s 14.GS: RPI89 c. 1 STATE OF ILLINOIS DWIGHT H. GREEN, Governor DEPARTMENT OF REGISTRATION AND EDUCATION FRANK G. THOMPSON, Director DIVISION OF THE STATE GEOLOGICAL SURVEY M. M. LEIGHTON, Chief URBANA REPORT OF INVESTIGATIONS — No. 89 WATER FLOODING OF OIL SANDS IN ILLINOIS BY FREDERICK SQUIRES and ALFRED H. BELL PRINTED BY AUTHORITY OF THE STATE OF ILLINOIS URBANA, ILLINOIS 19 4 3 DBRARY mmmam protection anFN CY STATE OF ILLINOIS y SPRINGFIELD, ILLINOIS^ STATE OF ILLINOIS DWIGHT H. GREEN, Governor DEPARTMENT OF REGISTRATION AND EDUCATION FRANK G. THOMPSON, Director DIVISION OF THE STATE GEOLOGICAL SURVEY M. M. LEIGHTON, Chief URBANA REPORT OF INVESTIGATIONS — No. 89 WATER FLOODING OF OIL SANDS IN ILLINOIS BY FREDERICK SQUIRES and ALFRED H. BELL PRINTED BY AUTHORITY OF THE STATE OF ILLINOIS URBANA, ILLINOIS 19 4 3 ORGANIZATION STATE OF ILLINOIS HON. DWIGHT H. GREEN, Governor DEPARTMENT OF REGISTRATION AND EDUCATION HON. FRANK G. THOMPSON, Director BOARD OF NATURAL RESOURCES AND CONSERVATION HON. FRANK G. THOMPSON, Chairman EDSON S. BASTIN, Ph.D., D.Sc, Geology ROGER ADAMS, Ph.D., D.Sc, Chemistry LOUIS R. HOWSON, C.E., Engineering WILLIAM TRELEASE, D.Sc, LL.D., Biology EZRA JACOB KRAUS, Ph.D., D.Sc, Forestry ARTHUR CUTTS WILLARD, D.Engr., LL.D. President of the University of Illinois GEOLOGICAL SURVEY DIVISION M. M. LEIGHTON, Chief SCIENTIFIC AND TECHNICAL STAFF OF THE STATE GEOLOGICAL SURVEY DIVISION 100 Natural Resources Building, Urbana M. M. LEIGHTON, Ph.D., Chief Enid Townley, M.S., Assistant to the Chief GEOLOGICAL RESOURCES Coal G. H. Cady, Ph.D., Senior Geologist and Head L. C. McCabe, Ph.D., Geologist (on leave) R. J. Helfinstine, M.S., Assoc. Mech. Eng. James M. Schopf, Ph.D., Asst, Geologist J. Norman Payne, Ph.D., Asst. Geologist Charles C. Boley, M.S., Asst. Mining Eng. Bryan Parks, M.S., Asst. Geologist Robert M. Kosanke, M.A., Research Assistant George M. Wilson, B.S., Research Assistant Henry L. Smith, A.B., Research Assistant Industrial Minerals J. E. Lamar, B.S., Geologist and Head H. B. Willman, Ph.D., Assoc. Geologist Robert M. Grogan, Ph.D., Assoc. Geologist Robert R. Reynolds, M.S., Asst. Geologist Oil and Gas A. H. Bell, Ph.D., Geologist and Head Frederick Squires, B.S., Petroleum Eng. Charles W. Carter, Ph.D., Asst. Geologist William H. Easton, Ph.D., Asst. Geologist Paul G. Luckhardt, M.S., Asst. Geologist Wayne F. Meents, Research Assistant Areal and Engineering Geology George E. Ekblaw, Ph.D., Geologist and Head Richard F. Fisher, M.S., Asst. Geologist Subsurface Geology L. E. Workman, M.S., Geologist and Head Arnold C. Mason, B.S., Assoc Geologist Merlyn B. Buhle, M.S., Asst. Geologist Frank E. Tippie, B.S., Asst. Geologist Margaret Sands, B.A., Research Assistant Ruth E. Roth, B.S., Research Assistant Walter R. Smith, Research Assistant Stratigraphy and Paleontology J. Marvin Weller, Ph.D., Geologist and Head Chalmer L. Cooper, M.S., Assoc. Geologist Petrography Ralph E. Grim, PhD., Petrographer Richards A. Rowland, Ph.D., Asst. Petrographer (on leave) Physics R. J. Piersol, Ph.D., Physicist B. J. Greenwood, B.S., Mech. Engineer Donald O. Holland, M.S., Asst. Physicist (on leave) GEOCHEMISTRY Frank H. Reed, Ph.D., Chief Chemist H. W. Jackman, M.S.E., Chemical Engineer James C. McCullough, Research Associate Coal G. R. Yohe, Ph.D., Chemist Carol Bartels, B.S., Research Assistant Industrial Minerals J. S. Machin, Ph.D., Chemist and Head Delbert L. Hanna. A.M., Asst. Chemist Fluorspar G. C. Finger, Ph.D., Assoc. Chemist Everett W. Maynert, B.S., Asst. Chemist X-ray and Spectrography W. F. Bradley, Ph.D., Assoc. Chemist Analytical O. W. Rees, Ph.D., Chemist and Head Howard S. Clark, B.A., Assoc, Chemist L. D. McVlCKER, B.S., Asst. Chemist P. W. Henline, M.S., Asst. Chemical Engineer -William F. Wagner, M.S., Asst. Chemist K. F. Bursack, B.A., Research Assistant Cameron D. Lewis, B.A., Research Assistant Mary Ann Winsche, B.S., Research Assistant Marjorie Winchester, B.S., Research Assistant MINERAL ECONOMICS W. H. Voskuil, Ph.D., Mineral Economist Douglas F. Stevens, M.E., Research Associate Grace N. Oliver, A.B., Assistant in Mineral Economics EDUCATIONAL EXTENSION Don L. Carroll, B.S., Assoc. Geologist PUBLICATIONS AND RECORDS George E. Ekblaw, Ph.D., Geologic Editor Chalmer L. Cooper, M.S., Geologic Editor Dorothy E. Rose, B.S., Technical Editor Kathryn K. Dedman, M.A., Asst. Technical Editor Alma R. Sweeny, A. 6., Technical Files Clerk Portia Allyn Smith, Research Assistant Meredith M. Calkins, Geologic Draftsman Leslie D. Vaughan, Asst. Photographer Oil and Gas Resources Special Staff to Aid in the War Effort Ground Water Geology Earle F. Taylor, M.S., Asst. Geologist M. W. Pullen, Jr., M.S., Spec. Asst. Geologist Arnold Eddings, B.A., Spec Research Assistant Virginia Kremers, B.S., Spec Research Assistant Margaret Parker, B.S., Research Assistant John A. Harrison, B.S., Spec. Research Assistant (on leave) Carl A. Bays, Ph.D., Spec. Geologist C. Leland Horberg, Ph.D., Spec Asst. Geologist Stewart Folk, M.S., Spec. Asst. Geologist Ernest P. DuBois, Ph.D., Spec. Asst. Geologist Paul Herbert, Jr., B.S., Spec Asst. Geologist Charles G. Johnson, A.B., Spec. Asst. Geologist Consultants: Ceramics, Cullen W. Parmelee, M.S., D.Sc, and Ralph K. Hursh, B.S., University of Illinois ; Mechanical Engineering, Seichi Konzo, M.S., University of Illinois. Topographic Mapping in Cooperation with the United States Geological Survey. This Report is a Contribution of the Oil and Gas Divison. May 1, 1943 Digitized by the Internet Archive in 2012 with funding from University of Illinois Urbana-Champaign http://archive.org/details/waterfloodingofo89squi CONTENTS PAGE Introduction .7 Acknowledgments 10 Preliminary investigations 7 Geologic description 13 Producing sands in the old Illinois fields . 15 Pennsylvanian system . . .' . . 15 Chester (Upper Mississippian) series . 15 Iowa (Lower Mississippian) series . 15 Natural floods 18 Partlow flood ......... 19 Flat Rock water invasion 26 Buchanan flood 34 Dennison Township McClosky flood . . 37 Applegate McClosky flood 48 Sandoval flood 55 Oakland City flood 59 Murphy, Dupo, Carlyle, Parker, and St. Francisville floods . 63 Summary .78 Accidental floods 78 Robinson sand floods 80 PAGE Accidental floods (Continued) Kraft flood 80 Titsworth and Newlin floods ... 84 Ben Folck flood 84 Harbison flood 84 Mefford flood 84 Lowrance and Funk — C. C. Baker flood 84 A. B. Coffman flood 85 L. C. Stewart flood 85 Thomas Gray and Chew floods ... 85 Biehl sand floods at Allendale .... 85 Tracey sand floods . . . . . . .88 Kirkwood sand floods 88 Conclusion 88 Applied floods in Illinois 94 Flood on the Kraft lease 94 Drake-Stifle controlled water flood . . 95 Carlyle purposeful flood 98 Sandoval intentional flood 98 Modern practice applied to Illinois . . . 100 Bradford technique 100 Siggins and Parker applied floods . . . 100 ILLUSTRATIONS FIGURE PAGE 1 Index map showing the producing areas that contain the various types of flooding described 8 2 Map of southeastern Illinois showing the location of oil-producing areas with reference to streams 9 3 Block diagram of southeastern Illinois 11 4 Geologic columns for the northern and southern parts of the Southeastern Illinois oil field 12 5 Diagram showing the position and relative water content of oil-producing pools in the older fields 15 6 Index map showing the locations of all the floods in the Southeastern fields discussed in this report 16-17 7 Partlow flood area, Clark County, structure map 19 8 Partlow flood area, fluid levels 21 9 Partlow flood area, pumping time of wells 22 10 Partlow flood area, relative yield of oil per acre 23 11 Partlow flood area, dissolved mineral content of water from the Lower Partlow sand . . 24 12 Graphs of Partlow oil production 25 13 Flat Rock flood area, wells and farm ownership 27 14 Flat Rock flood area, structure of the Robinson sand 28 15 Flat Rock flood area, pumping time of wells 29 16 Flat Rock flood area, relative yield of oil per acre 30 17 Graphs of Flat Rock oil production 31 18 Bridgeport sand area, Lawrence County fluid levels 32 19 Productive areas of each of the Lawrence County producing sands facing 32 20 Buchanan flood area, Lawrence County wells flooded out 33 21 Buchanan flood area, structure of the Buchanan sand 34 ILLUSTRATIONS— Continued FIGURE PAGE 22 Buchanan flood area, fluid levels 35 23 Buchanan flood area, graphs of oil production 36 24 Areas of McClosky production in Lawrence County 39 25 Dennison Township flood area, McClosky wells .... 40 26 Dennison Township flood area, structure of the McClosky "sand" 41 27 Dennison Township flood area, abandoned wells with date of abandonment 42 28 Dennison Township flood area, fluid levels 43 29 Dennison Township flood area, initial productions 44 30 Dennison Township flood area, relative yield per acre 45 31 Graphs of Dennison Township oil production 46 32 Index map of leases for which production graphs are shown 47 33 Applegate flood area, Lawrence County, structure of McClosky "sand" 49 34 Applegate flood area, structure of Kirkwood sand 50 35 Applegate flood area, fluid levels 51 36 Applegate flood area, initial productions 52 * 37 Applegate flood area, pumping time of wells 53 38 Graphs of Applegate oil production 54 39 Sandoval flood area, Marion County structure of Benoist sand 56 40 Sandoval flood area, abandoned wells with date of abandonment showing flood advance . 56 41 Sandoval flood area, initial productions ' . . 57 42 Sandoval flood area, fluid levels 57 43 Graphs of Sandoval oil production 58 44 Sandoval flood area, relative yield of oil per acre 59 45 Oakland City flood area, Gibson County, Indiana ; structure and producing areas ... 60 46 Stereogram of the Oakland City area 61 47 Graphs of Oakland City oil production 62 48 Murphy pool, Lawrence County, initial productions 63 49 Murphy pool, producing area and well abandonment 64 50 Dupo pool, St. Clair County, structure and fluid levels 65 51 Carlyle pool, Clinton County, wells and leases 51 52 Carlyle pool, structure .".,,. 67 53 Carlyle pool, pumping time of wells 68 54 Carlyle pool, fluid levels 69 55 Carlyle pool, abandonment of wells 70 56 Carlyle pool, relative yield of oil per acre 71 57 Graphs of Carlyle oil production 72 58 Parker pool, Crawford County, fluid levels 73 59 Parker pool, structure of the Robinson sand 74 60 Parker pool, pumping time of wells 74 61 Parker pool, thickness of the Robinson sand 75 62 Graphs of Parker oil production 76 63 St. Francisville pool, wells and leases, advance of flood 77 64 Graph of oil production on the Barnett lease, St. Francisville pool 77 65 Extent of accidental floods in Crawford County 79 66 Kraft flood 81 67 Locations of accidental floods and gas represjured areas in Crawford County . . . 82-83 68 Allendale pool, Wabash County, productive areas of Biehl sand 86 69 Delia Wright accidental flood at Allendale 87 70 Graphs of Allendale oil production 89 71 Combs and Smith accidental flood 90 72 Graphs of production influenced by accidental floods 92-93 73 Kraft and Rhine intentional flood 94 74 Drake-Stifle intentional flood 96 75 Graphs of Drake-Stifle oil production 97 76 Carlyle pool, relationship between flooded areas and total area 99 WATER FLOODING OF OIL SANDS IN ILLINOIS BY Frederick Squires and Alfred H. Bell INTRODUCTION An eventual shortage of oil has long been foreseen. For the last four years the amount of oil produced in the United States has been greatly in excess of new reserves discovered. The condition was aggravated when the United States entered the war. At the same time that the demands for oil for war pur- poses increased, the United Nations lost the important oil fields of the Dutch East Indies, and many oil tankers were lost through enemy action. In the light of these events all possible sources of additional oil supplies must be reexamined, not only from new potential producing areas but also by second- ary recovery of oil in old producing fields. The Illinois Geological Survey in 1931 began investigations of secondary recovery both by air and gas repressuring and by water flooding. Some of the data gathered have been published in various reports, most of which are now out of print. In order to make available to the industry the in- formation so far assembled in regard to re- covery of oil by water flooding, the present report has been prepared. This paper represents data on the suita- bility of Illinois oil sands for successful in- tentional water flooding in the older Illinois fields (fig. 1). Such phases of the geology of the older fields as have an intimate bear- ing on flooding are discussed and illustrated. It is shown that even where water en- croachment is slight but where the amount of water pumped with the oil is relatively large, oil production is prolific, and this fact is advanced as evidence of the flushing power of water. Natural flooding is described and it is shown that where edgewater encroachment is present, increases in oil production coin- cide with the advance of the flood. This procedure is so similar to artificial flooding that the results of edgewater encroachment imply that artificial floods would have a similar effect. The best evidence that Illinois sands are suitable for artificial flooding is the increased production due to accidental floods which has been observed in many localities. It was from observation of the favorable re- sults of