SPESociety of Petroleum Engineers

SPE 20755

A Field Case Study of Replacement Well Analysis:Guymon-Hugoton Field, OklahomaD.J. Ebbs Jr., A.M. Works, and M.J. Fetkovich, Phillips Petroleum CO.

SPE Members

Copyright 1990, Society of Petroleum Engineers Inc.

This paper was prepared for presentation at the 65th Annual Technical Conference and Exhibition of the Society of Petroleum Engineers held in New Orleans, LA, September 23-26, 1990.

This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author{s). Contents of the paper,as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author{s). The material, as presented, does not necessarily reflectany position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Societyof Petroleum Engineers. Permission to copy is restricted to an abstract of not more than 300 words. Illustrations may not be copied. The abstract should contain conspicuous acknowledgmentof where and by whom the paper is presented. Write Publications Manager, SPE, P.O. Box 833836, Richardson, TX 75083-3836. Telex, 730989 SPEDAL.

SUMMARY

This study analyzed the initial pressure data andthe performance results of replacement wellsdrilled in 72 sections of the Guymon-Hugoton Field,Oklahoma. Reservoir lateral continuity wasconfirmed and the current 640 acre well spacing wasdetermined to be adequate for effective drainage ofthe field.

INTRODUCTION

A review of the replacement well results in theGuymon-Hugoton Field was prompted by hearings and alater decision by the Kansas Corporation Commissionto permit increased well density in the KansasHugoton Field. The greater Hugoton Field (Fig. 1)has historically had 640 acre well spacing inOklahoma, Texas, and Kansas. It is considered thatinitial responses from replacement wells would beanalogous to infill wells unless the distances fromthe replacement wells and the original wells are soshort that the wells are, for practical purposes,twins. All identifiable replacement wells werereviewed to determine the expected results ofinfill drilling if it were pursued in the Guymon­Hugoton Field. Infill wells are often presented asa method of increasing effective drainage andrecovery plus accelerating the production fromtight and/or heterogeneous reservoirs, particularlyfor oil reservoirs and especially those undergoingwaterflooding. There have been reported oil fieldinfill successes',2 and efforts to either evaluateor investigate potential candidates for infilldrilling. 3 ,4 There has been very little publishedabout the potential for infill drilling gasreservoirs, although it has been recognized thatsome very tight reservoirs could benefit byaccelerating the rate of production where wells arein transient flow for extremely long times. S

References and illustrations at end of paper.

387

The replacement well review started with data fromour own wells, but was then expanded to include allreplacement wells within the Guymon-Hugoton Field.By using publicly available data bases, themajority of the identified replacement wells couldbe reviewed using production, official shut-inpressures, and official deliverability tests plussome completion records from scout tickets andother sources. Basic information to be determinedfrom the study included:

1. Were replacement wells encountering anyadditional reserves not recoverable from theoriginal well?

2. Were the replacement wells of higher or betterdeliverability than the original wells and ifso, why?

3. Are there selected areas within the Guymon­Hugoton Field that are more applicable tosuccessful replacement wells than otherareas?

4. Can replacement well results be used topredict the outcome of an infill well and thecombined production capability?

Utilizing mostly publicly available data, thequestions above have been answered ratherconclusively and when supplemented with additionalstudies,6,7.8,9 the decision to pursue infilldrilling within the Guymon-Hugoton Field can berejected without concern for a possible missedopportunity. This, of course, does not precludeother infill drilling possibilities in gas fieldswhere characteristics and circumstances aredifferent. It should be noted that an infill well,as used in this paper, denotes an additional wellcompleted within the same layer(s) and intendeddrainage area as another well or pattern of wells.In a review of oil field infill drilling results,Gould & Sarem3 point out that without a goodreservoir description, the risks of a successfulinfill project are high and even when there is goodpotential, you must know the reason for the

2A FIELD CASE STUDY OF REPLACEMENT WELL ANALYSIS

GUYMON-HUGOTON FIELD, OKLAHOMA SPE 20755

BACKGROUND

GENERAL FIELD DESCRIPTION

replacement. Ten such sections were found withmultiple wells but were not included in this study.Approximately 80 per cent of the replacement wellswere located more than 1000 feet away from theoriginal producer and definitely would not beconsidered a twin well (Table 2). In fact, morethan 70 per cent of the replacement wells are morethan 1500 feet away from the original well with theaverage distance between the original well and thereplacement well being 1657 feet. This averagedisplacement is between half the distance from thecenter of a square section to a side (1320 feet)and half the distance from the center to a diagonalcorner (1867 feet). Because those distances wouldrepresent probable relative locations for an infillwell, then more than 70 per cent of the replacementwell locations can be classified as similar toexpected choices for any infill locations.Approximately 49 per cent of the replacement wellswere located easterly from the original well,either NE or SE with the NE having a slightmajority. A similar number were located westerly,but the NW selection was twice as frequent as theSW. This probably has no significance other thanthe relative position of the section with regard tothe major portion of the field and with an operatoreither trying to increase productivity or toincrease the distance away from an offset well.

Just as it was previously mentioned regarding someoriginal wells that never produced, there are alsosix replacement wells that never produced. Theyare included in the study to illustrate that somealternative locations were tried but were notsuccessful. Of the eighty-six replacement wells,eighteen were converted unsuccessful deep teststhat utilized the new wellbores to either replace atroublesome existing producer or to provideproduction from a section where there was no longeran active producer.

The Krider and the commingled Herington-Krider arethe principal producing zones for both original andreplacement wells. However, the Winfield isproductive in selective locations, particularly inthe north part of the field, but generally becomeswet toward the south. The Herington, which wascompleted by itself in one original well and tworeplacement wells, is not very productive. Also,the Fort Riley which was commingled with otherzones in three original wells and one replacementwell is not a very productive zone. The Chase wasused to identify completion zones which were notdefined otherwise by the available records.

Thirty-nine of the 72 original wells werecompletely cased and perforated. The remaLnLngwells produced from openhole zones or a combinationof openhole and perforated zones. Eighty-one ofthe 86 replacement wells had casing set completelythrough the intervals and perforated. The otherfive were openhole for one or more layers.

