## Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

SIXTEENTH

ANNUAL

LOGGING

SYMPOSIUM,

JUNE

4-7,

1975

**THE DISTRIBUTION OF SHALE IN SANDSTONES AND ITS EFFECT UPON POROSITY
**

BY

E. C. Thomas Shell Oil Company New Orleans, Louisiana S. J. Stieber Butler, Miller & Lents Houston, Texas

ABSTRACT

A shale volume parameter derived from the response of the gamma ray log in shaly sands is often used to correct the responses of other logs for the effects of the shale. The correlation of a garmna ray parameter to shale volume is usually presented as one direct relationship. However, because the shale can be distributed through the sand in different ways, such as laminated, dispersed, structural, or any combination of these, one may expect vasying gamma ray responses dependent upon geometry. We show that each of these configurations can involve a different response from the gamma ray and this variable response can be used to determine the shale configuration. A gamma ray parameter vs. porosity crossplot can then be used to determine shale configuration, sand fraction and sand porosity; the derived equations can be used for implicit solution when processing digital log data. These configurational data can be combined with electric log parameters to determine the oil saturation of the net sand. These log derived parameters are herein verified by direct comparison to rubber sleeve core data.

The sparcity of porosity logs in many of the older South Louisiana fields often necessitates that other available log data be correlated to porosity. One log which is generally available is the gamma ray collar log which is used to perforate the well. Thus, empirical gamma ray response Density log porosity correlations in a particular reservoir are developed using recent penetrations, and this correlation is applied to the older gamma ray logs to determine a reservoir-wide porosity distribution. This simple approach is fruitful for reservoirs which are found in Tertiary sand-shale sequences and usually will not apply to more complicated mineralogies involving carbonates. An example of this type of correlation is shown in Figure 1.

-l-

T

The shale fractions we wish to determine are a function of volume. Within the interval investigated. The gamma ray responds to the number of radioactive events in a material and thus its mass. 3. There are only two rock types. we have chosen to use shale rather than clay minerals. Dispersed . 5. thus. a high porosity "clean" sand and a low porosity "pure" shale. thus. 2. to quantitatively determine shale content and distribution we have developed a simple mathematical model which relates gamma ray response to shale distribution and concentration. there is no change in shale type and the shale mixed in the sand is mineralogically the same as the "pure" shale sections above and below the sand. JUNE 4-7.shale on the sand grains. Thus. The observed in situ porosities are generated by mixing the two.SPWLA SIXTEENTH ANNUAL LOGGING SYMPOSIUM. The five main assumptions in the model are the following: 1. Laminated . if sorting or mineralization are the dominant factors in sand porosity variation. or pore filling. even though it is the clay -2- . 3.sand sized shale particles in load-bearing positions within the rock. 4. We assume for the Tertiary basins that both sands and shales have comparable grain densities. Of course there can be any combinations of these categories. There are three broad categories which describe how shale can be distributed in a sand: 1. However. then no simple gamma ray response correlation to porosity should be found because the gamma ray responds to the presence or absence of radioactive minerals. Constant background radiation will be assumed to be present in all measurements. The components of the system are sand and shale. Counting yields will not change as rock types are intermixed. the radioactivity will be proportional to volume. We must learn how the shale is distributed in the sand. 1975 It is assumed from the above that shale is the main destroyer of sand porosity and it is therefore reasonable to expect the gamma ray to correlate to porosity. 2.layers of shale within the sand. for this distribution governs the productivity. Structural .

