Even-textured pore space 500 mm Even-textured carbonate cement Carbonate cement 500 mm QAd4360(a)x 7100 7200 7300 3 Sequence L1 HFS L2.0, Sequence L2 1 2 3 4 ARFN 0 100 30 0 Depth (ft) 0 1000 0.1 1000 0.1 30 GR (API) Core porosity (%) Wireline log porosity (%) Core permeability (md) Rock-fabric permeability (md) Excellent match between measured core permeability (red) and permeability estimated using apparent rock- fabric numbers and total porosity (black). Capillary Pressure Models for Estimation of Original Water Saturation Discrepancies between predicted permeability and core permeability may be due to touching-vug pore system associated with karst breccia. Rock-fabric-specific capillary pressure models were developed by Lucia (1995, 1999). New data acquired in this study compare well with the existing Class 1 model. However, new data from ARFN 2 and 3 samples do not fit previous models. We developed new models for ARFN 2 and 3 rocks using the Thomeer method. Saturation was calculated for the reservoir using these capillary pressure models, wireline log porosity, ARFN framework, and reservoir column height. Overall excellent match supports use of capillary pressure model. Areas of poor match are the result of inaccurate resistivity tool response in intervals of low porosity. When available wireline logs are insufficient to distinguish interparticle porosity from vuggy porosity, total porosity must be used. Rock-fabric numbers defined by this approach (termed “Apparent Rock-Fabric Numbers” [ARFN]) are less accurate and may overstate permeability and saturation if vuggy porosity is present. Because most wells at Fullerton lack acoustic logs and usable resistivity logs (making it impossible to define interparticle porosity and vuggy porosity), the Apparent Rock-Fabric Number (ARFN) system was employed. We think the most accurate calculations of permeability and original water saturation are derived from good- quality wireline porosity data using transforms defined from rock- fabric relationships. Requirements of this technique are good-quality core analyses (to define porosity and permeability), thin sections (for rock-fabric typing), a cycle- based stratigraphic framework (for rock-fabric distribution), and complete log suites (including acoustic logs). Rock- Fabric Technique Water Saturation Model Comparison: Capillary Pressure vs. Log Resistivity Lower Clear Fork L2.2 Rock Fabrics/Petrophysics Lower Clear Fork L2.1 Rock Fabrics/Petrophysics Saturation Modeling Permeability Estimation ESTIMATION OF PETROPHYSICAL PROPERTIES INTEGRATING ROCK-FABRIC APPROACH WITH STRATIGRAPHY AND WIRELINE POROSITY Porosity Log Calibration Rock Fabric Distribution Following conventional log normalization and calibration, porosity logs were spatially normalized to minimize individual well acquisition and calibration errors. Maps below illustrate the effects of spatial normalization of sidewall neutron log values using various search radii for contouring average porosity data of entire reservoir interval for all wells. Differences between normalized values and actual well values were used to adjust log data over the entire reservoir interval. ARFN-based Permeability Estimation vs. Core Data Stratigraphic Distribution Data Quality Spatial Normalization of Porosity Logs The rock-fabric method of petrophysical characterization is based on relationships that exist between pore type, pore size, particle size, and sorting. Once rock fabrics are identified, they can be assigned to petrophysical classes that have rock-fabric-specific porosity/ permeability transforms (rock-fabric numbers [RFN]). The direct relationship between interparticle porosity and permeability (Lucia, 1995, 1999) allows permeability to be accurately calculated when interparticle porosity and rock- fabric number (RFN) can be defined (below). Core analysis data commonly contain suspect data, as evidenced by low-porosity/high-permeability data points (below). Many of these data are the result of poor sample cleaning. For this study we obtained new plugs that were carefully cleaned and analyzed. Note the difference in the two data sets from the same core. 3000-ft search radius: provides detail and reduces random error. This was used to normalize raw wireline log porosity data in the field. 