Wells were usually stimulated by either having thezones isolated and selectively treated or bytreating all of the zones at one time. Theoriginal wells were drilled during the period from1930 to 1974. All except three were completed by1955. Most of these wells were stimulated withacid ranging from 29 gals/ft to 1133 gals/ft withan average of 217 gals/ft. The replacement wellswere drilled from 1953 to 1987 with most of these

There are some circumstances,can improve the chances of a

well, but in general, thesecircumstances are not evident inField.

original behavior.which if identified,successful infillcharacteristics orthe Guymon-Hugoton

The Guymon-Hugoton Field is a gas field located inTexas County, Oklahoma (Fig. 1), which originallycontained an estimated 6.5 TSCF recoverablereserves and has produced approximately 5 TSCFthrough 1989. It is a portion of the largerHugoton Reservoir which extends into both Texas tothe south and Kansas to the north. The Guymon­Hugoton Field produces relatively dry gas(approximately 0.714 specific gravity) from theLower Permian Chase Group which is more fullydescribed by Siemers. 9 For the purpose of thisdiscussion, the important characteristics of theGuymon-Hugoton producing formations include thefollowing:

The Greater Hugoton Reservoir was first drilled in1922 in Kansas, and the first well in what is nowthe Guymon-Hugoton Field was the Home Developmentcompany, Allison No.1, drilled in 1923. Thegeneral development of the Guymon-Hugoton Field didnot take place until the late 1940's and early1950's. Wells completed in the Guymon-Hugoton aregenerally producing from depths between 2700 and3000 feet deep. As the ground elevation in thearea is more than 3000 feet above sea level, thereservoir is at, or above, sea level. Originalreservoir pressure was 475 psig. This valueconverts to approximately 445 psig at the wellheadand was used to evaluate whether any replacementwell encountered virgin or untapped reservoir gas.

1. interlayered carbonates2. layers of contrasting permeabilities3. no effective cross-flow between layers4. continuity of individual layers.

Eighty-six replacement wells were identified in theseventy-two sections specified above. There wereadditional sections in which more than one Hugotonwell was drilled, but if the original well neverproduced, it was not classified as a producer andtherefore the second well was not considered as a

Wells were replaced for a variety of reasons, butfor the most part, it was an effort to obtain abetter completion for increasing productivityand/or eliminating operational problems. A fewreplacement wells were drilled during the 1950'sand several more in the 1960's, but most of thereplacement wells have been drilled more recently(Table 1). Most of the seventy-two sections withreplacement wells (Fig. 2) had the original welllocated near the center of a 640 acre regularsection within the 'section, township, range' landdescription survey of Oklahoma. The field rulesfor the Guymon-Hugoton Field require that theacreage allocated to a well be contiguous. Onlyseven original wells were more than 500 feet fromthe center of the section and none were more than2000 feet from the center. The average distance ofthe original well from the center of the section,including all 72 wells, is 327 feet.

388

SPE 20755 D. J. EBBS, A. M. WORKS, and M. J. FETKOVICH 3

wells completed during the 1970's and 1980's. Theywere stimulated with 15% HCl acid or differenttypes of water based frac fluids with and withoutsand. The amount of acid ranged from 15 gals/ft to1000 gals/ft averaging 299 gals/ft. The waterbased frac fluids ranged from 111 ga1s/ft to 11,429gals/ft averaging 1587 gals/ft. The amount of sandranged from 214 Ibs/ft to 11,429 Ibs/ft averaging2009 Ibs/ft. Stimulation treatments of thereplacement wells were larger than the originalwell stimulations and more reflective of recentstimulation technology, however, many of theoriginal wells had excellent results whereas someof the more recent treatments were not aseffective. This may be partially a result of theprogressing differential depletion, particularly ifcommingled stimulation was attempted.

ANALYSIS OF REPLACEMENT WELL RESULTS

All of the wells cannot be shown, but selectedexamples will be described to illustrate variousresponses observed. During the discussion of eachof the examples presented, the section will beidentified by a reference number (RN) as shown onTable 3. Because of incomplete data in some cases,the explanation for a particular behavior may notbe specifically supported by the available wellinformation. However, other wells and calculatedresults from model studies8 ,10 have indicated theprobable cause of the effect observed. This willbe more obvious in specific examples and will bebrought to the readers' attention.

The Guymon-Hugoton Field has always been proratedin that the field delivery capability wasconsiderably greater than the nominated productionvolumes which defines the allowable. Because ofthis condition, the field has been producedessentially at a constant rate until the mid tolate 1970's, the beginning of the 'gas bubble',when production was restricted even further. Thisis important because the monthly rates are the mostreadily available data and are, in general, flatwith no character to analyze from a rate declineaspect. other data available from public sources,besides the monthly production, are the officialshut-in pressures and deliverability test results.It was found that on a few wells, even these datawere not easily obtained. In certain cases,completion and stimulation data are available fromscout tickets and/or state governmental agencies.Some production records, particularly before 1967,were not available either from regulatory agenciesor the commercially accessible databases. Analysisof the replacement wells was initially started withour own wells for which we had extensive data, butonly thirteen of the eighty-six identified fellinto that category. Our company records containsome limited amounts of data for other operatorsdue to our early position as both producer andgathering system operator. Even with thissupplemented data, there are still some gaps in theproduction and performance data plus many missingcompletion and stimulation details. For thisreview, it was not imperative that all of thisinformation be obtained, but it would have beenhelpful and would definitely be beneficial toanyone designing a replacement well or consideringan infill well.

389

One of the first questions to be answered was "Arethe replacement wells finding any new reserves?"Three plots were considered important to answerthat question for each of the sections containingreplacement wells. A material balance plot ofannual shut-in pressures (initially 72 hour shut­ins and beginning in 1975, 96 hour shut-ins) versuscumulative production was prepared for eachoriginal well. The pressures used were wellheadvalues as the reservoir is relatively lowpressured, shallow, and dry. This makes theextrapolation of wellhead pressure to reservoirconditions simply a function of average gas densityand unnecessary for this evaluation. Likewise,only pressure was plotted instead of theconventional P/z. As a continuation on this firstplot, the initial and subsequent wellhead pressuresfrom the replacement well(s) were plotted, withdifferent symbols, versus the section's totalcumulative production (Fig. 3). A semi-log plot ofthe section's average daily production versus timewas prepared using the monthly production values asreported to the State. A symbol was used toidentify replacement well production (Fig. 4). Thethird plot was backpressure data from thedeliverability test taken during official testing.Originally, a deliverability test was required eachyear, but current field rules only require flowtest every other year with shut-in pressures everyyear. Again, different symbols were used todistinguish between the wells and this plot wassuperimposed on the pressure vs. cumulativeproduction along with a section plat (Fig. 3).This section (RN70) is used to illustrate some ofthe concepts and it is not necessarily a typicalexample, although many are similar. There are manyvariations in the responses observed in thereplacement wells. Some of the apparently'abnormal' responses are quite explainable whensufficient information was obtainable. Forexample, some wells have water entry into thewellbore and the water restricts the gas flow plusit may cause lower than actual shut-in pressure tobe measured. If the replacement well has a bettercompletion, in regards to the water entry problems,the new well may exhibit a shut-in pressure higherthan the last pressure recorded for the originalwell and it may have indicated performance betterthan the last tests for the original well. Oneneeds to evaluate the entire performance history ofall of the wells on the section before making ajudgement. There are other conditions that cancause a higher reservoir pressure to be measured ina replacement well. In all replacement wells, notconsidered a twin well (one within a few hundredfeet of the original well), there will be aninitial reservoir pressure higher than the pressureat the old withdrawal point. This is due, ofcourse, to the pressure gradient established in thedrainage area while the well was producing. Thevalue of this delta pressure is a function of thedistance from the old withdrawal point, thereservoir rock, fluid properties, the rate of theoriginal producer before shut in, and how long theold producer had been shut in plus whether thereplacement well has been produced and for howlong. For practical purposes, the delta pressurevalues observed and calculated in the GreaterHugoton Field average about 12 to 15 psi. 6 ,7,8Another source of higher shut-in pressures is the