expressed as Ra* = Ya Aa (1) where Y is the counting yield. Dispersed shale (pore filling) Place some shale in the sand's pore space. except for heavy minerals. We define the fraction of the total volume occupied by this shale as Xb.) Thus. The grain volume fraction. 1975 which contains the bulk of the radioactive material.) Yb ’ fib . We will elaborate on the error caused by this assumption in Appendix 1. We define the subscripts a = sand and b = shale. Let's now mix together quantities of the porous sand and shale. the pure sand rock porosity. Express Va in terms of the fraction of the total volume of rock. (This is not rigorously true but. JUNE 4-7. Of course. we observe Rb = (I-&. The differentiation between sands and shales begins as the particles settle at differing rates according to their size and transport energy and not mineral type. 1. we introduce a void space. 5).SPWLA SIXTEENTH ANNUAL LOGGING SYMPOSIUM. Take a solid block of sand material and place it near an instrument capable of measuring the gamma rays emitted. we observe a counting rate. Now if we crush the solid and build a porous rock with the pieces. = (l-@a) Ya' Aa (2) where the prime was added to denote a change in counting yield because of a geometry change in the material counted. very vesicular minerals. grainosity. or colloids the differences in density are not dominant. There are three independent ways to place the shale in the sand block and these are treated separately. then V. Vt. If we repeat the same crushing and construction experiment using the shale material. Ya. and Xa = (l-@a). is then Xa. this will not be true for the diagenetic alteration of feldspars into clay within the sand stratum. the counting rate observed is R. We believe shaly sands and shales are mineralogically similar because both facies are derived from the same source material. we feel the porosity destroying material introduced into a sand stratum will be of the same composition as the shales above and below the sand stratum. - 3 - T ./Vt = @a. If we now repeat the gamma ray detection experiment using this new porous rock. Ra. A is the activity and the asterisk indicates that fla= 0. carried by the same river and emptied into the same basin. and assume the counting yield will not be significantly altered (Assumption No.

Ra For simplicity lets define Rb ' = Rb . which correlates to porosity. JUNE 4-7.(Ra + Xb Rb) Rb . (2) and (3) into (4) we get (4) If we substitute R=R. the bulk volume fraction occupied by shale. y. The value of y in a sand stratum is defined in terms of the above parameters as Rb - Y= R Rb . x @b. y and Xb then becomes xb = Y= (7) (8) The relation between l-y 5 when Xb is pore filling. One point to note is that the minimum porosity available to the dispersed model is when shale completely fills all the original pore volume of the sand or when Xb = @a. dispersed = 0. 1975 Now if we count the gamma rays. then Rb .Ra (6) where Rb is observed in the "purest" nearby shale. in terms Of Xb.Ra which is a measure of how radioactive the sands are.+ Xb Rb to (5) or simply the observed count rate increases proportional the amount of shale added to the pore systems. we observe R= (1-0a) Ya’ Aa + Xb (l-0b) yb' Ab.SPWLA SIXTEENTH ANNUAL LOGGING SYMPOSIUM. 0 min. Substituting (5) into (6). (10) - 4 - . and Ra observed in the nearest "clean" sand. If we are to apply this concept to borehole gamma ray measurements we must relate a defined parameter.

Thus R = (l-Ba) Ya'Aa . Thus. Laminated model Now when we add shale to the sand stratum we must replace sand and its associated porosity. get (12) = Rb - (1-Xb)(Rb-Ra)' (13) (14) (13) into (6).Xa (l-@b) Yb'Ab + Xa (I-8a) Ya'Aa. Or R= (l-@a> Ya'Aa . T -5- . Rb . then equation (4) for the counting rate is not valid because we must remove sand grains to add shale. the y to Xb relationship for structural shale is identical to that for the laminated model. Thus the porosity increases with the amount of shale porosity that we add in place of the solid sand grain. fla. Substituting (2) and (3) into (15) yields (15) R = Ra + Xb (Rb-Ra)* If we substitute (16) into (6) then . the sand fraction. added is Xb./= Rb . (11) Substituting (2) and (3) into (11) we R = Rb .Ra .SPWLA SIXTEENTH ANNUAL LOGGING SYMPOSIUM. JUNE 4-7.Xb (1-@a) Ya'Aa + Xb (l-@b) 'b'% (18) which is the same as (15). Using Xa = 1-Xb.(l-0. y = 1-Xb = Xa . Structural When we add shale in structural positions we remove sand grains only. 3. Thus. . 1975 When Xb > fla.with shale to keep a constant The amount of sand removed and shale total volume of material.Ra (16) (17) Thus for the laminated model y is the sand fraction. R = (1-0b) Yb'Ab .XaRb + XaRa. We derive this equation by starting with pure shale (Xb = 1) and work back to Xb = @a. (12) rearranges to R Substituting 2.) Ya'AaXb + (l-P)b)YbAbXb.Xb (Rb-Ra) = 1-Xb = X.

