spe 59738 detailed protocol for the screening and ... optimization/detailed... · spe 59738...

SPE 59738

Detailed Protocol for the Screening and Selection of Gas Storage Reservoirs

D. B. Bennion, F. B. Thomas, T. Ma and D. Imer, Hycal Energy Research Laboratories Ltd.

Copyright 2000, Society of Petroleum Engineers Inc.

This paper was prepared for presentation at the 2000 SPE/CERI Gas Technology Symposiumto be held in Calgary, Alberta Canada, 3-5 April 2000.

This paper was selected for presentation by an SPE Program Committee following review ofinformation contained in an abstract submitted by the author(s).Contents of the paper, aspresented, have not been reviewed by the Society of Petroleum Engineers and are subject tocorrection by the author(s). The material, as presented, does not necessarily reflect any positionof the Society of Petroleum Engineers, its officers, or members. Papers presented at SPEmeetings are subject to publication review by Editorial Committees of the Society of PetroleumEngineers. Electronic reproduction, distribution, or storage of any part of this paper forcommercial purposes without the written consent of the Society of Petroleum Engineers isprohibitted. Permission to reproduce in print is restricted to an abstract of not more than 300words; illustrations may not be copied. The abstract must contain conspicuous acknowledgmentof where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836,Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

AbstractGas storage reservoirs are used worldwide to store producednatural gas during periods of low demand for use during periodsof high demand. These formations are often depleted natural gasreservoirs. Proper selection of a gas storage reservoir isimportant to allow proper and economic operation of the projecton a long-term basis. This paper describes issues which need tobe taken into consideration from a reservoir perspective whenconsidering the development of a gas storage reservoir. Theseissues include the proper containment of the injected gas,maintaining injectivity and productivity over long-termoperations, and problems which may be associated with thepresence of free water or hydrocarbons in the storage reservoir(both mobile and immobile) as well as formation damage issuesthat often surround the drilling and completion of new wells inthe gas storage reservoirs for development purposes.

IntroductionGas storage reservoirs are used on a worldwide basis for thestorage of natural gas for use in periods of peak consumption,generally in the colder portions of the year when gas demand forheating is higher. Storage reservoirs are also used to bufferperiods of peak demand and prevent disruption of suppliesduring mechanical or other problems in producing fields.

Gas storage reservoirs generally consist of good to excellentquality formations which are often located spatially close to theultimate demand source (i.e. major population centers). Most of

these reservoirs represent natural gas pools which have beendepleted below their abandonment pressure during normalproduction operations, but are now used on a seasonal basis forgas storage.

For a reservoir to be a candidate for gas storage, the followingcriteria must be satisfied:1. Sufficient reservoir volume to allow for storage of the

required amount of gas without exceeding containmentpressure constraints and without requiring uneconomiccompression to high pressure levels.

2. Satisfactory containment of the gas by competent upper andlower sealing caprock.

3. Sufficient inherent permeability to allow injection andproduction at required delivery rates during peak demandperiods.

4. Limited sensitivity to reductions in permeability (andinjectivity/productivity) due to:S presence of in-situ water (mobile or immobile)S presence of liquid hydrocarbons (mobile or immobile)S plugging of the near injector region by compressor

lubricants or other introduced fluidsS reservoir stress fluctuations during successive pressure

cycles5. Absence of hydrogen sulphide gas (in-situ or bacterially

generated)6. We must be able to drill and complete additional wells in the

formation as required with causing severe formation damage(due to the highly depleted pressure condition which mayoften exist in these reservoirs).

The Typical Gas Storage ReservoirGas storage reservoirs are generally high permeability clasticsor carbonates (1000-10,000 mD in-situ permeability is common)existing at intermediate depths and temperatures. In general,these reservoirs are depleted formations which originallycontained dry (non-retrograde), sweet (no H2S) natural gas.Typically, these zones do not contain mobile water or active orpartially active aquifers, oil legs or residual liquid hydrocarbonsaturations, although this is not always the case.

