SPWLA 34th Annual Logging Symposium, June 13-16, 1993 A SIMPLE, CONVINCING MODEL FOR CALCULATING WATER SATURATIONS IN SOUTHERN NORTH SEA GAS FIELDS. ‘Steve Cuddy, Gareth Allinson, Richard Stecle, Abstract ‘A water saturation - height function (Swh) is used in the volumetric calculation ofthe hydrocarbons in place following tho valuation of wator saturation’ and Porosity from weil logs and cores, ‘A simple convincing function has been developed that calculates water saturation as a function of the height above the free water level alone, This function, which is virtually Independent of porosity and. lity of the rock, is based on the, bulk, volume'of "water (ie product of orasity and water saturation) as seen, bythe Conductivity measurement. Roaegendes reservolr rocks have a wide tange of pporoperm characteristics ranging from high quality dune Sandstones through lower porosity sandy sabkha to lighuy cemented fluvial sandstones, The model's predicted water saturations are largely unaffected by tock type, making it a powerful water saturation model for complex lithologies. ‘This surprisingly simple model for Swh has led to a better understanding of Souther North Sea gas fields. ‘The model uses the free water level as its base and helps explain the apparently varying gas waict coniacis. The ‘Swh model shows that in low porosity sandstones the rock is fully water saturated for some distance ahove the {ree water level consistent ‘with capillary pressure theory. Consequently, the apparent gas-waler contact varies from well w well depending on the average porosity of the rocks near the contact. ‘The derivation is simpler than with classical functions as there is no porosity banding. Consequently the ‘dewermination of the model does not require data from low porosity intervals and the resulting Swh curves do fol cross. The predictive qualities of the model have been verified by comparing the log derived water saturations with the “back-calculated” water saturations from the model at the well locations, ‘As the model describes the Souther North Sea Rotliegendes reservoirs extremely well, ithas been used as a predictor of the fluid contacts in’wells where the ‘contact 1s unclear, or where the contact 1s below the total depth of the well. A correlation program uses the model to determine the fluid contact by comparing the balk volume of water profile predicted by the model with bulk volume of water calculated by the electrical logs. Introduction ‘The saturation - height function (Swh) is required to calculate nyarocaroon reserves in ihe Tield, away trom the well control points. Although the Swh function is ‘important for computation of hydrocarbon saturatio there has been little work, to date, on how to derive BP Exploration, Glasgow, Scotland. Statoil (UK) Ltd, London, England. BP Exploration, Glasgow. these functions. The effect of this function on reserves ccan be large, causing radical changes in “book-values” ‘and argument over the form of the function has made ‘many equity negotiations difficult. ‘The petrophysical interpretation of a field consists of calculating porosity and water saturations. This analysis represents porosity within +/- 1 pu and water saturation +/- 10 su, at best, and only for a few feet surrounding the well-bore. Away from well locations, porosity can be estimated by porosity mapping. Water Saturation cannot be mapped as at depends on Tany factors including porosity and the height above the local fluid contact. The Swh function is intended to describe the variation of water saturations with these parameters so thatthe water saturation can be calculated in the reservoir model, The gas initially in place (GIP) can then be calculated and reserves determined. The fervor in reserves resulting from an equation that poorly describes the field can be large compared iy eur induced by poor, electrical tool ‘design of rock Daramelers derived from special core analysis studies. ‘The function has taken many forms over the years!-10. Listed below are examples of Swh functions: San 2 2a cfd.tidinwr) + 1038 10686 Sw = Swi + (1Swigg) 8" Suigg =D Bre log K a @ Sw= a4 Swi we loge a Sw = 24d Ore logh +40? + flog Wr # Blog h Sw = 14 (a1bQ) + {1.