0&& = 8a and when Xb = 8a. Substituting (9) into (19) (1-Y) 5 (19) 0dj.in this case we remove both the porosity and grainosity of the sand and replace it by the shale porosity and grainosity. we can construct a theoretical y-porosity relationship based on these three models. (22) *This value is derived from eq. when Xb > fla. (20). The porosity relation for the region when Xb > 0. See Appendix 1. When Xb = 0.xb (1-gb) for Xb < @a. (22) is limited as Ymin in eq.we start with the total porosity of a or @a. - 6 - . 0dis = 0b and this is a fixed end point.s = @a - (l-f&). Laminated . We then add shale in the pore space. 1. The end points of this equation are easily fixed and are independent of any proposed relation between Xb and y. The relation is linear with y. thus reducing the total porosity by the value of sand added. Ibdis = Ob . 1975 Now that we know the relation between y and shale fraction. thus decreasing the porosity by the amount of grainosity of the shale. the available space for shale.25. is derived just as is the radiation relationship by starting with pure shale and adding sand grains which have no porosity. When y = 0. @&a = 0a@b. The actual value of y at 0~1~ is difficult to define absolutely because of Assumption No. The error caused by this assumption is discussed in Appendix 1.xa0b = 0b . Substituting (14) into (21) yields (21) Q)dis = @b-Ylbb. 0dis = @a . JUNE 4-7. However this number is difficult to fix as stated before. (20) A reasonable number for 5 in South Louisiana sands is 1. Dispersed (pore filling) . (8) using Rb = 100 API units and Ra = 20 API units. 2. 5.(l-xb) @b.h Thus the relation between these end points is linear with y.SPWLA SIXTEENTH ANNUAL LOGGING SYMPOSIUM. ymax in eq.

This startling result is obvious upon inspection of the model and is the consequence of replacing the solid quartz grain with a corresponding volume of porous shale. These end points are independent of y-xb relation and are firm. and when y = 0. the counting yield will be different than for pure shale.SPWLA SIXTEENTH ANNUAL LOGGING SYMPOSIUM. then R = Xa (l-0b) Yb'Ab = (I-@a) Rb* (28) into (6) Rb (@a) Rb . and replace it with an equivalent bulk volume fraction of shale porosity. (23) Substitute (17) into (23) O1am = Y@a + (1-Y) 0b- (24) The end points are easily fixed when y = 1. However. JUNE 4-7.. 0str = Oa + xb @bSubstitute (18) or (17) into (25) Ostr = fla+ (l-y) 0b. 1975 ham = 0a . (25) (26) The maximum amount of structural shale which can be added equals the grainosity of the sand or Xmax = l_Ymin = l-P)a.) + xb (0b). See Appendix 1. only sand grainosity. But if we use Assumption No.in this case we remove no sand porosity. T 7 - . again the consequences of Assumption No.xb (8. (28) Then substituting Y" Rb . Also equation (24) demonstrates a linear relation between the two end points. (27) Notice that 0 increases linearly with decreasing y.) (30) Thus (27) and (30) are not equal.R. 5. Structural . 5. @Iam x P)b.(1-qa)Rb = Rb-Ra (29) Substituting Y = (8) into (29) * 5 (0. @lam T 0. The radiation at the maximum is a simple relation because no sand grains remain. 3.

