2

SPE 48952
Socbty of Petroleum Engineers

Net Pay Determination

for Primary and Waterflood Depletion Mechanisms

Inc.

W. M. Cobb, SPE, and F.J, Marek, SPE, William M. Cobb & Associates,

Copyright 199S, Sm,ie!y of Petroleum Engineers, Inc This paper was prepared for presentation 51 the 1998 SPE Annual Techm=l Exhibition lield in New Orleans, Louisiana, 27.3o September 1998, Conference and

This paper was selected for presentation by an SPE Program Committee following ravlew of information wntained b an abstract submitted by the author(s) Contents of the paper, as prasented, have not b%en reviewed by the Sociaty of Petroleum Engineers and are sublact to ~on by Ihe au~or7SJ The material, es presanted, does not necessarily reflect any positiOn of the Society of Petroleum Ewinaara, its titters, oc members. Papers presented at SPE ma-s are subject to publication rev~ew by Editorial Committees of tha Society of Petroleum Engineers. Electronic reprmfuction, distribution, or storage of any part of tfis paper for mmmercial purposes without the writtan mnsent of the Society of Petroleum Engineers is Pruhibled. Permission to raproduca in print is restricted to an abstract of not more than 300 -. ilfustra~ons may not be mpied The abstract must wntatn conspicuous acknowledgment of where and by whom the papar was presanted Write Librarian, SPE, P 0, B;x S33S3S, Rihardson, TX 7S083-38%, U.S.A, fax 01-972-952-9435.

variables including (1) fluid viscosity, (2) permeability distribution, (3) reservoir pressure differentials, and (4) reservoir drive mechanism (primary or waterflood). This paper provides guidelines for computing permeability cutoffs in primary and secondary recovery projects and defining those intervals which can be anticipated to be productive under various development scenarios. .41s0, improved techniques for converting permeability cutoffs to porosity cutoffs are discussed. Example calculations of permeability cutoffs for net pay determination in two water flood projects are presented. .. -. . . ... - ..
Introduction

Abstract

Net pay, as represented by thickness (h), is one of the most important parameters used in volumetric estimates of in-place hydrocarbons, well test interpretations, fluid injection analysis (water, gas, or enhanced recovery methods), reservoir engineering studies, or unitization procedures. Unfortunately, the petroleum engineering and geological literature provides very limited insight into how net pay should be computed. For a porous interval to be counted as net pay, the interval must contain hydrocarbons, be reasonably permeable, and for waterfloods, be lateraI1y continuous from injector to producer. The issue is what portion of this interval can be produced to yield meaningful quantities of hydrocarbons under the particular drive mechanism, Frequently, a permeability cutoff is selected and net pay is defined as being that part of the reservoir interval possessing permeability greater than the cutoff value. Due to the lack of core permeability measurements throughout all wells, the permeability cutoff cannot be directly used everywhere to differentiate between pay and nonpay rock. Consequently, net pay determination procedures usually relate the permeability cutoff to a log porosity cutoff. The resulting porosity cutoff, often coupled with a gamma-ray or other log cutoffs, is applied to available well log data to determine net pay on a well by well basis. It should be recognized that unless certain guidelines are followed, the porosity cutoff can result in very misleading net pay values because there is ofien not a strong direct relationship between permeability and porosity. This paper provides insight into the selection of the appropriate permeability cutoff to compute accurate porosity and other log cutoff values. Results indicate that the permeability cutoff for a particular reservoir is dependent on a number of 75

The goal of the Society of Petroleum Engineers[ definition of reserves is to identi~ those quantities of oil and gas which, by analysis of geological and engineering data, are anticipated to be economically recovered from known reservoirs from a given date forward. Based on this definition, it would seem logical that at a minimum, net pay should represent that portion of the reservoir containing oil and gas reserves which are anticipated to be economically recoverable. It is not the intent of this paper to discuss all the factors affecting the profitability of oil and gas production, but the analyst should recognize that economic profitability should be considered when estimating net pay. This paper focuses on technical factors which should be considered when computing producible net thickness as it relates to oil and gas reserves. A review of the petroleum engineering textbooks and technical literature shows that typically the distinction between pay and nonpay portions of a petroleum reservoir, whether containing gas or liquid hydrocarbons, is not discussed and is assumed to be known. For example, of seven basic petroleum engineering textbooks, only Pirson2 and CalhounJ have sections discussing pay determination methods. Pirsons text presents a 10-page discussion in a section entitled The Effective Pay. Both core and log methods are discussed. Two comments of particular interest are The two most important parameters that determine pay are fluid saturation and permeability, and of these two permeability is by far the controlling one. Yet it is difficult to select with assurance a permeability cutoff value; and its range in magnitude varies between O.I and 100 md, depending on the most common or average permeability value in the section, for the presence of high-permeability channels adjacent to otherwise tight zones may render the latter effective in yielding their oil. Pirson

2 .