5,000-ft search radius: masks too much small- scale variability 1000-ft search radius: too many bull’s eyes (well acquisition problems or calibration errors) Rock-Fabric Method Based on Total Porosity log k = [9.7982 – (12.0838 x log RFN)] + [8.6711 – (8.2965 x log RFN)] x log φ IP Rock- Fabric Porosity/Permeability Relationship 0.05 0.10 0.20 0.30 0.40 0.1 1.0 10 100 1000 Interparticle Porosity (fraction) 1.0 2.0 3.0 QAd2746c Permeability (md) Class 3 Field Class 1 Field 0.5 1.5 2.5 4.0 Rock Fabric Numbers Class 2 Field Rock-Fabric Method Based on Interparticle Porosity 0.01 0.1 1 10 100 1000 0.01 0.1 Permeability (md) Total porosity (fraction) Few suspect data New Core Analysis 0.01 0.1 1 10 100 1000 0.01 0.1 Total porosity (fraction) Permeability (md) Suspect data Old Core Analysis Estimation of Permeability and Original Water Saturation 0.08 0.08 0.06 0.06 0.04 CI = 1% SNP porosity 0 4000 ft 0 1000 m UNIVERSITY LANDS BLOCK 13 PSL BLOCK A-48 PSL BLOCK A-37 PSL BLOCK A-31 PSL BLOCK A-26 Clear Fork Unit PSL BLOCK A-47 PSL BLOCK A-32 0.070 0.065 0.060 CI = 0.5% SNP porosity 0.065 0.060 0.070 CI = 0.5% SNP porosity Nonreservoir ARF N GR (API) 0 100 Depth (ft) Wireline log porosity (%) 30 0 Rock-fabric permeability (md) 1000 0.1 Lower Clear Fork Wichita 123 L2.3 L2.2 L2.1 Rock- Fabric/Petrophysical Classification Stratigraphic Sequence L1 L2.0 6800 6900 7100 7200 7300 ARFN 3 ARFN 1 in most areas ARFN 2–2.5 in limestone areas ARFN 3 ARFN 2 in most areas ARFN 1 in some areas 7000 Type log showing general stratigraphic position of apparent rock-fabric numbers. Subtidal Peritidal Subtidal Peritidal Peritidal HFS L 2.2 is dominated by petrophysical Class 2 medium-crystalline anhydritic dolostone. Because of the differential effect of patchy, poikilotopic anhydrite on porosity and permeability, these rocks generally plot in the Class 1 field and are characterized by an ARFN of 1. Less common moldic grainstones and grain-dominated packstones are best represented by an ARFN of 2 to 2.5. 1 mm Grain-dominated Packstone ARFN1: Medium-crystalline Dolostones with Poikilotopic Anhydrite ARFN 2- 2.5 limestones Note tight clustering of data along ARFN 2.5 transform. Grainstone Total porosity (fraction) 0.1 0.1 Permeability (md) 0.1 1 10 100 1000 ARFN 2.5 TRANSFORM Spatial Distribution Permeability (md) Total porosity (fraction) ARFN1 TRANSFORM 0.1 1 10 100 1000 0.01 0.1 Cored well New core analysis and thin sections ARFN 2 Limestone ARFN 1 Dolostone ARFN 2.5 Limestone 1 mm 1 mm 1 mm 0 4000 ft 1000 m 0 Lower Clear Fork L2.2 Rock- Fabric 1 Dolostone 6800 1 GR (API) Core porosity (%) Wireline log porosity (%) Core permeability (md) Rock-fabric permeability (md) 1 2 3 4 ARFN 0 100 30 0 Depth (ft) 0 1000 0.1 1000 0.1 30 Lower Clear Fork L2.1 Wichita (L1 and L2.0) Lower Clear Fork L2.2 Rock- Fabric 2.5 Limestone 6850 2.5 GR (API) Core porosity (%) Wireline log porosity (%) Core permeability (md) Rock-fabric permeability (md) 1 2 3 4 ARFN 0 100 30 0 Depth (ft) 0 1000 0.1 1000 0.1 30 6900 7000 HFS L2.1 peritidal HFS L2.1 subtidal 3 2 1 2 3 4 ARFN 0 100 30 0 Depth (ft) 0 1000 0.1 1000 0.1 30 GR (API) Core porosity (%) Wireline log porosity (%) Core permeability (md) Rock-fabric permeability (md) ARFN 3 TRANSFORM 0.01 0.1 1 10 100 1000 0.01 0.1 Total porosity (fraction) Permeability (md) Tidal-flat facies Fine-crystalline dolostone Medium-crystalline dolostone New Wichita Data Typical Class 3 Fine-crystalline