accidental floods that Bradford Pennsylvania operators started the practice of artificial flooding with well known success. There have been several applied floods in Illinois, the favorable results of which are described as further evidence of the advisability of the practice. The location and volume of fresh water for flooding is mapped and described, and a geophysical method for locating water-bear- ing gravel beds is discussed. The use of brines as possible flooding fluids is suggested. A resume of present-day water-flooding technique is given. Preliminary Investigations Delayed production is a method used in the Bradford field to force water to travel through the tight as well as the open por- tions of a single stratum of sand. It was developed by Mr. Paul D. Torrey. Flood waters starting at corner wells finally meet at the center of the square, filling with oil and water first the open section of the sand and then, having nowhere else to go, the tighter part below or above the more perme- able streak. Not until after the whole sand body between the input wells is filled with fluid under pressure is the producing center well drilled. This principle may be applied to separate parallel sands divided by shale as well as to sand zones of different perme- ability in the same stratum. Knowledge of underground conditions is important. Many sands are irregular in thickness, porosity, and extent, and are further complicated by separation into lenses with irregular overlaps. To find WATER FLOODING OF OIL SANDS CHRISTIAN MONTGOMERY j_ A LEGEND ^ OIL FIELD WATER FLOODED OIL FIELD ( NATURAL OR ARTIFICIAL ) SCALE _1 5 10 15 20 MILES Illinois State Geological Survey Fig. 1. — Index map showing the producing areas that contain the various types of flooding described. (Oil and gas fields discovered since 1935 are not shown.) out in advance where and how flood water will travel in a proposed flood area, such information must be obtained. The best aids to a knowledge of un- derground conditions are production rec- ords, well records (giving initial oil, gas, and water productions and pressures), drill cuttings, cores of the sand, and rec- ords of resistance to and travel of injected gas, or liquid, for each vertical unit of the sand face. The core provides information as to porosity, permeability, and degree of sat- uration of the rock with connate water and oil, from which may be estimated the probable recovery of oil. The degree of permeability provides a basis for determin- ing the most efficient well spacing. Rec- ords taken at regular intervals are im- portant. It is necessary to know how much water is going into a flooding opera- tion because, until water actually reaches a pumping well, the amount of water and the thickness and porosity of the sand body are the only measures of the progress of the water through the sand. After flood production begins, the rate of input affects oil-water ratios from pumping wells. In- put pressures must be recorded as a check on input well leakage. If equal quantities of water are being pumped in daily, the pressure rises until the pumping well re- lieves it. If the pressure does not rise, there is a leak in the equipment. The approach of water to a pumping well is shown by rising fluid levels in the well and lengthening pumping time. These changes, even when gradual, are evident on a good set of records. Well records therefore take on added importance as the necessity of knowing more about underground conditions in- creases. Well spacing is determined by several factors of which the permeability of the sand is the most important, followed closely by depth, and the indications of porosity, oil saturation, and recoverable percentage of oil. These determine the cost of the operation and against them must be balanced the price of crude oil. Well patterns, whether five-spot or seven-spot, are usually determined by the sizes of single leases, the seven-spot pattern fitting better into the larger tracts. Operators now interested in Ilinois are choosing, first, territory in shallow pro- ducing formations because development costs depend upon depth of wells. At present prices for crude oil, if 500- foot wells are the economic limit for arti- ficial flooding, then 1000-foot wells will be within that limit for higher crude prices, and so on. This means that producers will locate first in Clark County on Casey sand, then in Crawford County on Robinson sand, and later in Lawrence County on Kirkwood sand, as the price increases. In- creasing prices would make it profitable to WATER SUPPLY FOR FLOODING Illinois Stale Geological Survey Fig. 2. — Map of eastern Illinois showing the location of oil- producing areas with reference to streams available as water supply for flooding. 10 WATER FLOODING OF OIL SANDS flood the deepest sands, were it practical to use only the wells already drilled. This seems feasible for many Illinois sands, judg- ing by results in the Allendale field. Graphs for well spacing, input well to oil well, have been worked out for Bradford sands in which sand permeability in milli- darcies is plotted against well spacing in feet. Such graphs plotted for Illinois sands, in general show wider spacing of wells than Bradford. The old Illinois field is drilled in a pattern of nine wells to 40 acres, with line wells 400 feet apart and other locations 460 feet apart. In some parts of the field the sand permeabilities are so great that, judging from the Bradford graphs, they would permit the use of staggered alternate old wells for water wells and pumping wells, making a distance of about 650 feet from water well to water well. The Patoka field was drilled with the intention of ultimately flooding it. Many wells were cored, and the permeability of the earliest cores settled the chosen producing well spacing. About 40 acres of this field is being prepared for flooding by drilling water wells in the center of each square, the cor- ners of which are formed by present oil producing wells. The result on production is awaited with interest. Water supply for flooding. — The map (fig. 2) shows the older Illinois oil fields and the streams in their vicinity. The Wabash and Embarrass rivers have water the year around. North Fork, a tributary of the Embarrass, is convenient to much of the northern oil territory and also car- ries water in quantity all the year. Rob- inson obtains its water supply from gravel beds near Palestine which is near the oil field. The Wabash River, from Allen- dale to a point due east of Robinson, is bordered by gravel deposits. All this terri- tory is convenient to the oil fields. The Drake-Stifle flood is operated with fresh water from a well in the glacial drift, which is good for a hundred barrels a day, just about enough for the single input well. If it were decided to "five-spot" this opera- tion, using alternate oil wells as input wells, the present fresh water supply would be inadequate. An alternative would be to use salt water from the upper water sand. This water, in a closed system not exposed to air, would not clog the sand, as is proved by the fact that all accidental floods in Crawford County are the result of under- ground salt-water invasion. The water, fresh or salt, must be kept from contact with the air in order to avoid clogging the sand through precipitation of iron oxides, or it must be chemically treated after ex- posure in order to remove the clogging material. Fresh water is generally considered the best flooding medium, and before extensive operations are begun the supply must be surveyed. The areas including existing water wells, and oil wells that penetrate fresh-water-bearing sand or gravel, should be mapped and the water output esti- mated. Streams and water-bearing gravel deposits are important sources of fresh water for flooding. Geophysical methods are now being used in advance of drilling to locate gravel deposits, which are then tested for water. As the Drake-Stifle flood used more than 50,000 barrels of water to produce 7,000 barrels of oil, it is evi- dent that a large water supply is essential. Acknowledgments Dr. George V. Cohee, formerly of the staff of the Illinois State Geological Survey assisted in preparing the section entitled "Geologic Summary." Much of the information in these pages has been local oil field knowledge for a long time. The authors did not discover it, they collected it. Their principal func- tion has been to ask questions and record the answers. These answers have involved a good deal of work for a great many people and it is to them that the authors wish to express their indebtedness. Among the many, stand out the names of Mr. C. C. Carroll and Mr. G. D. Farris and many others of the Ohio Oil Company, Mr. Wm. S. Corwin of the Tidewater, Mr. Alex U. McCandless of the Mahutska, Mr. Wm. Bell of Bell Brothers, Mr. Henry Lane of the Dinsmoor Oil Co., Mr. Paul Torrey and Mr. Mac Leighty. Our hope is that their efforts will be repaid by better returns from the old properties in which they are interested through development along the lines of this text and from a wider view point that American oil production may be increased at a time when oil is ammunition. BLOCK DIAGRAM 11 12 1TATER FLOODING OF OIL SANDS DEPTH IN FEET COLUMN SECTIO M UJ \- 00 > co z < z < > _l > CO z z LJ Q_ PRODUCTION ZONE U - 'o7o'*«'7 100 •^-Z>=Z- 200 zzszj--z~^r -zzrz^z^--=r -=^rz^r^rz^rz- -2—. —JT— 3-33-h: x x x * x * f^B^Br 300 SHALLOW GAS SAND, 295 FEET i'iii' f, f, ', ', i ' ) 'l ' l 'I u H 00 > CO z < CL CL co 00 00 00 / '/ i i l I / 1 / / / i /'/■/// / / / / -/ */-/ i/- ii — i +/-/ /_/ a.i-1 */ <-/ a./ -/ */- -zzzrz-Z7Z-z~z^- / / / / / -ziszjz-jz-— lol lol Id '/ 'to 1 lo / WESTFIELD LIME 400 '/ /III of o\ e\ o\ 334- TO 446 FEET 0\ 0\ 0| c ol ol ©I ol 'ol ol ol o\ ol ol ol ol ol ol ol ol ol o| ol ol ol o ol ol o| ol /-/-/-/-/-/ o\ ol ol o| / / / / ol ol o o / / } ) o\ ol ol o ol ol ol o| ol ol ol o ol 'ol ol ol ol o| ol o 600 ol ol ol o| ol 'ol 'ol 'ol ol ol ol o l«l ol o| ol DEPTH COLUMNAR PRODUCTION IN FEET SECTION ZONF Or 200 400 600 800 1000 1200 1400 1600 1800 ffi ss BRIDGEPORT BUCHANAN . .... , . ^ > — QC n z pa < CO McCLOSKY ^ LAWRENCE COUNTY CLARK COUNTY Illinois State Geological Surrey Fig. 4.— Geologic columns illustrating the differences in number of oil producing sands in the north- ern (Clark County) and southern (Lawrence County) parts of the Southeastern Illinois field. GEOLOGIC SUMMARY 13 Table 1. — Geologic Column for Southeastern Illinois Oil Fields System or series Group or formation, and lithology* Pleistocene Glacial drift and loess Pennsylvanian McLeansboro group — sh., ss., thin Is. and coal Carbondale group — sh., Is., ss., coal Tradewater group — ss., sh., and thin coal Caseyville group — ss., sh., and thin coal Chester (Upper Mississippian) series Menard — Is., sh. Waltersburg — ss. Vienna — Is., sh. Tar Springs — ss. Glen Dean — Is., sh. Hardinsburg — ss. Golconda — Is., sh. Cypress — ss. Paint Creek — Is., sh. Bethel — ss. Renault — Is., sh., ss. Aux Vases — ss. Iowa (Lower Mississippian) series f Levias — Is. Ste. Genevieve — Is. -< Rosiclare — ss. t Fredonia — Is. St. Louis — Is. Salem — Is. Warsaw — Is. ~\ Keokuk — Is. 1 ^ Burlington - Is. f Osage group Fern Glen — Is. J Kinderhook — sh., Is., ss. Mississippian and Devonian Chattanooga — New Albany sh. Devonian Limestone Silurian (formations undifferentiated) Dolomite Ordovician Maquoketa — sh. Kimmswick — Is. Plattin — Is. Joachim — Is. St. Peter — ss. Pre-St. Peter Unidentified Is. — limestone : ss. — sandstone ; sh. — shale. Geologic Summary The producing areas in the old South- eastern Illinois fields (fig. 1) occur along the LaSalle anticline, which is one of the major structural features of the State. The anticline consists of many axes in echelon arrangement, with cross-folds, and has a general southeast trend extending from northern Illinois through Edgar, Clark, Crawford, Lawrence, and Wabash coun- ties. In the Southeastern fields the fold plunges southeastward at a rate of 48 feet per mile, as. measured on the top of the St. Peter sandstone, which was encoun- tered at a depth of 2984 feet in northern Clark County and 5185 feet in central Lawrence County. In Clark County a small thickness of Pennsylvanian strata over- lies the eroded top of the Lower Mississip- pian beds. Lower Chester formations are present below Pennsylvanian strata in southern Clark, Crawford, Lawrence, and Wabash counties, with the Chester strata thickening southward where younger for- mations are present. The general struc- tural features in the old fields are shown 14 WATER FLOODING OF OIL SANDS Table 2. — Oil ang Gas Producing Strata in Southeastern and Other Old Illinois Fields Group or Formation Producing Strata* Pool County Approx. depth Pennsylvanian system — McLeansboro group Upper Siggins gas Siggins Cumberland, Clark 370 Bellair 500 Bellair Crawford, Jasper 560 Casey Casey Clark 450 Carbondale group Claypool North Johnson Clark 420 Lower Siggins Siggins Cumberland, Clark 560 Upper Partlow South Johnson Clark 490 Tradewater and Caseyville Bridgeport Albion Edwards 1570 groups Biehl and Jordan Allendale Wabash 1450 Bellair 800 Bellair Crawford, Jasper 800 j Biehl Keensburg Consol- idated Wabash 1740 Bridgeport Lawrence Lawrence 900-950 Buchanan Lawrence Lawrence 1250 j Robinson Main, Birds, Flat Rock, etc. Crawford 900-1000 Biehl Mt. Carmel Wabash 1490 Lower Partlow South