4A FIELD CASE STUDY OF REPLACEMENT WELL ANALYSIS

GUYMON-HUGOTON FIELD, OKLAHOMA SPE 20755

very nature of the pressure being measured. TheGuymon-Hugoton Field is composed of interlayered,no-crossflow carbonate formations with contrastingpermeabilities. 9 This combination sets up theconditions for differential depletion which resultsin pressures in the various layers not being equal.Differential depletion is not a new concept, but inthe context of such a large reservoir extent and ingas, it has not been seriously considered by manyuntil recently.l0 It is quite common for oil fieldwaterflood projects to be hampered by so called'thief zones' which are formations considerablymore depleted than other zones completed in thesame wellbores. The vertical communication betweenthese formations must be very restricted to set upthis condition. Unfortunately, the waterfloodcannot sweep the tighter, less depleted zone unlesssome way to isolate the 'thief zone' can be found.Almost invariably the zones are in communicationat, and only at the wellbores (sometimes behind thepipe). This can then result in very difficult, andoften expensive, workovers. The low permeabilitiesin these Guymon-Hugoton wells require stimulationwhich will often result in behind the pipecommunication of the producing intervals within thereservoir near the wellbore through the inducedvertical fracture. The importance of thissituation to the replacement well analysis is thatthe measured wellbore pressure, during bothproduction and shut in, will more nearly reflectthe more permeable layer pressure as described byFetkovich, et al. 10 This brings up the point aboutmeasuring a considerably higher pressure in areplacement well. There are a few wells in thereplacement well analysis that appear to havereservoir pressures in excess of fifty psi abovewhat would be estimated based on the original wellhistory. Although these higher pressure values aresignificantly lower than the field discoverypressure (indicating considerable depletion), theyrequire some additional investigation. Notsurprisingly, some wells exhibiting higher pressureare wells close to either edge of the field. Ofthe 72 sections involved in the study,approximately one-half could be classified as edgewells. The reservoir quality declines as eitherthe east or west field boundaries are approached9

and because of this poorer rock quality, theproducing capabilities in these areas are lower.This causes less depletion of even the morepermeable of the zones when compared to the levelof depletion toward the center of the field. Anexample of the possible response observable in somewells can be seen on Fig. 5 and Fig. 6.

There is another problem involved besides theflowing gradient in estimating what pressure toexpect in a replacement well. That is the timedelay between the last measured pressure on theoriginal well and the first pressure available fromthe replacement well. In this last example, thetime between the measurements is at most a year. Asimilar time interval applies to about fifty-sevenper cent of the replacement wells, but four hadintervals in excess of twenty years (Table 4). Anapproximation for this study has been used whichinvolves the average field pressure as plotted fromdata obtained from the Kansas ConservationCommission (KCC) (Fig. 7). This plot is theannual, arithmetic average pressure versus timecalculated from the individual wellhead shut-inpressures as reported to the Oklahoma Corporation

390

Commission (OCC). Although the data wereoriginally reported to the OCC, the KCC hasconsistently maintained a record of the values.Assuming that the Guymon-Hugoton Field is, forpractical purposes, in pseudosteady state duringnormal production periods, the pressure at everypoint should decline by a like amount. Therefore,the pressure relative to any point in the reservoirwould reflect a similar pressure decline as thatindicated by the average. Of course, this wouldnot necessarily hold true for short time periods atspecific wells due to local proration, linepressure changes, or any number of conditions thatmight disturb a given drainage area or region andnot immediately effect the field as a whole. Inreview of the usefulness of these assumptions, theinitial pressure for each of the replacement wellswas estimated and compared to the actual measuredpressure. This was done by calculating thedifference between the last measured shut-inpressure from the original well and the averagefield pressure for the same time. This differencewas then added to the average field pressure at thetime the replacement well was tested. Table 5presents the results of such estimates and showsthe per cent error between the estimated values andthe measured values. Nearly half (44%) of theestimates were within ± 10% of the measured value.This would represent about 10 to 20 psi during thetime most replacement wells became active (1970­1989). Thirty-four wells had estimated pressureswithin ± 15 psi of the measured values. Recallthat this value has significance with regard to theflowing pressure gradient within the reservoir andthe distance from the original well to thereplacement well. Major differences in theestimated pressures and the measured pressures wereinvestigated to account for any greater thanexpected error.

As referred to earlier, there are severalconditions that can cause well pressures to begreater than the estimates and RN59 illustrates oneof these conditions (Fig. 5 and Fig. 6). Theoriginal well was completed in the Herington andthe Krider formations and was acidized as acommingled completion. After producing forapproximately thirty-six years, the replacementwell was drilled. It is unknown by us why it wasdecided to replace the original well, but possiblyit had mechanical problems. Note the location ofthe original well was in the approximate center ofthe section (Fig. 5) and the replacement well is inthe approximate center of the northwest quadrant.The stimulation of the new well was somewhatdifferent than the previous producer, in that, theHerington and Upper Krider were acidized separatelyfrom the Lower Krider. Under normal circumstances,one would expect that to be a better stimulationmethod. As the Lower Krider is frequently morepermeable, commingled stimulations may permit mostof the acid to enter and react in that zone,leaving the tighter intervals with only minorstimulation, if any. The differential depletion asobserved in the Guymon-Hugoton Field can cause thispreferential stimulation situation to be even morepronounced as the pressure differences between thelayers becomes greater. The measured pressure inthe replacement well was 115 psi higher thanestimated by the method described earlier. If onestops there, the replacement well looks to be asuccess. However, the productive capability of the

SPE 20755 D. J. EBBS, A. M. WORKS, and M. J. FETKOVICH 5

replacement well is only about 5 per cent of theoriginal well (Fig. 5). Based on limited specificdata from the well and the expected behavior of no­crossflow layered reservoirs, it appears that thestimulation of the tighter, less depleted zones wassuccessful, but that the more permeable zone wasnot stimulated and maybe even still have somecompletion damage. The higher pressure hasdeclined during annual testing until it is near theprevious trend line for the original well and verylittle additional production has been realized.Because the commingled well will reflect thepressure of the most permeable layer, or moreprecisely, the layer with the greatest q(max)/Gjvalue,10 the new location and the stimulationprocess must have reversed the roles of the layersin this well. This replacement well has nosignificant additional reserves beyond what wasproduced by the original well, even though therewas a higher pressure measured in the well. Theindicated higher pressure in the tighter zone isexpected and as predicted by the method discussedby Fetkovich et al. 10 They present that thevertical distance (in this case delta pressure onthe P vs. Gp plot) between the total system value(volumetric average pressure here) and each layervalue is inversely proportional to their volumeratios. Although the volumetric average pressureis unknown, it would be near, but higher than themeasured pressure which is more nearly reflectingthe higher permeability layer (probably Krider).The difference between the Herington pressure andthe volumetric average pressure would be three tofive times greater than the Krider difference andopposite in sign. In this particular case (RN59),the replacement well appears to have lost reservesbased on the trend of the pressure versuscumulative production plot (Fig. 5). Becausethere is not clear evidence as to why thereplacement well was drilled, the preceding commentmay not apply, plus further production and/oradditional stimulation may improve the wellperformance. RN59 illustrates that the assumptionregarding the predictability of the replacementwell pressure will not be valid if the productionzones are not the same or if stimulationsignificantly alters the q(max)/Gj relationshipsbetween layers. It should be noted that there wasless than one month between the last production ofthe original well and the first production of thereplacement well in the RN59 section.

