s 14. GS: cir /aef c. 3 Gud SucyuUJ 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 CIRCULAR No. 125 FLOOD TIDE IN ILLINOIS FREDERICK SQUIRES Reprinted from The Producers Monthly, July. 1946 PRINTED IiV AUTHORITY OF THE STATE OF [LXJNOIS URBANA, ILLINOIS 1946 'S STATE GEOLOGICAL SURVEY 3 3051 00004 6643 Flood Tide in Illinois By Frederick Squires :: Introduction This article gives the highlights on three applied water-floods now operat- ing in Illinois oil pools, points out earlier conditions which encouraged * Petroleum Engineer, Illinois State Geo- logical Survey, I'rhuiin, Illinois. Printed by permission or the Chief, Illinois State Geological Survey, I r- bana. Illinois. tlnforniation on this operation was taken from "Water Flooding of Oil Sands in Illinois," R. I. No. :tT7S, August 1044, Bureau of Mines, p. 1(!0. SHell. Alfred H. Further data on natural water-flooding; in Illinois oil fields. Papers on "Improved Methods of Ex- ploring for and Recovering Petroleum in Illinois. '• presented at the Second Annual Petroleum Conference, June 1. i.n. 111., pp. 24-25, figs. 10 and 11. their application, and suggests territory favorable for water-flooding. It proposes a method of flooding certain old terri- tory, without redrilling, and recom- mends flooding in the early life of cer- tain pools. Three Contrasting Floods The three areas being flooded are shown in figure 1. These three opera- tions are on sands of different ages and depths. They vary in the distances be- tween wells and in the kind of water used. The only similarity is their success. Highest in the geologic column is the Siggins flood, operated by the Forest Oil Corporation on a Pennsylvanian sand. Lower is the Patoka flood, conducted by the Pelmont Corporation and the Sohio Oil Company on the Bethel sand, and last is the McClosky flood, operated by the Pure Oil Company on their ex- tensive McClosky lime holdings. These three floods work on the state's pro- ductive sands of greatest areal extent. The Siggins Operation Under an agreement between the Ohio Oil Company and the Forest Oil Corporation, a flood was undertaken by the latter in the Siggins pool in Union Township, Cumberland County, and KEY MAP SHOWING LOCATIONS OF SIGGINS, PATOKA, AND M c CLOSKY FLOODS CONTOURS ON SIGGINS FIRST SAND, SIGGINS POOL Fig. 2. Well locution map of the Siggins pool. This slmiv* at Fig. .'!. Contours on Siggins first sand, Siggins pool. Square A In figure 1. near the center shows the location of the No. 2 flood. This Is at the top of the structure. (Bull. 54, PI. XXVII) Fig. 5. Section and plan of Kraft flood. This was the first applied flood in Illi- nois. It made use of salt water injected _ —, .,_ _. from the surface as a flooding medium. SIGGINS FLOODI It is the predecessor in Pennsylvanian ■id to Hi .«_.»„» AA i of the Slggins floods \o. 2 and No. 3. Note the complete redrilling and iig of the No. 2 flood compared with the wider spacing and the drilling of input wells only in the No. 3 flood. SAND THICKNESS DIAGI Fig. d oil production was best in the areas of highest initial production Casey Township, Clark County, on part of the wells shown in figure 2, con- toured on the top of the Siggins first sand in figure 3. The operation, called by them the No. 2 flood, covers 40 acres of the Chrystler farm, which oc- cupies the SW. Vi of the NE. Vi of Sec. 13, T. 10 N., R. 10 E. All the old wells were abandoned and plugged, and the 40 acres were redrilled in a pattern of five water wells on each boundary line filled in as a grid, making 25 water wells in all, with 16 producing wells drilled at the centers of squares formed by each set of four water-input wells (fig. 4). All the input and output wells were tubed with 1 Ms -inch cement- lined pipe on cemented packers for the purpose of flowing the oil-producing wells. The first 40 acres are not a commer- cial success because water flows too PERMEABILITIES IN THREE ADJACENT WELLS %\ 1 J m KEY PLAN 1 m | I I 11 r freely between the input and output wells, due either to overburden lifting from excessive early water pressures or from bypassing for the same reason. The greatest daily oil production from flood No. 2 was about 80 barrels, settling down to 40 barrels, accompanied by a volume of water that increased as the volume of oil diminished. Before the Siggins operation started, sufficient evidence was available to lead to the belief that Pennsylvanian sands could be successfully flooded. Many ac- Fig. !t. Permeabilities in three adjacent w Oil Corporation. These are representative They are from wells of highest initial ai average sand thickness. The lis; determinations a of the better parts 1 cumulative produc are high on the st id graphs by Fores* >f the Siggins pool, ion and better than ILLINOIS STATE GEOLOGICAL SURVEY CUMULATIVE NATURAL PRODUCTION PER ACRE BY FARMS SIGGINS POOL Fig. S. Cumulative natural production per acre by farms Siggins pool. High productions correspond more closely to high initial ructure than great sand thickness. The permeability profiles shown in figure 9 are taken in areas of high initial and cumulative production and greater than average sand thickness. The permeability profiles shown in figure CS ' ial and cumulative production. 