2014 RMS-AAPG Luncheon, Oct 1st, Denver, CO A Comprehensive Deterministic Petrophysical Analysis Procedure for Reservoir Characterization: Conventional and Unconventional Reservoirs 2014 RMS-AAPG Luncheon, Oct 1st, Denver, CO Presented by: Michael Holmes, Antony Holmes and Dominic Holmes Digital Formation, Denver, Colorado, 2014

Outline Introduction Procedures Conventional and Unconventional reservoir petrophysical models Procedures 1. Standard shaley formation petrophysical model 2. Unconventional reservoir petrophysical model Four porosity components model TOC calculations Standard vs. shale only density/neutron comparisons Free and adsorbed hydrocarbons

Outline Procedures Cont. Examples 3. Fracture analysis 4. Relative permeability model 5. Rock physics model and mechanical properties – brittle vs. ductile 6. Comprehensive petrophysical model Examples Niobrara, Colorado Barnett Shale, Texas Antrim Shale, Michigan Shale Gas, Western Canada Bakken, Montana Tight Gas, Colorado New Albany, Illinois

Conventional Reservoirs Introduction Conventional vs. unconventional reservoir petrophysical models Conventional Reservoirs Shale Matrix Effective Porosity Water Oil/Gas The Reservoir

Unconventional Reservoirs Introduction Unconventional Reservoirs Free Shale Porosity PhiFS Water Adsorbed Hydrocarbon ? Free Hydro-carbons Total Organic Carbon-TOC Free Hydro-carbons Bound Water Free Water Clay Fluids 𝐏𝐡𝐢 𝐂𝐥𝐚𝐲 Total Porosity 𝐏𝐡𝐢 𝐭 𝐏𝐡𝐢 𝐞 Four Porosity Components 𝐕 𝐒𝐇 𝐕 𝐦𝐚𝐭𝐫𝐢𝐱 Solids Effective Porosity Phie Non Shale Matrix Silt Clay Solids *Note: Components not to scale

Procedures In the following discussions an example from the Niobrara (Colorado) is used to illustrate procedures

Procedures 1 – Standard Shaley Formation Analysis Shale Matrix Porosity Bulk Fluid Volumes Grain Density Raw Data Fluids Lithology Pay RWA Perm Core data symbols/heavy line

Procedure 2 – Unconventional Reservoir Petrophysical Model Four Porosity Component Model The goal is to calculate the four porosity components from the unconventional reservoir model Effective Porosity Phi e Total Organic Carbon TOC Clay Porosity Phi Clay Free Shale Porosity Phi FS   TOC Phi Components Phi e FS Clay TOC Phi e Clay FS

TOC Calculation TOC Passey et al Hot colors indicate increasing TOC Note Mismatch Comparison of core TOC (Illustrated by thick black line) with petrophysically determined TOC from each porosity log TOC Passey et al Responses in Organic – lean intervals TOC = Total Organic Carbon

TOC Calculation TOC Schmoker Note Mismatch TOC Schmoker Schmoker has three different correlations of RhoB with TOC: High Appalachian correlation Low Appalachian correlation Williston Basin Bakken

Regular Density/Neutron Cross Plot Calculations are: Total porosity Phit Shale volume VSH Effective porosity Phie Matrix volume Vma Fluid saturation in effective porosity – oil, water, gas Our preference is to use a density/neutron cross plot for total porosity, to minimize influences of changing matrix and fluid properties

Shale Only Density/Neutron Cross Plot Subtract the non-shale components are: Effective porosity – account for fluid content Matrix Volume Total Organic Carbon – as a volume fraction Determine porosity from the shale only density/neutron cross plot Calculate clay porosity as the product of cross plot porosity and 𝑉 𝑆𝐻 The plot can also be used to estimate clay mineral species

Free Shale Porosity and Free Available Porosity Free Shale Porosity = Total Porosity – Effective Porosity – Clay Porosity Free Available Porosity = Free Shale Porosity + Effective Porosity Clearly free shale porosity is zero or greater. If negative values are calculated it might be a consequence of incorrect estimates of TOC, an incorrect assumption of TOC density, or an incorrect calculation of shale volume. Free Shale Porosity PhiFS Slight Mismatch

Free vs. Adsorbed Hydrocarbons Free hydrocarbons are located in the free available porosity element, and are calculated using standard approaches Publications on calculating adsorbed hydrocarbon volumes are sparse. Empirical relations are: Gas – Published Relation Adsorbed G.I.P. (SCF) = 1359.7 X Area X Thickness X RhoB X (16 X TOC) Oil – Suggested Relation Adsorbed O.I.P. (Bbl) = S2 X 0.0007 X RhoB X h X Area X 7758 S2 = Hydrocarbons generated by thermal cracking

Procedure 2 – Unconventional Reservoir Petrophysical Model 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Raw Data Clean Formation Shale Formation 1 Gr, SP 4 Saturations 7 Porosity Comparison 10 Net Pay – Shale 13 2 Porosity 5 Bulk Volumes 8 Permeability 11 Shale Model 14 TOC Comparison 3 Resistivity 6 Lithology 9 Net Pay – Clean 12 Porosity Comparisons

Procedure 3 – Fracture Analysis from Standard Open-Hole Logs The methodology involves examining rates of change of curve magnitude with depth Criteria are established by the interpreter for “abnormally rapid” change If the change to higher porosity is deemed to be too rapid for normal sedimentary processes, then open fractures are suggested If the change is to lower porosity, closed (healed) fractures are suggested Results can be compared with image logs, and there is usually quite good comparison with this petrophysical methodology Calculations involve all available logs Porosity Resistivity Calculated Matrix Curves Caliper Density Correction

