positive effect of faulting and fracturing on production is a
reasonable expectation.  The evidence available for this
research did not yield such information, for the trends of
large initial production shown in Figure 53 are few.
Altogether they suggest an eastward- to northeastward-striking
mosaic of fractures.

           MAPPING OF FRACTURE TRACES AND LINEAMENTS


     Aerial photographs, topographic maps and satellite imagery
were used to map lineaments and fracture traces.   Aerial
photographs were "standard black-and-white," at the scale of
3.2 in. per mi (5.1 cm per km).  Fracture traces and lineaments
were mapped on clear overlays, reduced  in scale and posted to
7.5-minute topographic maps by careful  registration (Figure
54).  Black-and-white thematic-mapper satellite imagery
covering 115 mi square  (71.5 x 71.5 km) was chosen because the
Band-7 data increase resolution of linear features.  Images
were winter scenes with no cloud cover, at the scale of
1:125,000.

     Fracture traces and some lineaments  (Figures 55, 56, 57,
and 58) were mapped on aerial photographs, mostly under
stereoscopic inspection with magnification.  Lineaments were
mapped also on satellite imagery  (Figure 59).  Geomorphic
features interpreted as fracture traces and lineaments included
uncommonly straight segments of streams, straight-line
topographic features  (many of them escarpments, some of them
swales), and alignments of vegetation  (Figure 54, 55, 56, 57,
59).  Fracture traces and lineaments are abundant where bedrock
is composed of the hard, brittle strata of the Arbuckle,
Simpson, Viola and Hunton Groups; they  are fewer where the
thick Pennsylvanian shales are bedrock.  As shown by Figure 54,
mapping  from aerial photographs under stereoscopic viewing with
                                97

-------
                                           R. 7 E.
vo
00
                                             Miles
                                        0  Kilometers  1
                Figure 52. Thickness of Frisco-Bois  d'Arc Formations, eastern
                           part of Fitts Pool.   Contour interval 25 ft  (6.1
                           m).  Upper part of Hunton Group was eroded before
                           deposition of Woodford  Shale.
                                                                                  T. 2 N.

-------
                                        R. 7 E.
10
vo
                                      0  Kilometers  1

               Figure  53.  Initial-potential  production  of wells  completed in
                          Hunton Group before  1938,  eastern part of Fitts
                          Pool. Contour interval  1000 bbl per day.   Broad
                          black line  is 4000 bbl-per-day contour.   "NP11
                          signifies that  no  potential was reported.  Broad
                          dashed lines show  trends of large production that
                          do not vary with thickening of Frisco-Bois d'Arc
                          Formations.  Because Frisco-Bois  d'Arc is the most
                          productive  part of Hunton  Group,  divergence between
                          thickness and production suggests the  contribution
                          of fracturing to production.
                                                                               T. 2 N.

-------
                                               R. 7 E.
H
o
o
                                                                                  T. 2 N.
                                                                                   t
                                                                                   N
                     Figure 54. Topographic map, east-central part of T. 2 N., R. 7
                                E., showing fracture traces mapped from aerial
                                photographs (short, straight, narrow lines with
                                index numbers) and lineaments mapped from satellite
                                imagery  (broad black lines).  At the surface, the
                                extensive Stonewall Fault  (Figure 51) strikes
                                eastward through the central part of Section  9,
                                into northern part of Section 11, but is not  shown
                                directly by fracture traces.

-------
  0
Miles
 1     o      Kilometers      i
_l     i  ,  ,  ,  ,  i  ,  ,  i  ,  I
Figure 55. Lineament (large opposed arrows) and numerous
           fracture traces (some shown by small opposed
           arrows).  Bedrock chiefly is sandstone of  the
           Simpson Group.  Lineament is mapped as a fault
           (Gillert, 1952); north side is downthrown.
           Fracture traces and lineament shown mainly as
           alignments of vegetation.  Lineament extends
           through northern parts of Sections 33 and  34,  T.  1
           N., R. 7 E.
                             101