W.M.Cobb, F.J. Marek

SPE 48952

goes on to quote a 1952 paper on carbonates by Archie that indicates Archie considered a 0.1 md permeability cutoff appropriate but the corresponding porosity cutoff ranged from 5 to 15 percent depending on the type of limestone and its pore structure. Calhouns text presented a single methodology drawn from a 1949 University of Oklahoma M,S. thesis. The methodology is based on creating plots of cumulative flow and storage capacity as fbnctions of permeability and porosity, respectively. A cuto~ was then chosen that arbitrarily eliminated 2.5 percent of both capacities, and for the example reservoir discussed, resulted in a permeability cutoff of 50 md and a porosity cutoff of six percent. While the latter cutoff appears reasonable, the former suggests some reasonably high permeability rock intervals are arbitrarily being excluded from further reservoir engineering calculations. The technical literature has discussed porosity and permeability cutoffs for specific reservoirs. Two technical papers45 have qualitatively discussed net pay, but there is no systematic method which describes how porosity or permeability cutoffs should be computed for different reservoirs under different drive mechanisms. In their important 1978 paper, George and Stiles6 describe their experience using a 0.1 md net pay cutoff for several west Texas carbonate waterflood projects. While this paper presents several innovative concepts for estimating net pay, there does not appear to be any technical justification for the 0.1 md cutoff. Also, it would appear the 0.1 md value is an air permeability rather than an oil permeability and is likely to result in an optimistic calculation of net pay for waterflood processes. In this paper, a variety of considerations for the selection of the producible net pay cutoff parameters are presented and discussed; however, this paper does not address the particulars of the net pay selection required for unitizations. While the principles discussed here can provide some general guidelines, historically each unitization has had its own unique choice of net pay cutoffs that resulted from negotiations.
Fundamental Considerations Determinations for Making Net pay

conditions. This range is also dependent on the reservoir pressure gradient (drawdown) and skin factor. While air permeabilities are the data generally avaiIable in large quantities, permeabilities at in-situ conditions (rocks saturated with brine and hydrocarbons and compacted at reservoir conditions) should be used when selecting the permeability cutoff. Significantly, air permeabiIities can overstate in-situ hydrocarbon permeability (such as (ko) Swir ) and usually lead to unrealistically low PorositY cutoffs which then yield optimistic values of net pay. Rocks within the reservoir interval which have permeabilities like those of the overlying and underlying rock intervals should be classified as nonpay. If the overlying rock is of such a quality that it is considered to be the reservoir seal (or nonproductive), then that quality and type of rock within the reservoir interval should be treated as nonpay. Pressure gradient is also important in estimating producible net pay. For example, a reservoir with a 2000 psi drawdown into the wellbore may produce reservoir fluids in significant quantities; whereas a reservoir with similar rock and fluid properties and similar drainage radius but only a 200 psi drawdown might produce negligible quantities of fluid. Once the interrelationship between fluid mobility, reservoir pressure gradient, and skin factor are understood, permeability should be used as the basis on which producible net pay and nonpay footage within the reservoir interval is determined. The use of other criteria such as porosity, water saturation (as a proxy for pore size distribution), or Vclay cutoffs comes only from the application of the permeability cutoff value to log data. The portion of the reservoir interval considered to be producible net pay may be different for various recove~ processes such as primary depletion versus waterflooding. This will be discussed in detail later in this paper. The sensitivity of the reservoir engineering calculations to the selected net pay criteria sho~ld be d~termined. For some reservoirs, the difference in original oil in place (OOIP) resulting from changing the permeability cutoff by a factor of two or more could be less than five percent; however, for other reservoirs this twofold change in cutoff might have the effect of altering the 00IP by 20 to 50 percent or more.
Mobility - The Appropriate Determination Starting Point for Net Pay

In its simplest form, producible net pay can be defined as those portions of a reservoir that are porous, reasonably permeable, and contain producible hydrocarbon reserves. To develop producible net pay cutoff criteria for a particular reservoir, Darcys law must be considered. That is, fluid mobili~ (permeability/viscosity), pressure gradient, and wellbore skin factor must be evaluated. These basic principles are discussed as follows. * The mobility (permeability/viscosity) of fluids within the rock is a major criteria for developing the net pay cutoffs. This means that gas reservoirs should have a lower net pay cri~~ * .by a factor of 100 or m-, * &l rwrvoirs because of the much lower viscosity of natural gas. A reasonable, but yet somewhat arbitrary, range for the mobility net pay cutoff is 0.5 to 1.0 md/cp at reservoir 76