Another example, RN10, illustrates moredefinitively the above described response (Fig. 8 &9). The replacement well on this section wasdrilled because of a casing leak in the originalwell, which was completed as commingled Herington­Krider. Performance of the original well was quiteconsistent until a few years before the leak wasidentified: The replacement well was completedonly in the Winfield which was previously notproducing in this section. Initial testing of thereplacement well revealed the Winfield was about277 psi depleted at this location when thereplacement well was completed. It had a wellheadshut-in pressure of 168.1 psig. This is about 50psi higher than would be predicted had thecompletion intervals been the same as the originalwell. Note that the difference in the pressurebetween the Winfield and the Krider-Herington (50psi) is not as much as the apparent difference (115psi) between the Herington and the Herington-Krider

391

as indicated in the previous example (RN59). Thisis due to the better rock quality, especially thepermeability, 9 of the Winfield as compared to theHerington.

The preceding two examples had replacement wellpressures significantly higher than estimated usingthe original well and these results were explainedby the difference in productive zones. RN40 alsohad higher measured pressure (196 psig) for thereplacement well than predicted (82 psig) using thelast reported pressure from the original well.Upon examination of the shut-in pressure (Fig. 10)and production (Fig. 11) data, it appears that theoriginal well was suffering from an increasingliquid accumulation in the wellbore. This isparticularly noticeable upon reviewing the last fewyears of production history and the last shut-inpressure. If one uses the next to last shut-inpressure to estimate the replacement well's initialpressure, the estimate is 193 psig compared to themeasured 196 psig. It should be noted that forthis example there was a four year non-productiveperiod between the last production of the originalwell and the first production of the replacementwell, whereas the previous three examples had, atmost, three months of no production from thesection. The RN40 original well was an open holecompletion with a slotted liner producing from theKrider and Winfield formations. Although the wellwas plugged back through the Winfield, the sourceof water could have been from there, as it is verydifficult to successfully isolate a zone in acompletion such as this. The replacement well wascompleted higher and only in the Krider. It is notknown whether the water shut-off was completelysuccessful, but it appears to be based on thelimited data available. However, the replacementwell's productive capacity is not nearly as big asindicated for the original well prior to theapparent water problem. Even though the originalwell was a commingled well and the replacement wellwas completed in only the Krider, the expectedpressure for the replacement should be closelyrelated to the estimated pressure using originalwell values. These measured pressures will tend toreflect the Krider pressure because of itsgenerally higher permeability values and thecontrast between the permeabi1ities of the Kriderand Winfield are not as great as between the Kriderand the Herington. The loss in productivity isprobably related both to the loss of the Winfield,as indicated by a reduction in apparent ultimatereserves (Fig. 10, P vs. Gp )' and an increased skinon the Krider as a result of partial completionthrough that zone.

The following example section, RN21, had noproduction between April 1969 and October 1975,over 6 years. An open hole completion in the"Chase" which was acidized with 15% HCl (18000gallons) is what is known about the completion. Areview of the production data indicates no majorproblems until the mid-1960's. The performanceappears to have dropped off quite badly based onthe two backpressure points furtherest to the lefton the plot (filled circles) (Fig. 12). The lasttwo values on the pressure versus cumulativeproduction plot appear to either be reflectingpressures too low (possibly liquid in wellbore)and/or indicating some drainage is occurring ascompared to previous data. The replacement well

6

A FIELD CASE STUDY OF REPLACEMENT WELL ANALYSISGUYMON-HUGOTON FIELD, OKLAHOMA SPE 20755

was completed with selected, limited perforationsin the Herington and Krider. Utilizing the methodpreviously mentioned to estimate the replacementwell's initial pressure resulted in an estimate of153 psig as compared to the measured pressure of160 psig. The value of the replacement well'spressure indicates that its drainage area continuedto decline at approximately the same rate as thetotal field average during the 6 and one-half yearsof non-production from this section. It alsoimplies that production is from the same intervalsproducing in the original well. This is confirmedby the pressure versus cumulative production plotcontinuing on a parallel trend to the early portionof the plot for the original well. Note that thebackpressure points imply some reducedproductivity, but with indicated performanceimprovement for the replacement well since therecent rate increase (Fig. 13) due to a higher gasallocation for the Guymon-Hugoton Field. However,even with the improvement, the productivecapability is not as good as the original well.This is quite understandable with the limitedperforations (only eight one-foot intervalsperforated over the Herington and Krider layers).This could cause the total skin to be less negativeas compared to the original well even though thereplacement well was stimulated with 8000 gallonsof 15\ HCl.

The final Guymon-Hugoton Field example is RN63.More has been produced from this section than anyof the other examples. The original well producedapproximately 13 BSCF before it was replaced in1982, probably for mechanical reasons (it was anopen hole completion). Only about nine monthselapsed between the last production of the originalwell and the first production of the replacementwell. The estimated pressure was 151 psig comparedto about 171 psig measured. The difference is onlyslightly larger than one would expect, so untilcombined with additional information it would beassumed that the replacement well was producingfrom the same interval as the original well.Comparing the backpressure data, it is obvious thatthe replacement well has more productive capacitythan the original well which, of course, could be aresult of the completion and stimulation. However,a review of the pressure versus cumulativeproduction plot shows that a different drainagevolume is involved and it is smaller. Withoutdetailed information regarding these two wells, itis not possible to exactly determine the cause ofthis response. It could be the result of apartially damaged, more permeable zone being betterstimulated, but at the same time losing theproduction from a lesser permeable zone (withsignificant hydrocarbon volume) because of thecased hole with limited perforations. This wouldaccount for the slightly higher than expectedpressure and it could also account for theindicated increased productivity plus the decreasedultimate recovery.