-------
Figure 56. Lineament (opposed arrows) in upland terrain of
           carbonate rocks, Arbuckle Group.  Lineament  is
           swale in grassland (Figures 57 and 58).  Southern
           boundary, Franks Fault Zone, is on trend with
           single arrow.  Dark pattern, northeastern corner,
           shows truncated, folded strata north of fault.
           Lineament is in Sections 14 and 23, T. 1 N., R. 6 E.
                              102

-------


Figure 57. View northeastward along lineament shown in Figure
           56.  Lineament expressed physiographically as swale
           in grassy uplands.  Viewer's position was in county
           road, near the place marked "A".   Small tree in
           central part of horizon is in the lineament.
                              103

-------
Figure 58. View southwestward along lineament shown in Figure
           56.  Lineament expressed physiographically as
           poorly defined, broad swale.  Viewer's position was
           in county road, near the place marked "A".
           Lineament extends through cluster of trees located
           behind and to the left of crooked, middle fence
           post.
                              104

-------
                           Kilometers
10
Figure 59. Satellite imagery, western part, Franks Graben.
           Scale, 1:250,000.  Large opposed arrows mark
           general extent of Franks Fault Zone, a prominent
           lineament.  Southwestern part of image is
           Arbuckle Uplift.  Square shows location of
           Stonewall.  Small opposed arrows are at ends of
           small lineament that probably is part of Stonewall
           Fault.
                              105

-------
magnification produces much data — an advantage for the study
of a few square miles, but perhaps too much detailed
information when the objective is to evaluate a region.

                   INJECTION-SENSITIVITY MAP
     Figure 60 is an injection-sensitivity map of Franks
Graben, constructed on the assumption that fluids would be
emplaced in pre-Pennsylvanian strata.  The entirety of the
Franks Fault Zone is regarded as being injection-sensitive,
because of the brittleness of lower Paleozoic rocks within the
zone, the many large faults and more small ones, and the
associated general fracturing of rock. Altogether these
circumstances lead to the inference that within the fault zone,
many faults are likely to be unsealed.

     A questionable Zone of Sensitivity is mapped south of the
Stonewall Fault (Figure 60).  This fault penetrates from the
deep subsurface to the surface, but because it bounds the more
downthrown side of the graben, subsurface geologic evidence
close by is scarce.  Plans for disposal wells in proximity to
the fault should take geologic evidence into careful account,
especially if fluid is to be injected under extraordinary
pressure.

     Because of the abundance of lineaments mapped in the
interior of the graben (Figure 61), all but small parts of the
province are classified as being in a Zone of Caution.
Clearly, the thick sequence of shaly Pennsylvanian confining
beds sealed Fitts Pool, but the many lineaments suggest that a
system of fractures involves the confining beds.  The
proposition of injection into pre-Pennsylvanian rocks should
take into account the nearby geologic circumstances,
particularly if plans include the emplacement of more fluids
than have been withdrawn from the reservoir, or injection at
pressures larger than the original formation pressure.

                          CONCLUSIONS
      (1) Mapping of lineaments from satellite imagery is
 effective and quicker than mapping from aerial photographs.
 Different observers can produce interpretations that are
 strongly similar  (Figure 61).

      (2) Thematic-mapper data have good resolution. Band-7,
 black-and-white imagery tends to emphasize linear features
 moreso  than  does  false-color imagery. Winter scenes were
 the more useful.  Purchase of film-negative imagery offers the
                                 106

-------
    R. 5 E.
                      R. 6 E.
                                              R. 7 E.
                                                             T. 2 N.
                                                             T. 1 N.
                              Miles
                            Kilometers
10
Figure 60. Injection-sensitivity map,  Franks Graben,  based on
           the assumption that  injection  zones would be in
           pre-Pennsylvanian rocks.  Extensively faulted rocks
           in Franks Fault Zone are  in Zone of Sensitivity
           (S). The area near the Stonewall Fault is suspected
           to be injection-sensitive,  in  the general absence
           of subsurface geologic data. Most of the interior
           of the graben is in  Zone  of Caution (C).
                               107

-------
      R. 6 E.
        R. 7 E.
                                                        T. 2 N.
                                                        T. 1 N.
                            Miles
          0
Kilometers
10
Figure 61. Map  of lineaments in Franks Graben.   Dashed and
           solid lines are interpretations  of two observers.
           Interpretations are in general agreement.
                            108

-------
option of working with images at several scales.