While it is obvious no one would consider as pay a microdarcy reservoir containing an oil with a reservoir viscosity of 100 cp or more, there are a number of gas reservoirs currently being produced where permeabilities are in the microdarcy range. The reason for this is twofold: first, th~isc~si~ of gas, being on=e order of 0.02 cp, allows for significant gas mobiIity in very tight rocks with overall fluid mobility on the order of 0.5 md/cp. For the 100 cp oil, the rock permeability would have to be 50 md or more to result in the same fluid

SPE 48952

Net Pay Determination for Primary and Waterflood Depletion Mechanisms

3. :..

mobility. The second reason very tight gas reservoirs may be productive is that the production mechanism is fluid expansion and natural gas is by far the most compressible of the fluids found in hydrocarbon reservoirs.
Air Permeability Overstates In-Situ Permeability

yield some primary production. Yet due to such reservoir factors as permeability variation, relative permeability effects, or high gas saturations, these low permeability intervals will frequently not produce secondary oil.
Secondary (Floodable) Net Pay. For seconda~ operations, whether or not the tighter intervals contribute is no longer dependent on pressure reduction but is now dependent. on the ability of fluids to move into and pass through the rocks and reach the producing well. Under these conditions, lower permeability rocks which had contributed to production during primary depletion may not receive significant water injection volumes and hence are effectively nonpay. The permeability cutoff criteria for computing floodable net pay is dependent on a number of variables including: + Water cut economic limit + Permeability distribution (Dykstra-Parsons coefficient, V) + Gas saturation at the start of water injection q Injection and producing well pressure (pressure gradients) + Average reservoir pressure at the start of injection + Injection and producing well skin factors (restrictions to infectivity) + Mobility ratio q Distance from injector to producer (well spacing) q Lateral continuity between the injector and producer A review of the waterflood literature reveals a number of technical articles which use words such as pay, net pay, permeability cutoff, and porosity cutoff. These articles ofien provide numerical values for porosity cutoff such as 10 percent or permeability cutoff such as 1.0 md. Moreover, several articles4SG7 suggest that net pay is dependent on the reservoir drive mechanism (primary or secondary depletion). Unfortunately, there appears to be no systematic method or listing of important considerations in the determination of net pay cutoffs for either prima~ or waterflood depletion recoveries. In predicting infectivity and productivity for waterflood operations, it has been suggested the gross reservoir thickness be used in a multi-layered numerical simulation analysis. The idea is that the simulator will differentiate between the net pay and the nonpay. While this approach has merit in some cases, there are many waterfloods where numerical simulation has not been nor will be performed. More significantly, simpler reservoir engineering calculations for waterflood screening, feasibility, and surveillance procedures require accurate knowIedge of floodable reservoir pore volume (producible waterflood net pay). Cobb and Marek7 showed that in computing volumetric sweep efficiency in mature waterfloods, reliable estimates of floodable pore volume are essential. It is imperative that accurate producible net pay cutoffs be developed to monitor waterflooding activities. This paper describes two methods for estimating permeability cutoff to differentiate between producible net pay and other rock intervals in a waterflood for laterally continuous zones without inter-layer crossflow. The first method, referred to as the gas fillup time approach, is simple and can

Air permeability measurements of high permeability rocks (greater than 100 md to air) usually overstate the effective permeability to oil measured at irreducible connate water saturation at overburdened conditions by 10 to 30 percent. Part of this reduction is called the Klinkenberg effect and the remaining reduction is due to the presence of multiple fluids (water with oil or gas). At low air perrneabilities in the range of one to Hve-md, the reduction factor is significantly greater and may exceed a factor of five or more. Figure 1 is a plot of liquid to air permeability ratio versus air permeability for a number of sandstone core samples from two reservoirs. Figure 2 is a graph simiIar to Figure 1 except the permeability ratio is plotted on a logarithmic scale. It is noted that those core samples with an air permeability of about one md or less possess significantly lower Iiquid permeabilities.
Rock Properties of Overlying the Minimum Net Pay Criteria Intervals Should Establish

In most reservoirs, some core is taken from above and/or below the reservoir interval. The core analysis of these rocks provides a starting point, or minimum permeability and porosity, for developing the net pay criteria. A higher net pay criteria may be developed by reviewing the core analysis of the rocks within the reservoir interval.
Pressure Gradient is an Important Producible Net Pay Determination Consideration for