2 D.B.BENNION, F.B.THOMAS, T.MA, D. IMER SPE 59738

Containment of Reservoir GasFor an effective gas storage reservoir, the injected gas mustobviously remain in place, possibly for an extended period oftime, without escaping through permeable channels in theoverlying or underlying reservoir seal. We refer, in thissituation, to competent barriers that separate the storage zonefrom other zones and may act as permanent bleed zones for gaslosses which cannot be recovered. These connected zones mayalso contain undesirable fluids (gases containing liquids,hydrogen sulphide, water zones, oil/hydrocarbon zones, etc.).Typically thick (3-4 meters or more) dense shales are present inthese intervals to act as an impermeable seal for the reservoirgas. To verify that the cap and base rock have sealingcompetency, a cap rock permeability test is generally conductedin the laboratory on a sample of preserved cap rock which istaken via coring from the penetration of the storage formationand the cap and base rock.

Two different cap rock competency tests are generallyconducted and are required to be satisfied for sealingcompetency to be present. These are:

1. Absolute liquid permeability measurement2. Threshold intrusion pressure to gasCore material must be properly handled and preserved for

the proper analysis of caprock. In general, a low invasion coringprogram with an inhibitive water-based or oil-based mud systemshould be used to obtain the core material. The cap rock samplesshould be preserved on site using PROTEC or CORESEAL andshould not be subjected to conventional solvent extraction anddrying protocols that can remove water of hydration from theclays and shales present and permanently destroy the caprockmorphology, which may result in obtaining erroneously highpermeability and intrusion pressure measurements.

Absolute Liquid Permeability Measurements. Fig. 1 providesa schematic illustration of the equipment used for caprockpermeability measurements. These tests are almost alwaysconducted on vertical full diameter core (oriented in the verticaldirection). In some cases, vertical plugs (cut from the long axisof a vertically oriented core (Fig. 2) are used for the tests. Testsshould not be conducted on horizontally oriented core plugs, asstreamlines of flow in this situation run parallel to naturallyoccurring bedding planes, and may result in anomalously highpermeabilities in comparison to those expected to beencountered by the stored gas which will be intruding into theoverlying caprock in a vertical fashion.

Caprock samples should be subjected to X-ray or NMRanalysis prior to testing to ensure that they do not contain coringinduced fractures or other features which may representpermeability conduits that are not reflective of the reservoir.Natural open fractures existing in the caprock are obviousdetriments to the gas storage process and should be carefullyevaluated from a core analysis, seismic and geotechnical stressanalysis of the subject formations, as these will eliminate thezone almost immediately as a gas storage candidate.

The caprock test apparatus allows the preserved state sampleto be confined at reservoir overburden pressure conditions in auniaxial cell and also allows full reservoir conditions oftemperature and pore pressure to be applied to the sample toprecisely duplicate downhole conditions. A positivedisplacement pump/pressure source is used to apply a netpressure differential of 7000 kPa (approx. 1000 psi) to the coreface using formation water (or another brine known to bechemically compatible with the caprock). A digital capacitancetransducer is used to measure the applied pressure dropprecisely. Flow rate through the caprock sample is monitoredover an extended time period (generally 7-14 days) and aneffective fluid permeability to the formation brine is determined.

For effective caprock, the measured brine permeabilityshould be less than 1 nanodarcy (1 x 10**-06 mD or0.000000001 Darcy). Caprock permeability higher than thiswould indicate, on a long-term basis, that expulsion of connatewater from the caprock could occur which may allow theintrusion and production of gas.

Multiple samples of caprock should usually be tested,particularly if multiple lithologies are present, but may not bespatially continuous throughout the reservoir. This isillustrated in Fig. 3 where we observe that although lithology 1is good sealing caprock, it is not regionally continuous over theentire reservoir and in places only lithology 2 or 3, which maynot be competent, are present as a seal.