     (3) Band-7 data reduce the likelihood of misinterpreting
cultural features as lineaments.  Posting of lineaments to
topographic maps further reduces the probability of such error.

     (4) Lineaments are significantly more numerous in terrain
underlain by limestone, dolomite and sandstone than in terrain
underlain by shale.

     (5) Mapping of fracture traces and lineaments from aerial
photographs is affected rather strongly by differences in cover
of vegetation.  The number of detectable fracture traces is
markedly less in grassland than in forested land.

     (6) In the complexly deformed Franks Graben, mapping of
lineaments is valuable for construction of injection-
sensitivity maps.  In the interior of the graben a system of
fractures in Pennsylvanian confining beds is suggested by the
fabric of lineaments.
                                109

-------
                    SELECTED REFERENCES
Alpay, 0. A., 1973, Application of aerial photographic
interpretation to the study of reservoir natural fracture
systems: Jour. Petroleum Tech., v. 25, p. 37-45.

Amsden, T. W., and Rowland, T. L., 1967, Geologic maps and
stratigraphic cross-sections of Silurian strata and Lower
Devonian formations in Oklahoma: Okla. Geol. Survey
Map GM-14.

Andresen, M. J., 1962, Paleodrainage patterns: Their mapping
from subsurface data, and their paleogeographic value: Amer.
Assoc. Petroleum Geologists Bull., v. 46, no. 3, p. 398-405.

Arbenz, J. K., 1956, Tectonic map of Oklahoma: Okla. Geol.
Survey Map GM-3.

Azimi, Esmaeil, 1978, Use of remote sensing for fracture
discrimination and assessment of pollution susceptibility of
a limestone-chert aquifer in northeastern Oklahoma: Unpub.
Master of Science thesis, Okla. State Univ., 98 p.

Barker, J. c., 1951, The geology of a portion of the Lawrence
Uplift, Pontotoc County, Oklahoma: Tulsa Geol. Soc. Digest,
v. 19, p. 169-191.

Bates, R. L., and Jackson, J. A., (ed.), 1984, Dictionary of
geological terms: Anchor Press/Doubleday, 3rd ed., 571 p.

Beardall, G. B., Jr., 1983, Depositional environment,
diagenesis and dolomitization of the Henryhouse Formation, in
the western Anadarko Basin and Northern Shelf, Oklahoma:
Unpub. Master of Science thesis, Okla. State Univ.,  127 p.

Benoit, E. L., 1957, The Desmoinesian Series, Edmond area,
central Oklahoma: Shale Shaker Digest, v. 16, p. 338-350.

Berger, Zeev, 1986, New technique for structural analysis of
low-relief basins: Amer. Assoc. Petroleum Geologists Bull., v.
70,  no. 5, p. 564.

Berger, Zeev, 1988, Detection and analysis of basement
structures in low-relief basins using integrated analysis of
Landsat data: Proceedings, Sixth Thematic Conference on Remote
Sensing for Exploration Geology, Environmental Research Inst.
of Michigan, v. 1, p. 111.

Bingham, R.  H., and Bergman,  D. L.,  1980, Reconnaissance of


                                110

-------
the water resources of the Enid quadrangle, north-central
Oklahoma: Okla. Geol. Survey Map HA-7.

Bingham, R. H., and Moore, R. L., 1975, Reconnaissance of the
water resources of the Oklahoma City quadrangle, central
Oklahoma: Okla. Geol. Survey Map HA—4.

Blanchet, P. H., 1957, Development of fracture analysis as an
exploration method: Amer. Assoc. Petroleum Geologists Bull.,
v. 41, no. 8, p. 1748-1759.

Bogli, Alfred, 1980, Karst hydrology and physical speleology:
Springer-Verlag, 284 p.

Brasier, Francoise, 1987, Injection of hazardous wastes: EPA
overview, in  Proceedings of the International Symposium on
Subsurface Injection of Liquid Wastes, New Orleans, La., U.
S. Nat. Water Well Assoc., p. 1-2.

Brown, H. A., 1967, Structural control of Canadian River in
western Oklahoma: Okla. Geol. Survey, Okla. Geology Notes, v.
27, p. 135-149.