To calculate ariticiptied recoverable reserves, the pressure drop from the reservoir to the wellbore for a particular drainage radius should be considered. Pressure drop divided by distance, such as log(re/rw ), along with fluid mobility and skm factor determine whether hydrocarbons can be produced from a particular interval of rock in a reasonable time.
Net Pay Criteria Can Vary Depending on the Recovery

Mechanism
Primary Depletion Net Pay. For primary production, the net pay cutoff criteria will be lower than or equal to the cutoff criteria for waterflooding operations. Because primary production is usualIy based on pressure depletion concepts (except for strong water drive systems), any rock interval undergoing a pressure reduction wiIl contribute fluids resulting in production at the wellbore. This is why high mobility gas with its high compressibility can result in a considerable contribution to production from rock intervals that possess very low permeability at in-situ conditions. Further, low permeability oil reservoirs with low oil viscosity (high mobility fluids) will

77

W.M. Cobb, F.J. Marek

SPE 48952

referred to as the water cut approach, is technically more correct but requires additional data and is more complicated. Both calculations require engineering and economic judgments, Permeability Cutofl Based on Gas Fillup Time. To understand this method, consider Figure 3 which depicts a crosssection of a single layer between an injection and production weIl early in the life of a waterflood. This figure illustrates the saturation distribution between an injection and production well prior to the time of gas fillup. This saturation distribution has been described in detail in References 8 and 9. Oil production resulting from water injection in this layer does not commemce until the Ieading edge of the oil zone reaches the producing weI1. When the oil front initially reaches the producing well, gas fillup is achieved. Figures 4 and 5 depict an areal view of waterflood performance before and at the time of gas fillup for a waterflood project implemented with a free gas saturation, Sg, present at the start of injection. The time to fEiC~-~Up, tf, is dependent on the barrels of water injected to fillup, Wif, and the water injection rate>iW. That is: tf = Wi+365 * i\v ........................................................(l) where: * sg ...........................c....................(2) Vp () layer To compute a permeability cutoff, the engineer/geoscientist must select the maximum time he would be willing to wait for fillup to occur. For example, if the engineer/geoscientist is willing to wait for a period of 15 years from the start of water injection to initial secondary production response, a minimum permeability value which will yield production during the time period can be computed from Equation 1, This minimum permeability becomes the permeability cutoff. All intervals possessing permeability greater than the calculated permeability cutoff would respond in less than 15 years. fimple Calculation. The fillup time method of permeability cutoff is illustrated using a five-spot pattern and the other parameters presented in Table 1 along with the oil/water relative permeability curves shown in Figure 6. A maximum geological layer fdlup time of 15 years is selected. Permeability cutoff is calculated using Equation 1 and the water injection rate described in Equations E.5 and E.8 of Appendix E by Craigg. Table 2 summarizes the permeability cutoffs for several values of gas saturation, pattern size, and injection wellbore skin factor. These cutoffs are the permeability to oil measured at the irreducible connate water saturaA review of this table clearly shows that the tion, (ko)swk. Wif= permeability cutoff depends on a number of reservoir variables. The producibility cutoff is independent of the layer thickness since thickness appears in both the numerator and denominator of Equation 1 and therefore cancels out. This technique is simple and straightforward. However, it does not take into account the injection and production performance of the more permeabIe layers, In fact, it is possible that injection into the more permeable layers will have caused 78

the producing well to reach a water/oil ratio or water cut that exceeds the economic limit before the limiting filIup time (such as 15 years) is reached. Therefore, the permeability cutoff based upon fillup time should be considered as a preliminary estimate. Also note that if the waterflood is initiated prior to the development of a free gas saturation (Sg = 0), this method is not applicable and the water cut method described in the next section should be used. Permeability Cutoff Based on Water Cut. A more complete method for computing permeability cutoff considers waterflood performance for all layers. Consider Figure 7 which depicts a three-dimensional view of one quarter of a noncommunicating multi-layered five-spot pattern. For convenience, permeability decreases from the top layer to the bottom layer. It can be seen that water breakthrough has occurred in the top two layers. For layers three through eight, gas fillup has occurred and possibly water breakthrough has taken place, whereas neither gas fillup nor water breakthrough has been achieved in layers nine and ten. At this instant in time, the water cut from the producing well has reached an assumed economic limit of 95 percent. The permeability cutoff can be defined as the permeability of the lowest permeable layer which has not produced a meaningful amount of waterflood oil at the water cut economic limit. For example, assume that at the water cut economic limit, a layer must produce at least five percent of its mobile waterflood oil pr it will not be counted as watertlood net pay. An estimate of the permeability cutoff using the water cut method can be calculated by setting up a simple multi-layered waterflood model. The model can be an analytical model such as the steady-state model developed by Craig-GeffenMorse (CGM) described in Appendix E of SPE Monograph 3 or it can be a more sophisticated black oil simulator. Water is injected into the multi-layered model until the economic limit water cut is reached. Each layer can be analyzed to determine if it produced significant amounts of waterflood oil. A permeability below which no significant waterflood oil is produced becomes the permeability cutoff. This method takes into account most of the factors which affect waterflood behavior including stratification, areal sweep, fractional flow, variable infectivity, mobility ratio, and skin factor. By using a numerical simulation model, it would be possible to consider other variables such as inter-layer crossflow, counter-current imbibition, and layer position within the geological interval. The permeability cutoff using the fiample Calculation. water cut method is illustrated with the five-spot pattern described in the previous example (Table 1). In the water cut method, 20 equal thickness layers were incorporated into a steady-state CGM model. The average reservoir permeability is 20 md and the gas saturation is 10 percent. Waterflood analysis is performed using Dykstra-Parsons permeability variations of 0.60, 0.70, and 0.80. The permeability cutoff is the value below which less than five percent of the waterflood moveable oil would be recovered from the layer at a total pattern water cut of 95 percent. The cutoffs presented in Table 3