Table 1 illustrates the results of typical “satisfactory” and“unsatisfactory” cap rock evaluations.

Gas Intrusion Testing. Fig. 4 illustrates the experimentalapparatus used for a caprock gas intrusion test. This test isgenerally conducted on the same sample used for the liquidpermeability measurement, described previously, if satisfactoryresults from the liquid permeability measurement test areobtained.

The equipment consists of the vertically mounted fulldiameter vertical core sample maintained at bottomholetemperature and pressure conditions. In this test, however, watersaturated natural gas is exposed to the face of the core at a 7000kPa pressure differential. A highly accurate production burette(0.01 cc accuracy) is attached to the production end of the coresample which allows one to determine over a long-termexposure period (14-21 days) if any production of fluid from thesaturated pore space (indicating gas intrusion into the sample)occurs. Any appreciable fluid intrusion occurring over thisperiod indicates a failure and once again suggests that thesample may not be competent caprock from a sealingperspective.

In-Situ Permeability and Permeability Variations withStressAs mentioned previously, gas storage reservoirs typically havehigh inherent permeability to allow for easy gas injection andrapid delivery of large gas volumes during peak demand

SPE 59738 DETAILED PROTOCOL FOR THE SCREENING AND SELECTION OF GAS-STORAGE RESERVOIRS 3

periods. The threshold of required permeability variesdepending on the amount of pay present and the type of wellbeing used (vertical, fractured vertical or horizontal). In general,in-situ permeability in gas storage reservoirs exceeds 1 Darcyand is often substantially higher than this.

Core analysis permeability (gas permeability measured onclean, dry core samples extracted from the formation using airor nitrogen under a nominal (1378 kPa confining pressure))may not be reflective (are often higher) of those present in thereservoir, due to the fact that the routine samples do not containa irreducible water saturation (which will be present in thereservoir), and are generally not measured under a confiningoverburden pressure (which may substantially reduce thepermeability of the formation due to grain compression effects).This is particularly significant in high permeability formationscomposed of conglomeritic grains or by relying on pervasive,open micro-fracturing for the high inherent permeability. Properpermeability evaluations can generally be obtained from in-situcondition core tests (under proper overburden pressureconditions and with the correct initial water saturation in place),or via pressure transient analysis of the in-situ permeability inthe reservoir (often a more accurate technique if fractures ormacro-scale heterogeneities exist in the reservoir which cannotbe adequately represented in small core samples tested in thelab). Typical permeability water saturation and permeability-overburden pressure variation plots are shown as Figs. 5 and 6respectively. These are generic representations only and thespecific configuration of these curves will be highly dependanton reservoir pore system lithology and morphology.

Gas storage reservoirs are also susceptible to significantcyclic variations in overburden stress through the annualpressure cycling operation. This is schematically illustrated asFig. 7. It can be seen that at the peak of the storage cycle, thereservoir pressure will be at its greatest level resulting in thepresence of the least amount of effective stress on an in-situbasis, which usually corresponds to the highest permeability andfacilitates improved permeability and injectivity. When thereservoir is depleted during peak production, the bottomholepressure falls and the net overburden stress increases which mayresult in reductions in permeability and potentially indeliverability. This issue should be evaluated prior to design ofthe project as certain types of formations (as mentionedpreviously) may be highly sensitive to this phenomena. Well-consolidated competent intercrystalline sandstones andcarbonates tend to be the least affected by this phenomena inmost circumstances.

Presence of Mobile or Immobile Liquid SaturationsIn general, the optimum gas storage reservoir does not containany substantial free mobile water. Mobile water results inreduction in productivity, due to relative permeability effects,and can result in severe hydrate problems at high productionrates due to Joule Thompson expansion effects. The presence ofmobile water contacts in the base of the reservoir can also result

in cyclic trapping of a portion of the injected gas due to cyclichysteresis effects when water-gas or water-oil contact advancesand retreats in the same reservoir volume over a period of time.This phenomena has been discussed in the literature (Ref. 1)and can result in some cases in substantial permanent losses ofinjected gas due to trapping effects. Figs. 7 and 8 illustrate thisphenomena.