Brown, R. 0., and Forgotson, J. M., Jr., 1980, Predicting the
orientation of hydraulically created fractures in the Cotton
Valley Formation of East Texas: Soc. Petroleum Engineers,
AIME, SPE 9269, 12 p.

Bruce, L. G., 1989, A method for predicting fracture-enhanced
permeability in regions of "flat-lying" strata (draft): Unpub.
Doctor of Philosophy dissertation, Okla. State Univ.

Burchfield, M. R., 1985, Map of Oklahoma oil and gas fields:
Okla. Geol. Survey Map GM-28

Cardott, B. J., and Lambert, M. W., 1987, Thermal maturation
by vitrinite reflectance of Woodford Shale, Anadarko
Basin, Oklahoma: Amer. Assoc. Petroleum Geologists Bull.,
V. 71, no. 7, p. 898-899.

Carr, J. E., and Havens, J. S., 1976, Records of wells and
water quality for the Garber-Wellington aquifer, northern
Oklahoma and southern Logan Counties, Oklahoma: U. S.
Geol. Survey open-file report 76-619, 32 p.

Carr, J. E., and Marcher, M. V., 1977, A preliminary appraisal
of the Garber-Wellington aquifer,  southern Logan and
northern Oklahoma Counties, Oklahoma: U. S. Geol. Survey
open-file report 77-278, 23 p.
                                Ill

-------
Caylor, J. W., 1957, Surface geology of western Garfield
County, Oklahoma: Unpub. Master of Science thesis, Univ. of
Okla., 78 p.

Christenson, S. C., and Parkhurst, D. L., 1987, Ground-water
quality-assessment of the central Oklahoma aquifer, Oklahoma:
project description: U. S. Geological Survey open-file report
87-235, 30 p.

Clark, L. H., 1983, Briefing and policy review: Underground
injection of hazardous waste (draft): U. S. Environ. Protect.
Agency.

Clemons, R. R., 1984, The remote sensor exploration of the
Ardmore and Marietta basins of Oklahoma: Unpub. Doctor of
Philosophy dissertation, Texas Tech Univ., 452 p.

Cooley, M. E., 1986, Divisions of potential fracture
permeability, based on distribution of structures and
lineaments in sedimentary rocks of the Rocky Mountains-High
Plains region, western United States: U. S. Geol. Survey
Water-resources Investigation Rept. No. WRI 85-4091.

Cox, J. C., and Harrison, S. S., 1979, Fracture-trace
influenced stream orientation in glacial drift —
northwestern Pennsylvania: Can. Jour. Earth Sci., v. 16,
no. 7, p. 1511-1514.

Dennis, P. E., 1954, Water-loss investigations: Lake Hefner
studies, technical report: U. S. Geol. Survey Prof. Paper 269,
p. 10-16.

Dickey, P. A., 1979, Petroleum development geology: Petroleum
Publishing Co., 398 p.

Dix, 0. R., and Jackson, M. P. A., 1981, Statistical analysis
of lineaments and their relation to fracturing, faulting, and
halokinesis in the East Texas Basin: Bureau Econ. Geol., Univ.
Texas at Austin, Rept. Investigations No. 110, 30 p.

Dolan, R., and Howard, A., 1978, Structural control of the
rapids and pools of the Colorado River in the Grand Canyon:
Science, v. 202, p. 629-631.

Elkins, L. F., 1969, Internal anatomy of a tight, fractured
Hunton Lime reservoir revealed by performance — West Edmond
Field: Jour. Petroleum Tech., Feb., p. 221-232.

Eliason, J. R., 1984, A technique for structural geologic
analysis of topography (digital, Idaho, Oregon, Washington):
                                112

-------
Unpub. Doctor of Philosophy dissertation, Washington State
University, 196 p.

Evans, D. M., 1966, Man-made earthquakes in Denver: Geotimes,
v. 10, no. 9, p. 11-18.

Feder, A. M., 1984, Contemporary remote sensing for
hydrocarbon exploration and development: Oil and Gas
Jour., v. 82, pt. 9, p. 150-151.