Fracture Analysis Example Note Fracture concentration in Niobrara C and Ft. Hays Individual Log Responses Stacked Data Pink O = Open Fractures – ? Low stress Blue C = Closed Fractures – ? High stress

Procedure 4 – Relative Permeability Model Reservoir 𝐒 𝐰 > 𝐒 𝐰𝐢 𝐒 𝐰 Buckles Relation Phie X S wi =Constant Holmes Adaptation Phie Q × S wi =Constant Slope = Q 𝐒 𝐰𝐢 Reservoir at 𝐒 𝐰𝐢 Solve the Corey relation S we = S w − S wi 1 − S wi K rw = S we 4 Water K rh = 1− S we 2 1− S we 2 Hydrocarbons

Relative Permeability Example Oil Well KEFF = 𝐊 𝐫 ×𝐏𝐞𝐫𝐦 Reservoir Components Fluid Volume Relative Permeability Effective Permeability 𝐒 𝐖 > 𝐒 𝐖𝐢 Pay Flag Water/Oil Ratio

Relative Permeability Example Gas Well Low water adjacent to pay High water adjacent to pay Reservoir Components Fluid Volume Relative Permeability Effective Permeability 𝐒 𝐖 > 𝐒 𝐖𝐢 Pay Flag Water/Gas Ratio

Procedure 5 – Rock Physics Model and Mechanical Properties – Brittle vs. Ductile To calculate mechanical properties, the following measurements are required Acoustic compressional Acoustic shear Density Often acoustic shear is not available but can be estimated from other logs. The example shows pseudo curves based on the Krief geophysical model (Dipole Sonic not run in the Niobrara example). Dipole Sonic

Rock Physics Model and Mechanical Properties Raw Log DT DTS/DT DTS Density Neutron

Young’s Modulus vs. Poisson’s Ratio Brittle Ductile

Procedure 6 – Comprehensive Petrophysical Model A Standard Template is Used for All Examples Clay Porosity 1 2 3 4 5 6 7 8 9 10 11 1. Fluid Saturation 4. Permeability 7. Water/Oil Ratio – Oil Reservoirs Water Bbl per MMCF – Gas Reservoir 10. Porosity Types – Phie and shale porosity 2. Bulk Volume – non shale fraction 5. Pay Flag – Clean Formation Yellow = Gross “Sand” Red = Net “Sand” Green = Pay 8. Brittle vs. Ductile 11. Porosity Types – Free Shale Porosity and TOC 3. Lithology 6. Pay Flag – Shale 9. Fractures

Niobrara, Colorado – Oil Clay Porosity Niobrara benches are brittle Very little shale contribution Niobrara shales are ductile Variable free shale porosity Fractures in Niobrara & Ft. Hays Fair to good core/log Correlation

Bakken, Montana – Oil High free shale porosity Very high TOC Clay Porosity High free shale porosity Very high TOC Water production from lower Three Forks

Barnett, Texas – Shale Gas Zone 1 – 4 Shale Higher free shale porosity than zone 5 Shales show variable ductile/brittle responses Good correlation core/logs

Antrim, Michigan – Shale Gas Pay contribution from most of the shales Shale shows variable brittle/ductile responses High TOC and free shale porosity Fractures sporadic Good correlation core/logs

Western Canada – Shale Gas Major contribution from shales Shales are entirely brittle High values of free shale porosity Good correlation core/logs

Piceance Basin, Colorado – Tight Gas Clay Porosity Minor shale contribution Fractures common Very low TOC and free shale porosity Sand intervals are brittle

Bakken, North Dakota Oil-in-place in shale: 10.2 MMBO per 640 acres Clay Porosity Oil-in-place in shale: 10.2 MMBO per 640 acres Carbonate: 9.6 MMBO High free shale porosity and TOC in both shale intervals

New Albany, Illinois Clay Porosity Good correlation Core/Log

References Michael Holmes, Antony Holmes, and Dominic Holmes “A Petrophysicial Model to Estimate Free Gas in Organic Shales”, Presented at the AAPG Annual Convention and Exhibition, Houston Texas, 10-13 April, 2011. Michael Holmes, Antony Holmes, and Dominic Holmes “A Petrophysical Model for Shale Reservoirs to Distinguish Macro Porosity, Micro Porosity, and TOC”, Presented at the 2012 AAPG ACE, Long Beach, California, April 22-25. Holmes, Michael, et al. "Pressure Effects on Porosity-Log Responses Using Rock Physics Modeling: Implications on Geophysical and Engineering Models as Reservoir Pressure Decreases." Prepared for the SPE Annual Technical Conference and Exhibition held in Dallas, Texas, USA, 9-12 October (2005). Michael Holmes, Antony Holmes, and Dominic Holmes “Petrophysical Rock Physics Modeling: A Comparison of the Krief And Gassmann Equations, and Applications to Verifying And Estimating Compressional And Shear Velocities” presentation at the SPWLA 46th Annual Logging Symposium held in New Orleans, Louisiana, United States, June 26-29, 2005 James W. Schmoker “Use of Formation-Density Logs to Determine Organic-Carbon Content in Devonian Shales of the Western Appalachian Basin and an Additional Example Based on the Bakken Formation of the Williston Basin”, Petroleum Geology of the Black Shale Eastern North America 1989. Q.R. Passey, S. Creaney, J.B. Kulla, F.J. Moretti, and J.D. Stroud “A Practical Model for Organic Richness from Porosity and Resistivity Logs”, AAPG 1990.