43 ■ from wells of lo' cidental floods, in which water in a stratum above had broken through the casing and traversed the oil sand, had greatly increased oil production. One of them, the Kraft farm operation, which started as an accidental flood, was converted by the writer into a commercially successful applied flood by injecting salt water from the surface. The operation is shown in plan and section in figure 5. The Siggins operation was continued in flood No. 3, using a better method dictated by experience gained in the earlier operation. Two hundred acres north of the original 40 were developed with new input wells, drilled as five- spots at the centers of the squares formed by four old pumping wells. This procedure resulted in much wider spac- ing, water well to water well, and re- quired the drilling of only a fraction of the earlier number of wells. Water pres- sure was kept below the danger point of 300 pounds per square inch, and the wells were pumped. These changes spelled success. The Siggins first sand, as shown by the contour map, is an elongated dome having a maximum relief of about 140 feet. Structure, however, had no bearing on the method or success of the flood- ing operation. Sand thicknesses (fig. 6), when con- sidered along with oil content, had a greater influence. The map shows that Fis'. 11. Lahorn loii for Sintiins operation. This i Die Forest Oil Corporation office at Casey, I equipped and operations in it follow all drilling i Paul I'hillippi. the company's technician, is Fiji. 12. Booster pump for Sijifiii.s operation. Water is deli to the pump from the source well pump and raised by the desired input pressure. the south part of the No. 2 flood started in an area of relatively thin sand and that it and the No. 3 flood moved into areas of thicker and richer pay. Re- sponse to flooding closely followed these changing conditions. Figure 7 shows initial well productions in the Siggins area. In general, the best response to water- flooding corresponds to areas of greatest cumulative natural production (fig. 8). It is interesting to note that all this evidence agrees with findings made from core analyses and therefore would have been, even without core information, a reliable guide in the choice of the Sig- gins pool for flooding. Figure 9 shows a series of permeability profiles, made for the most part from examinations of sand chips. They show high permeabilities and considerable variation from top to bottom. The oper- ators use these profiles, also, as a guide in the location of packers. These high permeabilities prove that the extension of the well spacing for the No. 3 flood was logical. Figure 10 shows the above-ground in- take-well assembly. Oil content was de- termined in some cases by colorometric analyses. After selective shooting, caliper runs have been made to determine changes in well-bore profiles in order to find out to what extent the heavy shoot- ing of dense sections increased their circumference over unshot or lightly shot open sections and in order to equalize water inlet. The greatest di- ameter obtained was 30 inches. Figure 11 is a photograph of the corn- is. 13. Natural \ pool. This NATURAL WATER ENCROACHMENT AND DIRECTION, SANDOVAL POOL MAPS SHOWING NATURAL FLOOD ADVANCE AND ENRICHMENT OF DENNISON FLOOD Fig. 14. Maps showing nati d advance and enrichment of Deiinison flood. This territory in shown also at CC figure 1. advanced from the southwest toward the northeast as -liiiwn by the contours and enriched each farm in its path with umulated before. It demonstrates that pumping wells at the front edge of a flood ma] be by-passed by oil and is an argu- ment in favor of five-spotting. (111. Geol. Survey Kept. Inv. Xo. S», figure 27, p. 42, and figure 30, p. 45) pany laboratory, and figure 12 is a photograph of the booster plant. Water is obtained from a well in the glacial drift in the valley of