Flanagan, K. D., 1989, Selected attributes of confining beds,
Burbank oil field and nearby areas, north-central Oklahoma:
Unpub. Master of Science thesis (draft), Okla. State Univ.,
114 p.

Gillert, M. P., 1952, The geology of the Simpson Group on the
northeast flank of the Arbuckle Mountains, Oklahoma: Unpub.
Master of Science thesis, Univ. of Okla., 76 p.

Gordon, Wendy, and Bloom, Jane,  1987, Deeper problems: Limits
to underground injection as a hazardous waste disposal method:
Proceedings of the International Symposium on Subsurface
Injection of Liquid Wastes, New Orleans, La., U. S. Nat. Water
Well Assoc., p. 3-50.

Graham, L. L., and others, 1985, Protection of public water
supplies from ground-water contamination:  U. S. Envir. Prot.
Agency Seminar Pub., EPA/625/4-85/016, 182 p.

Hagen, K. B., 1972, Mapping of surface joints on air photos
can help understand waterflood performance problems at North
Burbank Unit: Unpub. Master of Science thesis, Univ. of Tulsa,
77 p.

Ham, W. E., 1969, Regional geology of the Arbuckle Mountains,
Oklahoma:  Okla. Geol. Survey, Guidebook 17, 52 p.

Harris, S. A., 1970, Bends of the South Canadian River of
Oklahoma as related to regional geomorphology: Shale Shaker
Digest, v. 6, p. 160-175.

Harris, S. A., 1975, Hydrocarbon accumulation in the "Meramec-
Osage"  (Mississippian) rocks, Sooner Trend, northwest-central
Oklahoma: Amer. Assoc. Petroleum Geologists Bull., v. 59, no.
4, p. 633-664.

Harvey, R. L., 1972, West Campbell (Northeast Cedardale) gas
field, Major County, Oklahoma: Amer. Assoc. Petroleum
Geologists  Mem. 16, p. 568-578.
                                113

-------
Hendricks, T. A., 1937, Geology and fuel resources of the
southern part of the Oklahoma Coal Field; Part 1, the
McAlester District: U. S. Geol. Survey Bull. 874-A, 90 p.

Hyatt, D. L., 1936, Preliminary report on the Fitts Pool,
Pontotoc County, Oklahoma: Amer. Assoc. Petroleum
Geologists  Bull., v. 20, p. 951-974.

Jennings, J. N., 1985, Karst geomorphology: Basil Blackwell,
Inc., New York, 293 p.

Johnson, K. S., 1969, Mineral map of Oklahoma, exclusive of
oil and gas fields:  Okla. Geol. Survey, Map GM-15.

Johnson, K. S., Burchfield, M. R., and Harrison, W. E., 1984,
Guidebook for Arbuckle Mountain field trip, southern Oklahoma:
Okla. Geol. Survey Special Publ. 84-1, 21 p.

Jordan, Louise, 1962, Geologic map and section of pre-
Pennsylvanian rocks in Oklahoma, showing surface and
subsurface distribution; Okla. Geol. Survey, Map GM-5.

Judson, S. and Andrews, G., 1955, Pattern and form of some
valleys in the Driftless Area, Wisconsin:  Jour. Geol., v. 63,
p. 328-336.

Kennedy, L. G., 1989, Confining layer study in the West Edmond
oil field area, using subsurface, remote sensing and
geochemical methods: Unpub. Master of Science thesis (draft),
Okla. State Univ., 173 p.

Knepper, D. H., 1987, Remote-sensing studies of the Anadarko
Basin: Okla. Geol. Survey, Okla. Geology Notes, v. 47, no. 3,
p. 117.

Komar, C. A., Such, L. Z., Overbey, W. K., Jr., and Anderson,
T. O., 1973, Delineating a subsurface fracture system in
a petroleum reservoir — an experiment: Jour. Petroleum
Tech., V. 25, p. 531-537.

Kreitler, C. W., 1985, Hydrogeology of sedimentary basins as
it relates to deep-well injection of chemical wastes:
Proceedings of the International Symposium on Subsurface
Injection of Liquid Wastes, New Orleans, La., U. S. National
Water Well Association, p. 398-416.