SPE 48952

Net Pay Determination for Primary and Waterflood Depletion Mechanisms .-.
q

..

show how they depend on Dykstra-Parsons permeability variation and gas saturation. Lateral Continui@. It is well recognized 234 that for a permeable interval to be included as part of waterfloodable net pay, the interval must be laterally continuous between the injection and production wells. Reservoir continuity is generally less than 100 percent with carbonate reservoirs frequently showing more discontinuity than sandstone. Reference 6 indicates that for several West Texas carbonates, lateral continuity is in the range of 70 to 80 percent for well spacing as low as 20 acres per well. One of the more difficult tasks of computing net pay in waterflood operations is estimating lateral continuity. There is no simple remedy for this problem, but estimates of reservoir continuity may require detailed geological cross-section analysis, reservoir outcrop studies, pressure transient te~ing, material balance calculations, and regular injection and production testing during the waterflood. Barbe and SchnoebelenlJ review a field study in which they utilize geological correlations, pressure transient testing, and regression analysis of infill drilling performance to estimate continui~ in a carbonate water flood.

Relationship Between Permeability and Porosity Cutoffs

Permeability cutoff is usually related to a porosity cutoff for net pay determination in reservoir applications. Porosity cutoff is usuaIly obtained from a semi-log graph which relates permeability to porosity such as the graph shown in Figure 8. UnfotiateIy, even if the appropriate permeability cutoff is computed, this graph usually presents two important analysis difficulties. First, the permeability values usually represent air permeability, ka, and consequently overstate reservoir permeability. It is recommended that the permeability data used in preparation of this type graph should be (ko)s Wir because this usually represents the reference or base permeability used in relative permeability curves. Moreover, this permeability value, when coupled with relative permeability data and waterflood principles, can be used to compute water injection rates. Figure 9 presents a graph of (ko) Swir versus porosity. Figure 10 presents a comparison of ka and (ko)Stir versus porosity. Using a 10 md permeability cutoff, the porosity cutoff varies from 15.6 percent when using ka to 18.3 percent when using (ko)s
wir

A second complication related to the use of graphs such as

those shown in Figures 8 or 9 is data scatter. George and Stiless provide two significantly different techniques for eliminating data scatter. Because of their importance and for completeness, the methods are reviewed in this paper, Both techniques assume core data representative of the reservoir are available and a reliable permeability cutoff can be obtained. The first technique results in a single porosity cutoff value. The method can be described according to the following procedure. * Step I - Define actual pay as being all cored footage possessing a permeability greater than the permeability cutoff.