DISCUSSION OF RESULTS

The last five examples described in the previousreview were selected to illustrate some specificdeparture from expected results. The first exam­ple, RN70 (Fig. 3 and Fig. 4), had initialresponses that we expected for a Guymon-Hugoton

392

replacement (or even infill) well that wascompleted in the same layers with approximately thesame simulation results. There was less than twomonths, at most, between the last production of theoriginal well and the first production of thereplacement well. Therefore, the pressure trendshould continue, the well capability should besimilar, and the indicated ultimate recovery shouldbe consistent. All of these items are fulfilled inthis first example, but many other wells could alsohave been used for this illustration as indicatedby the considerable number (34) of replacementwells exhibiting less than ± 15 psi variance fromthe estimated pressure as compared to the measuredpressure. Table 5 indicates that there was approx­imately as much overestimation of pressure asunderestimation, at least for estimated pressureswithin ± 30 per cent of the measured values. Thisrepresents 75 per cent of all the values that couldbe compared. This tends to confirm that thereplacement well pressures were not biased towardpressures higher than could be estimated from thegeneral field pressure decline. In no case wasthere ever an observed shut-in pressure close tothe 445 psig needed to represent virgin or originalreservoir pressure. Only three replacement wellshad measured pressures in excess of 300 psig withthe maximum 350.7 psig. Earlier, the comment wasmade that approximately one-half of the replacementwells could be considered edge wells. This was, ofcourse, referring to the edge of the reservoir andtoward the portions on the east or west sides ofthe field where the reservoir rock qualitydegrades. There are also edges of the Guymon­Hugoton Field along both the Kansas and Texas statelines. These are not physical edges or boundariesas regards the reservoir, but there are replacementwells near these field 'edges' in similar densitiesas found near the reservoir edges (Fig. 2). Incontrast, there are no identified replacement wellsin the 10 to 12 townships covering the middleportion of the field.

Near the west edge of the field is a section (RN36not shown) which has had three different producingwells. The original well was completed in 1953 andwas produced until 1967. This original completionhad a combination Herington and Krider perforatedinterval and was a small producer (less than 200MMscf cumulative production. The first replacementwell was reportedly completed in the Krider andWinfield in 1977 (10 years after original wellshut-in). This well had an initial pressure higherthan the last pressure of the original well, butthen it was also completed in different layers.For some reason, the well produced less than oneyear after producing for short intervals and atvery small rates (maybe wet). The secondreplacement or third well was completed in 1979 andwas also perforated in the Herington and Kriderlayers, similar to the original well. Initialpressure value (167 psig) indicated that thedrainage area of this well had declined 116 psi inthe thirteen years since the original well ceasedproduction. The average field pressure haddeclined approximately 101 psi over the same periodwhich would indicate that this area is incommunication with the major portion of the field.The higher pressure in the Winfield also confirmsthat no-crossflow is taking place in this area, atleast between the Krider and the Winfield.

SPE 20755 D. J. EBBS, A. M. WORKS, and M. J. FETKOVICH 7

The exponent in this equation (0.85) is the valueused by the acc for the official deliverabilityevaluation and is fairly representative as indi­cated by the performance of most wells. Rearrangethe backpressure equation and solve for 'C'

This re-introduces the question of whether betteror higher deliverabilities are obtained fromreplacement wells. Given that most of the originalproducers were completed during the 1940's and1950's, it seems reasonable that completion andstimulation techniques would have improved overthis period and newer wells might have betterproductivities compared to the older wells. Thisdoes not appear to be correct based on our reviewof results. As a measure of the relative wellproductivity, the 'c' value was calculated for eachof the replacement wells and original producers.

Based on 1) the observed pressure declinethroughout the field being in general agreementwith the average field pressure decline, 2) thelack of any observed original or virgin pressure inany replacement well, and 3) the explainableresponses of higher than expected observedpressure, it has been concluded that no replacementwell has encountered gas reserves not already incommunication with existing producers. This wouldimply no additional reserves have been found by thereplacement wells unless the replacement well canaccelerate the gas production into an earlierperiod and that an increased recovery can beproduced before the rate declines below an economiclimit.

(3)

the original and anyof the replacement to

can be evaluated.

Creplacement

CoriginalCR

Using 'C' values for bothreplacement wells, a ratiothe original well 'C' ratio

C ratios of all magnitude occur in all portions ofthe field having replacement wells, so there doesnot appear to be a preferential portion of thefield to drill a replacement well to obtain afavorable 'C' ratio. There are, of course, areaswhere one is more likely to have a largeproductivity well, but for this study a significantportion of that area is devoid of replacementwells. Reference is made to the middle of fieldwhere reservoir rock quality is better as opposedto the edges of the field where the reservoirquality decreases. Our review indicated higherpressures existing toward the field edges, butproductivities were generally very low. Even therethe pressure decline essentially reflects the

Figure 16 has a plot of both the frequencydistribution and cumulative frequency percentagesof the calculated 'C' ratio values for thereplacement wells. Note that approximately eightyper cent of the samples are less than or equal to 2and that the most probable value is 1.02. Theaverage value is 1.94, but it is biased by somesingle points that are greatly exaggerated. Thefour points with a ratio seven or greater wereconsolidated into one grouping for convenience toshorten the plot. These four points had largecalculated 'C' ratios because the original wellstested extremely small productivities. Thereplacement wells had only small to mediumproduction capabilities with expected recoveriesfrom three of these sections being considerablyless than 1 BSCF, while the other has alreadyslightly exceeded 1 BSCF. The point is, that witha small well, even apparently dramatic improvementcan still result in a small well with low recoverypotential and the possibility that it is noteconomic. If the five smallest and the fivelargest values of 'C' ratio are eliminated, thearithmetic average value drops to 1.31. This is amore reasonable average and may reflect theimprovements available by not completing open holewith commingled stimulations. It should still benoted that the probable 'C' ratio for a replacementwell is 1.02 or a similar productivity as theoriginal well. Because the exact intervals ofcompletion are not always clearly identified, noeffort was made in calculating the C ratio toseparate out replacement wells that were apparentlycompleted in layers different than the completionof the original well. Some of these effects werediscussed with the various examples and will not berepeated here. Suffice it to say, it may bepossible to increase the productivity from asection by drilling a replacement well, but theprobability is that it will be similar to theoriginal well unless an identifiable problem(mechanical, different layers, water entry) isbeing corrected. The same comments would apply toan infill well, that is, it should be expected thatan infill well would be similar to the existingwell unless they are producing from entirelydifferent layers, which using our terminology isnot an infill well.

the

(2)

(1)

calculated by using

QC

A situation similar to that described above, but onthe east edge of the field, is RN43 (not shown).This section also had three different wells. Theoriginal well was an openhole, liner completionwith probably the Herington and Krider producing.Being a very small well, it only produced for alittle over one year (1955-1956). Another well,the first replacement well, never produced after itwas completed in 1974 as a cased, perforated'Chase' completion. In 1983, the third well(second replacement) was completed in the Krider.It, too, was a very small well plus it had aninitial reported shut-in pressure of about 122 psig(240 psi decline). The field average pressure haddeclined approximately 192 psi since the originalwell stopped producing, therefore this section,like the one just previous, has declined more thanthe field average. Because this last replacementwell was completed only in the Krider, it should beexpected that the shut-in pressure would besomewhat smaller than if it were a Herington-Kridercompletion, such as the original well. Combinedproduction on this section is only slightly over100 MMscf. Both RN36 and RN43 information plusexhibits were presented to the acc as part of astudy to answer questions posed by the accregarding the advantages and disadvantages ofinfill drilling in the Guymon-Hugoton Field.