Hurricane Creek, pumped six miles through 6-inch pipe and boosted at the property by this pump to the desired input pressure. Un- treated and unfiltered water has proved satisfactory because the system is en- tirely closed against oxidation, and the water is filtered by the gravel from which it is pumped. The Screening Test The success of the No. 3 flood will start a hunt for favorable producing strata on which to extend the process. In order to reduce the chances of fail- ure, a screening test must be applied to Pennsylvanian and all other produc- ing strata. Territory producing large quantities of water must be eliminated. Two operations in Illinois, one on sand which produced water from the day the first well was drilled and the other on sand which had become water-logged through badly abandoned wells, have been disappointing. A second rule is to select rich terri- tory. Richness can be checked by core analysis, by lease history, and by a com- bination of both. Figures 6, 7, and 8 give such a lease history for the Siggins field. A study of a new field, by applying the tests shown in these drawings for the Siggins pool, provides a good screening test when supplemented by a determina- tion of oil-water ratios from the wells under consideration. A permeability test, by electric logging old wells, will greatly help this screening test. Core analyses and oil production during flooding at Siggins have corroborated deductions made from preliminary screening. Parts of the North Johnson pool, the first sand at Bellair (the Sussanah Smith Farm for example) as well as the New Hebron area, large parts of Oblong and Martin Townships in Craw- ford County that produce from the Rob- inson sand, and parts of the old Allen- dale field in Wabash County would stand up under such screening tests. Many areas in the new fields, no doubt, will do the same. The Patoka Operationf The wells included in the Patoka flood are located in Sees. 20, 21, 28, and 29, T. 4 N, R. 1 E., Marion County, Illinois, near the town of Patoka, from which the pool takes its name. Test-flooding was done at the north end of the field, in Sec. 21, the success- ful results of which led to a campaign in 1944 to drill intake wells and to flood '& 'JMSCLOSKY FLOODS of McClosky floods umbered Oil this plan etail in figure B».l7 -----W-^ — i S M'CLOSKY FLOODS , Fig. 1«. Key plan of MeClosky floods south area. Pools numbered on this plan are shown in detail in figure test, ig the whole area. The sand is the Bethel which occurs here at a depth of about 1,400 feet. It is the same sand as that producing in the nearby Sandoval pool, which was flooded successfully by edge- water encroachment (fig. 13). The most striking feature of the operation is the long spacing of 660 feet from intake well to intake well and from producer to producer. This can be compared with the Siggins No. 2 flood, where the spac- ing was 330 feet, and with the Siggins No. 3 flood, where the spacing was 440 feet. The injection water is Tar Springs salt water, which is pumped from two wells, aerated, chemically treated and filtered. This procedure is in contrast to the Siggins operation, which uses un- treated and unfiltered fresh water. The Patoka pool has been a pioneer in several ways: It was one of the first structures in the state to be located by seismograph, rotary drilling found its first Illinois application there, and it was the first to use wide well spacing and salt water for flooding. The Bethel sand, on which this Pa- toka flood operates, is especially adapted to water injection. It has responded to water encroachment at Sandoval and St. Prancisville and to accidental flood- ing from upper water on the Johnson farm in northern Lawrence County. Such Bethel sand pools as Dix are ex- tremely promising. The (Cypress) Kirkwood wells of the old Lawrence County field produced from about the same depth as the Bethel sand at Patoka; they are spaced 9 wells to the 40-acres; the producing sand covers large areas with a thick uniform sand body; and the area has been unusually rich and free from water. The large initial productions of the wells indicate high sand permeability. All these factors favor flooding. The de- ductions from them are reinforced by the record production due to accidental flooding of wells on the Combs and Smith farms and by increased produc- tion due to water encroachment at Oakland City. McClosky Operation The third flooding process has taken place on McClosky lime in areas marked "Basin McClosky" on the general plan. It covers pools which are owned almost entirely