Johnson Clark 600 Buchanan South Lawrence Lawrence 1370 Pennsylvanian Westfield Clark 290 Pennsylvanian Warrenton-Borton Edgar 160 Chester (Upper Mississippian) j! j |j | ; j; J | series — Cypress ss. Cypress Allendale Wabash 1920 Kirkwood Lawrence Lawrence 1400 Bethel ss. Bethel Allendale Wabash 2010 Tracey Lawrence Lawrence 1560 Bethel St. Francisville Lawrence 1840 Benoist Sandoval Marion 1540 Iowa (Lower Mississippian) r.:';:.L series — Ste. Genevieve formation Fredonia member McClosky "lime" Allendale Wabash 2290 McClosky "lime" Lawrence Lawrence 1700 Oblong "sand" Main Crawford 1340 St. Louis Is. Martinsville "sand" Martinsville Clark 480 Westfield Is. Westfield Clark 330 Salem Is. Westfield Is. Westfield Clark 380 Osage group Carper Martinsville Clark 1340 Carper Casey Clark 1280 Carper Westfield Clark 910 Devonian system Devonian Is. Martinsville Clark 1550 Ordovician system "Trenton" Is. Dupo St. Clair 500 "Trenton" Is. Martinsvillef Clark 2680 "Trenton" Is. Waterloo^ Monroe 410 "Trenton" Is. Westfield Clark 2260 * Sandstone unless otherwise noted. t Abandoned. % Abandoned : revived 1939. in the block diagram (fig. 3). Table 1 is a geologic columnar section for this area. Figure 4 shows columnar sections for the Clark County and Lawrence County areas. The uppermost Chester formations are not present in the old fields. Production in these fields, as shown in table 2, has been obtained largely from Pennsylvanian, Chester, and Lower Mis- sissippian strata and in small amounts from Devonian and Ordovician strata. PRODUCING SANDS 15 Producing sands in the old Illinois fields pennsylvanian system Sandstones of Pennsylvanian age have produced more than half the oil in the old Southeastern fields, which include parts of Edgar, Coles, Clark, Cumberland, Jasper, Crawford, Lawrence, and Wabash counties. The area of Pennsylvanian pro- duction in the Southeastern fields is ap- proximately 65,000 acres. Pennsylvanian sandstones producing oil in Illinois are composed of fine- to medium- sized, angular quartz grains cemented with silica. Many of the sandstones are more or less micaceous, carbonaceous, and pyritic. In the cores examined by the Survey the sands appear to be physically suitable for flooding. In general the sands average from 18 to 20 per cent pore space with permeability ranging from 0.0 to 2,430 millidarcys. The average total thickness of the produc- ing sands is approximately 29 feet and the depth below the surface in the producing areas ranges from 160 feet in Edgar County to 2,000 feet in Wabash County. In Crawford County, the largest area of Pennsylvanian production, the average depth of the producing formation is 950 feet. CHESTER (UPPER MISSISSIPPIAN) SERIES Sandstone formations of the lower Ches- ter series are productive in the Lawrence County pool, the Bellair pool, and north of Oblong in the Crawford County Main pool in the old Southeastern fields. The greatest area of Chester sand production in the old pools was in Lawrence County where the Cypress (Kirkwood) sandstone was productive in an area of 16,000 acres and the Bethel (Tracey) sandstone in an area of 4,000 acres. Chester sandstones are composed of very fine to fine angular quartz grains cemented with silica and calcite. The sandstones may be micaceous and locally pyritic and may have numerous shale partings in the upper and lower parts of the formation. The formations are fairly uniform in texture. The pore space of the sandstones is usually 19 or 20 per cent, and the perme- ability ranges from 1 to 1120 millidarcys. The average thickness of the sands as recorded in driller's logs is about 23 feet, but in some areas the thickness may be as much as 30 or 40 feet. IOWA (LOWER MISSISSIPPIAN ) SERIES Oil production from Lower Mississip- pian strata in Illinois has been principally from the McClosky "lime" in the Fredonia member of the Ste. Genevieve limestone. In the old pools the McClosky "lime" is Illinois State Geological Survey Fig. 5. — Diagram showing the position and relative water content of oil-producing pools in the older fields. From it may be decided which of the pools contains too much water to be benefited by applied water flooding and which are promising. 16 WATER FLOODING OF OIL SANDS LOCATIONS OF FLOODS 17 >c\i -£>r^ — ""O * J3 'J7>T3 -9 Cfl -T3 rt G O 8^ - DRY HOLE ^ ABANDONED PRODUCER ^'"CONTOUR ON TOP OF LOWER PARTLOW y SAND, DATUM SEA- LEVEL SCALE O V& 'M 3 /8 V 2 MILE Illinois State Geological Survey Fig. 7. — Structure map of Partlow flood area, South Johnson pool, Clark County, Illinois. (Bell, A. H., Paper presented at Second Annual Petroleum Conference of Illinois, 1933 : 111. Geol. Survey and Ill.-Ind. Petr. Assn., fig. 8, p. 22, 1934.) the Murphy, Dupo, Carlyle, Parker, and St. Francisville floods in Illinois. The map (fig. 6) shows the location of the floods in the old Southeastern field. Partlow Flood History. — The South Johnson pool (fig. 2) was outlined by drilling in 1907 and 1908 and largely drilled up by 1910. A 20 WATER FLOODING OF OIL SANDS few more wells, estimated less than 5 per cent of the total, were drilled later from time to time. The average initial daily production from 76 wells in the Lower Partlow sand was 150 barrels, the majority ranging between 50 and 250. An excep- tionally large well was the Z. E. Brant No. 8 which flowed 1500 barrels in the first 24 hours. A number of the early wells reported water below the oil in the Lower Partlow sand, and water seems to have been pres- ent in this sand from the beginning in the whole area of the pool. No data are available on the quantity of water pumped in the early years of production. Location and extent. — The Partlow flood is located in the South Johnson pool, near the south edge of Clark County, and coincides with the productive area of the Lower Partlow sand which occuoies about one square mile in sees. 26, 27, 34, and 35, T. 9 N., R. 14 W. 1 Stratigraphy and structure. — The sur- face material is alluvium and glacial drift, and there are no outcrops of bedrock in the Partlow flood area. The underlying bedrock consists mainly of shale, sand- stone, thin beds of limestone, and coal of the Pennsylvanian system. The oil sands are the Upper Partlow at an approximate depth of 460 feet and the Lower Partlow at an approximate depth of 560 feet in the central part of the area (both in the Pennsylvanian system). The Lower Part- low sand is much the more prolific source of oil. The South Johnson pool is located along an anticline, the axis of which extends south by west from the top of the Martins- ville dome into the Bellair pool in the northwest part of Crawford County. Con- tour maps on both the Upper Partlow and Lower Partlow sands show a closed dome centering in the SW. ]/\ sec. 26 and SE. ]//\. sec. 27. Figure 7 is a structure con- tour map showing the elevation of the surface of the Lower Partlow sand. Initial production. — The initial produc- tion of wells is not known for the pool as a whole. 1 Mylius, L. A., Oil and gas development and possi- bilities in E-ist Central Illinois: Illinois Geol. Survev Bull. 54, pi. 30 in map case and pp. 158 and 477-8, 1927. Fluid levels. — At the Survey's request, the Ohio Oil Company measured the fluid level in 15 wells in the Partlow flood area. Figure 8 illustrates the results of these tests. An area of higher pressure in the central part of the field grading downward to the east and west is shown. This indi- cates a tendency for flow outward from the central area. Well pumping times. — An indication of the amount of water produced in the South Johnson pool is given by the pumping times of the wells, which are shown in figure 9. The general picture is that of a central area of high water production along the high part of the structure with less water to the east and west. There is an east and west streak of relatively low water production which appears not to be related to the structure. Oil production. — Figure 10 shows the relative yield of oil per acre for the various leases. The heavier shading shows the area where the greater amount of oil per acre is produced. Note that the 40 acres that has produced the most oil consists of the R. E. Deverick No. 2 and the Blanken- becker leases (fig. 7) which includes the area in which water is believed to enter the Lower Partlow sand from below. Water-oil ratios. — The upper figure for each lease (fig. 10) indicates the water-oil ratio in 1925, that is, the number of barrels of water produced per barrel of oil. It will be noted that there is a general relationship between the amount of water and the amount of oil produced. In other words, the greater the water production, the greater the oil production. In order, if possible, to determine the source of the flood water, samples of water from key wells were tested for amounts of dissolved minerals. It has been observed that dissolved mineral content generally in- creases with depth. The evidence indicates that the flood water entered from a lower horizon at the place where the water showed the greatest mineral concentration (fig. 11). The most striking feature of this flood is the fact that water has entered the Lower Partlow sand from below and at a place near the center of the pool and has spread outward. The oil production was greatly benefited (fig. 12). NATURAL FLOODS 21 s bO' LEGEND WELL IN WHICH FLUID LEVEL WAS MEASURED BETWEEN APRIL 21 AND JULY 6, 1934 ALTITUDE OF TOP OF FLUID COLUMN CONTOUR ON FLUID LEVEL - INTERVAL 20 FEET DATUM SEA-LEVEL BOUNDARY OF PRODUCTIVE AREA SCALE % ENCLOSED BLANK AREA PRODUCING AS OF MARCH 1933 SCALE Va Vz MILES Illinois State Geological Sun ey Annual Petroleum Conference of Illinois 1933 111 cL i ac// ° 6~ny "30 23 'I 2 *1 M. Mc CROSK1 IS O 19 16° PROPERTY MAP OF McCLOSKY FLOOD AREA PART OF DENNISON TOWNSHIP LAWRENCE COUNTY, ILLINOIS MC CLOSKY ( PAST O MC CLOSKY WELL NUME 'RESENT) ND NON-PRODUCER (SEE APPENDIX) State Geological Si. Fig. 25.— Map of the McClosky wells in the Dennison Township flood area. (Bell, A. H., and Piersol, R. J., Illinois Geol. Survey Illinois Petroleum No. 22, fig. 2, p. 3, 1932.) NATURAL FLOODS 41 ZAP OF CLOSKY SAND PART OF DENNISON TOWNSHIP LAWRENCE COUNTY, ILLINOIS O MC CLOSKY SAND PRODUCER ( PAST OR PRESENT) -©- MC CLOSKY SAND NON - PRODUCER /" CONTOUR ON MC CLOSKY SAND ELEVATION BELOW SEA -LEVEL CONTOUR INTERVAL - 10 FEET SCALE eo/oyical purvey Fig. 26. — Structure map of the Dennison Township McClosky area contoured on top of McClosky "sand." It is interesting to compare this map with figure 27 which shows water invasion over the axis of the structure. (Figure reprinted from Illinois Geol. Survey Illinois Petroleum No. 22, fig. 4, p. 5, 1932.) 42 fVATER FLOODING OF OIL SANDS MAP SHOWING FLOOD ADVANCE INDICATED PLUGGING BACK RECORDS PART OF DENNISON TOWNSHIP LAWRENCE COUNTY, ILLINOIS O MC CLOSKY SAND PRODUCER (PAST OR present) ■e- MC CLOSKY SAND NON- PRODUCER CELINE SHOWING APPROXIMATE 13 1913, YEAR WHEN WELL WAS PLUGGED BACK PB. WELL PLUGGED BACK, DATE UNKNOWN Ab. WELL ABANDONED Fig. 27. — Wells abandoned because of advancing crest of the flood mark a path across the axis of the LaSalle anticline. This should be compared with the path of the Applegate McClosky flood which has about the same inclination although it flows in the opposite direction. (Illinois Geol. Survey Illinois Petroleum No. 22, fig. 8, p. 10. 1932.) NATURAL FLOODS 43 Fig. 28. — The map showing fluid levels provides a picture of the direction of water movement outward from a high area. Note the similarity with the progress of the flood as shown in figure 27. (Illinois Geol. Survey Illinois Petroleum No. 22, fig. 9, p. 11, 1932.) 44 WATER FLOODING OF OIL SANDS M IAP OF McCLOSKY FLOOD AREA DENNISON TOWNSHIP LAWRENCE COUNTY, ILLINOIS SCALE LEGEND LINE SHOWING INITIAL PRODUCTION IN BARRELS PER DAY 6 AVERAGE THICKNESS OF MC CLOSKY SAND IN FEET FOR A LEASE OR FRACTION OF A LEASE State Geological Survey Fig. 29. — Contour map of initial productions of McClosky "sand" wells in Dennison Township flood area. (Illinois Geol. Survey Illinois Petroleum No. 22, fig. 5, p. 6, 1932.) NATURAL FLOODS Illinois State Geological Survey Fig. 30. — The map showing the relative yield per acre indicates, by the increasing richness of oil pro- duction in the direction of the flood advance, that oil was moved by the flood across property lines: (Bell. A. H., Paper presented at Second Annual Petroleum Conference of Illinois : 111. Geol. Survey and Ill.-Ind. Petr. Assn., fig. 11, p. 25, 1934.) 46 WATER FLOODING OF OIL SANDS '"" 1913 1915 'Ear ,913 1915 -C J BOBDEt, >13 1915 1920 SOUTHWEST 70 NCRTHEAS D 1925 193 LEIGHTY fc» W. A. GOULD l\ V- H H GOULD i i ""-, 930 1913 1915 PRODUCTION CURVES SHOWING PROGRESS OF WATER-FLOOD IN McCLOSKY FLOOD AREA DENNISON TOWNSHIP, LAWRENCE COUNTY, ILLINOIS (vertical scale varies for different leases) Fig. 31. — Production curves showing progress of water flood in Dennison Township flood area by relative dates of secondary production peaks. (Illinois Geol. Survey Illinois Petroleum No. 22, fig. 6, p. 8, 1932.) NATURAL FLOODS 47 MAP OF McCLOSKY FLOOD AREA DENNISON TOWNSHIP LAWRENCE COUNT Y, ILL INOIS 1/2 SCALE LEGEND 1915 DATE OF MAXIMUM INCREASED PRODUCTION DUE TO FLOODING 1912 DATE OF MAXIMUM PRODUCTION - , NO APPARENT INCREASE FROM FLOOD Illinois State Geological Survey Fig. 32. — Index map of leases for which production graphs are shown in figure 31 (Illinois Geol. Survey Illinois Petroleum No. 22, fig. 7, p. 9, 1932.) 