The 'c' value wasbackpressure equation

393

8

A FIELD CASE STUDY OF REPLACEMENT WELL ANALYSISGUYMON-HUGOTON FIELD, OKLAHOMA SPE 20755

average field pressure decline indicating depletionof the gas in those areas whether active wells arepresent or not.

pressures that are significantly differentwhich indicates the layers are not in verticalcommunication.

ACKNOWLEDGEMENT

NOMENCLATURE

REFERENCES

- backpressure equation coefficient• original gas-in-place, MMscf [std m3]• cumulative gas production, MMscf

[std m3 ]wellhead shut-in pressure, psia [kPa]wellhead flowing pressure, psia [kPa]

= surface gas flow rate, Mscfd[std m3 /d]

- calculated absolute open flow, Mscfd[std m3 /d]

- gas compressibility factor,dimensionless

1. Barber, A. H. Jr. , George, C. J. , Stiles,L. H. , and Thompson, B. B. : "Infill Drillingto Increase Reserves - Actual Experience inNine Fields in Texas, Oklahoma, and Illinois,"JPT (August 1983) 1530-1538.

z

5. The current 640 acre spacing for the Guymon­Hugoton Field provides efficient and effectivedrainage of these gas reserves. Productioncould have been accelerated by largerallocations and allowables as indicated by thelong plateau period.

We wish to thank Phillips Petroleum Company forpermission to publish this paper. We also wish tothank D. R. Wier and the Great Plains Regionpersonnel for supplying data and insight for thisreview. Finally, we wish to thank our supportstaff for their aid and assistance in preparingthis paper.

As indicated earlier, it is not anticipated thatinfill wells in Guymon-Hugoton would findsignificant gas volumes not already being depletedor well productivities that would be effectivelygreater than the existing producer on a givensection. Should one wish to evaluate the potentialfor infill drilling to accelerate recovery, theremust first be a determination as to whether theexisting well could accomplish the same purpose.Are all available layers perforated and effectivelystimulated? Is the current production rate atmaximum well capacity? Is the pipeline or deliverypressure as low as feasible: Are mechanical orliquid entry problems resolved or non-restricting?Assume the answers to all these questions are yesand there is a reasonable chance to have anallowable greater than what can be produced by theexisting well. Then an estimate can be made byusing the material balance plot (P vs. Gp ) combinedwith the backpressure curve, similarly to what wasdone to evaluate the replacement wells. This wouldnot provide an entirely accurate forecast becauseof drainage and interference between the originalwell and the infill well plus all other nearbyproducers. It would, however, permit theevaluation of whether an infill well can beeconomically completed in a section by providing anapproximate total section withdrawal assumingconditions around that particular section wereunchanged. This simple material balance methodshould be useful to eliminate cases where an infillwell does not, either increase recovery over agiven span or have acceptable incrementaleconomics. For situations not eliminated by theabove approximations, it is recommended that amultiwell, multilayered mode15 ,9 be used toforecast that total section recovery with the modelincluding all surrounding wells for the historymatch portion and then adding the infill wells forpredictions.

CONCLUSIONS

Based on our review and analysis of each of the 86replacement wells completed in the Guymon-HugotonField, the following conclusions are presented:

1. Replacement wells' relative location withintheir own sections were appropriate forevaluating reservoir continuity overreasonable areal distances.

No replacement well exhibited virgin reservoirpressure. No indications were found thattrapped or undrained reserves were encounteredby a replacement well.

2. Barbe, J. A. and Schnoebelen, D. J.:"Quantitative Analysis of Infill Performance:Robertson Clearfork Unit," JPT (December 1987)1593-1601.

4. Wu, Ching H., Laughlin, B. A., and Jardon,Michel: "Infill Drilling Enhances WaterfloodRecovery," JPT (October 1989) 1088-1095.

6. MCCoy, T. F., Fetkovich, M. J., Needham,R. B., and Reese, D. E.: "Analysis of KansasHugoton Field Infill Drilling, Part I: TotalField Results," SPE 20756, 65th AnnualTechnical Conference New Orleans, LA September23-26, 1990.

3. Gould, T. L. and sarem, A. M. Sam: "InfillDrilling for Incremental Recovery," JPT (March1989) 229-237.

5. Hinn, R. L. Jr., Glenn, J. M., and McNichol,K. C.: "Case History: Use of a MultiwellModel to Optimize Infill Development of aTight-Gas-Sand Reservoir," JPT (July 1988)881-886.

in differenthave shut-in

wells completedthe original well

Replacementlayers from

Reservoir characteristics, as indicated by thebackpressure curve and the material balanceplot, are generally similar between theoriginal well and the replacement well. Thereis, of course, field-wide variations inreservoir properties.

2.

3.

4.

394

SPE 20755 D. J. EBBS, A. M. WORKS, and M. J. FETKOVICH 9

7. Fetkovich, M. J., Needham, R. B., and McCoy,T. F.: "Analysis of Kansas Hugoton InfillDrilling, Part II: Twelve Year PerformanceHistory of Five Replacement Wells," SPE 20779,65th Annual Technical Conference New Orleans,LA September 23-26, 1990.

9. Siemers, W. T. and Ahr, W. M: "ReservoirFacies, Pore Characteristics, and Flow Units ­Lower Permian Chase Group, Guymon-HugotonField, Oklahoma," paper SPE 27057, 65th AnnualTechnical Conference New Orleans, LA September23-26, 1990.

8. Fetkovich, M. J. , Voelker, J. J. , and Ebbs, 10. Fetkovich, M. J. , Bradley, M. D. , Works,D. J. : "Development of a Multiwell, A. M. , and Thrasher, T. S. : "DepletionMultilayer Model to Evaluate Infill Drilling Performance of Layered Reservoirs WithoutPotential in the Guymon-Hugoton Field, Crossflow," paper SPE 18266, 63rd AnnualOklahoma," SPE 20778, 65th Annual Technical Technical Conference Houston, TX October 2-5,Conference New Orleans, LA September 23-26, 1988.1990.

TABLE 1

DRILLING & COMPLETIONSUMMARY

Years Drilled Original Wells Replacement Wells

1930-1939 1 01940-1949 35 01950-1959 33 51960-1969 1 91970-1979 2 381980-1989 _0_ .......1.L-

Totals 72 86

Open Hole Completions 32 5

Acidized (only) 60 36Hydraulic Fractured* 6 30

No Stimulation Record 6 20

* Most wells were acidized during a portion of the stimulation.