Step 2- Compute the pore volume ($h) for the actual pay in the core data base. + Step 3- Through trial and error methods, compute the porqsity cutoff value, when applied to the total core data base, which gives a pore volume equal to the pore volume of the actual pay from Step 2. q Step 4- The trial and error method can be more easily utilized by arbitrarily selecting porosity cutoffs such as two, four, six eight, etc. percent. For each value of porosity cutoff, compute the pore volume and compare the calculated pore volume with the actual pore volume from Step 2. George and Stiles second technique for eliminating scatter requires greater effort but does not utilize a single porosity cutoff. Instead, the method utilizes a weighting factor which ranges from zero to one, depending upon the scatter in the permeability versus porosity relationship. The method is summarized as follows. Step 1 - Subdivide the core into porosity ranges such as zero to two percent, two to four percent, four to six percent, etc. Step 2- Within each porosity range, compute the fraction of core samples within that range which have a permeability greater than or equal to the permeability cutoff which is defined as a weighting factor. Step 3- Plot the fraction of core samples within the porosity range (weighting factor) versus the average porosity within the porosity range. Some scatter will likely exist within this plot, but a straight-line relationship can usually be fit through points. Step 4 - Each foot of rock within each well can be weighted depending on the porosity of that foot by using the straight-line weighting factor relationship. The total weighted feet for each zone or each well can then be obtained. To illustrate this second technique for eliminating data scatter, consider the core data presented in Figure 9. A 10 md permeability cutoff has been calculated for this oil reservoir to compute waterfloodable pay (the 10 md cutoff appears to be high until it is recognized that the overall average reservoir permeability is about 125 red.) An inspection of the data indicates none of the data points with a porosity less than 16 percent have a permeability greater than the 10 md cutoff. On the other hand, all of the core data with porosity greater than 24 percent have permeability values greater than 10 md. The porosity between 16 and 24 percent is subdivided into intervals shown in Table 4. Within each porosity interval, the fraction of core samples having a permeability greater than 10 md is computed. This fraction is referred to as a weighting factor. Weighting factor is plotted versus porosity as shown in Figure 11. For each foot of rock with a porosity less than 16 percent, the weighting factor is zero. For each foot of rock with porosity greater than 24 percent, the weighting factor is unity. For intermediate values of porosity, each foot of rock is weighted based on the weighting factor from Figure 11. For example, if a foot of rock possesses a porosity of 20 79

W.M. Cobb, F.J. Marek

SPE 48952

percent, the foot is weighted by a factor of 0.57 and is therefore assigned 0.57 net feet. Consider the data in Figure 9. After applying a 10 md cutoff, the computed pore volume is 75.6 percent of the computed pore volume without a cutoff. Moreover, the 10 md cutoff pore volume is 77.6 percent of the pore volume obtained when using a I md cutoff. Simple calculations show the reservoir rock with less than 10 md will yield primary production, but those same rocks will not contribute any appreciable production during waterflooding. This difference in producible net pay gives rise to a drainable net pay versus a floodable net pay.J14
Reservoir Engineering Calculations Should Be Made to Determine the Sensitivity to the Selected Net Pay Criteria

Nomenclature

A= pattern area, acres net pay, feet h= water injection rate, barrels per day iw ka = permeability to air, md (ko)Swir= permeability to oil at irreducible connate water,
k50 kro kW P

Pwi Pwrf Si
Sp Sg so Swc g wbt

. median reservoir permeability, md relative permeability to oil, dimensionless . relative permeability to water, dimensionless average reservoir pressure at start of injection, psi . bottomhole injection pressure, psi bottomhoIe producing pressure, psi injection well skin factor, dimensionless production well skin factor, dimensionless gas saturation at start of waterflooding, fraction oil saturation, fraction connate water saturation at start of waterflood,

md

Whatever permeability cutoff is used to compute net pay, the sensitivity of the OOIP (and subsequent recovery) calculations to the chosen permeability cutoff should be computed. For some reservoirs a considerable range of net pay cutoffs wilI result in essentially identical recovery estimates. However, for some reservoirs with a much narrower distribution of permeabilitim and also generally composed of poorer quality rocks, the recovery estimates wiII be significantly affected by the net pay permeability cutoff criteria,

fraction
average water saturation in water zone up to gas filIup time, years Dykstra-Parsons permeability variation, pore volume of geological layer, reservoir barrels shale volume, fraction volume of injected water necessary to reach gas porosity, fraction water viscosity, cp oil viscosity, cp

breakthrough, fraction

tf
v Vp

dimensionless

Conclusions A search of the technical literature indicates little informa-

clay Wif o pw Po

tion is available regarding guidelines for selecting net pay criteria. The mobility (permeability/viscosity) of fluids within the rock is a major criteria for developing the net pay cutoffs. This means gas reservoirs should have a much lower net pay criteria, perhaps by a factor of 100 or more, than oil reservoirs because of the much lower viscosity of natural gas. Permeability shouId be the basis which distinguishes between net pay and nonpay and defines those portions of the reservoir which wilI be producible under a particular development scenario. Net pay cutoffs will depend on the reservoir drive mechanism (primary versus waterflooding) and will generally result in lower vaJues for primary depletion. Primary net pay cutoff criteria should give consideration to factors such as fluid mobiIity, reservoir pressure gradient, and well completions. Secondary net pay cutoff should give consideration to mobility ratio, permeability variation, relative permeability, gas -mtion, injection and production well pressures, skin factor, well spacing, lateral continuity, and water cut economic limit. Two new procedures are presented which provide guidance and valuabIe insight in the determination of producible net pay permeability cutoff for use in waterflood screening, feasibility, and surveillance calculations. 80

fillup, barrels

Acknowledgment

We thank Dr. H.R. Warner, Jr. of ARCO for valuable discussions and helpful comments in the preparation of this manuscript,
References