Trapped hydrocarbon saturations may also be problematic incertain circumstances due to solubility and swelling effects.Trapped immobile liquid saturations may be present in some gasreservoirs due to accumulation of condensate liquids (if thereservoir initially contained a retrograde condensate gassystem). Injection of gas may result in a solubility increase(dissolution of a portion of the injected gas into the hydrocarbonliquids) which results in an increase in the apparent liquidsaturation (swelling of the liquid). This may result in areduction in the gas phase permeability as illustrated in Fig. 9.In some gases, swelling and viscosity reductions in theimmobile hydrocarbon phase may combine to actually result inmobile hydrocarbon production, generally highly undesirable fora gas storage reservoir application.

Maintaining Injectivity and ProductivityAll benefits of the gas storage operation will be lost if theoperator is unable to inject or produce sufficient gas to meet thedesign and demand criteria of the project. On existing wells(those generally present from the original production life of thereservoir) which have good initial productivity, the greatestproblem is often reduced injectivity/productively due tocompressor lubricant carryover.

Many compressors, particularly older models, can consumelarge volumes (sometimes many liters per day) of lubricants.This material is often carried as a finely atomized mist into theformation, where it can gradually accumulate in the nearwellbore region, resulting in a trapped extraneous phase whichmay plug or severely reduce productivity over a period of time(as illustrated in Fig. 10). Properly designed and maintainedcompressors and filtration / precipitation equipment to removeor reduce the volume of atomized lubricant can be useful in thissituation. Many lubricants are oil-based and result in theintroduction of an extraneous phase into the reservoir. In somecases, water-based lubricants may be used, which may have lessaffinity for plugging due to a natural solubility in connate waterwhich may be present in the reservoir. This, however, is notalways effective as, in most mature gas storage injectors, little,if any, connate water saturation remains in the near wellboreregion as the large volume of dehydrated dry natural gasinjected through this zone often removes virtually all waterpresent over a period of time by dehydration and desiccationeffects. This can actually result in some long-term increases ininjectivity, depending on the configuration of the relativepermeability curves for the porous media (Fig. 11). This mayalso make wells of this type susceptible to reductions ininjectivity if water-based kill or workover fluids are used as the


introduced water will rehydrate into the “dry” cylinder of rocksurrounding the injection well and will re-establish theirreducible water saturation which may take an extended periodof time to revaporize. High salinity brines, therefore, are apotential concern in this situation as, due to subsequentdesiccation effects caused by extensive dehydrated gas injection,precipitation of the suspended solids in the pore system bysupersaturation effects may also contribute to near wellboreplugging issues.

Care must also be taken to avoid the introduction of viablebacteria with workover or completion fluids, which maypossibly result in plugging or souring of the wells (although thisis generally not a problem if no substantial water saturation ispresent in the near wellbore region).

Formation Damage EffectsIn many situations, to properly develop a reservoir for gasstorage, recompletion of existing wells and drilling of new wellsis required. This is particularly the case in recent years wherehorizontal wells have been used increasingly for gas storageapplications, as a single horizontal well can replace multiplevertical penetrations, or, due to increased reservoir exposure,allow formations which may not have been considered to havesufficiently high permeability for development with verticalwells to now be viable gas storage candidates.

Two factors can combine to create significant formationdamage in these situations:1. The formations are generally highly pressure depleted at the

time of completion of the wells. This means that extremeoverbalance pressures will be present with conventionallyweighted drilling or completion fluids, and that the potentialfor lost circulation and severe invasion and mechanicaldamage of the high permeability pore system, due to wholemud invasion, may be an issue.