Dolowackestone Fabrics Thin-section descriptions are mostly Class 3 fine-crystalline wackestones with little vuggy porosity. Therefore, most of the Wichita can be chararcterized using a rock-fabric number and an ARFN of 3. The rare class 2 grain- dominated dolopackstone and vuggy tidal-flat fabrics in the Wichita will not be properly classified using this method (see crossplot below). Wichita (L1, L2.0) Rock-Fabrics/Petrophysics Fenestral pores 1 mm 1 mm 1 mm Total porosity (fraction) 0.1 Subtidal Cycles Intrasequence Variability: Lower Clear Fork L2.1 Spatial Trends within Subtidal Cycles of L2.1 ARFN 2 Limestone and Dolostone ARFN 1 Dolostone ARFN 2 Dolostone ARFN 1 Dolostone ARFN 2 Limestone ARFN 2 Limestone and Dolostone Grain-dominated Packstone Medium-crystalline Dolostone 1 mm Medium-crystalline Dolostone with Poikilotopic Anhydrite Moldic Grainstone Subtidal facies consist of complex assemblages of Class 2 dolostones and limestones and Class 1 dolostones ARFN 3: Fine-crystalline Dolostone Sequence-top peritidal facies consist of Class 3 dolostones. These rocks have limited vuggy porosity and are characterized by an ARFN of 3. Transgressive (subtidal) and highstand (peritidal) legs of high-frequency sequences contain different rock fabrics and thus require different porosity/permeability transforms. Class 2 dolostones with abundant anhydrite behave as ARFN 1 rocks because of their patchy porosity (see L 2.2 above). ARFN 1 Dolostone This group includes Class 2 dolostones and limestones having little vuggy porosity and Class 1 moldic limestones that have significant vuggy porosity. Peritidal Cycles Limestones Dolostones HFS L2.1 GR (API) 0 100 Depth (ft) Wireline log porosity (%) 0 30 6900 7000 Peritidal Subtidal ARFN 1 TRANSFORM Permeability (md) 0.01 0.1 1 10 100 1000 0.01 0.1 Total porosity (fraction) ARFN 2 TRANSFORM 0.1 1 10 100 1000 0.01 Permeability (md) Class 2 0.1 1 10 100 1000 Total porosity (fraction) Permeability (md) 0.01 0.1 ARFN 3 TRANSFORM 1 mm 1 mm 1 mm 1 mm 1 mm 0 500 1000 1500 2000 0 20 40 60 80 100 25% 20% 15% 10% 05% Injection Pressure (psia) Air Saturation (%) ARFN 2 (New Thomeer Model) Sw(Class 2) = 1-{2.71828^(-0.2/(Log(15.5064*H/Phi^(-2.8394))))} Gamma Ray GAPI 0 150 6900 7000 7100 7200 Depth (ft) Porosity V/V 0.4 V/V 1 0 Water saturation from inverted electric logs V/V 1 0 0 Water saturation from capillary pressure model Wichita Peritidal Lower Clear Fork Peritidal Subtidal L 2.0 Subtidal L 2.1 L 2.2 Pore space Grains Dolomite crystals Rock- Fabric Petrophysical Classes Wackestone Note: bar is 100 mm Crystal size >100 mm Dolomite Crystal size <100 mm Crystal size >100 mm Limestone Limestone Crystal size 20–100 mm Dolomite Crystal size <20 mm Packstone Grainstone Packstone Mudstone MUD-DOMINATED FABRIC GRAIN-DOMINATED FABRIC QA15772(a)cx Class 1 Class 2 Class 3 Effect of Patchy Cement on Petrophysical Class Interparticle porosity (percent) A B 10,000 1000 100 10 1 10 20 30 40 Permeability (md) Class 1 Class 2 Class 3 Equal pore-throat size Increasing pore-throat size C Lime mud ESTIMATION OF PETROPHYSICAL PROPERTIES INTEGRATING ROCK-FABRIC APPROACH WITH STRATIGRAPHY AND WIRELINE POROSITY Poikilotopic sulfate Patchy anhydrite cement 500 mm Sw(Class 1) = 0.02219 x H^-0.316 x f^-1.745 0 500 1000 1500 2000 0 20 40 60 80 100 Phi = 20% Phi = 10% Phi = 5% Injection Pressure (psia) Air Saturation (%) ARFN 1 (Lucia, 1995, 1999) 0 500 1000 1500 2000 0 20 40 60 80 100 Injection Pressure (psia) Air Saturation (%) Sw(Class 3) = 1-{2.71828^(-0.1/(Log(0.3827*H/Phi^(-1.9717))))} ARFN 3 (New Thomeer Model) 25% 20% 15% 10% 5% 0.05 0.10 0.20 0.30 0.40 0.1 1.0 10 100 1000 Total Porosity (fraction) 1.0 2.0 3.0 Permeability (md) Class 3 Field Class 1 Field 0.5 1.5 2.5 4.0 Class 2 Field Apparent Rock Fabric Numbers Bureau of Economic Geology Cored well New core analysis and thin sections 0 4000 ft 1000 m 0