TABLE 2

DISTANCE BETWEEN REPLACEMENT WELL(S) AND ORIGINAL PRODUCER

Distance. Feet

200 - 499500 - 999

1000 - 14991500 - 19992000 - 24992500 - 3099

1657

395

Number of Wells

1087

3221

8average

SfE 20755

TABLE 3

REPLACEMENT WELL REFERENCE TABLE

Ref. Location operator Lease Ref. Location Operator Lease Ref. Location Operator LeaseNo. Sec-T-R No. Sec-T-R No. Sec-T-R

1 36-1-10 Getty Reynolds #1 25 15-2-11 Cotton Gear #1 49 26-4-13 Panhandle E. Smith #1First Nat'l Clark #1 Wallace Folkers #1 Keenan A #4

2 21-1-12 Phillips Gayle #1 26 23-2-11 Tascosa Mons #1 50 8-4-14 Panhandle E. Burris B #1Gayle #2 Cotton Sweet #1 OK State F #1Gayle #3 Wallace Sweet #1 Burris B #2

3 24-1-12 Tascosa Muller #1 27 6-2-12 cities stonebraker AG #1 51 25-4-14 Mobil Bartels #1Phillips Jones BB #1 Wallace CSC #1 Bartels #2

4 27-1-12 Phillips Krull #1 28 7-2-12 Phillips Gerald #1 Bartels #3Krull #3 H & L Davison #1 52 33-4-14 Cities Ziegler #1

McKenzie Hant1a #l 29 31-2-12 Phillips Orv #1 Kaiser F. Eaton #136-1-12 Kerr McGee Seright #1 Orv #2 53 10-4-18 Kansas Neb. Pauls #1

Apache State #1 30 16-2-13 Kerr McGee OWen #1 Cotton Sneed #16 1-1-13 Phillips McFadden #1 JNC Hiebert #1 54 35-4-18 Kansas Neb. Schroeder #1

McFadden #2 31 29-2-13 Amoco Jefferies #1 Texas Energies Friesen #27 5-1-13 Amoco Boston #1 Jeffer ies #2 55 10-5-12 Panhandle E. Bevan #1

Boston #2 32 20-2-17 Kapca Grimmer #1 Bevan #2B 7-1-13 Amoco Burrows B #1 Vanderwork #1 56 26-5-17 Santa Fe Parham #1

H & L Etling #1 Vanderwork #2 Parham #lA9 8-1-13 Amoco Dick A #1 33 35-2-17 Amoco Randles #1 57 2-5-18 Santa Fe Duncan #1

Dick #1 Stinson #2 Duncan #lAH & L Hill #1 34 22-2-18 Kansas Neb. Tharp #1 58 10-5-18 Texaco Phillips #1

10 11-1-13 Phillips Folder #1 Tharp A #lX Phillips #2Folder #2 35 32-2-18 Texola Rhodes #l H & L Phillips #1

Co> 11 19-1-13 Amoco Hungerford #1 Rhodes #2 59 22-5-18 Santa Fe Chrispens #1COen Hungerford #2 36 5-3-12 Cities Enz A #1 Chrispens #lA

12 29-1-13 Amoco Draper #1 Tenneco Enz #1 60 34-5-18 Santa Fe Hiebert #1Draper #2 Wallace Enz #1 Hiebert #lA

13 24-1-14 Phillips Line #1 37 10-3-12 Cities Robbins A #1 61 25-6-13 Mobil Weeks #2Line #2 Robbins A #2 Weeks #2-1

14 35-1-14 Phillips Jerome #1 38 3-3-13 Panhandle E. Messinger #1 Weeks #2-3Jerome #2 Messinger #2 62 25-6-14 Mobil Pearl Davis #l

15 10-1-15 Phillips strat #1 Anabaco Messinger #1 Pearl Davis #3strat #2 H & L Messinger #1 63 18-6-15 Mobil Parker #1

16 31-1-15 Phillips Percy #1 39 9-3-13 Cities Stonebraker H #1 Parker #3Percy #2 H & L Quilty #l 64 26-6-15 Mobil Isham #1

17 4-1-17 Santa Fe Wall #4 40 4-3-14 CanoeD Smith #1 Isham #3Wall #4A Kaiser F. Myers #1 65 35-6-15 Mobil Miller #1

18 7-1-17 Tascosa Charles #1 41 32-3-17 Cabot Casto #5 Miller #3Phillips Norton C #1 Casto #14 66 16-6-16 Mobil CUrtis #1

19 15-1-17 Phillips Atkins A #2 42 18-3-18 Panhandle E. Enns #1 Curtis #2Atkins A #3 Anadarko Voth 8 #1 67 18-6-16 Mobil Cope #1

20 16-1-17 Phillips Atkins A #1 43 19-3-18 Panhandle E. Fletcher #1 Cope #2Atkins A #lR Anadarko Voth C #1 68 24-6-16 Mobil Ebersole #1

21 20-1-17 Kerr McGee Edith #1 Voth C #2 Ebersole #2H & L Wagner #1 44 14-4-11 Panhandle Dev. Hart #1 69 33-6-16 Mobil Guest #1

22 24-1-17 Phillips Ingles #1 Hinkle D. Hart #1 Guest #2M & L Lewis #1 45 8-4-12 Cities Potts #1 70 20-6-17 Mobil Hoobler #1

Lewis #lA H & L Hitch #1 Hoobler #223 21-1-18 Kansas Neb. Prewett A #1 46 13-4-12 Panhandle E. Haley #1 71 33-6-17 Santa Fe Dore #1

Prewett A #3 Shaffer B #2 Blankenship #124 14-2-11 Cotton Mayer #1 47 24-4-13 Panhandle E. Gable #1 72 23-6-18 Plains Cain #1

Wallace McCall #1 Elrod #1 Mapco Cain #248 25-4-13 Panhandle E. Keenan #1

Keenan #2

TABLE 4

TIME ELAPSED BETWEEN ORIGINAL WELL'S LAST PRODUCTIONAND REPLACEMENT WELLS FIRST PRODUCTION

Time, Yrs No. Wells % of Wells Cum %

< 1 34 39.5 39.5

~ 1 < 2 15 17.4 56.9

~ 2 < 3 4 4.7 61.6

~ 3 < 4 2 2.3 63.9

~ 4 < 5 4 4.7 68.6

~ 5 < 10 6 7.0 75.6> 10 < 20 10 11.6 87.2

> 20 4 4.7 91.9Never produced _7_ 8.1 100.0

86

TABLE 5

± PER CENT DIFFERENCE BETWEEN REPLACEMENT WELL'S INITIALESTIMATED AND MEASURED PRESSURES

ESTIMATED - MEASURED

seE 2075;

% ERRORMEASURED

*100

~ abs (10)> abs (10) ~

> abs (20) ~

> abaMeasured not

AlgebraicNo. Wells Range of error, % Average, %

38 + 9.6 to - 9.6 0.7abs (20) 13 + 14.9 to - 19.6 2.5abs (30) 6 + 23.8 to - 24.8 1.6(30) 19 + 82.9 to - 66.4 - 11.6available .....!Q...

86

Fifty per cent of estimates were within ± 10% of measurements available.

397

SfE 20755

".,...•..•..

....