<(C*EMe*O(dst @))) a 6) Where: ‘Sw = Water saturation Swirr = Irreducible water saturation o sit K Formesbiiy h Height above contact ewe Gas water contact Abedig = Constants ‘These functions have varying complexity, with litle in common between them.” They are simply curves fits through the observed data sets and are not hased on rock property physics. Figure 1 shows a typical Swh plot from the Southern North Sea. The highest porosities give the lowest water saturations and the Swh curve that represents high porosity rock is asymplotic to the free water level. The Feason the free watt level (FWL) is used rather than ie gas water contact (GWC) is important and is, explained later.SPWLA 34th Annual Logging Symposium. June 13-16. 1993 The Swh function is important for good reservoir modelling and must satisfy the following criteria * Correctly describe the hydrocarbon distribution hroughout the field * Sumple'to derive and apply. Why this 1s important is explained later. * A convincing function is valuable as it ends argument, especially in equity negotiations. The Classical Function ‘The conventional method of determining the Swh function is by porosity band. This has a separate Swh curve for each of several porasity bands as shown by Figure 1 Deriving functions for uhese porosity bands can be very sic, Fra "the data from the fel sno eualy distributed between the different porosity bands. ten as in the Hyde case, there is litle data in the lower Porosity bands that makes the curve fing difficult Second, the Swh data show that in low porosity sandstones the rock is fully water saturated for some distance above the FWL, such that the FWL can be moch’depe han te GWG. "Thi is Consist capil sure Uieory, Consequently, the apparent au8 water Contact varies between wells defending on te average porosity of the rocks near the contact. ‘This Phenomanam, Enown ac “Threrhold Heights ie the ‘emical offset between the Swh curves, The offset must be known before the Swh curves can be calculated, hich, in practice, 8 dificul. A different Swh function for cach porosity band can be ‘an unconvincing model for a field especialy if the Swh Curves cross. Banded Swh curves by porosity are also mathematically unconvincing as the predicted water saturations are fixed for the range of porosities in the porosity band. This can result in the Stair-stepping of the calculated water saturation, Problems with the classical method are: + Defining the threshold height can be very difficult and it must be known for each porosity band before Swh curves can be calculated * Sufficient data in each porosity band is required in to derive sensible curves. * Porosity banded Swh curves are visually and ‘mathematically unconvincing, ‘The Leverett J-Function In the early 1940's, M,C. Leverett? described a dimensionless J-function of the wetting phase saturation that uses plysical propertics of the took and fluid. It was originally proposed as a means of converting all Capillary presoure da to a universal eurve. Although his function has survived the tests of ime, more recently. itis thought it should be limited to single reservoirs. sw) = 5 ) LLeveret’s J-function (dimensionless) ‘Water saturation (fraction) Pe = Capillary pressure (dync/cm”) = Interfacial tension (dynefem) K = Peemesbility (em?) © — = Porosity fraction) ‘To include wettability effect, he contact angle (8) has been incorporated into Eq. 6: usm = PVE ‘The J-function is recognized to be nearly linear® on a log-log scale and a good fit can be achieved by Eq. 8 ‘which is derived from core measurements where cand are constants. ” 4S™) = 5p @) ‘The capillary pressure is related to the height above the fice water level by: Pe = 8(PwPeheWwL 9) where: 1B ~ Acceleration of gravity (avce/oce) Pw = Density of the water phase(gicm3) Pg Density of the gas phase(g/em’) heWL = Height above the free water level (feet) Combining Eqs. 7, 8 and 9 gives: acces. [B SWB = ppghew VK «19, Now the bulk volume of water, BVW, is: BYW = Sw @ ay Combining Eqs. 10 and 11 A = Yrs _./9 BW = V gowephrwt VK a ‘The permeability predictor forthe Southem North Sea: Log = 16931-2022 co) “Therefore combining Eqs. 12 and 13: «aay B 4 {29% BYW=\V Zpwepiiws V 1TE95T BORE) © ‘Typical values