1, Petroleum Reserve Definitions, Society of Petroleum Engineers and World Petroleum Congress, approved March, 1997, SPE,Richardson,TX. 2. Oil Reservoir Engineering-2nd Edition, S.J. Pirson, McGrawHill Book Co. Inc., New York (1958) 441-451.
3. Calhoun, J.C. Jr.: Fundamentals of Reservoir Engineering, University of Oklahoma Press, Norman, Oklahoma ( 1960) 168-171. 4. Snyder, Richard H.: A Review of the Concepts and Methodology of Determining Wet Pay , paper SPE 3609 presented at the 1971 SPE 46th Annual Fall Meeting of AIME, New OrIeans, Oct. 3-6. 5. Vav~ C. L., Kaldi, J. G., and Sneider, R. M.: Geological Applications of Capillary Pressure: A Review, AAPG Bulletin Volume 76, (June 1992) 840-850. 6. George, C.J. and Stiles, L. H.: Improved Techniques for Evaluating Carbonate Waterfloods in West Texas, JPT (November 1978) 1547-1554.

SPE48952

Net Pay Determination for Primary and Waterflood Depletion Mechanisms

7. Cobb, W.M. and Marek, F. J.: Determination of Volumetric Sweep Efficiency in Mature Waterfloods Using Production Data. paper SPE 38902 presented at the 1997 SPE Annual Technical Conference and Exhibition, San Antonio, Oct. 5-8, 8. Craig, F. F., Jr,: The Reservoir Engineering Aspects of Waterjlooding, Monograph Series, SPE, Dallas (1971)3. 9. WiI1hite, G.P.: Waterflooding, Textbook Series, SPE, Dallas (1986) 3. 10. Craig, F. F., Jr., Geffen, T.M. and Morse, R. A,: Oil Recovery Performance of Pattern Gas or Water Injection Operations from Model Tests, Trans. AIME (1 955) 204. 1 I. Ghauri, W.& Osborne, A. F., and Magnuson, W. L.: Changing Concepts in Carbonate Waterflooding - West Texas Denver Unit Project - An II1ustrative Example, JPT (June 1974) 595-606. 12. Stiles, L.H.: Optimizing Waterflood Recovery in a Mature Waterflood, The Fullerton Clearfork Unit, paper SPE 6198 presented at the 1976 SPE Annual Fall Technical Conference and Exhibition of AIME, New Orleans, Oct. 3-6. 13. Barbe, J-A. and SchnoebeIen, D.J.: Quantitative Analysis of Infill Performance: Robertson C1earfork Unit, JPT (December 1987) 1593-1601.

14. Christman,P.G.: Modelingthe Effects of Infill Drilling and Pattern Modificationin DiscontinuousReservoirs,SPE Reservoir Engineering (February 1995) 4-9.

81

8 W.M. Cobb, F.J. Marek SPE 48952 __________________________________________________________________________________________________________

TABLE 1 HYPOTHETICAL FIVE-SPOT PATTERN USED TO ESTIMATE PERMEABILITY CUTOFFS Pattern Area p pwi wi pwf wf Sii Sp p w w o o 80 Acres (40 Acre Well Spacing) 1000 psi at start of water injection 3600 psi 400 psi 0.0 0.0 15 percent 0.6 cp 3.0 cp

TABLE 2 PERMEABILITY CUTOFF USING THE GAS FILLUP METHOD AND OTHER FACTORS FROM TABLE 1 FOR A 15-YEAR RESPONSE TIME

Gas Saturation percent 5.0 10.0 15.0

(ko)Swir, md oS wir 80 Acre Pattern 40 Acre Pattern (40 Acre Spacing) (20 Acre Spacing) Sii = 0 Sii = -4 Sii = 0 Sii = -4 1.25 2.50 3.77 0.64 1.27 1.92 0.59 1.20 1.80 0.29 0.58 0.87

SPE 48952 Net Pay Determination for Primary and Waterflood Depletion Mechanisms 9 ___________________________________________________________________________________________________________

TABLE 3 PERMEABILITY CUTOFF USING THE WATERCUT METHOD AND OTHER FACTORS FROM TABLE 1 AT A 95 PERCENT WATER CUT ECONOMIC LIMIT

80 Acre Pattern k50 = 20 md, Sii = 0 50 Dykstra-Parsons, V 0.60 0.70 0.80 Sg = 10% g 1.10 3.30 5.60 Sg = 0% g 0.24 0.71 1.20