Kuhleman, J. H., 1951, Mississippian and Lower Pennsylvanian
stratigraphy of portions of Stonewall and Atoka quadrangles,
Oklahoma: Tulsa Geol. Soc. Digest,  v. 19, p. 192-213.
                                114

-------
Lattman, L. H.,  1958, Technique of mapping geologic fracture
traces and lineaments on aerial photographs: Photogrammetric
Engineering, v.  24, no. 4, p. 568-576.

Lattman, L. H.,  and Parizek, R. R., 1964, Relationship between
fracture traces and the occurrence of groundwater in carbonate
rocks:  Jour. Hydrology, v. 2, no. 2, p. 73-91.

Lawson, J. E., and Luza, K. V., 1988, Oklahoma earthquakes,
1987: Okla. Geol. Survey, Okla. Geol. Notes, v. 48, no. 1, p.
54-63.

Lillesand, T. M., and Kiefer, R. W., 1987', Remote sensing and
image interpretation: John Wiley and Sons, 612 p.

Low, J. W., 1957, Geologic field methods: Harper and Brothers,
489 p.

Luza, K. V., and Lawson, J. E., Jr., 1979, Seismicity and
tectonic relationships of the Nemaha Uplift in Oklahoma: U.S.
Nuclear Regulatory Commission, NUREG/CR-0500, 66 p.

Luza, K. V., 1981, Seismicity and tectonic relationships of
the Nemaha Uplift in Oklahoma; Part III: U.S. Nuclear
Regulatory Commission, NUREG/CR-1500, 70 p.

McGee, D. A., and Jenkins, H. D.,  1946, West Edmond oil
field, central Oklahoma: Amer. Assoc. Petroleum Geologists
Bull., V. 30, no. 11, p. 1797-1829.

Maarouf, A. M. S., 1981, Morphostructural analyses of space
imagery in the central Colorado Plateau: Unpub. Doctor of
Philosophy dissertation,  University of Utah, 130 p.

Mann, Wallace, 1957, Subsurface geology of the Franks graben,
Pontotoc and Coal Counties, Oklahoma: Unpub. Master of
Science thesis,  Univ. of Okla., 62 p.

Manni, F. M., 1985, Depositional environment, diagenesis, and
unconformity identification of the Chimneyhill Subgroup in the
western Anadarko Basin and Northern Shelf of Oklahoma:
Unpub. Master of Science thesis, Okla. State Univ., 133 p.

Melton, F. A., 1929, A reconnaissance of the joint-systems in
the Ouachita Mountains and central plains of Oklahoma: Jour.
Geology, v. 37,  p. 729-746.

Melton, F. A., 1955, Photo-geology in flatland regions of low
dip: Shale Shaker, v. 6, no. 3, p. 5-8, 11-12, 15-20, 39.
                                115

-------
Melton, F. A., 1959, Aerial photographs and structural
geomorphology: Jour. Geology, v. 67, no. 4, p. 351-370.

Miser, H. D., 1954, Geologic map of Oklahoma: Okla. Geol.
Survey.

Morgan, G. D., 1924, Geology of the Stonewall quadrangle:
Okla. Bureau of Geology Bull. 2, 248 p.

Morgan, G. D., 1924, Boggy unconformity and overlap in
southern Oklahoma: Okla. Bureau of Geology Circ. 2, 8 p.

Morton, R. B., 1981, Reconnaissance of the water resources of
the Woodward quadrangle, northwestern Oklahoma: Oklahoma Geol.
Survey Map HA-8.

Naff, J. D., 1962, Geology and paleontology of the upper Boggy
drainage area, Coal and Pontotoc Counties, Oklahoma: Unpub.
Doctor of Philosophy dissertation, Univ. of Kansas, 470 p.

Nelson, R. A., 1975, Fracture permeability in porous
reservoirs: An experimental and field approach: Unpub.
Doctor of Philosophy dissertation, Texas ASM Univ., 171
P-

Nelson, R. A., 1985, Geologic analysis of naturally fractured
reservoirs: Gulf Publishing Co., 320 p.

Okonny, I. P., 1981, Geologic analysis of remote sensing
imagery of the eastern Niger delta, Nigeria: Unpub. Doctor of
Philosophy dissertation, Purdue Univ., 294 p.