48 WATER FLOODING OF OIL SANDS Applegate McClosky Flood History, location, and extent. — The Ap- plegate pool, which is located in sees. 12, 13, and 24, T. 4 N., R. 13 W., and sees. 7, 18, and 19, T. 4 N., R. 12 W., Petty Township, Lawrence County, was drilled up with more than 100 wells. They pro- duced less than the wells in the Dennison Township pool and the flood began later than the Dennison flood. It was first be- lieved locally that all the water was and would stay east of the main north and south road through the pool. The flood obviously started there but has since advanced across the pool in a southwestern direction. It is interesting to observe that the paths of the Applegate and Dennison floods are parallel across the LaSalle anticline but the floods moved in opposite directions. Structure. — The productive area lies on the north and west flanks of the LaSalle anticline ; on the crest the McClosky is dry although the Kirkwood sand is productive. There is an outlier on the west parallel to but almost separated from the main pro- ductive area. At the time of this study no wells had been flooded out. Structure con- tours on the Kirkwood sand are shown in figure 33 and on the McClosky "lime" are shown in figure 34. Fluid levels. — The contours showing fluid levels (fig. 35) make a very regular picture of a water movement from the northeast to the southwest. Even the out- lier shown in figure 34 fits into the regular succession. Initial production. — Figure 36 shows high initial production and therefore great permeability of sand at the place where flooding started and throughout the path of the flood. Pumping ti?nes.^A\ though as yet no wells have been abandoned, records showing longer pumping time (fig. 37) closely cor- respond to the areas of higher fluid levels. Production graphs. — The production records as shown on figure 38 indicate a high oil yield, but there are no marked secondary production peaks. The fact that the whole length of the flood is less than a mile, which is a shorter distance than pro- duced the most impressive flood peaks in the Dennison Township flood, may account for their absence here. Relative yield. — As there are no produc- tion peaks, the only way to appraise the value of the flood on oil production is to compare the production per acre with the average production of unflooded McClosky sand. It is many times that average. At the time this study was made, many wells in the flood area were making 12 barrels a day. The total area of McClosky production is shown on figure 19. NATURAL FLOODS R .13 W R. 12 W. 49 ,00 LEGEND • OIL WELL -•- ABANDONED WELL -§- DRY HOLE -^r GAS WELL y CONTOUR ON TOP OF KIRK WOOD SAND ' DATUM SEA- LEVEL SCALE Va Va. 3 /e '/ 2 mile Illinois Slate Geological Survey Fig. 33— Structure map of the Applegate flood area contoured on top of the Kirkwood sand. 50 WATER FLOODING OF OIL SANDS R.I3W. R.I2 W. % Vz SCALE I 2 Miles ZOO ^CONTOUR ON TOP OF MC CLOSKY SAND DATUM SEA- LEVEL PRODUCTIVE AREA Illinois State Geological Survey Fig. 34. — Structure map of Applegate flood area contoured on top of McClosky "sand." Unpublished map by P. S. McClure. The Applegate maps show an interesting relation of production to struc- ture. The McClosky is dry on the highest part of the LaSalle anticline and parts of the east flank but is productive far down the west flank. The water invasion started in the high part of the structure and advanced downdip until its effect had been felt clear across the structure. NATURAL FLOODS 51 LEGEND • MCCLOSKY PRODUCER -y- MCCLOSKY DRY (§) WELL IN WHICH FLUID LEVEL WAS MEASURED JANUARY TO MARCH 1936 -700 ALTITUDE OF TOP OF FLUID COLUMN ^---"CONTOUR ON FLUID LEVEL ; DATUM SEA-LEVEL ..••■ BORDER OF MCCLOSKY PRODUCING AREA SCALE '/a '/4 3 /B '/ 2 MILE Illinois State Geological Survey Fig. 35.— Contours on fluid columns of equal height for the Applegate area indicate direction of fluid movement. Figures 35, 36, and 37 present corroborating evidence of a flood across the La Salle anticline from northeast to southwest. LEGEND J45 WELL HAVING INITIAL PRODUCTION OF 145 BARRELS PER DAY -•- ABANDONED WELL -$- DRY HOLE y CONTOUR SHOWING INITIAL PRODUCTION SCALE '/8 ^4 3 /8 V 2MILE l Illinois State Geological Sur I IG u ;7"^ ap of A PP le ? ate flo °d area contoured on initial productions of McClosky "sand" wells It should be compared with figures 35 and 37 to demonstrate that water under higher head seeks the line of least resistance, which in this case is the line of higher permeability LEGEND • WELL PUMPED 6 HOURS PER DAY -- DRY HOLE / LINE SURROUNDING WELLS PUMPED ' 24 HOURS PER DAY LINE SURROUNDING WELLS PUMPED 12 HOURS OR MOPE PER DAY SCALE ■£ Illinois State Geological Sunt Fig. 37.— Map of Applegate flood area showing pumping time of wells. Areas of longer pumping times correspond to areas of higher sand permeabilities. 54 WATER FLOODING OF OIL SANDS a.r. applegate tr. < 6 J L PERRY KING NO. 2 J 1 1 L W. APPLEGATE 1920 1925 1930 Illinois State Geological Survey Fig. 38.— The production graphs for the Applegate area show no secondary peaks. Water invasion must have followed so closely the removal of oil and gas, due to highly permeable sand, that its effect on every well was continuous and there was no lag. These wells cover 320 acres and have produced 2,476,676 barrels to date at the rate of 7,740 barrels per acre. NATURAL FLOODS 55 Sandoval Flood History, location and extent. — The dis- covery well of the Sandoval pool, the L. Stein No. 1, was drilled in 1909. Within five years the production limits had been defined by 80 wells, and the field was drilled up soon afterward with a total of 116 pro- ducing wells. The pool is located in sees. 4, 5, 8, and 9, T. 2 N., R. 1 E., Marion County. The producing formation is the Benoist sand (Bethel), Chester series. Structure. — Figure 39 shows the lease names, well numbers, and structure con- tours on the Benoist sand. The principal feature is an elongate dome; its axis trends a little north of east, and it has a closure of about 50 feet. As may be seen from the map, production was closely confined to the structure, wells off structure going into the water which entirely surrounds the oil- bearing area. Dates of abandonment. — As gas and oil were removed from the higher parts of the structure, the reservoir pressure there was lowered and the water surrounding the dome moved inward to establish a new equilibrium — never quite reached, due to the constant removal of the fluids. The out- side wells were soon pumping too much water to pay and so were abandoned. The contours on figure 40 show dates of aban- donment at five-year intervals. They follow structure contours with considerable regu- larity. Due to a low price for the oil, the outer circle of wells were shut down as it no longer paid to pump so much water for so little oil. The operators were surprised to find that the next inner string of wells increased their oil production. Pumping water from the outer circle of wells had reduced the motive power flushing oil to the inner wells. This was restored when the water power was no longer removed. Water is now over the top of the struc- ture and all wells are being pumped at high speed with long stroke in order to get the last of the flood-produced oil. Fluid levels. — The contours showing structure, dates of abandonment, and fluid levels tell a striking story. Before drilling, the dome of porous Benoist sand was com- pletely filled with oil saturated with gas and under pressure from a high head of surrounding water. After drilling, the oil and gas were rapidly released, creating an area of lower pressure. Water then began to invade the lower edges of the dome to restore equilibrium. Its lag in doing this can be measured by the decreasing heights of the fluid columns in wells from the border to the center of the dome, and by the time interval between drilling of wells and the secondary production peaks. Figure 41 shows this with contour lines, the arrows pointing in the directions in which the water is moving. The surface represented by the fluid level contours is in general the inverse of the upper surface of the domed Benoist sand. Production. — Initial productions are plot- ted on figure 42. The Benoist leases in the geographical center and on the high part of the structure were the last to show flood production peaks. These occured in 1927 and 1928 and measure a lag of about 20 years between drilling peak and flood peak. The Stein lease reached its peak in 1924, the Dean in 1921, and the Warfield in 1920. Edge leases as a rule show less lag and less distinct production peaks as is to be expected because they are nearer edgewater and so are affected by it sooner than more central farms and before pres- sures drop much. Differences in sand permeability caused some irregularity in water advance. Graphs (fig. 43) illustrate these statements. Relative production per acre for each lease from 1916 to 1931 is shown on figure 44. The yields of oil are indicated by the degree of shading. As production figures prior to 1916 (which includes all the flush production of the field) could not be ob- tained, it is impossible to estimate the aid given to production by water action. This of course was present and effective from the first. 56 WATER FLOODING OF OIL SANDS Illinois State Geological Survey Fig. 39. — The structural dome in the Benoist sand at Sandoval is well shown by contours. Water encroachment was influenced by this structure but its regularity was interrupted by differences in permeability of the sand (fig. 40). (Bell, A. H., Paper presented at Second Annual Petroleum Conference: 111. Geol. Survey and Ill.-Ind. Petr. Assn., fig. 4, p. 18, 1934.) R.I E Illinois State Geological Survey Fig. 40. — Water did not advance by regular increments of level up the slopes of the dome but was influenced also by paths of lesser resistance due to greater permeability of the sand. Compare figures 39 and 40. (Bell, A. H., Idem.) NATURAL FLOODS R.IF. 57 Illinois State Geological Survey Fig. 41. — Permeability of the sand, as indicated by initial production of wells, is shown to be in- dependent of structure; this accounts for an irregular water invasion. (Bell, A. H., Idem.) R.I E. Illinois State Geological Survey Fig. 42. — A competition between gravity and sand permeability is further illustrated in the map show- ing fluid-levels. This dumb-bell shaped set of contours bears a resemblance to the inverse of the structure contours but shows undulations at the east and west which are the results of greater water penetration following paths of greater permeability. (Bell, A. H., Idem.) 58 WATER FLOODING OF OIL SANDS EAST HALF 1920 1925 1930 Illinois State Geological Survey Fig. 43. — Sandoval production graphs show secondary peaks due to lag between initial production under gas pressure and later production under water drive. The lag between these peaks increases with increasing distance of the lease from the original water-oil contact. (Bell, A. H., Op. cit, fig. 7, p. 20, 1934.) NATURAL FLOODS DIP. 59 Illinois State Geological Survey Fig. 44. — Due to lack of early production records, the relative yield per acre cannot be plotted ac- curately. Nevertheless this map shows a general enrichment toward the middle of the pool. (Bell, A. H., Op. cit, fig. 5, p. 19, 1934.) Oakland City (Indiana) Flood History, location, and extent. — The Oak- land City field was discovered in 1909 and drilled up with more than 200 wells that developed nearly 2000 acres in sees. 11, 14, 22, 23, 26, 27, and 35, T. 2 S., R. 8 W., Patoka and Monroe townships, Pike County, Indiana. The producing sand is the Oakland City, correlated with the Kirk- wood sand of Illinois (the Cypress forma- tion of Chester age). The lease names and lease well numbers are shown on figure 45 (left). Structure. — Figure 45 (left) is also a structure map, with contours showing the elevation of the top of the Oakland City sand. They indicate an anticline the axis of which trends generally north and south but which plunges to the south and south- west. A stereogram (fig. 46) shows struc- ture and sand thickness. The sand body is a single lens and is unusually uniform. Abandonment. — This flood is nearing the last stages, a large percentage of the wells having been abandoned. Figure 45 (right) shows the original boundaries of production, the wells still producing as of 1933, and the abandoned area between. Present pro- duction in general is confined to the highest parts of the structure, corresponding closely to production at Sandoval. Fluid levels. — Not enough fluid levels have been measured to provide sufficient information for plotting fluid-level contours but it is known that water rises to great heights in temporarily shut-down wells in the flood area. Production. — Initial productions were small, and not enough records were obtained to use them in a study of sand permeability. All wells pump straight time. A striking characteristic of the Oakland City pool is the fact that flood production peaks were often as high as the early flush production peaks, and in the John Yager lease it was higher. Almost all leases showed flood peaks, some of them abrupt and strongly marked. The Southfork Fee showed the shortest interval of time from flush production to flood production ; it was drilled in 1909 and reached its secondary production peak in 1916. The John Yager well, drilled in 1909, not influenced by water drive until 1927, marks the longest time lag. 60 WATER FLOODING OF OIL SANDS LEASES, WELLS, & STRUCTURE ^^CONTOUR SHOWING ELEVATION OF TOP OF OAKLAND CIT-i SA NO (KIRKWOOO C CORCORAN , CO.HOUCHINS &--,-- ---h ONROE TWP. \ ♦ /♦'IsiPPLE \ 2 /'/ 14 i /S. THOMPSON/ / BU ■■-/■■Mi ILLINOIS STATE GEOLOGICAL SUPVE / / * V^nA. PUMPING TIME DATES OF ABANDONMENT \ 17 1917 Y~\ _I2 PUMF PING TIME PER HOURS) -» — —BOUNDARY OF ORIGINAL PRO- ♦ j ♦ ! ♦ 'i'* '*\ OUCTIVE AREA __l ■>-- -r-- V ._— -80UNDARY OF * /?