in the Southern North Sea area for the ‘constants are: 74 dynefem 0.06 =231 eoRaThe BYW funtion vatulated using Eas 14 for dhe range of porosities found in the Southern North Sea area are tabulated below for three typical heights above the free water level: Porosity 10" 100" 300" Opa Oss 0057 0.0385 20 0154 00570085 ‘The BYW is the same at 10 and 20 porosity units with nly slight variation at other porosities in this range. ‘This suggests thatthe bulk volume of water ata certain height above the froo water level io. virtually independent of rock properties such as porosity and permeability. Observation of Southern North Sea data Supports these results Data from the Southern North Sea Gas Fields ‘A Swh function was developed for Southem North Sea fas fields based on the Leverett J-function’s prediction that the bulk volume of water is related simply to the height above the free water level. This approach also axidreses the problems ofthe classical Swh function. Figure 2 shows the water saturation data set from field S. This data show the porosity banding of the Swh Function and the dareshold height at the base of the low porosity data. The variation of the bulk volume of pater depends almost entirely om the heisht above the free water level as shown by Figure 3. consistent with Levereit’s equation in the for Equation 14 3 of ‘There is a good correlation between the bulk volume of iter with the height above the FWL that is independent of porosity. implies that a Swh Function based on the bulk volume of watcr avoids the problems associated with porosity banding and the variance af the GWC from the FWL. This type Swh function has been called the “FOIL” funcuon, Only net reservoir dara, drat is more whan one metre from a bed boundary, is displayed, and used for Tine fitting. ‘The data within one metre of eloctrofacies’ bed boundaries are eliminated to reduce data that could be affected by the different vertical logging tool responses. Data that is non-net, as indicated by a porosity cutoff, is not used as the Tock matrix in these Srchnne da nat have snfficient prrnsity to hold the minimum volume of water as predicted by Equ. 14, ‘These intervals compute water saturations that are ‘greater oF equal to 100 su. ‘A function derived from this data can easily be used in hydrocarbon reserve calculation by dividing the bulk volume of water by the mapped porosity: Bulk Water Volume Porosity Water Saturation as) ‘The Southern North Sea Rolliegendes reservoir rocks? have a wide range of poroperm characteristics ranging from high quality dune sandstones, through lower 'SPWLA 34th Annusl Logging Symposium, June 13-16, 1993 porosity sandy sabkha, to tightly cemented fluvial sandstones. The model's prodicied bulk volume of water is largely unaffected by rock type making it a Powerful water saturation model for complex lithologies. Figure 4 shows how the data are independent of rock permeability as the high and low permeability points are randomly distributed throughout the plot. There are four free water levels in this field and the function is shown to be independent of the field area and the local FWL by the Figure 5. A simple function describes the correlation: BY 1-A*HAB (16) Where: BWW Sw * 0 OM) Sw ‘Water saturation o Porosity H Height tbove te ree water level A.B = Constants The form of this function was verified by plotting the logarithm of BVW against the logarithm of Height. This isthe form of the straight line equation y = mx+v. Logie RVW = R#LogigH + LogioA (17) AAs the dataset is reasonably linear with log axes, as shown by Figure 6, a power function should correctly describe the correlation. The constants A (0.01619) and B (0.85771) were determined by least squarcs regression, with BVW the independent or predicted variable, Figure 7 shows the resulting single FOIL curve for field. The classical water saturation curves for the field are easily derived from this function by dividing bulk volume of water by porosity. These ‘Swh curves give the threshold height for each porosity bband as expected. Figure 8 shows the Swh curves for the field ‘The function was shown w be a good predictor of yas saturations for the field, by comparing the log derived zonal gas saturation