----

RI8ECM

I

-

R17ECM

---~---

R,5ECM RI6ECMR14ECMR,3ECM

._----.--------.-.--~ - -. . ,'. _: -------~---- .. ----------~----- ---:----~-- ---: :-

•-

R12ECM

- .. :-!•• c ..~I·."-

RllECM

,,,,,-1···---- ···-.--------r·--l. ;-: :-,: :- -

-~---------i--------..---..---------i----.----------- ~. ---: --------~ ',,~. ~ l"-. -• ••

Rl0ECM

T6N

T1N

T5N

T3N

T4N

,I

~~ FINNEY

"""-T<W~ KEAANY ,.--

) I--

OTt ~LLGRANTl

wofroo STEVENS .J...DCOLORADO ]' KANSAS

GUYMON~ T!XAS~HUGOTON-~ OKLAHOMA

~-,r- TEXAS

~-~

~

.•.............

.'.............•......

.'.....'

Fig. 1-Hugoton field regional map. Fig. 2-Guymon-Hugoton replacement well map.

SEC 20-BN·17ECM• ORIGINAL WELL LI

~lfJ-

!J REPLACEMENT WELL

• • • 2640

•!J 1-'••

• .2• D-O 2640 6280•

•8ACKPRESSURE CURVE •• •1000 ••N •« ••0;

•...•o 100 ••0 .,0x •N

"flos: 10

~U 0... ,

'00 1000 10000'0RATE. MCFD

,1991

ORIGINAL WELL

REPLACEMENT WELL

1979 1983 198719751955 1959 1963 1967 1971

YEARS1951

1 T

1947

10

1000

10000:J I

~Cl~ 100U~

500040002000 3000

Gp, MMCF1000

oo

Ii!500

450

400

350

~ 300(/)a..0.." 250en:c~ 200

150

100

50

Fig. 3-RN70 pressure va. cumulative production and baekpressure plot. Fig. 4-RN70 production rate va. time plot.

ORIGINAL WELL

REPLACEMENT WELL

20 751

1982 1986 1990 1994

~1"")1="

19781970 1974

YEARS

1 , lit I I I , ii'1950 1954 1958 1962 1966

100003:J1----------------------~--=~::....---.i

1000

~Cl~ 100U::E

10

400030002000

Gp, MMCF1000

SEC 22-SN-IBECM

I~ ORIGINAL WELL LI

'~'tE" REPLACEMENT WELL #lA

"• • 2B40• • #1

•0

0 2840 62BO

•• • 0

BACKPRESSURE CURVE •• "•10000 • •...« •~ 1000 ".0 • ....·•• 000 100X "~ "! '0 ~

" . )UlL

1100 1000 100001 10

RATE, MCFD

oo

60

460

160

400

600

100

350

~ 300~0: 260tiS::I:s: 200

Fig. 5-RN59 pressure YB. cumulative production and backpresBure plot. Fig. 6-RN59 production rate VB. time plot.

600040002000 3000Gp, MMCF

1000

SEC 11-1N-13ECM

I: ORIGINAL WELL LI mornREPLACEMENT WELL#1

• 264: ,,#2••• •• 0 2640 5280••'.•".IACKI'lIEIIURE CURVE•

" 1000

'. ...«•• iii'. lLo 100. " 0 ,.. 0'0 0

/..... ODD X~ 10

JulL ,

100 '001 10000'0

RATE MCfD

oo

60

100

400

460

600

360

~300

0: 260en::I:~200

160

19901986198019761960 1966 1970

YEAR196619601946

o " I' I ,;; I ' " I' I 'i "I I I I' I I I ,:", ,

1940

6001j-----------------------------.,(!l

tiS 460a..

~ 400:::::>enU) 360wa::a.. 300ZII- 250::::>::I:U) 200w(!l<! 150c::W~ 100

9 60wu::

f8

Fig. 7-Guymon-Hugoton field average shut-In pressure VB. time plot. Fig. 8-RN10 pressure VB. cumulative production and backpressure plot.

SPE 2075510000:J ,

ORIGINAL WELL

REPLACEMENT WELL

15001200600 900

Gp, MMCF300

SEC 4-3N-14ECM

I~ ORIGINAL WELL LI ""Bj#2<J REPLACEMENT WELL ; <J

•• 2640• • #1• •• D-

• 0 2640 5280•

•8ACKPRESSURE CURVE •1000..

\0(

in..0" 10000)( D • •~ ~II]Cb •lD

,jU..

1IDD 1000I ID

RATE, MCFO

oo

50

150

100

460

500

400

350

S2 300~c: 250en:::I:~ 200

19931988198319781968 1973

YEARS196319581953

1 j

1948

10

1000

~cLt 100U::E

Fig. 9-RN10 production rate YB. Urne plot. Fig. 10-RN40 pressure Y8. cumulative production and backpres8ure plot.

g

1000

10000 3 ,

500040002000 3000

Gp, MMCF1000

SEC 20-1N-17ECM• ORIGINAL WELL LI ""EB<J REPLACEMENT WELL

#12640

#1•<J

•D-•

0 2640 5280• • •••••••

8ACKPRESSURE CURVE •••..1000

"'in D..o 100

DDo00

ITb'llbx ..~ •10 <J • D'r •~

~..1,. lDD 1000 10000

RATE, MCFD,oo

500

450

400

350

Q 300(I)a..c: 250en:::I:~ 200

150

100

50

199119871983197919751967 1971

YEARS1963195919551951

ORIGINAL WELL

REPLACEMENT WELL

1 j

1947

10

~ClLt 100U::E

Fig. 11-RN40 production rate VB. time plot. Fig. 12-RN21 preSSUfe VB. cumulative production and backpressure plot.

SPE 207 'j 5

10000

ORIGINAL WELL

REPLACEMENT WELL

1000

~Cli:t 100U~

10

1 I

1951 1955 1959 1963 1967 1971 1975

YEARS1979 1983 1987 1991

Fig. 13-RN21 produetlon rate VB. time plot.

SEC 18-6N-15ECM• ORIGINAL WELL LI

~"EE'J REPLACEMENT WELL

#1•

'. 2640 'J

• #3•

D-O 2640 6260

••• •

••BACKPRESSURE CURVE •

1000 •• •... •« •iii- •• !:J...

ei 100 II]

0 ,0 ,X~ ~

0'0

~ • [J]

u...I10 '00 1000 10000

RATE. MCFD

I

500

450

400

350

C!J 300U5a..0..' 250V5::c3!: 200

150

100

50

oo 5000 10000

Gp, MMCF15000 20000

Fig. 14-RN63 pressure VS. cumulative production and backpressure plot.

401

SPE 207 c;,

10000::r----------------------------...,ORIGINAL WELL

R_~P.!-ACEMENT WELL

1000

~Clu:: 100U~

10

1 I

1947 1951 1955 1959 1963 1967 1971

YEARS1975 1979 1983 1987 1991

Fig. 15-RN63 production rate vs. time plot.

8

17_0

<196

6

Frequency JCumula!ive Freq~~

• Average= 1- 9

C 50 % Probability= 1-0

3 4"CO Ratios

2

•

-_/-----,/-/

('I

JI

),

j(I

!I

I

100

90-

80

'* 70>-(.)<:Ol:>0- 80Olu:Ol>-~

<0"5 50E:>

U

" 40>-(.)<:Ol:>0-Ol 30u:

20

10

Fig. 16-Frequency distribution of "e" ratios.

402