/?/?"* j -» I 19 33 PRODUCTIVE ♦A9 | L AR EA !♦ \ j — * LEGEND PRODUCING OIL WELL ■ DRY HOLE SCALE Illinois State Geological Survey Fig. 45. — (Left) Map of the Oakland City area, Gibson County, Indiana, showing farms, wells, and structure contours drawn on top of the Oakland City (Kirkwood) sand. (Right) The outline of the original field and the field producing in 1933 should be compared with the structure map as a demonstration of the fact that water coming in at lower levels on all sides has raised the remaining oil into the top of the structure. These maps show an unusually perfect case of encroachment regulated by structural elevation: (Bell, A. H., Paper presented at First Annual Petroleum Conference, 1933: 111. Geol. Survey unpublished ms., fig. 10.) NATURAL FLOODS 61 *0 bo *0 5 *S? .G u en » 1) > T3 G > '/I o s o 3 G n *n CO U> o 1—1 o a; X) o 03 Ui en V) G G O J* U u -G £ G G » G "nj < G +j en 3 'i- .2 rt en CU en >►, CU u u T3 G 03 CU ^ as r Offi rci CU jl: PQ bn G en £ en CU o G .G ^ a S ,G nJ —, u bn o of V G cfl U! XI 1 o CU vn <5f G CJ o re Uh u o 62 WATER FLOODING OF OIL SANDS 1915 1920 192b 1930 1935 1925 1930 1920 1925 1930 FRED A. BUTLER 1910 1915 1920 1925 WARRICK MASON 1915 1920 1925 1930 920 1925 1925 1930 1925 1930 R H. HENNIG 1920 1925 '915 1920 1925. 1930 1915 1920 STEPHEN COOF 1915 1920 1925 1930 Illinois State Geological Survey Fig. .47. — The Oakland City production graphs give a striking picture of the results of the water drive. All show secondary peaks and all show time intervals between the primary and secondary peaks that are related closely to the distance of the wells from the original water-oil contact. For example, the centrally located John Yager has a time lag of 18 years between peaks. The well now producing on this lease is making more oil than when first drilled. The way in which structure controls water encroachment is illustrated by the south end of the field. Here leases on the very top of the structure are still producing, although water had to travel twice as far to attain the same elevation on the west as on the east. Water movement has been slow, as evi- denced by the fact that in some places it has taken more than ten years to travel a mile. Figure 47 contains production graphs. Production records are incomplete, but the Oakland City data demonstrate that water moved oil tow T ard pumping wells and that geologic structure determined the position of the movement. It suggests a very uni- form permeability of the sand. Figure 19 shows the area of Cypress (Kirkwood) sand to which the conclusions drawn from this flood may be applied. NATURAL FLOODS 63 ft ABANDON £ * CAS WELL f .V CAS FIELD r C ™ L S ^ i Illinois State Geological Survey Fig. 48. — Contours on wells of equal initial production in the Murphy pool. Production fell below the economic limit and the wells were abandoned during the time lag between initial and flood pro- duction peaks. Wells drilled later produced flood oil. Murphy, Dupo, Carlyle, Parker, and St. Francisville Floods Murphy flood. — The Murphv flood, sees. 5, 6, 7, 8, and 31, Ts. 2 and 3 N., R. 11 W., Dennison Township, Lawrence Coun- ty, is illustrated by maps showing the initial production (fig. 48) and water encroach- ment and abandonment (fig. 49). Wells were abandoned before the benefit of water encroachment had been felt, but later wells drilled into the flood area were benefited. Dupo field. — The Dupo field in the "Trenton" limestone is interesting because the water movement started at the high part of the structure and advanced outward toward the lower structural levels. This is shown by structure contours and contours on wells of equal fluid level (fig. 50). The installation of pumps capable of handling large quantities of fluids has in- creased oil production. Carlyle flood. — At Carlyle a modest water invasion occurred which moved oil so that greater productions were obtained at increased distances from the origin of the encroachment. The pool is located in sees. 2, 3, 10, and 11, T. 2 N., R. 3 W., Clinton County, as shown on figure 51. The structure is shown on figure 52. Evidences of the direction and extent of the water movement agree in all the tests of pumping time (fig. 53), fluid levels (fig. 54), dates of abandonment (fig. 55), relative per acre yield (fig. 56), and pro- duction peaks shown on production graphs (fig. 57). Water entered from the south- east and moved a little east of north across the producing area. The production was light and showed some water from the beginning, and the effect of water movement on oil production has not been large. Some leases showed mild flood production. An attempt to flood this pool artificially is de- scribed on page 97. 64 WATER FLOODING OF OIL SANDS A.B.JORDAN LEGEND • OIL WELL -if- DRY HOLE ABANDONED WELL LIMIT OF MC CLOSKY PRODUCTION ^J PRODUCING AREA 1935 3**= WATER ENCROACHMENT ^%) ABANDONMENT BY YEARS J- & Illinois State Geological Survey Fig. 49. — Map of Murphy pool, Lawrence County, Illinois, showing limit of McClosky production, producing area as of 1935, water encroachment and well abandonment by years. •A • • '• / y'k\ "•-. \ / •" // 293 \ '• * / •' \ # '• '• / : L A •\ '• *'■ / • K • \ _ *• * / : y/\284 V^- / ••' / V°° \ '•-•. y " •' yy ^ # \ \ '• x ""• ; /f\\\V_ \ / ;/• \ 3 5°^C^ \ '•• / o o .// s' / \ \ v' \ / '■ I' */ \ \ \ / * p • /\ / DYROFF \ o0 V "/ ^<*. • / •'. / .> \ / \^V *. '*•/ / '' z' '~\ y^ s. '/ * * / '• # ^?^ * V ?==== ^^ == \ REICHERT \ •• 2 //• '• • / °. y\ i / If v \ • 11 '• *. / X-A •• // °^ \ 387 '• // \ \ ' h / \465 J? TARLTON ° > ^- DRY HOLE ^S* CONTOUR ON FLUID LEVEL * WELL IN WHICH FLUID LEVEL WAS ^ y STRUCTURE CONTOUR ON TOP OF MEASURED JAN. TO APR. 1934 ••'* "TRE NTOn" ; DATUM SEA-LEVEL Illinois State Geological Surrey Fig. 50. — Contours on fluid levels January to April 1934 and on elevation of the top of the "Trenton" limestone in the Dupo pool, St. Clair County, Illinois. 66 WATER FLOODING OF OIL SANDS R.3W. BECKER MC CABE / II 6 8 ( / 2*^ /,« // / tURPH v / / 9 >/ ,*' •8 5 . J 3 7^ l(T 6" 2 13 .8 SCHLAFLY SCHOMAKER DONEWALD ,-> 12 3 4 14 i? x / rv" — —-=-- *IsV_^/ 6^ I 6* 7 e 8 ^ l2 V Jl 12 | " • ('DETERS •_ I 8 7 • 9 6 10 ♦ DETERS A/C2 • J3 JO .7 1 .3 J« .13/ 0.L_. \ 7* 9 # I2*Y 4 «* 7* 1 1* *2 ,8 .10 /^ • 8 J 3 . 4 5 • DETERS A/CI # 2 6* 14 10. '• DIEPENBROCK .3 5 ^?| -hi— ~ .3* IS .7*/ 14 J8 / */ \ J X GURDES 7 N, ROGERS \2 \ KARHOFF / 4 / * A''l 6 5 4+ 3' r l2 HEMPEN .A/C2, II 7 6. 9 JO 5 -®" 5 \ *. i *3 # 2 *. ->^« -2| .4 5^ HEMPEN I 7 A/C I 3^ ! ^ i H. WILKINS x U 6 •$- "9- ♦ 2* 13* 20* 9.^ 7 23 ^ i 22 *, ..♦ ._- . I M 6" / / / '♦/ \ 17 19 <" ^- " 18 -f MARIA SMITH r" V--7- -^ >^6 J.B.WILKINS • OIL WELL -*- ABANDONED WELL -£- DRY HOLE /BOUNDARY OF PRODUCING AREA SCALE Va '/a MILE /P33 Illinois State Geological Survey Fig. 51. — Map showing wells and leases in the Carlyle pool, Clinton County, Illinois. (Bell, A. H., Paper presented at Second Annual Petroleum Conference of Illinois : 111. Geol. Survey and Ill.-Ind. Petr. Assn., fig. 2, p. 14, 1934.) NATURAL FLOODS 67 Illinois State Geological Survey Fig. 52. — Structure map showing a slight doming in the Carlyle pool. (Bell, A. H., Idem.) 68 WATER FLOODING OF OIL SANDS R. 3 W. Illinois State Geological Survey Fig. 53. — Map showing pumping times of wells in hours per day (1933) indicating relative amounts of water produced with oil in the Carlyle pool. (Bell, A. H., Idem.) NATURAL FLOODS 69 Illinois State Geological Survey Fig. 54. — Contour map showing heights of fluid columns in wells in the Carlyle pool. These levels show the direction of the water invasion, and the evidence is corroborated by the contours on water production shown in figure 53. (Bell, A. H., Idem.) 70 WATER FLOODING OF OIL SANDS R.3 w. 1 /. .\ v 6 ,-; ."! • • . /" 2i . • ,• •.:*•* . /. : 1 "• ' /■■"• : • • •... 1 • • • i I • ,•■•• A 2'l\ • ] ] .-•■ 20/ ! ,/ -j; ,\ / 19-- ! / T. ! /*' 1 \ 19 19 < /* • • • i •'•. ♦ ♦ j/: 1 .- j I ♦ / N. IV •.••■["■*""••• i9* i • • ; \ ! ^ • 1 '•. 1 19 : N ♦'■<>■ • 1 • •,...,•••■ /'■•";". • i • > 20 20 19 I9-I 4 / r*~ ^s 19 19 ** 1 N >* ♦••£ 1 / / a ■••''" ! 27 ! •-•• "'•■■ 20 20 20 ^ ! *_*i. _tfv. (T !*'■•••....•{• • .■••• !'%>•• •♦ ^-4-_ ■■•■.! .••"'22, \ ^— ■ - "~ • OIL WELL -•" ABANDON-ED WELL i 4' V ^"BOUNDARY OF PRODUCING AREA | J 12 i i i ! 21 1921 "' BOUNDARY OF PRODUCTIVE AREA AS OF FEBRUARY 1933 SCALE !/4 V* MILE 1933 Illinois Stale Geological Surrey Fig. 55. — Relative time of abandonment of wells, due to water. This evidence strengthens the indica- tions of the direction of fluid movement as shown by water volume and pressures in figures 53 and 54. (Bell, A. H., Idem.) NATURAL FLOODS 71 Illinois State Geological Survey Fig. 56. — The relative yield per acre for leases shows that water has moved oil with it in its north- westward passage, enriching increasingly each lease in its path. (Bell, A. H., Idem.) 72 WATER FLOODING OF OIL SANDS ' 20 - NORTHEAST PART \\V \ ...l^HUMACHER \. N \ .--' %^ \" N -t% / ..c.. c -^--.^ DETER %3 100. O D Q 2 60 o. < Z Z 40 < ..X- SOUTHWEST PART ^f ^v5T- ss*> ^s \ 'vs<^— ' 1 ISr 1 - 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 YEAR Illinois State Geological Survey Fig. 57. — Carlyle area production graphs show mild secondary flood peaks which occur after proper time lags in relation to the distances of the leases from the origin of the water drive. (Bell, A. H., Paper presented at Second Annual Petroleum C onference of Illinois : 111. Geol. Survey and 111.- Ind. Petr. Assn., fig. 3, p. 15, 1934.) NATURAL FLOODS 73 -®- JQHN R WEGER Illinois Stale Geological Survey Fig. 58.— Contours on fluid levels for wells in the Parker pool, Crawford County, Illinois, present a picture of water movement from northwest to southeast. Parker pool. — The Parker pool, sees. 9, 10, 11, 14, 15, 16, T. 5 N., R. 12 W., Honey Creek Township, Crawford County, which had always produced great volumes of water with its oil, shows evidence of water movement with a few production peaks due to flooding. Water drive helped its oil production. Conditions are illustrated in figure 58 for fluid levels, 59 for struc- ture, 60 for pumping time, and 61 for sand thickness. Its production is shown in figure 62 by means of graphs. St. Francisville pool. — The Jones and Barnett farms at St. Francisville attracted attention by marked increases in oil produc- tion due to water encroachment. The producing sand is the Bethel. St. Francisville and Sandoval pools are the only instances to date of edgewater en- croachment on this formation. An applied flood on the Bethel sand at Patoka is con- templated. The pool is located in sees. 19, 20, 28, 29, T. 2 N., R. 12 W., Dennison Township, Lawrence County, and is illustrated by figures 63 and 64. Initial productions of the wells were small and the flood increases are therefore more noticeable. Water invaded from the southeast and moved in a northwestern direction. A graph of the oil production of the Barnett lease is shown. The entire area of Tracey production is shown on figure 19. 