2. High natural permeability, which is almost always aprerequisite for a gas storage zone, combines with thisoverbalance effect to increase the potential for lostcirculation and near wellbore damage effects. Although highpermeability formations are generally fairly forgiving withrespect to many classical formation damage mechanismssuch as fines migration (due to large pore throats and oftenvery little mobile material), phase trapping and blocking(due to very low, favorable capillary pressure) and clayrelated damage (due to generally excellent reservoir qualityand very little reactive clay in most situations), mechanicaldamage caused by the loss of large volume of fluidcontaining drill solids, corrosion products, clay, improperlysized bridging agents, etc., may be severely damaging. Thisis particularly the case if an open hole completion iscontemplated, as even a very shallow zone of mechanicaldamage of this type may severely compromise theproductivity of the wellbore.Tables 2 and 3 illustrate the results of whole mud leakoff

tests, conducted at only a moderate (1379 kPa) overbalance

pressure in a high permeability conglomeritic gas storagereservoir candidate. It can be seen from examining this data thatuncontrolled fluid losses occurred, resulting in deep invasionand large reductions in permeability due to permanentmechanical entrainment of drill and mud solids. Fig. 12illustrates the appearance of the pore system in such a situationwhere significant plugging of the high permeability matrix byinvaded mud solids is clearly apparent.

Proper lab testing (Ref. 8) can allow one to determine if thedrilling and completion practices proposed for a gas storagereservoir will provide suitable results prior to the actualexecution of the operation in the reservoir. If damage withconventional drilling and completion practices appears to besevere, several options are available to reduce the damage.These include:1. Underbalanced drilling and completion. If properly

executed, a UBD application may eliminate invasiveformation damage even in a pressure depleted highpermeability reservoir. Problems with this approach centeraround the ability to maintain the underbalanced conditionon a constant basis throughout the entire drilling andcompletion operation, as it has been demonstrated that evenrelatively short periods of overbalance pressure (such asthose associated with pipe connections, survey jobs, bit trips,frictional flow effects, etc.) can result in significant invasivedamage to the formation where all or a portion of the benefitof the UBD application may be lost (Ref. 9). In some cases,if very low bottomhole pressures are present, maintaining anunderbalanced condition with a nitrified fluid system maynot be possible. In this situation, even gas or mist drillingmay be overbalanced due solely to frictional backpressureeffects associated with high circulation rates required tomaintain sufficient annular velocity to facilitate adequatehole cleaning.

2. Specially designed overbalanced fluid systems containingcustom-designed bridging and fluid loss agents to rapidlycreate a sealing, competent filter cake on the formation faceto limit fluid invasion. The filter cake must be designed to beeasily removed by backflow or a non-invasive completiontreatment, or be shallow enough to be penetrated by amechanical stimulation treatment (such as open holeperforating, etc).

3. Repressurization of the reservoir by using existing gasinjection in wells (with the caveat that this may take anextended period of time and not be practical due to the type,location, condition and number of wells available) toincrease bottomhole pressure to a much higher level in orderto reduce overbalance pressure conditions. This may allowfor a more conventional drilling application, or easierapplication of one of the methods discussed in Point 1 or 2above.

ConclusionsNot all reservoirs are good candidates for gas storage. This


paper has outlined a number of the reservoir criteria that shouldbe evaluated and potential concerns which need to beinvestigated when considering if a reservoir is a suitablecandidate for gas storage. Issues that have been discussedinclude:

S Reservoir qualityS Competence and stability of sealing caprockS Reservoir stress issuesS Reservoir issues causing reduced injectivity and

productivityLab testing and evaluation using core analysis has been

illustrated to be an effective technique in diagnosing many ofthese reservoir issues and allows the accurate determination ofthe suitability of a reservoir as a gas storage candidate prior tothe cost and risk of actual project implementation.

AcknowledgmentsThe authors express appreciation to the management of HycalEnergy Research Laboratories for permission to publish thispaper and to Vivian Whiting for her assistance in thepreparation of the manuscript, figures and tables.