TABLE 4 FRACTION OF CORE SAMPLES WITH PERMEABILITY GREATER THAN THE PERMEABILITY CUTOFF FOR DIFFERENT POROSITY RANGES Fraction of Core Samples with Permeability Greater Than 10 md (Weighting Factor) 0.00 2/6 = 0.333 5/10 = 0.500 11/16 = 0.688 12/16 = 0.750 1.00

Porosity Range percent Less than 16 16 to 18 18 to 20 20 to 22 22 to 24 Greater than 24

10 W.M. Cobb, F.J. Marek SPE 48952 __________________________________________________________________________________________________________ FIGURE 1: RATIO OF LIQUID/AIR PERMEABILITY VERSUS AIR PERMEABILITY FOR TWO SANDSTONE RESERVOIRS
Liquid / Air Ratio 0.8 Reservoir A Reservoir B 0.7

0.6

0.5

0.4

0.3

0.2

0.1

0 0 200 400 600 Air Permeability, md 800 1000 1200

FIGURE 2: RATIO OF LIQUID/AIR PERMEABILITY VERSUS AIR PERMEABILITY FOR CLASTIC AND MIXED LITHOLOGY CORES
Liquid / Air Ratio 0.8

0.7

0.6

0.5

0.4

Clastic Mixed

0.3

0.2

0.1

0 0 200 400 600 Air Permeability, md 800 1000 1200

Injector

FIGURE 3: CROSS-SECTION VIEW OF A SINGLE LAYER DEPICTING FLUID SATURATION DISTRIBUTIONS PRIOR TO GAS FILLUP

Producer

Water Zone

Oil Zone

Unaffected Zone

So = 1.0 - S wbt o

So = 1.0 - S wc So = 1.0 - S wc o o - Sg

SPE 48952 Net Pay Determination for Primary and Waterflood Depletion Mechanisms 11 ___________________________________________________________________________________________________________ FIGURE 4: AREAL VIEW OF FLUID SATURATIONS FOR A SINGLE LAYER WITHIN A 5-SPOT PATTERN PRIOR TO GAS FILLUP

Producer

Sg So S wc
Injector

So = 1.0 - S wbt o So = 1.0 - S wc o

FIGURE 5: AREAL VIEW OF FLUID SATURATIONS FOR A SINGLE LAYER WITHIN A 5-SPOT PATTERN AT GAS FILLUP

Producer

Injector

So = 1.0 - S wbt o

So = 1.0 - S wc o

12 W.M. Cobb, F.J. Marek SPE 48952 __________________________________________________________________________________________________________ FIGURE 6: EXAMPLE OIL/WATER RELATIVE PERMEABILITY CURVES

Relative Permeability 1.2

0.8

k ro
0.6

0.4

0.2

k rw

0 0 0.2 0.4 0.6 0.8 1 1.2

Water Saturation, percent

FIGURE 7: THREE-DIMENSIONAL VIEW OF A NON-COMMUNICATING MULTI-LAYER 5-SPOT PATTERN WITH A PRODUCING WATER CUT OF 95 PERCENT PRODUCER AT 95 PERCENT WATER CUT

INJECTOR

WATER ZONE

OIL BANK

UNAFFECTED (GAS) ZONE

SPE 48952 Net Pay Determination for Primary and Waterflood Depletion Mechanisms 13 ___________________________________________________________________________________________________________

FIGURE 8: AIR PERMEABILITY VERSUS POROSITY FOR RESERVOIR C

Air Permeability, md 2500

2000

1500

1000

500

0 0 500 1000 Porosity, percent 1500 2000 2500

FIGURE 9: OIL PERMEABILITY AT IRREDUCIBLE WATER SATURATION VERSUS POROSITY FOR RESERVOIR C
Oil Permeability, md 2000

1500

1000

500

0 0 500 1000 Porosity, percent 1500 2000

14 W.M. Cobb, F.J. Marek SPE 48952 __________________________________________________________________________________________________________

FIGURE 10: COMPARISON OF AIR AND OIL PERMEABILITY VERSUS POROSITY FOR RESERVOIR C

Permeability, md 2500 Air Permeability Oil Permeability

2000

1500

1000

ka (ko) S wir

500

0 0 500 1000 Porosity, percent 1500 2000 2500

FIGURE 11: FRACTION OF CORE SAMPLES WITHIN A POROSITY RANGE (WEIGHTING FACTOR) WITH PERMEABILITY GREATER THAN CUTOFF VALUE OF 10 MD FOR RESERVOIR C

Fraction of Core Samples Greater Than 10 md 0.8

0.7

0.6

0.5

0.4

0.3 16 17 18 19 20 Porosity, percent 21 22 23 24