74 WATER FLOODING OF OIL SANDS LEGEND J CONTOUR ON TOP OF y robinson sand ; .40° DATUM SEA -LEVEL .gV* - DEPRESSION CONTOUR -^ .,■ •■ BOUNDARY OF PARKER POOL o SCALE IA * Illinois State Geological Survey Fig. 59. — Structure of the Robinson sand in the Parker pool. Jllinois State Geological Survey Fig. 60. — Map showing the pumping time for wells in the Parker pool. NATURAL FLOODS 75 Illinois State Geological Survey Fig. 61. — Thickness of the Robinson sand in the Parker pool. S SEACHRIST 1915 1920 1925 1930 Illinois Slate Geological Survey Fig. 62. — Oil production graphs for the Parker pool show several secondary peaks. NATURAL FLOODS 11 Illinois State Geological Survey Fig. 63. — Map showing wells, leases, and line of furthest advance of flood water in 1933 in the St. Francisville pool. The increase in production on the Jones and Barnett leases is particularly interesting as it is the only known edgewater encroachment on the Bethel sand in the Southeastern Illinois field. Fig. 64. — A production graph for the Barnett lease (two producing wells) in the St. Fran- cisville pool. 1925 1930 1935 78 WATER FLOODING OF OIL SANDS Summary The high lights of the natural floods are these : Water came from a lower formation to cause the Partlow flood. At Flat Rock water brought in so much oil from outside the area that the drilled part of the sand produced more oil than it contained orig- inally. The Buchanan flood is noteworthy for its record-breaking production. The Dennison territory may have been made permeable as a weathered "high" on a cross-fold. Applegate flood water moves unaccountably in the opposite direction from that of the otherwise closely related Den- nison flood. Oakland City is unusual in that the flood peaks attained and sometimes surpassed the peak heights of the original flush production. The Murphy flood was abandoned during the period of lag before the flood reached the pumping wells. The Dupo flood water showed the highest fluid levels at the top of the structure. As far as this paper is concerned, natural floods have no value unless their evidence points to the probable success or failure of applied floods on other Illinois sands. It may be argued that natural floods move under their own power and that the only areas available for purposeful flooding are sands now devoid of all reservoir fluid move- ment. But, as has been stated, all sand and lime oil-producing strata in Illinois are under some edgewater pressure, and edge- water encroachment differs only in speed. Given enough time, all Illinois producing sands would experience edgewater encroach- ment. Even Robinson sand wells, when left unpumped for a year, fill up 200 feet or more with fluid. Only edgewater pressure could maintain such heights of fluid column. Then how do these natural floods indicate a probable success for artificial floods? First, they demonstrate that water moves oil. Second, they demonstrate that many Illinois sands are continuously permeable over large areas. Third, when a natural flood at Sandoval was supplemented by an artificial flood introduced within it, this ap- plied flood further increased production. It was proved that each kind of flood produced the same kind of result in increasing oil production. The efficiency of accidental floods which occur in sands under edgewater drive, is further evidence that applied floods would be successful in edgewater encroachment territory. Because of these facts, it is logical to recommend that purposeful flooding in advance of natural water-drive be applied in pools that are already under edgewater pressure. ACCIDENTAL FLOODS Instances of accidental floods in Illinois have been sought out and studied and are here described because they are the best evi- dences short of actual trial that similar sands will produce similar or better results under purposeful flooding. Accidental floods at Bradford, Pennsyl- vania, drew the attention of oil operators to a new production possibility and were the starting point from which the whole process of flooding as a production method has developed. Illinois presents the same kind of evidence in a more conclusive manner. This evidence will be found in the following pages. Where water is present in sands above the oil-producing zone, in many instances it has broken through the casing and entered and traversed the lower oil sand. This has resulted in accidental floods in the Robin- son, Kirkwood, Tracey and Biehl sands, in locations which are indicated on figure 6, pages 16-17. The Survey has developed the following technique for the study of these accidental floods: The location of each flood is shown on an index map and given an index number. Wherever information is available, a sep- arate map of each flood area is made giving farm name, location, lease owner, well lo- cations and numbers, structure, water spread, contours on equal sand thickness, equal initial productions, equal pumping times, and abandonment dates. The heights of static fluid columns are measured and shown by contours. Flooding wells and flooded wells are designated. Sand thick- ness, permeability, porosity, and grain size are described. The time of beginning of the flood and distance travelled by it are given, and a graph is made on which is shown esti- mated normal production and total produc- tion. By subtraction of normal from total, flood production is estimated. Water analy- ses are made as an aid in discovering the source of flooding water. As it is not pos- sible to cover all these points in this dis- cussion only the more important features are described. ACCIDENTAL FLOODS 79 51 + 'IS* • • A3 • • 1 § A fc • C ^ .* i • • 1 • • • • • * ;l: • • w * CD (J 3 8 ♦ o * o Sp 4- i s i si ®UJ i < x ' <§>■ A a a: *-8 A 8 L n«s if < - — \ > 7»3 v — < DC .O $ ^O @o • • • • cc ft UJ • - • • N N 5>- < CC n" :V # *, u. ?P • "I • .^H X) ■ ~ o «! _ +-• rt G tu •V xj +j C o n$ ro „.° u a>vo n 4> <-M a! o o c« "^^ +J — VO c (U CD ^ bf) X o - / » 3 r" XI c 3 O U J-l aj X> u u hn u >, U Ih Uh i-. U o 1-1 ft O Ui tA in c c/i O 3 IT) 1h O o o o -d a? 'J, ri ni >. m C 4-> rrt If, 1; E -d ^ ■M (Ti a; 0) n J3 o tjj: a; tfi a o • iii ox -C CH (U bt ^ n £ ^ 2 o 80 WATER FLOODING OF OIL SANDS ROBINSON SAND FLOODS Twelve examples of typical Robinson sand accidental floods are shown in plan on figure 65. The Kraft example is important because although it started as an accidental flood it was continued as a purposeful flood by using first one and later two wells as inputs for brine. It was the first purposeful flood in Illinois and was successful in increasing oil production on a commercial scale. The typical way in which water enters Robinson producing sands from upper or lower water sands is shown in figure 66. As there are no analyses of sand cores from the neighborhood of any of these accidental floods, and as the amount of water entering the sand is not known, there is no way of determining the exact perme- ability of the sand. An approximation may be made in the Crawford County area based on the sand's permeability to air in many repressured areas in the vicinity of the acci- dental floods. For this purpose a map (fig. 67) has been made showing repressured and accidentally flooded areas. The amount of gas delivered and the pressures required for each part of the territory have been published. 2 KRAFT FLOOD The 40 acres which is the SW. 34 °f the NE. y 4 of sec. 1, T. 7 N., R. 14 W., Licking Township, Crawford County, was leased in two parts by different lessees. One of the leases was on two acres in the northeast corner and the other was on the remaining 38 acres. Well No. 2 on the 2-acre lease was drilled very near the line so that offset wells were drilled nearby to the south and west. Kraft well No. 3 was abandoned due to faulty casing, the aban- donment being carelessly done so that the water from the shallow gravel bed in the glacial drift which covers the area was admitted to the producing sand and soon 2 Bell, A. H., and Squires, F., Preliminary summary of results obtained from a survey of repressuring opera- tions in the southeastern Illinois oilfield: Illinois Geol- Survey 111. Pet. No. 23, 1932. travelled through it, flooding out Rhine No. 9 and later Rhine No. 8, without in- creasing oil production from either well. That the water came from the glacial gravel was proved by the fact that a new well encountered a vacuum in the gravel. This can be explained only by the loss of water from the gravel to the oil-producing sand, thus reducing the pressure on the gravel below atmospheric. In 1922 the oil production of the Kraft lease showed its first increase which was found to be from Kraft No. 8 pumping well. After a short decline a considerable in- crease in production was obtained beginning in 1924, mounting until the end of 1926, and holding above the original unflooded production until after 1930. The wells were reached by water from Kraft No. 3 in the following order: Rhine No. 9, Rhine No. 8, Kraft No. 2 two acres, Kraft No. 8, and Kraft No. 1 two acres, Kraft No. 1, and last, Kraft No. 10 and Kraft No. 5. The spread of water is shown by contours on figure 73. After water reached Rhine No. 8 Mr. Paul Torrey was consulted and on his recom- mendation this well was used as a water input well, using all the salt water pumped from the Kraft and Rhine producing wells. Later Kraft No. 8 was added as a water input well, using 100 barrels of salt water per day. Water from the gravel bed con- tinued to reach the oil sand from the abandoned well during the entire time. The flood was commercially successful. Over a period of nine years after the flood began the Kraft farm produced 38,500 bar- rels of flood oil. About seven acres was flooded, producing approximately 5500 bar- rels to the acre, without increasing operat- ing expenses. This is the first instance of intentional surface flooding in Illinois. The operators were the Remlik Oil Co., members of which, in association with the Dinsmoors, introduced gas repressuring in Illinois on the Mumford farm. On this farm was also tried conjoint use of air and water to in- crease oil production. ACCIDENTAL FLOODS 81 FARM - KRAFT & RHINE SEC. I ,T.7N. R.I4W. KRAFT 9 FLOODED LOWER SALT SAND AT BASE OF POTTSVILLE- SECTION TAKEN ON LINE A-A' ROBINSON SAND FLOOD Illinois State Geological Survey Fig. 66. — A cross-section showing how water first entered the Robinson sand from the gravel on the Kraft lease. In a general way it illustrates what happens in most Robinson sand floods, namely the release of salt water from upper sands through holes in the casing, allowing it to enter and traverse the Robinson sand moving oil ahead of it to adjoining pumping wells. On the Titsworth and Newlin farms, water was purposely released from the lower water sand from which it rose to a head high above the Robinson sand which it entered as upper water would do. 82 WATER FLOODING OF OIL SANDS ILLINOIS STATC G£OLOCICAl iUIXfEY ACCIDENTAL FLOODS 83 Fig. 67. — Map showing the oil fields and locations of both accidental floods and gas repressured areas in Crawford County. The pressures required to force given vol- umes of gas into known thicknesses of sand depend upon the permeability of sand in the neighborhood of the input well. The variations in permeability indicated by the differences in required input pressure should be considered in connection with plans for water-flooding. For data on in- put pressures, see Illinois Geol. Survey Illinois Petroleum No. 23. 84 WATER FLOODING OF OIL SANDS TITSWORTH AND NEWLIN FLOOD The Titsworth and Newlin flood, sec. 6, T. 7 N., R. 13 W., Crawford County, is valuable as a demonstration of a second source of water for flooding the Robinson sands. Titsworth well No. 5 was cased on top of the Robinson sand and drilled into water in the "Salt sand." Water stood in this well at 578 feet above the bottom of the Robinson sand. Water invaded the prop- erty to the point where it stood at 478 feet above the bottom of the Robinson sand in Newlin well No. 2, a drop of only 102 feet from the source well. Due to the death of one of the owners this test was discontinued before oil pro- duction had a chance to increase. BEN FOLCK FLOOD A spectacular flood occurred on the Ben Folck lease in sees. 14, 15, T. 6 N., R. 12 W., Crawford County (figs. 6, 65). Three wells were drilled and lost on the Highsmith lease, one of which was an offset to Ben Folck well No. 4, which made a large increase in oil. Number 3, a location directly south of No. 4, was also benefited. The farm is drilled in a single line of five wells along the west edge, and the three southern wells are ordinary pumpers and have not been reached by water. Mr. O'Hara, who drilled the wells that did the flooding, reported an extremely high water-level in the 600-foot sand in this lo- cality. Wells to this sand east of the Ben Folck lease flow water as soon as the drill penetrates the sand. HARBISON FLOOD The Harbison, sees. 3, 4, 10, T. 6 N., R. 13 W., Crawford County, is one of the best known of the Robinson sand accidental floods. The increase in production comes from one well. The farm northwest of this, the D. Davidson owned by Fitzgibbon, has been drilled to the salt sand, which gave some oil production with a great deal of water under pressure. The flood may be caused by this water or water from an abandoned well in the Robinson sand on the Dee farm directly west of Harbison No. 5. Mr. Yargar, Ohio Oil Company super- intendent, made the following statement with reference to the Harbison flood. His production figures do not compare exactly with pipe-line runs, but they show that the man in charge attributed the increase in production to water. "June 21, 1932. Pro- duction on the D. T. Harbison farm was running along at about 30 barrels per week in September of 1928, and in October of 1928 a water-drive started on well No. 5 and production started up September 22, 1928. The production was 40 barrels for the week and from that time on it came up a little each week to the 11th of May, 1929. It was 90 barrels for the week and then ran along around 80-72-60 barrels per week until September 1930. Got to 53 barrels per week and we put a vacuum on No. 5 well but did not raise the production but kept it to and around 50 barrels per week to 55 up until the present time. It is now around 45 to 50 barrels per week." B. Yargar. MEFFORD FLOOD Mr. Harry Ferriman noticed a flood in- crease from his Robinson town-lot wells. The Mefford flood (sec. 28, T. 7 N., R. 12 W., Crawford County near the Ferri- man wells made a large increase as shown by the production graph. LOWRANCE AND FUNK C. C. BAKER The territory around Hebron (sec. 21, T. 6 N., R. 12 W., Crawford County) shows the most uniformly good results from accidental flooding. Repressuring near these floods has demonstrated that the sand has less than average permeability since a com- pressor requires 125 psi to force 30,000 cu. ft. of gas into 20 feet of sand, about the average of the Kraft and higher than the Robinson sand average for a similar vol- ume. Three Hebron floods have shown a high recovery of oil, Folck 2620 barrels in 3 years, Lowrance and Funk 5060 bar- rels in 15 years, and C. C. Baker a large per acre