O'Leary, D. W.,  Friedman, J. D., and Pohn, H. A., 1976,
Lineament, linear, lineation: Some proposed new standards
for old terms: Geol. Soc. America Bull., v. 87, no. 10, p.
1463-1469.

Paine, J. W., 1958, Subsurface geology of T. 4 N., R. 4 and 5
E., Pontotoc County, Oklahoma: Unpub. Master of Science
thesis, Univ. of Okla., 46 p.

Parizek, R. R.,  1975, On the nature and significance of
fracture traces and lineaments in carbonate and other
terranes, in Karst hydrology and water resources, v. 1:
Proceedings of the U. S.-Yugoslavian Symposium, Dubrovnik, p.
3-1 - 3-62.

Peters, D. C., Speirer, R. A., and Shea, V. R., 1988,
Lineament analysis for hazard assessment in advance of coal
mining: Proceedings, Sixth Thematic Conference on Remote
                                116

-------
Sensing for Exploration Geology, Environmental Research Inst.
Of Michigan, v. 1, p. 253-260.

Pittman, E. D., 1981, Effect of fault-related granulation on
porosity and permeability of quartz sandstones, Simpson
Group (Ordovician), Oklahoma: Amer. Assoc. Petroleum
Geologists Bull., v. 65, no. 11, p. 2381-2387.

Podwysocki, M. H., Moik, J. G., and Shoup, W. C., 1975,
Quantification of geologic lineaments by manual and machine
processing techniques,in NASA Earth Resources Survey
Symposium: Volume 1-B, Geology, information systems and
services: Nat. Aeronautics and Space Admin., NASA-TM-X-58168,
v. 1-B, p. 885-903.

Probst, G. L., 1988, Predicting subsurface joint trends in
undeformed strata: Proceedings, Sixth Thematic conference on
Remote Sensing for Exploration Geology, Environmental Research
Inst. of Michigan, v. 2, p. 423-436.

Puckette, J. O.,  1989, Evaluation of confining-bed integrity,
Pontotoc County,  Oklahoma: Unpub. report, School of geology,
Okla. State Univ., 33 p.

Rausch, R. W., and Beaver, K. W., 1964, Case history of
successfully water flooding a fractured sandstone reservoir:
Jour. Petroleum Tech., v. 16, p. 1233-1237.

Ray, R. G., 1960, Aerial photographs in geologic
interpretation and mapping: U. S. Geol. Survey Prof. Paper
373, 230 p.

Richason, B. F., Jr., ed., 1978, Introduction to remote
sensing of the environment:  Second edition, Kendall-Hunt Pub.
Co., Dubuque, Iowa, 582 p.

Sabins, F. F., Jr., 1978, Remote sensing, principles and
interpretation: :  W.H. Freeman and Co., 426 p.

Schridder, L. A., Watts, R. J., and Wasson, J. A., 1970, An
evaluation of the East Canton oil field waterflood:  Jour.
Petroleum Tech., v. 22, p. 1371-1378.

Shelton, J. W., Ross, J. S., Garden, A. J., and Franks, J. L.,
1985, Geology and mineral resources of Payne County, Oklahoma:
Okla. Geol. Survey bull. 137, 85 p.

Shoup, R. C., 1980, Correlation of Landsat lineaments with
geologic structures, north-central Oklahoma: Unpub.  Master
of Science thesis, Univ. of Okla., 106 p.
                                117

-------
Stearns, D. W., Berger, Zeev, Hopkins, H. R., and Nelson, R.
A., 1988, The contribution of remote sensing data to
exploration of fractured reservoirs: Proceedings, Sixth
Thematic Conference on Remote Sensing for Exploration Geology:
Environmental Research Inst. of Michigan, v. 1, p. 113-122.

Stohl, L. P., 1988, Photogeologic-geomorphic mapping provides
clues to structural features beneath glacial terrain:
Proceedings, Sixth Thematic Conference on Remote Sensing for
Exploration Geology, Environmental Research Inst. of Michigan,
V. 1, p. 373-382.