by the Pure Oil Company, the block being so complete that it amounts to a unit operation with all the advan- tages of such a layout. Nature taught the lesson. The Mc- Closky natural water encroachments in Dennison and Petty Townships, Law- rence County (fig. 14), § showed that water moved large quantities of oil and that permeable horizons were continuous over wide areas. Flooding the McClosky was recom- mended by the Illinois Geological Sur- vey in 1937 in Circular 23, "Contribu- tions of the Fifth Annual Mineral Industries Conference," October 9, 1937, "Recent Developments in Water Flood- ing in Illinois Oil Fields," by Frederick Squires, page 123: "Flood characteristics of the McClosky are these: Water travels fast, oil pro- duction is always benefitted. Only part of the area has been flooded. Therefore DIAGRAMS SHOWING ADVANCES IN M c CL0SKY FLOODS IN THE DUNDAS CONSOLIDATED POOL CONTOURED AT 2-MONTH INTERVALS CAMPBELL DIAGRAMS SHOWING ADVANCES IN M c CL0SKY FLOODS IN THE CLAY CITY POOL CONTOURED AT 2-MONTH INTERVALS The eontour line represents the front of the the obvious deductions are: (1) That the remaining McClosky should be inten- tionally flooded; (2) that this can be done without new drilling, by using al- ternate wells for flooding and pumping; and (3) that recoveries, if at all like natural floods, will amount to millions." The Pure Oil Co. has started a series of floods shown in figures 15 and 16. They are using 32 old wells as input wells and produce from other old wells. The Basin McClosky was drilled in gen- eral with one well to 20 acres, although smaller areas had one well to 10 acres and still smaller tracts, drilled after Pearl Harbor, had one well to 40 acres. It is obvious from the map that almost all the 20-acre patterns have unequal distances, either 1,320 or 660 feet, be- tween adjoining wells. The 10-acre pat- tern is a uniform 660-foot spacing, and the 40-acre pattern a uniform 1,320-foot spacing. Water for the earliest injection in the southern end of the tract was obtained from the Pure's Little Wabash River pumping station. It was used without treatment and only a small amount of filtering. The McClosky lime is so open that nothing plugs it, making chemical treatment unnecessary. It should be re- marked here that the areas under flood had reached their economic limit of yield under primary recovery methods before resorting to this secondary proc- ess. In later installations, the casing in input wells was perforated opposite the water-bearing Cypress, which flooded the McClosky under natural head. Twenty-eight separate floods have af- fected more than 3,800 acres to a greater or lesser degree. Large volumes of water were used. The speed and direction of water ad- vance has varied as shown by the dated contours surrounding each input well (figs. 17 and 18). Invasion is extremely rapid. An oil production graph is given for two adjoining wells, shown here as fig- ure 19. They are the first and second locations from the water-input well. The second location is usually more produc- tive as is shown here. The total oil production obtained to date from the 3,800 acres affected indi- cates that the production from all the McClosky acreage shown on the two McClosky maps will be very great when subjected to flooding. Judging by the success already achieved in flooding the McClosky sand, a large proportion of the remaining ter- ritory will respond favorably. Such pools as Johnsonville seem certain to succeed. The method now in use may be im- proved after longer experience. Five- spot well spacing and line flooding may be tried and the results compared, in- sofar as possible, with the circle floods now in operation. Input water from the Cypress sand or other upper sands may be metered and regulated as conditions require. If and when new McClosky pools are found, the probability of later flooding should in- fluence the location of wells. Certainly neighboring wells should be equidistant. A glance at figures 17 and 18 shows that permeabilities, always high, vary wide- ly. A 40-acre spacing should work in such areas as 3 and 6 (on figs. 17 and 18). After developing on 40-acre spacing and noting the permeabilities encoun- tered, closer spacing would be possible if necessary. The next question to arise would be the time to flood. Reservoir pressure could be maintained by flooding as soon as the field was outlined. This would be a logical procedure. Summary The one thing that these three floods have in common is success (fig. 20). IOO 80 They vary in every other important re- spect — the sands, the flooding water, the age of the wells, the well spacing, and the number of new wells required (fig. 21). This is in striking contrast to flood- ing in Pennsylvania, New York, Kansas, and Oklahoma, where the operations have a much greater similarity. Experi- ence gained in these fields cannot be applied to all the problems in Illinois, where each sand is a law unto itself. The principal difference between Illi- nois sands and out-of-state sands is permeability. The average high permea- bilities for 50 wells each was 545 milli- darcies for the Cypress, 221 for the Bethel, and 1,858 for the McClosky. In- asmuch as degree of permeability deter- mines well spacing, Illinois sands with their greater permeability permit wider well spacing than do out-of-state sands, as is demonstrated in the spacing in the three fields just discussed. The Siggins No. 2 flood started with 330 feet between like wells (the Bradford technique); the < 6 1 t l, _ 7 J \ < 1 \ UJ CD '1 Q O O Jl -NO. 2^- L L I ' 1 ! i ' 3 j i 1 1 H — \ NO. l-> ^ «V i i i i p-i — * ^S ri- \ . \_ i i \ \ N ! \ /WWi 20 2 4 6 8 10 MONTHS GRAPH OF PRODUCTION . Graph of Pr ■ i McClosky flooded ell from the flood in, lolled by water first. g or at least count DIAGRAM SHOWING SOURCES OF FLOODING WATER SIGGINS PATOKA BASIN M BASIN M'CLOSKY DIAGRAM SHOWING RELATIVE DISTANCES WELL TO WELL ill i iMisiic iMhsii A 1r.11.11n1 nsitihinu Fig. 20. Total ill. McClosky floods. It includes the prodm months. All the properties are 'is-. 21. I)i:i(ir.'iiii showing sources of floodii rain showing relative distances well to w* atoka and McClosky floods. The notewort rnwiiii; is the difference in every ease. 1 ave demonstrated the probability of sncce three most widespread sands In Illinois. ispeet of this three floods flooding the Siggins No. 3 increased this amount to 440 feet; Patoka went to 660 feet; and the Basin McClosky expanded it to an average of 1,000 feet. By these figures it is demonstrated that the permeability of some oil sands in Illinois is so great that the minimum well spacing could be the same as (or greater than) the standard old field spacing of 400-440 feet. Therefore, such old fields can be flooded without re- drilling but by using old wells only. Large areas of the Cypress, Bethel, and McClosky sands of the old fields come into this category. When their permea- bilities would permit spacing wider than 440 feet between like wells, the only effect of using old wells is that their spacing, which is closer than is neces- sary, makes for a quicker clean up. The early belief was that flooding should be applied only in the old age of wells, like those in the Siggins sand. The wells in the Patoka and McClosky floods are comparatively young; yet to- gether they show even better results in flood production and have a far greater salvage value. The logical step appears to be to flood, or to conjointly repres- sure and flood most fields soon after drilling, and to flow the wells from start to finish. Engineering Problems Involved The petroleum engineer should be encouraged to reinvestigate, since a rule need not be true just because it is old. Bradford sands were once thought floodable only because they were tight. Now they are believed to be floodable in spite of their tightness. Illinois flooding presents many new problems, some of which are listed below. A meter should be provided for meas- uring and a valve for regulating the water which leaves an upper sand to flood a lower sand. Both devices should record above ground. A side-wall sampler should be de- signed to operate in shot holes in order to get permeability test samples from old wells and to aid in screening them for flooding. Permeability should also be deter- mined in the well itself. The weakest spot in any screening method is the lack of permeability data. Without it, there is no sure measure for well spac- ing. Without it, there is no sure way of telling whether old wells, without new drilling, may be flooded out within an economic time limit. Permeability information might be de- termined in old wells by measuring fluid injections as the sand bore is filled up, foot by foot, with a removable seal. It might be obtained also by meas- uring the intake rates of two liquids of different conductivity under equal pres- sure when the interface is located by an electric pilot. Permeability to liquids might be computable from the results of tests on permeabilities to gases enter- ing uniformly reduced sections of pro- ducing sand. The use of alkaline flooding water was mentioned in a