thus giving misleading results. y = 0 at 15%. while a corresponding disparity appears in the lower laminated section.85 point.85. The porosity of the sand laminae anywhere along a given line is read at its intercept on the dispersed shale model line. x @b or 0. Figure 4 shows the oil saturation and porosity calculated for the entire interval. 24. Note the good sand at the top and the extremely laminated material at the bottom. but of course it can be done algebraically. 20 for y. We can then construct a y-density porosity crossplot triangle limited by these data.(20). This is done by fixing the y = 1 point at 8 = 33%. Isoporous lines are added to Figure 2 by constructing lines which connect the y = 0 apex to the dispersed shale model line.588. we start at the o.SPWLA SIXTEENTH ANNUAL LOGGING SYMPOSIUM.0 apex. 1975 In reality we do not have any of these pure models but rather we have a combination of the three. - 8 - . where clean sand porosity is 33% and pure shale porosity is 15%. then proceed up an isopor to the intercept on the dispersed shale model line of Id= 29. shown in Figure 2.4%. using 5 = 1. The FDC 8 is calculated using measured grain densities and fluid densities from the core adjusted to reservoir temperature and pressure. sand porosity and shale distribution.0495. Using Figure 2. and(24). The percent laminations is then read from the laminated model line as the percentage of the laminated model line substended by the isolaminous line measured from the y = 1. Figure 3 shows the petrophysical log section across the reservoir. a FDC log in a thick sand reads 0 = 28% and y = 0. Solution of eq. Eq. Figure 5 shows some of the core photographs. A line drawn between these points is the laminated model limit expressed by eq. which is on an isolam corresponding to 10% shale laminations. This is an artifact of biased sampling. yields y = 0. y = . using eq. The good agreement in the clean sand is apparent. We recently had the opportunity to test this hypothesis when we cored a well in South Louisiana which penetrated this same reservoir.25. Isolaminous lines are added to Figure 2 by constructing lines parallel to the dispersed shale model line. One simplification is to assume the amount of structural shale is too small to be significant and remove this possibility from the model. A line connecting this point to the y = 1 point is the graphical approximation of eq. 22. As an example we choose a Miocene reservoir in South Louisiana.28. This simplification permits the response of a gamma ray-density log to be solved (graphically or algebraically) for sand fraction. JUNE 4-7. 10 as 8. (22). for this example the bed is composed of 10% shale laminations and the porosity of the interbedded sand is 29. This is the same reservoir shown in Figure 1. Thus. We have illustrated this solution graphically. For a specific example. Figure 6 shows the corresponding FDC 8 plotted versus stressed core porosity. demonstrating the laminated nature of the sand. The minimum flfor the dispersed is given by eq. (26) is also presented. Only the sand between the laminations was plugged for core analysis.4.

if one knows he is in a laminated environment he can apply the laminated resistivity model. from the gamma ray-density log porosity crossplot. the average porosity as seen by the density log is reproduced. SYMBOLS ARYXY- activity. ACKNOWLEDGMENT The authors wish to thank Shell Oil Company for granting permission to publish this work. We also thank B. f. Ausburn of Shell Oil Company who allowed us to use Figure 4 which he had prepared earlier. is read from the induction log. 1975 Using the core photographs. with an overlay showing the change in calculated So in the heavily laminated zones. and Rsh takes from the induction log in a nearby shale zone. Thus.= R. The effect shown here increases as the percent laminations increase to the point where the shale resistivity dominates the apparent resistivity. the oil saturation increases. 1 fsd b . an experimental constant which relates the true activity to the number of events counted by the detector a .h Ra The value of R. We appreciate the many discussions with our collegues who helped solidify many of the concepts presented. Conversely.d and fsh are obtained from the crossplot. void space volume fraction of material counting yield. The effect of using the true sand 0 and Rt to calculate So is shown in Figure 7. the sand fraction and sand porosity matches that counted from the photographs and measurements in the lab. In most cases. number of radioactive events counting rate observed at a detector or the Resistivity.SPWLA SIXTEENTH ANNUAL LOGGING SYMPOSIUM. Then the actual value percent of Rsd and Rt of the sand can be calculated.subscript for clean sand - T 9 - . This is a repeat of Figure 4. JUNE 4-7.d x R. thereby condemning a zone which could produce hydrocarbons. the amount of sand and shale can be counted precisely and when a clean sand porosity of 31% (the average 0 from stressed core analysis) and pure shale porosity of 15% are assigned to the laminations. E.