recovery in 15 years, which is the best and most consistent record of any group of neighboring accidental floods in Illinois. Second to this is the area around the town of Oblong where a great many floods have occurred. The sand is more permeable than it is in the Hebron district, as shown by its lower resistance to gas entry on Oblong district gas-repressured properties. ACCIDENTAL FLOODS 85 A. B. COFFMAN FLOOD A. B. Coffman farm in Pike County, Indiana, near Petersburg made a striking increase in production following the aban- donment of the Rufus Morgan well (fig. 72). The sand is correlated with the Illi- nois Buchanan. The increase was due to water flooding. L. C. STEWART FLOOD In sec. 30, Pike County, Indiana, half a mile due south of the town of Bowman, L. C. Stewart well No. 2 started to increase production in 1923 due to water and in 1933 made the largest production of any year since the lease was drilled (fig. 72). The Stewart well No. 2 does not produce any water but its offset well, Fowler No. 3, pumps 144 barrels a day. The sand is the Oakland City, equivalent to the Illinois Kirkwood, and is reached at 1266-1271 feet. THOMAS, GRAY, AND CHEW FLOOD These Indiana leases show the highest fluid level yet encountered. The water is most abundant on the northwest, Gray Nos. 1 and 2, Thomas No. 13, and Chew No. 5 pumping 24 hours with a 40-inch stroke, 17 per minute. When shut down for some time, the first three flow from the tubing, from a depth of 1100 feet. Pumping times range from 24 hours on the north to 5 hours for the most southern wells. The sand is the Oakland City. The In- diana sands are especially productive under flooding conditions and there is high water pressure in nearly every pool studied. BIEHL SAND FLOODS AT ALLENDALE Significant in the old Illinois field were the accidental floods on the Biehl sand at Allendale. The map (fig. 68) shows the productive areas and the locations of the floods. The discovery well was completed in 1912, and by 1926, some 200 wells had been drilled, followed since by as many more. Of the total, 300 are still pumping. These wells have an average age of four- teen years and an average production of 2500 barrels to the acre. The average initial production of 70 barrels a day has declined to two barrels. The average thickness of the sand is 29 feet. The wells make very little water. The sand is almost gasless and is gas-pumped to increase the fuel supply which even then is inadequate so that in most cases oil engines are required for power. Repressuring has been tried on two properties with good results. At Allendale when water breaks through the pipe which cases off the water sand, as often happens, it rises in the well to an average height of 1200 feet and floods the Biehl sand under a pressure of nearly 600 pounds. The direction of the water movement from well to well is shown on plans and cross-sections (figs. 66, 69, and 71) and demonstrates that permeability, not struc- ture, is the controlling factor. The Biehl sand varies in thickness, but the flood often takes the direction of thinning rather than thickening sand. Many floods go faster up structure than down. The average rate of travel is one well location in three months, about the same as observed in the McClosky sand at the beginning of water invasion. To date, no flooding well has influenced the oil production of more than a single pumping well, except on the Jake Smith and Alice Biehl leases. There a deliberate flooding program has been adopted and sev- eral flooding wells are being used to move oil to each producing well from several di- rections simultaneously. This is described in the section on "Applied Floods." Figure 70 is a set of graphs showing the oil production resulting from six floods. They illustrate four conditions: (1) flood production increasing ; (2) flood production past its peak and declining because the pumping well is making an increasing amount of water; (3) flood production de- clining because the pumping well has been abandoned; and (4) flood production de- clining because flooding well has been re- paired. These graphs show productions totalling more than 58,000 barrels, with three pro- duction curves going up and only one back to normal, but even under these conditions, there is an increase of more than 6,000 bar- rels for each of the nine wells. If the whole field of 300 wells were flooded, with al- ternate flooding and pumping wells, and the same rate of increase held good, the field would produce a million barrels of flood oil. This does not take into consider- ation the oil recovered by water-drives which would operate on every pumping well from its other three sides. This is usually estimated at three times the results from a one-sided flood. 86 WATER FLOODING OF OIL SANDS ACCIDENTAL FLOODS 87 WRIGHT— +-LEIGHTY FARM-DELLA WRIGHT SEC.8.T.IN. R.I2W. WRIGHT I UNFLOODED WRIGHT 3 WRIGHT 5 WRIGHT 7 FLOODED FLOODING ABANDONED x SURFACE— r SECTION TAKEN ON LINE DO' TYPICAL BIEHL SAND FLOODS Illinois State Geological Survey Fig. 69. — Plan and cross-section showing the flooding and flooded wells, and the watered areas in the Delia Wright accidental flood at Allendale. WATER FLOODING OF OIL SANDS In the flooded-out well shown on the graph, the oil came down-dip to the pump- ing well faster than the pumping equipment could handle it. Provision should always be made to measure and control the amount of flooding water admitted to the sand. Like the McClosky, the Biehl sand is so permeable that new drilling is unnecessary. This permits the most profitable of all flood- ing programs. Water is available from the nearby Wa- bash River and in parts of the field from inexhaustible water-bearing gravel beds. TRACEY SAND FLOODS The H. C. Johnson lease, sec. 25, T. 5 N., R. 13 W., Lawrence County, experi- enced an accidental flood originating from a defective w 7 ell over the line on the lease adjoining it on the east. One well was affected, and an area of 2j/ 2 acres was flooded. The increase in oil production be- gan in 1933 and in four years had amounted to 1250 barrels of flood oil. The area and conditions are shown in the plan (fig. 71) with a plan and cross- section through the flooding and flooded wells. The extent of the Tracey sand produc- tion in Lawrence County is delineated on figure 19. Applied floods in the Tracey sand would probably be profitable. KIRKWOOD SAND FLOODS The Oscar Smith and T. C. Combs farms, sec. 20, T. 4 N., R. 12 W., Law- rence County, experienced an impressive accidental water flood. The location is shown on figure 71 together with a cross- section through the Combs well. The flood involved three pumping wells, covered an estimated ten acres, lasted for seven years on the Smith farm and eleven years on the Combs farm, and increased normal produc- tion by 36,689 and 60,606 barrels, respec- tively, which is a recovery of 7,338 and 12,121 barrels per acre for the whole flood period and 1,048 and 1,102 barrels per acre for each year the property was flooded. The Kirkwood sand has a wide extent in Lawrence County, as shown on figure 19. Production from the sand has been prolific and long-lasting. Conclusion Figure 72 shows graphs of the production of wells from accidental floods. Table 3 shows that accidental floods have added more than 400,000 barrels to Illinois oil production. They occur on the Robinson, Tracey, Kirkwood, and Biehl sands, the last two showing the best results. It is noteworthy that none of the sands have had a drilling pattern with closer spacing than one well to 4^2 acres which would indicate that applied floods in Illinois may be de- signed for spacing wider than average at a consequent decrease in cost. The fact that Clark County sands have not produced accidental floods is due to the fact that no upper water is available, and it does not condemn them for purposeful floods. The data shown in table 3 are good evi- dence that other areas of the same sands will produce under applied floods. There is no difference except intent between an accidental and an applied flood. Use of gas and liquid conjointly, which has given favorable results in certain areas at Bradford, may find a place in future Illinois practice. ACCIDENTAL FLOODS 89 5000 4 500 4000 3500 Fl .OOD G UCTIOh FLO od beg A 1 NORM AL - **- PROD JCTION^" ESTIM UTED NORM> \L PRC 3DUCTI ON-S~ | , 3500 I 3000 o ^ 2500 i/> _J £ 2000 a. CO — 1500 z o £ 1000 Q o £ 500 3500 / ~*-FLOi 1DING 'PAIR 01 stoppe ' FLOOD o ay WELL / \ FLOOC (EST BEGAA MATED <) 1 / V -FLOO PRO DUCTI DN / \ / \ ~~\ ?~ ESTIM ATED NORMAL PRC EDUCTION ^ 1 FLOOD K / / y FLOOD BEGAN FLOOD PRODUCT IO / NORMAL PRODUCTION ESTIMATED 1 NORMAL PRODUCTION''^ 1 1 1 FL .OOD B FLOO D PRO DUCTIC )N W 1 FLOOD BEGA / V p UMPING STOPP El 1 WELL PULLED OUT 1 ' / *— , ~^: E " 3TIM ATED NORM-t L PRODUCTIC )N Fl .OOD J i -FLOO D PRO DUCTIC )N ^ NORM PRODI. AL CTION > "5 vo rj- i-h co co t^ CM CM co lo •**■ t-h co CM CM CM vO iO co i-i t-i ,-h io i-h vo i— i O Ui J-i »— T 1— I i—H i— I i-4 CT) CM COO ON CO lOTt VO »— " O O co ininfo co"cm~i-<~ cm'io" O " t) ioloio 1010 loioio io ^'-Q io • u ~§ r^ CM CM CM to CM* CM iO CM CM CM* O CM CM K CM tJ- io CM lOi^ui lO CM J>* CM u-j eg P 3 faK ■"• O Si CvCMOCMCMt^cocoCMvooOOrJ-vooCMiOvo^H S?—- "g ">> ^ OrHCvjK0000NN00'i-'HiO0\VDinv0V00\O\ OnOn CM CM*" ^ 3 C ^ CviTt-iO^iors^NONOOrHOio^VOVOKu)^ g^g ^ vq^OO CMy^io CMJ» O^C» co^CMjO^vo ^ u l'~l, rN L ( ^V i ^ — « o \n u cn~cm~lo~^~cm~°o~cm~cm~i^^oo"oo~cm~cm~^~^ LLj J-i CD CM r-i ,-h ,-h t— I i— I CO Oh ft CM »-H-* CO CMt>^ O ON co ip 00 o CM vo CO i-4 vo"©" coco ooooo ,-tcoOOO oo^o^o^oo^vq co~CM"co~CM*'t>.* OCKONN t>.iO co voco p CO Q OJ O ^ ^ l-lrHCMrHiHCMrHrHr-lCMrHCVlfO'-lrHi-liH ,-H O T3 a £ So OfOOvOrt-M^OOOOOOOMDNOOONN-H rrt OM-iOHHNimoa>OO^OiOOOvOKiOiO „, O CD ^^^^^CM^OO^O\OO^co^CM^O^vo^co^^^r^ON rt •^ .2 H ocT cm*~u^^io" , oo" n cm*~u^^^oo"vo"cm"^co" , co"'o'^vo"' ^"B^N ^h ^ _ _ CM i-< ,-h CO I 'O" +H m U X - 0\OrHV0rcOrt-Tl-N)Ov0V0^C000 l O' ! tCviN <-> cp O 3 CM i— • i— it-hi— i i—i OiOovO Tf CM co co CM CM CD r^J CO r-4 r~ °-2 CM 1-hi-h CM CMCM CM CM CMCM^H CM '-h CM CM CM cOr-nCM c -1 Mi CNOvOvOnOnOvOvonOvOsONONOvOnonOvOvONOv O ONVO iO 000 CM vovo t-4 vo"~o" CO vo u->vp CM CM OS ON iHfOrHCV)r-l ONOtN COtN. O OMO K o r y\ o\y\ cm^ tvTW co' cm" co" CMt>* CM CM t-h CM «-< vovo vovo VO COCO CO CO *0 LO 10 10 10*0 co CO co co <0 On Ov On on On O i-n CD On .en 3 oJ o o-g CD Cti .3 O ^2 .3 3 u O OJO en .3 bJO bC 3 en cu cd ^ 3 g . ' .. - rt 3 ^ ^ 00 -T -T ^ ri ^O -H CM t^ ^ CM J. J. -5 co - | r-v. '-'"Sb ^cos:"xj ■•'O'O £25 ^ H IS 1 ^8^ o ^ o 13 "Iro — -I co' J^fe N NIO VO 1 . ^CM ^^.^> ■ ^ CO CM ^ ^ * CM 3 ^ N „ d cl> CM c^ a .St .-3 ^^ 5^33 cd v; ^«cm ^ 3^ ^ U LO en _, • 3 rt cd .^ CD iHMCO^lOVONOOCKOrHCNjeO^-inVONOOa o ^-h cm ffi rh CQ f_ >bJ CM CMCM • J " 1 '— '•"'-' H-l 92 WATER FLOODING OF OIL SANDS — - m 1 m -/- 5 * =T^ < 1— „, .^ _Lo |Z o / /< 2 / « ./ ^^ /~o: - V a o ,/* 2 J a / / / $ / / \ < g v u 2 / ^ §~s *^- ^« / \ / V i / 8 s I 2 \. / V / a ^ 2 \ / / 1 / J / a " .> / _ -V- 7^" _ / a g S w - 3 u. ;; 8 i 2 ! \ \ \ \ ! | 3 1 8 3 «V3A b3d ST3«UV9 ^3A b3d SH3«bva ACCIDENTAL FLOODS 93 d3d s"i"3aavg av3A «3d sisaava Fig. 72. — Graphs of production influenced by accidental floods. They represent increases much greater than have been obtained by air and gas repressuring in their respective localities (fig. 19) and point to flooding as being the most promising secondary method yet devised for Illinois sands. (See fig. 6 and table 3.) 