with the back-calculated gas saturations calculated from the volumetric computation, ‘That is to say that the model calculated gas saturations at the well locations are similar 10 those derived from the electrical logs as shown by Figure 9. ‘The intervals with the greatest hydrocarbon pore height GECPH are clearly the most important and give the HEPH = (1-Sw)*@*H gy ‘The function can be used to calculate water saturations in wells without resistivity curves. There were n0 resistivity curves in the well $1 due wo well problems, ‘The predicted water saturations compare well with the core water saturations as shown by the Figure 10. If a banded Swh function was used, the water saturation ‘curve would be constant for a range of porosities and then would jump or stai-step to the next constant waterSPWLA Mth Annual Logging Symposium, June 13-16, 1993 ‘The Free Water Level and the Gas Water Contact ‘The gas water contact (GWC) is often referred to as the at the base of the gas column where the logs Indicate a sharp increase in water saturations to near 100 su. The free water level (FWL) is the depth where this sharp increase would occur if the formation had very high poroperm characteristics. Consequently, the FWL may not be seen in any well na field with poor rock properties and the GWC would vary between wells Aepending on the local rock quality. Using the FWL rather than the GWC has the following, advantages: * The GWC is often unclear as the gas column can be considered to be a continuous transition tone even where the reservoir is very thick. ‘The GWC needs an agreed convention. Many equity determinations have foundered on arguments to whether the GWC should be, say, at Sw = 40 su or 60 su..‘The FWL, by contrast i the limit (Sw = 100 su) of the Swh curve in the highest porosity {formation Itis more convenient for reservoir modelling and ‘understanding field compartmentalization to have a common depth forall Swh functions. Formation pressure and capillary pressure data are related to FWL not the GWC. ‘The FWL is, therefore, used rather than the GWC as this unifis the fluid contact predicted from electrical logs, formation pressures and capillary pressure data. Using the FOIL Function in the Hyde Field ‘The Hyde Field is located immediately to the north of West Sole, some 55 km to the east the North Humberside coast and is predominantly in Block 48/6. The reservoir is in the Lower Leman Sandstone Formation (Early Permian age) and is part of the Rodiegendes Group. ‘The reservoir was deposited in a desert by aeolian, fluvial, and lacustrine. processes. Three major lithofacies ,associations have been recognized from core studies Aeolian Dune. Aeolian sandstone have we Dest permeabilities by virtue ofits better sorting and lack ‘of dewital clays. Clean aeolian dune sandstones give highest porosities in the reservoir with an average porosity around 16 pu. Dune base sandstones (wind. pple) give a lower average porosity of 12-14 pu as they are less well sorted. * Sabkha._ Sandy sabia has good porosity bu he presence clay enhances compaction effects End thus reduces primary porosity. Muddy sabiha Porostes and permeabiiies are very Tow with no Fesorvlr porn, Fluvial ‘The fuvil sandstones often have poorer permeabilities (<0.3 mD) and porosities (<10 pu) than the sandy sabkha sandstones. Their porosity is Gependont on tho dotrital clay content and pore filling cements. In addition, in all facies, diagenetic overprint of pervasive fibrous illite ‘clays severely reduces as permeabilities. Only in the well sorted grain-flow facies which has a’ macro-porous network are moderate a ‘permeabilities The Hyde reservoir has been divided into four zones from Zone 1 at the base to Zone 4 at the crest. Zones 1 and 3 are predominately fluvial sandstone and the reservoir quality is poor. Zones 2 and 4 are mostly seolian dune with some sabkha. Zone 4 sequence is made up of an allemation of dry erg facies with either damp or dry sand sheets. ‘The acolian sandstones have the beat average reservoir propertics in the Field duc to their grain-flow stratification and consequently Zone 4 is the principal producing zone in the development area. Zone 2 is composed almost entirely of acolian