(22 predicts y = . 1975 bfY- subscript for pure shale fraction of bulk volume. The Ib error introduced by uncertainty in y is less than +.67. thus the y = 1. defined by eq. We can calculate y = .5. thus the posi ion of this line is firm within the confines of the other The minimum 0 available is fixed at @a&. 8 = t$. (30). grainosity. The y = 1 point is fixed by the uncertainty in determining 0a from field measurements. 0 = @a apex is firm. 6 using counting rates in pure shales and clean sands. Eq. JUNE 4-7. Equation (24) demonstrated that the equation which connects these two apexes is linear.part of a rock expressed as a volume fraction of the total bulk volume.SPWLA SIXTEENTH ANNUAL LOGGING SYMPOSIUM. cement type. We feel justified in using assumption (5) because we are also assuming no fluid radioactivity and constant background in spite of unknown changes in borehole size and rugosity. (27) and y = . It is often the sand fraction. 8 and is a measure of sand radioactivity. the void space of a rock expressed as a volume fraction of the total bulk volume. arrive at the simple equations presented in the text. apex.5. However. The value of @max iS fixed at 8a + Xa@b or 0a + (l-oa) fib = . porosity. and placement. used in Resistivity models defined by eq. For the structural shale model we have similar uncertainties in y at the @max point. (20) predicts y = -59 while Eq. and detector type and size. thus the max error for fldis in the working area of the triangle (8 apparent > @b) is +1. value of y at this point is less well known. - 10 - .33 from eq.43. We feel this is acceptable for any practical application. By assigning a few numbers to the equations we can explore the potential errors caused by assumption (5) and demonstrate the small error introduced by its acceptance. Let's refer to Figure 2. the solid. the assumptions. Similarly for the y = 0. But for the theoretical model we can set this number exactly. <- 0X- APPENDIX1 We realize that errors have been introduced to the model by However.5 and the mean error about +.41 from eq. The actual value is probably between t ne two. these assumptions are necessary to many of our assumptions.

.

SPWLA SIXTEENTH ANNUAL LOGGING SYMPOSIUM. 1975 I Y '0 - 12 - . JUNE 4-7.

FRAC.t 5 PROFILE FIGURE 4 .?O I 2 0 0 I . 011 SATURATION. FROM DEEP INDUCTION AND 151 mv FIGURE 0.. PV 20 -30 I .1 ohm-meters 1. .30 I POROSITY.50 I I .60 I .40 I .2b . BV ...24 I I .22 I 20 1 .0 3 PETROPHYSICAL SECTION FOR RESERVOIR IN SOUTH LOUISIANA..z i % rl - I I I I 87950 _J I + I Boo0 .32 I .S.API 0 I UNITS 60 I 0 jo I . . CORED INTERVAL SHOWN BY BAR IN TRACK I.OIL SATURATION-POROSITY DENSITY LOGS.2B I FRAC.

.

32 .-_ JAND EFFECTS FIGURE 7 OIL SATURATION-POROSITY PROFILE FROM DEEP INDUCTION DENSITY LOGS.0 .lO OIL SATURATION..30 .80 1 34 I .24 I r. . CORRECTED FOR SHALY SAND ON INDUCTION & DENSITY SHOWN DOTTED LINE.26 I .70 I -----I _-_..20 . . .6O I I POROSITY.___- 7900 - l------l I r___*____-_ I I I L__ :---- ! I L__. SOLID LINE.30 I r ----.- FRAC BV .M .28 1 .a 1 .40 FRAC PV . FIGURE 6 FDC POROSITY DISCREPANCY & STRESSED CORE BETWEEN ANALYSIS POROSITY METHODS DEMONSTRATING APPARENl CAUSED BY LAMINATIONS.