94 WATER FLOODING OF OIL SANDS Illinois State Geological Survey Fig. 73. — Plan of the Kraft and Rhine wells. This drawing shows the property on which the first intentional water flood in Illinois was developed. Here also was tried for the first time conjoint use of air and water flooding. A vacuum found on the gravel bed proved it to be the source of the water producing the accidental flood. APPLIED FLOODS IN ILLINOIS Purposeful water flooding has long been successful at Bradford, Pennsylvania, and Bolivar, New York. More recently, the shallow sands of Oklahoma and Kansas have been flooded to increase production. The Berea sand in Ohio has been worked with good results and considerable work has been done in Texas. A flooding opera- tion is underway in West Virginia. Illinois has presented evidence, by the results of edgewater encroachments and ac- cidental floods, that the process will be successful here. To date the only applications of these lessons in Illinois have been made in two operations on the Robinson sand, one on the Cypress sand, one on the Bethel sand, one on the Biehl sand, and two on the Gas and Siggins sands. Water has been introduced from the sur- face under controlled conditions into the Robinson sand in Crawford County on the Kraft and Drake-Stifle leases, into the Cypress sand in the Carlyle pool in Clinton County, and into the Benoist sand at San- doval. The Biehl sand in the Allendale field in Wabash County has been purposely flooded from below the surface with salt water from upper sands. Two important operations are under way in the shallow areas in the Parker and Siggins pools under the direction of the Forest Oil Corporation, experts in this method of production. Flood on the Kraft Lease The property is 40 acres in the SW. %. of the NE. y 4 sec. 1, T. 7 N., R. 14 W., Licking Township, Crawford County. The wells were reached by fresh water from the gravel through Kraft No. 3 in the following order: Rhine No. 9, Rhine No. 8, Kraft No. 2 (two acres), Kraft No. 8, and Kraft No. 1 (two acres), Kraft No. 1, and last, Kraft No. 10 and Kraft No. 5. After water reached Rhine No. 8 in 1924 this well was used as a water input well, APPLIED FLOODS 95 the input water being the total water pumped from the Kraft and Rhine pro- ducing wells. Later Kraft No. 8 was added as a water input well, also using water pumped from the oil wells on this and the adjoining lease. This is illustrated in figure 73. This use of water as a flooding medium under controlled surface injection was the first attempt to produce oil in Illinois by intentional water flooding. Over a period of nine years after the flood began, the Kraft farm produced 38,500 barrels. The seven acres flooded produced 5500 barrels to the acre. Operat- ing costs were not increased and the process was profitable. Drake-Stifle Controlled water flood On June 8, 1933, flooding of Illinois oil sands was legalized by an act of the legis- lature. Within three months, water was introduced into Drake No. 4 well in sec. 12, T. 7 N., R. 14 W., Oblong Township, Crawford County, on a lease owned by the Tidewater Oil Company. This area was selected because it was near successful accidental floods on the Livingston and J. McKnight farms and because water was available from a shallow fresh water well in the glacial drift. The extent of the flood as of May 6, 1935, is shown in figure 74. Water is pumped into Drake No. 4 at the rate of 85 barrels a day under 45 pounds pressure plus a static head of 1000 feet. In the field it was found that Stifle No. 5, previously a "head" well, began pump- ing "straight time" and that Drake well No. 6 was pumping longer than usual. Each was making flood water and oil. No other well showed any effect of flood. With only two wells affected, flood pro- duction has already exceeded the normal production of the entire lease for the same time, and the present daily average flood production is about four times the daily average normal production. Operating ex- penses have not been increased. If the same amount of water is introduced daily through one well into an oil sand, the pressure will increase (due to greater re- sistance as the water advances farther into the sand from the input well) until relief is obtained by removing some of the water through pumping wells. If the pressure does not increase or if it declines, as it did in this case, it can be explained only by leakage. If the packer leaks, the water will normally overflow the input well, and the water will show at the surface. But here, part of the water, leaking past the packer, moved up the hole and entered an upper sand, far more permeable than the lower, which drained off all the excess water under less head above it than the distance to the sur- face. Its travel through the upper sand was discovered at a nearby shallow well. More than 50,000 barrels of water was pumped into the input well at an average pressure on the sand of more than 400 pounds and a volume of about 80 barrels a day. The Drake-Stifle flood demonstrates that the process pays (fig. 75). It is the cheapest possible flood because it requires no new wells. On the other hand, it is slow and will take altogether too long to influence the next row of wells. It has been proved by the Drake-Stifle flood that 400 feet from water well to oil well in sands of equivalent permeabilities is not excessive. It is sug- gested that, for some Illinois conditions, flooding alternate wells in each direction in the customary drilled pattern will be ade- quate. This would result in a series of contiguous five-spots, each pumping well surrounded by four input wells and would multiply, by the number of input wells, the speed of the Drake-Stifle circle method. It would require no new drilling which is the principal expense in most water-flooding op- erations. 96 WATER FLOODING OF OIL SANDS R.I4W. ® .a ' LEGEND ( OIL WELL DRY HOLE INPUT WELL CONTOUR ON TOP OF ROBINSON SAND', DATUM SEA -LEVEL APPROXIMATE BOUNDARY OF WATER -FLOOD AS OF MAY 19 35 10 II Illinois State Geological Survey Fig. 74.— Plan of the Drake-Stifle flood. Water comes from a well in the drift and is introduced in a closed system to the Robinson sand. Dotted line shows the spread of water in the Robinson sand. APPLIED FLOODS 97 OIL PRODUCTION FOR 18 MONTHS TOTAL £2 UNDER NORMAL DECLINE K53 DUE TO FLOOD 7932 BARRELS 3715 BARRELS 4217 BARRELS < I a. z a- z> . o QUARTER YEAR A 4 J / V » r 1 4 2 3 4 QUARTER YEAR ^%* mp iff P « ;M^ « iff 9 w & Up §p ip 4 1 2 3 4 i 1934 1935 Illinois State Geological Survey Fig. 75. — Graps showing oil production, pump pressures, and water input for the Drake-Stifle flood. 98 WATER FLOODING OF OIL SANDS Carlyle Purposeful Flood The Carlyle oil field, sees. 2, 3, 10, 11, T. 2 N., R. 3 W., Clinton County, was discovered in 1911, and has produced more than three million barrels of oil. It had a productive area of 915 acres, and an average per acre production of 3400 barrels at the end of 1931. The field was rapidly drilled up during the early years of its life. The producing horizon, known as the Carlyle sand, is the Cypress formation of the Chester series, and occurs at depths of approximately 1000 feet. It is the equivalent of the Kirkwood sand in Lawrence County and the Oakland City sand in Indiana. = In general, the sand appears to be slightly domed in the central part of the field and to dip in all directions away from the cen- tral area. A number of well logs report water in the oil sand below the oil, particularly near the southeastern edge of the field. The average initial production of 50 barrels per well has declined to one. The average sand thickness is 17 feet. The developed field has 190 wells, of which about half are still pumping. Experimental air repressur- ing has been tried. The wells were gas- pumped to increase the scanty gas supply, oil engines replacing gas engines for power. An attempt was made to flood the entire field to increase oil production. A half million barrels of water were pumped in, under pressures up to five hundred pounds, plus a static head of a thousand feet. The greatest input was about a thousand barrels a day. The project was abandoned in favor of a return to repressuring with air which has shown no beneficial results to date. In figure 76 water distribution in the sand is proportional to the size of the circle around each input well. Each circle indi- cates the total amount pumped into that particular well. The sand, 17 feet thick with 15 per cent porosity, would have 2 # V2 feet of voids and each circle represents the area of the top of a cylinder, 2^2 feet high, which would accommodate the total water- input. The large circle indicates the top of a cylinder that would contain all the water pumped into the whole property, and the smallness of this circle, compared to the total area of sand, explains the lack of re- sults, because this operation has returned only 600 barrels of water per acre into a sand from which had been taken out 3400 barrels of oil. Sandoval Intentional Flood This field is a conspicuous example of successful edgewater encroachment, and it was at first believed that the natural flood supplied all the water required for the great- est possible recovery of oil. In the later life of this pool, Devonian production was obtained in the same area. Large oil wells were drilled which produced along with the oil such quantities of water as to establish a serious brine disposal problem. This was solved, in part, by introducing Devonian brine into the Bethel oil produc- ing sand which was already under a natural water drive. The unexpected result was a tripling of Bethel oil-production. Experts have, in the past, eliminated fields producing oil under natural water drives from consideration for intentional con- trolled surface flooding. Sandoval offers evidences that the opposite is apt to be right, in other words, that successful edgewater encroachment is evidence in favor of the possible success of applied floods in the same area. Such a deduction is especially perti- nent in view of the fact that all Illinois sands are under some degree of edgewater pressure. APPLIED FLOODS 99 • PRODUCING. OIL WELL •»- ABANDONED OIL WELL + DRV HOLE (•) FLOOD WELL; TOTAL WATER INPUT TO 1937 IN 3600 BARRELS; AREA OF CIRCLE IS PROPORTIONAL TO VOLUME OF WATER ■••'' *''* BOUNDARY OF PRODUCING AREA 0^ SCALE 'At '/t >A» 'A "ILE Illinois State Geological Survey Fig. 76. — This plan compares flooded area with total area of the Carlyle pool. (Squires, F., Paper presented at the Fifth Annual Illinois Mineral Industries Conference : Illinois Geol. Survey Cir. 23-B, fig. 19, p. 57, 1938.) 100 WATER FLOODING OF OIL SANDS MODERN PRACTICE APPLIED TO ILLINOIS The latest flooding operations in Illinois have taken advantage of the improved tech- nique developed in Pennsylvania and prac- ticed there and in Oklahoma and Kansas. These improvements are said to produce three times the oil obtained by accidental floods in comparable areas. Since Illinois has adopted these improvements a brief summary of the latest methods is here pre- sented. In addition, certain suggestions are made which may be applied to special Illinois con- ditions. Illinois flooding water resources are discussed. Bradford Technique Modern flooding patterns at Bradford have evolved by observation and practice from the accidental flood through the cir- cle, the line, the four-, five-, and seven-spot patterns. At present, properties are usually developed with wells in one of two patterns, the five- or seven-spot, with the five-spot the most often used. The seven-spot, since it is harder to fit into property lines, is best adapted to large farms. Controlled patterned flooding has tripled the oil recoveries over the amounts obtained from accidental uncontrolled operations. Crude oils are analyzed and studied in order to know their gravities, viscosities, and tendencies to emulsify. The sand for- mations are examined for cementing mate- rial, as it is known that silica cement is pref- erable to lime and iron oxide. Gas and water-bearing portions of the formations and zones of unusually high permeability are cemented off. Siggins and Parker Applied Floods Applied floods have been developed in the Siggins and Parker pools. Forty acres of Christman Brothers farm in sec. 13, T. 10 N., R. 10 E., Cumberland County, Illinois, and forty acres in sec. 21, T. 11 N., R. 14 W., Parker Township, Clark County, havt been drilled up and prepared for intensive flooding by the Forest Oil Corporation, Bradford, Pennsylvania. The depth of the producing sands in the field ranges from 465 to 600 feet in one case and from 230 to 350 feet in the other. The pattern chosen is, for both, the five-spot, with five input wells drilled on each line enclosing the Siggins tract, and three in the Parker, and the line of input wells extends in a row across the forty acres. In the center of each square, the corners of which are formed by input wells, a producing well has been drilled. This makes a pattern of 330 feet from water well to water well and the same dis- tance from producing oil well to producing oil well in the Siggins pool and 660 feet in the Parker pool. The Siggins forty con- tains 25 input wells and 16 oil producing wells and the Parker contains 9 water input and 4 oil wells. Both input and oil producing wells are completed in the same way. The well is drilled, using 8 % -inch pipe in the upper part and an 8-inch hole through the rest of the sand. The well is cased with 6^-inch casing where it is necessary to shut off cave below the 8-inch pipe. The well is then shot and cleaned out. Next, a string of 1^-inch pipe with a packer is lowered into the hole, the packer is set and cemented in, the 6%-inch pipe having previously been withdrawn. At the surface, the 1^-inch pipe is fitted with a reducer, a nipple, a valve, a nipple, and a tee, in the order named. Into the top of the tee is screwed a plug drilled for a pressure gauge, and into the side of the tee is placed a short nipple and all are connected to a section of pipe that in turn is connected to a meter, if it is an input well. The input wells are subjected to suffi- cient water pressure to cause the oil pro- ducing wells to flow. There is no brine disposal system since all brine is recycled, being readmitted in turn to the oldest part of the flooded territory. MODERN PRACTICE 101 Water for the projects is piped through a six-inch line to the operations from a drilled well in the valley of Hurricane Creek, 5^ miles away. At present, the main water pump is lo- cated near the water well so that the entire water line is under pressure. It is believed that the gravel formation which supplies the water filters it sufficiently that no other filtering will be necessary. The water is kept from the air by a closed system which it is hoped will preclude the necessity of any chemical treatment. Some of the wells are ready to flow, and the results on production are being awaited with great interest by all producers in the old fields as it is no longer profitable to operate much of this territory. They hope for profitable operation through some such secondary method as this Siggins pool de- velopment. Illinois State Geological Survey Report of Investigations No. 89 1943 DBRARY ENVIRONMENTAL PROTECTION AGENO& STATE OF ILLINOIS SPRINGFIELD, ILLINOIS