Sandstone. The cored sections show a predominance of wwind-rippic stratification compared to the predominant ‘grain-flow stratification of Zone 4, so permeability is ‘Poorer overall in Zone 2. ‘The fluid contacts in Hyde have presented a problem to petrophysicists during the ficld’s exploration and “pprassal. It was recognized early in the appraisal phase that the varying GWC could be explained, through ‘capillary pressure theory, as the variation in the height above the FWL that could support water saturations of 100 su,_ Various so-called GWCs were unified when it was realized that they resulted from porosity variations throughout the field through a phenomenon known as threshold height. Although this helped with the understanding of the fluid contacts, it compounded the problem of mathematically deriving water saturation height curves. ‘The set of curves that were derived for Hyde are shown by Figure 1. These curves were difficult to compute by powosity band as there are not equal amounts of data i each porosity band. These curves are not visually or mathematically convincing as it can be seen that they potentially cross. In addition, the function did not describe the reservoir in the flanks of the field as well s ‘development area, although a similar rock distribution and. zonation’ scheme was present youghout the structure. ‘This caused concern that gas in pans of the field was not being fully recognized. ‘This lead to the search for a better Swh function and the discovery of the FOIL function, ‘The POM. function is clear in the data af Figure 11 and avoids the problem of the threshold height and the need for low porosity well data. There is a slight difference between tne flank aid core area wells.” ine reason tor this lies in the geological structure of the field as shown by Figure 12. ‘The formation is a classic “bow shape”. The aeolian sandstones have tne best reservotr properties in the Held due wo their grain-flow stratification and consequently, Zone 4 is the principal producing zone, In the core evelopment area of the field, Zone 4 is high above the FWL, whereas in the flank areas Zone 4 cuts the FWL. ‘The FOIL function varies rather than by area. The function for each zane is, however, i of rock porosity and permeability. ‘A plot of the bulk volume of water vs. height by zone type is shown by Figure 13. ightly by zone (facies)‘The Hyde FOIL function is similar to field $ in showing porosity independence although the FOIL function varies slightly by facies type. Although two functions are used to describe the gas distributions in Hyde, the extremes in rock types between dune sands with high characteristics and very poor fluvial ‘of lacustrine rock result in only a slight perturbation in the FOIL function. Predicting the Free water level ‘As this function describes the gas column well, it can be used as a predictor of the free water level in wells ‘where the contact is unclear. or where the contact is below the total depth of the well. The method works by deriving the function from the wells where there is a clear vontact suchas wells with good. poroperm characteristics in the neighbourhood of the contact. ‘Wells that are not highly deviated are used. as these have the greater confidence in the true vertical heights. ‘The derived function is then with the derived ‘ull volume uf waicr (BVW) prutile i Ure wells wid Poot contacts by using acomelation algorithm. The FOIL function is used to compute a BVW profile assuming a FWL 50 feet above the estimated contact and compares this with the actual BVW profile by computing the quality of fit using the following algorithm: 419) Quality of fit of FWL(i) = : Levels Where: | BVW() = The BYW calculated by the logs at height above he FWLG) ‘the BVW calculated from the FOIL function at height j above the FWL) The sum of the squared. differences between the log calculated BVW and the BW prediciod by the FOIL function, for Pwo. ‘The notional FWL is then moved down by 1/2" wo 49.5" and the procedure repeated to calculate, another Correlation coefficient. ‘Ths is for 200 levels luni the notional contact is 50° below the estimated Conlact. The correlation coefficients are plotted against ‘depti-as shown by Figure 