- CL 1 Intro Petrophysics
- 323170959-Ley-de-Archie
- RPA Simm 2007
- TLE_AVO Classification of Lithology and Pore Fluids Constrained by Rock_Avseth
- Some Simple Methods for Refining Permeability Estimates From NMR Logs And Generating Capillary Pressure Curves
- Reservoir Zonation and Permeability Estimation
- 17154442-Basic-Petroleum-Geology-BOOK-by-HALLIBURTON.pdf
- 04-551.pdf
- ip tutorial
- Artigo Solo Cimentado Artificialmente
- SCA1987-09
- Logging While Drilling
- Reservoir Characterization
- Development and Assessment of Protective Winter Covers
- Determination of the Vuggy Porosity of Some Oolitic Building Limestones and the Practical Implications
- Presion Capilary Historia de La Saturacion
- 25 - matlosz1995
- Diagenesis Climate & Porosity
- Tight Gas Sandstone Reservoirs in China, Characteristics and Recognition Criteria - Zou Et Al. 2012
- Jotazinc.pdf
- Deveolpment Criteria For Geotextiles Granular Filters
- download-22887697
- Matlab Solution to Diffusion-reaction Problems
- Shashank Bishnoi Master of Engineering Thesis
- Production of Porous Calcium Phosphate
- 10 Electrical
- chap3_603
- Est Pore Throat Size in SST From Routine Core-Ana
- ANSYS FLUENT 12.0 User's Guide - 7.2 - Porous
- PP 58-62 Effect of Depth Ratio on the Thermal Performance of Double Flow Packed Bed Solar Air Heater Satendra

Skip carousel

- tmp863E
- tmpC56D
- tmp6FE8.tmp
- tmpF0AE
- tmpD8A4
- tmpADD4.tmp
- tmp4BFE
- tmp2AEF.tmp
- tmpD4E1
- Tmp 1682
- tmp1B97
- tmp3D6F
- tmp4E3E
- tmpDA34
- tmpB867
- tmp914D
- Tmp 6506
- tmpC9F0.tmp
- tmpE3B
- An Image Analysis Technique to Estimate the Porosity of Rock Samples
- tmp5B8E.tmp
- tmp9A6
- tmp3A22
- Open Graded Friction Course by Marshall Mix Design Approach and Its Performance Studies
- tmp6C7B.tmp
- tmp74F7
- tmp4F03
- As 1774.5-2001 Refractories and Refractory Materials - Physical Test Methods the Determination of Density Por
- tmpBB67
- As 2331.3.6-2001 Methods of Test for Metallic and Related Coatings Corrosion and Related Property Tests - Ele

Skip carousel

- National Trust natural childhood report
- 50170_1940-1944
- Little Miracles Swarowski Beaded Earrings
- Untitled
- Turlock Electric Rates - Schedule IT
- CULTURE AND BEHAVIOR
- Tucker - The Lost Key
- Untitled
- The Bay State Monthly — Volume 1, No. 3, March, 1884 by Various
- WITHOUT HER HANDS IN MINE.doc
- Hinton - A New Era of Thought
- 2012 Movie Review
- Rate Sheet
- 128 Totally Spaced Out
- Eversource-Energy-(Formerly-Connecticut-Light-and-Power-Co)-Rate-39-Large-Interruptible-Service
- Montana-Dakota-Utilities-Co-Irrigation-Power-Service-(WY)
- Tri Phasic White Mask
- 69938_2005-2009
- Energy-Australia-Basic---Home-(ActewAGL)
- Tillamook Peoples Utility Dist - Residential Service
- City-of-Hingham-LARGE-GENERAL-SERVICE
- Untitled
- 4-County Electric Power Assn - Residential
- 02-07-17 edition
- tmp7D7D.tmp
- Maui-Electric-Co-Ltd-Purchases-From-Qualifying-Facilities-100-KW-or-Less-(Molokai)
- 2016 West Nile virus surveillance in Iowa
- Central-Maine-Power-Co-Residential-Service---Time-Of-Use
- A Challenge to America
- Lake-Country-Power-Rates-and-Rebates

Sign up to vote on this title

UsefulNot usefulClose Dialog## Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

Close Dialog## This title now requires a credit

Use one of your book credits to continue reading from where you left off, or restart the preview.

Loading