Swesnik, R. M., 1948, Geology of West  Edmond oil field,
Oklahoma, Logan, Canadian, and Kingfisher Counties,
Oklahoma: Amer. Assoc. Petroleum Geologists, Structure of
Typical American Oil Fields, A Symposium, v. 3, p. 359-398.

Sullivan, K. L., 1985, Organic facies variation of the
Woodford Shale  in western Oklahoma:  Shale Shaker, v. 35,
no.4, p. 76-89.

Thomas, E. P.,  1986, Understanding  fractured oil reservoirs:
Oil and Gas Jour., v. 84, pt. 7, p. 75-79.

Thornbury, W. D., 1956, Principles  of geomorphology: John
Wiley & Sons, Inc., 618 p.

Thornhill, M. R., 1989, Lineaments  and fracture traces,
eastern Pontotoc County, Oklahoma  (draft): Unpub. Master of
Science thesis, Okla. State Univ.,  126 p.

Tillman, J. E., 1983, Exploration for reservoirs with
fracture-enhanced permeability: Oil and Gas Jour., v. 81, pt.
2, p. 165-179.

Trantham, J. C., Threlkeld, C. B.,  and Patterson, H. L., Jr.,
1980, Reservoir description for a surfactant/polymer pilot
in a fractured, oil-wet reservoir - North burbank Unit Tract
97: Jour. Petroleum Tech., v. 32, p. 1647-1656.

Walsh, S. J., and Mynar, F., II, 1985, Comparison of Landsat
digital enhancement techniques for  lineament detection:
Geol. Soc. America- 98th Ann. Mtng., Abstracts with
Programs, p. 743.

Warner, D. L.,  Davis, S. N.,  and Syed, Talib,  1986,
Evaluation of confining layers for  containment of injected
wastewater: Proceedings of International Symposium on
Subsurface Injection of Liquid Wastes,  New Orleans, La., U.
S. Nat. Water Well Assoc.,  p. 417-446.
                                118

-------
Watson/ D. F., and Philip, G. M., 1987, Neighborhood-based
interpolation: Amer. Assoc. Petroleum Geologists,  Geobyte,
v. 2, no. 2, p. 12-16.

Watts, K. R., 1977, Assessment of Landsat imagery  for  the
investigation of fracturing in an unconfined chert and
carbonate aquifer: Unpub. Master of Science thesis, Okla.
State Univ., 85 p.

Weber, N. V., 1974, A quantitative analysis of the
relationship between geologic structure and drainage line
orientation  in a neotectonic region: Dept. of Geology  and
Geography, Indiana State Univ., Prof. Paper  No. 6, p.  41-55.

Whiteside, R. F., and Raef, S. F., 1986, Mechanical integrity
of Class 1 injection wells: Proceedings of International
Symposium on Subsurface Injection of Liquid Wastes, New
Orleans, La., U. S. Nat. Water Well Assoc., p. 57-76.

Wilkinson, W. M., 1977, Fracturing in Spraberry reservoir,
West Texas,  in Kostura, J. and Ravenscroft, J., ed.,
Fracture-controlled production:  Amer. Assoc. Petroleum
Geologists   Reprint No. 21, p. 40-55.

Withrow, P.  C., 1972, Star-Lacey Field, Elaine and Kingfisher
Counties, Oklahoma: Amer. Assoc. Petroleum Geologists  Mem.
16, p. 520-531.

Wood, P. R., and Burton, L. C., 1968, Ground-water resources
in Cleveland and Oklahoma Counties: Okla. Geol. Survey
Circular 71, 75 p.

Woodruff, J. and Parizek, E. J., 1956, Influence of underlying
rock structures on stream courses and valley profiles  in  the
Georgia Piedmont: Assoc. Amer. Geogr. Annals, v. 46, p. 129-
139.

Woodruff, J. D., Talley, J. H., and Miller, J. C., 1974,
Selection sites for high yielding wells in the Delaware
Piedment: Geol. Soc. America, Northeastern Section,
Abstracts with Programs, p. 87.

Zall, L. and Russell, O.,  1979, Ground water exploration
programs in Africa: Proceedings, 5th Ann. William  T. Pecora
Memorial Symposium on Remote Sensing, p. 416-425.
                                H9         * US GOVERNMENT PRINTING OFFICE 1989- 648-163/00330

-------