14. ‘The minimum voutclation voefficicat gives the FWL. In the best wells, this computer program pin-points the ‘contact down to the nearest 1/2". In the wells with poor indications of the FWL the method can greatly improve the estimation of the contact. It is expected that the FWL will be constant Uugugh-out the field oF at least through-out pans ofthe field In realty, the FWL can ‘vary up to 30° due to errors in the deviation survey and the lg depth measurement tis best that he technique is used, without bias, to determine the FWLs and that Commduaticy Hy uot assumed. “The wells ate drei grouped with probable common contacts allowing {greater errors in wells with the higher deviations. FOILG) FWL() ‘This method is demonstrated by application to two ‘ther Southern North Sea wells. Well S2 (Figure 15) shows a fairly clear contact around 8000" TVDss. Notice that the bull volume of water shows a clean ‘sweep whereas the water saturation is less smooth because of porosity variations. The field derived Swh SPWLA 34th Annual Logging Symposium, June 13-16, 1993 function (FOIL) is compared to the log derived BVW using the technique deceribed above. "The correlation coefficients are calculated for eve ible contact between 7950" and 8050" TVDss. results plotted against depth as shown, The minimum correlation coefficient gives the FWL. In the best, wells this ‘computer program pin points the contact down to the nearest 1/2", In well $3 the contact is less clear as shown by Figure 16. Although the program has difficulty in giving the precise depth of the contact it gives a clear range. The ‘method has been used in wells where the contact is below the total depth of the well when a FOIL profile i apparent in the gas column, ‘This technique is very useful for equity determinations where the FWL is determined by a precisely defined ‘mathematical method rather than by “eye”. This method ‘requires a function that describes the formation well and works very poorly where porosity banding has been Conclusions A simple Swh function has been developed that acurately predicts water saturations in Southem North Sea gas fields. The function correctly describes the gas distribution in the field and reduces the error on the hydrocarbon reserves. This is important, as the errors resulting from an equation that poorly describes the ficld can be large compared to cxrors induced by electrical tol design of rock parameters derived from special core analysis, ‘The predicted water saturations are largely unaffected by rock type, making it a powerful model for complex luologies. The visually and mathematically convincing fimetion uses the free water level as ite hase and explains the apparently varying gas water contacts in several North Sea fields. Its derivation is simpler than with classical functions as there is no porosity banding and no threshold height problems. Consequently the determination of the mode] does not require data from low porosity intervals and ime resulting Swn curves do mot cross. ‘The mode! Nas been used as a predictor of the fluid contacts. ‘Theory and observation indicate that a unit volume of rock holds a fixed amount of water dependent on the Yertical height above the free water level and virtually independent of rock properties such as permeability and porosity. ‘The function can, therefore, be derived from the best. near-vertical wells with ‘good poroperm characteristics that have a fluid contact and later applied to infill or flank wells that may be more deviated and have poorer rock ies and may not have a fluid contact. As the function is uoed throughout the field there may be acot saving as the tol hat measure the ‘water saturation in these later wells may not need to be as acura. Acknowledgment ‘The authors wish to thank Statoil and BP who granted their permission to publish the data presented here. We would like to thank them for providing the time and SeSPWLA 34th Annual Logging Symposium, June 13-16, 1993 resources to write this paper. As the work on the Southern North Sea is ongoing, the authors would ‘appreciate comments on the concepts raised in this paper. References 1, Brown, ILW. 1951. Capillary pressure investigations petroleum transactions of AIME 192:67-74. 