References1. Bietz et al, “Gas Storage Reservoir Optimization

Through the Application of Drainage and ImbibitionRelative Permeability Data”, CIM 92-75, presented at theCIM 1992 ATM, Calgary, AB. June 7-10, 1992.

2. Craft, B.C., Hawkins, M.F., Applied Petroleum ReservoirEngineering, Prentice-Hall, Inc. (1959).

3. Bennion, D.B., Thomas, F.B., “Recent Improvements inExperimental and Analytical Techniques for theDetermination of Relative Permeability from UnsteadyState Flow Experiments”, paper presented at the SPETechnical Conference at Trinidad and Tobago, June 27,1991.

4. Buckley, S.E. and Leverett, M.D., “Mechanisms of FluidDisplacement in Sands”, Trans., AIME, Vol. 146 (1942)107.

5. Welge, H.J., “A Simplified Method for ComputingRecovery by Gas or Water Drive”, Trans., AIME, Vol.195 (1952), 91.

6. Archer, J.S. and Wong, S.W., “Use of a ReservoirSimulator to Interpret Laboratory Waterflood Data”,SPEJ (Dec. 1973) 343.

7. Sigmund, P.M. and McCaffery, F.G., “An ImprovedUnsteady-state Procedure for Determining the RelativePermeability Characteristics of Homogeneous PorousMedia”, SPEJ (Dec. 1973) 343.

8. Bennion, D.B. et al, “Recent Advances in LaboratoryTest Protocols to Evaluate Optimum Drilling,Completion and Stimulation Practices for LowPermeability Gas Reservoirs”, SPE 60324 to be presentedat the SPE Rocky Mountain Regional Meeting, Denver,CO, Mar. 12-15, 2000.

9. Bennion, D.B. et al, “Underbalanced Drilling: Praisesand Perils”, SPEDE, Nov. 1998.


TABLE 1–SUMMARY OF CAPROCK GAS THRESHOLD PRESSURE TESTS

DirectionalOrientation

Intrusion Pressureto Gas (MPa)

Effective Permeabilityto Fluid (mD)

Comments

VerticalVerticalVerticalVertical

11.41Higher than 18.0

818

0.00005<0.000001

0.00021<0.000001

FailPassFailPass

TABLE 2–UNDERBALANCED/OVERBALANCEDMUD LEAKOFF TEST IN A GAS STORAGE RESERVOIR

CORE AND TEST PARAMETERS

Stack Length (cm)Diameter (cm)Effective Flow Area (cm2)Bulk Volume (cm3)Porosity (fraction)Pore Volume (cm3)Routine Air Permeability (mD)Test Temperature (EC)Gas Viscosity (mPa•s)Fixed Initial Water Saturation (fraction)Net Overburden Pressure (kPag)Mud Overbalance Pulse Pressure (kPag)Mud Underbalance Pressure (kPag)Rock Microfine Concentration (kg/m3)Rock Microfine Size (micron)

5.142.545.08

26.100.135

3.525888

460.0190

0.16026400

1379690

30<38

TABLE 3–UNDERBALANCED/OVERBALANCEDMUD LEAKOFF TEST IN A GAS STORAGE RESERVOIR

PERMEABILITY SUMMARY

TestPhase

Permeability(mD)

RegainPermeability (%)

Initial Gas Permeability @ Swi (Direction #1)

Underbalance Mud Circulation (Direction #2)

Regain Gas Permeability (Direction #1) @ 7 kPa Drawdown14 kPa Drawdown28 kPa Drawdown

Overbalance Pulse - 5 Minutes (Direction #2)

Regain Gas Permeability (Direction #1) @ 7 kPa Drawdown14 kPa Drawdown69 kPa Drawdown

3035

–

231125603035

–

774821860

*

–

7684

100

–

262728

* Baseline


Net Overburden Stress

Net Overburden Stress

spe 59738 detailed protocol for the screening and ... optimization/detailed... · spe 59738...

Documents