patent 1,238,355 is- . sued to me many years ago. The method was tried only on sands, where it was a failure, probably due to its action on clay. Alkaline water gives astonishingly good sand-washing results on small specimens in the laboratory but dismal failures in the field whenever clay is present in the sand. It should be demon- strated by field test whether or not the McClosky lime, which is always free from clay, will permit more successful flooding with alkaline than with neu- tral water. A reliable selective plugging method is needed. Aquella, with its one-micron particle sizes, might be made to enter and plug sand pores that are too perme- able. Gas injected in a five-spot water- flood might provide a retarding effect by blocking off the zones that are too permeable, thereby checking too rapid flooding. Alternating pressure and re- lease of gas, effected through the output well of a water-flood five-spot, might check the advance of water through more permeable sections because of the speed differences of gas and liquid movement. Horizontal drilling might be used in new ways. Horizontal holes in unwork- able coal veins might be used as burn- ing chambers provided with electric ignition, steam, air and gas currents in which gas is produced by incomplete combustion of coal. This lean gas could be enriched by passing it through oil sands. In many places, coal and oil strata are layer-caked in the Illinois coal basin. The process of well cementing is open to improvement. Well casing might be formed with cement alone and without pipe. A mud sheath is formed every time a rotary well is drilled. Dr. Swann, inventor of the Swann under-reamer, made a good start on a method of form- ing a cement sheath. The introduction of heat and chem- icals, such as gaseous HC1 into oil bear- ing lime formations should be thorough- ly investigated. Flooding needs rustless pipe with threaded couplings. A prac- tical coupling must be devised. Almost every method for reservoir control depends on unification. It can be done for the farmer by a single clause in the lease. It needs to be sim- plified for the operator. An example of such forward-looking problems and their solutions is illus- trated in the electric logging permea- bility test made by the Survey on an old well in the Siggins No. 3 flood. Pig- COMPARISON OF PERMEABILITY BY GEOPHYSICAL AND CORE ANALYSIS AND ORIGINAL WELL LOG CORE ANALYSIS WELL K-O-7 CORE ANALYSIS WELL J-O-8 CORE ANALYSIS SECTION ON B-B' Fig. 23. Comparison of pcrineal>ilii \ l«> geophysical i well log. The in 1 <- 1 |> i < < .< I i < > n oiadc lij Dr. !!;<>-. from liis geophysical log 4 here checked 1>> core analyses from the four surrounding wells and hy the driller' log of the well itelf. II..- slight variations in such items as bottom of casing do no discredit the geophysical interpretation and (he siinilnrH.i of the driller's log witl it strengthens its accuracy. ure 22 shows the mobile laboratory in the background and the well logging equipment at the well. Figure 23 shows the geophysical log by C. A. Bays, and for comparison, permeabilities obtained from cores taken from neighboring wells and from the drilling log of the well itself. The owners were satisfied with the results of the test because they used this well as a water input, and it has proved satisfactory for this purpose. This method of electric logging of old wells for relative permeabilities will be of great help in choosing old wells for flood possibilities as it fills out the screening program in the only test which cannot be determined by lease histories. Conclusion These three floods, conducted under widely different methods on as widely different sands, have reached the com- mon goal of success. They cover only a small proportion of the area of the sands on which they operate and a far tinier fraction of the total area of the sands of Illinois. Each flood has proved to be a separate problem, for the solu- tion of which out-of-state operations present little precedent. A tremendous opportunity for expansion lies ahead of the petroleum engineer. Acknowledgments The writer takes this opportunity to acknowledge the help of M. M. Leigh- ton, Chief of the State Geological Sur- vey, who designated the research proj- ect, the help of A. H. Bell, who read the report critically, Carl A. Bays, who directed the electric logging, and M. W. Pullen, Jr., and Robert N. M. Urash, who assisted in the field work. Particu- larly important to this project has been the broad-minded cooperation of the com- panies whose work has been described.