2 Ibrahim A, Bassiouni, Z, Desbrandes R 1992. Determination of relative permeability curves in tight gas sands using log data. The Transactions of the Society of Professional Well Log Analysts 33rd ‘Annual Logging Symposium. June 14-17. 3. Heseldine, G. M. 1974, A method of averaging api pressure curves. ‘The Transactions UF die Society’ of Professional Well Log Analysts 15th Annual Logging Symposium, June 2-5. 4. Lee, S. T, 1989. Capillary-gravity equilibria for hydrocarbon fluids in porous media. 64th Annual SPE Tech Conference SPE-19650. 5. Leverett, M. C. 1941, Capillary behavior in porous solids. Petroleum Transactions of AIME 142:152-169 6 Ma, S. Jiang, M. X. Morrow, NR_ 1991. Correlation ‘of Capillary pressure relationchipe and calculation of permeability. 66th Annual SPE Technical Conference SPE-22685. 7. McCardell, W. M. 1955 review of the physical basis for the use'of the J-function. The Fighth oil recovery ‘conference transactions 8. Prickett, H. D. Bremer, K E. 1989. Improved initial water saturation distribution for a three-dimensional model. 6th SPE Middle East Oil Show. SPE-17958. 9. Steele, R.P, Allan, R. M, Booth, A. J. 1993. in press. Hyde: A proposed field development in the Southern North Sea using horizontal wells, Proceedings of the 4th Conference on the Geology of Northwest Europe. 10 Xie, X. 1991. A formulation for the capillary pressure relationship and a statistical description of pore size distribution, Academia SPE-21890. About the Authors Steve Cuddy is the Chief Petrophysicist for BP Exploration in Glasgow . He received a B.Sc. degree in physics from Sussex University in England. Before ing BP in 1984. he worked for Schlumberger for 10 years. Gareth Allinson is the Senior Geologi ‘based in London, England. Before joi ‘worked for BP for 18 years. Kichard Steele was BP's Southern North Sea ‘Sedimentologist, He worked with BP for 11 years until recently joining Shell in Oman.SPWLA 34th Annual Logging Symposium, June 13-16, 1993 Figure 1: Classical Swh curves for the Hyde field Porosity Bands (18-6pu) 100 80 — z 60 + H z & s £ ay s a | =z 40 : ‘= i > 2 3 20 0 0.2 04 0.6 0.8 1 Water SaturationSPWLA 34th Annual Logging Symposium, June 13-16, 1993 HEIGHT ABOVE FWL, HEIGHT ABOVE FWL 950 15 280 245 210 178 40 105, 70 38 Figure 2: Water saturation verses height data aa Te f es Bulk volume of water showing @ independenceHEIGHT ABOVE FWL SPWLA 34th Annual Logging Symposium, June 13-16, 1993 Figure 4: Bulk volume of water showing permeability independence 2 8 140 || 109 | 70 i 0.000 0.016 0.090 0.045 0.080 0.075 0.090 0105-020 VOLUME OF WATERSPWLA 34th Annual Logging Symposium, June 13-16, 1993 Figure 6: Bulk volume of water vs. height. Log scales 3 = . - BULK VOLUME OF WATER Figure 7: Bulk volume of water (FOIL) function ee 3 Height above FWL (feet) 8 0.04 on oa on Bulk Volume of Water (%)Field Derived Water Saturation 'SPWLA 34th Annual Logging Symposium, June 13-16, 1993 Figure 8: Swh curves 20 pu a 2 Ga = 2 200 100 - L o : - Co) Oy ry) i Ce re ‘Wane Saration Figure 9: Comparison of field and well calculated Sw 1 High HCPH © Med’ HCPH & Low HCPH os | Oe os | he bo ae . ee + — aoecenene os 08 ox a ‘Well Derived Water SaturationSPWLA 34th Annual Logging Symposium, June 13-16, 1993 Figure 10: Analysis of well S1. A well without resistivity logsFigure 11: Bulk volume of water verses height function by well for Hyde ve FWL Height a0 80 70 60 50 40 30 0.00 0.03 SPWLA 34th Annual Logging Symposium, June 13-16, 199% 0.06 0.09 Bulk Volume of Water'SPWLA 34th Annual Logging Symposium, sune 13-10, 1993 Figure 12: Hyde’s geological structure Flank well Core well 100 90 80 ft 70 Ewe 60 50 abo 40 30 beigh 20 0.00 0.03 0.06 0.09 Bulk Volume of Water osCorrelation Coefficient SPWLA 34th Annual Logging Symposium, June 13-16, 1993 Figure 14: Change of FWL correlation coefficient with depth 0.12 0.10 0.08 0.06 0.04 0.02 0.00 8940 8950 8960 8970 8980 <- Free Water Level 8990 $+} +++ 9000 —+— 9010 9020 Depth FtTVDssSPWLA 34th Annual Logging Symposium, June 13-16, 1903 jure 15: Analysis of well S2 showing correlation plotSPWLA 34th Annual Logging Symposium, June 13-16, 1993 Figure 16: of well S3 showing correlation plot poafo a gs : rs t iy