A PETROPHYSCIAL MODEL TO ESTIMATE FREE GAS IN ORGANIC ALES Michael Holmes, Digital Formation, Inc. Dominic Holmes, Digital Formation, Inc. Antony Holmes, Digital Formation, Inc. Copyright 2011, held by the submitting authors. This paper was prepared for presentation at AAPG Annual Convention and Exhibition, Houston Texas, 10-13 April, 2011. ABSTRACT A method is presented whereby conventional open hole logs density, neutron, Pe (desirable but not essential),, and resistivity can be used to quantify the volume of free gas in organic shale. The calculations involve determining silt and clay mineral volumes in the shale fraction of the rock. associated with the clay minerals is subtracted from total porosity, and the difference remaining is silt porosity. porosity is added to any, usually very small, amounts of clean formation porosity which might exist when shale volume is less than 100%. This summed porosity is then combined with water saturation to determine free gas volumes. A summation of free gas-filled shale porosity can then be compared with cumulative adsorbed gas volume to yield a comprehensive petrophysical analysis. INTRODUCTION Petrophysical interpretation of shale gas reservoirs is not straight forward. Conventional reservoirs can be analyzed to define porosity accessible to hydrocarbons (often termed effective porosity), and its contained fluids water, oil, and gas. By contrast, petrophysical evaluation of shale gas reservoirs is in its infancy. Procedures applicable to conventional reservoirs cannot be applied, and new approaches need to be developed. gas is comprised of two quite different types: a. Gas adsorbed onto the rock surface, and concentrated in the (total organic carbon) fraction of the shale. This gas is only released quite slowly as reservoir pressure is reduced. b. Free gas located in the small to very small volumes of porosity dispersed within the shale reservoir. It is this type of gas termed free gas that assuming a successful stimulation, will flow toward the wellbore, and perform in an analogous manner to conventional reservoirs. In this paper, we will examine methodologies to quantify both adsorbed and free gas volumes. For adsorbed gas, we used techniques that have already been published notably Passey et al. For the free gas, the techniques are novel, developed over the past few years. NOMENCLATURE RhoB NPhi Original raw density gm/cc Original raw neutron limestone, fractions RhoBAdjusted RhoB corrected for gas effects replace gas with water NPhiAdjusted NPhi corrected for gas effects replace gas with water Volume Convert from weight percent by multiplication RhoB/1.25 Rho NPhi Volume Clay 1 Volume Clay 2 Volume Adjust RhoB to reflect shale response only Adjust NPhi to reflect shale response only Determined from cross plot of Rho vs. NPhi chosen by interpreter Determined from cross plot of Rho vs. NPhi Clay 1 point chosen by interpreter Determined from cross plot of Rho vs. NPhi Clay 2 point chosen by interpreter 1
Clay 1 Value of porosity to be assigned to Clay 1 chosen by interpreter FLOW DIAAM OF ANALYSIS Analysis Passey, et al, 1990 Petrophysics Standard y Formation Analysis Clay 2 Value of porosity to be assigned to Clay 2 chosen by interpreter Cross plot porosity of vs. Rho NPhi in weight % Volume of Adsorbed Gas Correct the density and neutron logs for gas effects Using the corrected logs, determine component volumes: Matrix in Volume Effective Volume Volume Saturation Clay (Clay 1 Volume x Clay 1 ) + (Clay 2 Volume x Clay 2 ) Free Clay From density and neutron responses in the shale volume, determine the volumes of: Clay #1 Clay #2 From input porosity values of Clay #1 and Clay #2, determine: Clay Phi E Effective (non-shale fraction) Free Available Phi + Free E. where: Is the calculated greater than or equal to Clay? Yes Calculate: Free = - Clay Free Available = PhiE + Free No Water Saturation (S ) W R = Water Resistivity W R = Apparent Water Resistivity ( ) Wa Phi = Total T m = Cementation Exponent R = Formation Resistivity t Free Gas Free Available 1 TRADITIONAL ALE PETROPHYSICAL MODEL Non-Clay Matrix Organic Matter Matrix Clay Minerals Clay Bound Water Mobile and Capillary Bound Water Hydrocarbons Effective Pore Space Total Pore Space Calculate: Water Saturation Free Gas-Filled PETROPHYSICAL CALCULATIONS ADSORBED GAS The starting point of the analysis is the identification of organic rich shale sequences, contrasted with organic lean shales. Using the R technique of Passey et al, a value of resistivity in organic lean rocks is chosen, and then equivalent choices are made for porosity log responses. The second step is to calculate values. Calibration to the specific reservoir is required by an appropriate choice of level of organic metamorphosis (LOM), which is directly related to vitrinite reflectance (Ro). The calibration is best achieved if sample measurements of are available. can then be converted to volumes of adsorbed gas. For the Barnett, the transform is: Gas content, SCF per ton = 16 x (wt%) The multiplier (16) is probably reservoir specific, but there is little published data. For metric systems the gas content is converted to SCM per ton by dividing the above calculation by 35.3147 ft 3 /m 3. 2
Adsorbed gas-in-place can then be calculated using the following equation: Adsorbed gas = Water Multiplier x Area x h x RhoB x Gas Content 0 100 200 300 400 0 100 200 300 400 Units are MMCF (or MCM) Water Multiplier = 1359.7 tons water per acrefoot (or 11,023.3 tons water per hectare-meter) Area is measured in acres (or hectares) h in feet (or meter) RhoB in gm/cc R Base 0.2 2 20 200 2000 Deep NPhi Base 0 0.1 0.2 0.3 0.4 NPhi 0 100 200 300 400 0 100 200 300 400 RhoB Base 2 2.2 2.4 2.6 2.8 3 RhoB Dt Base 200 160 120 80 40 0 DT Hot colors indicate increasing 0 GAPI 400 V 0 V/V 1 0 1-0.01 wt 100-0.01 wt - Sonic 100 0.01 wt 100 0.01 wt 100 Comparison of core (illustrated by black squares) with petrophyscialy determined 1:100 MD in ft 1420 1440 1460 1480 1500 1520 1540 1560 1580 0 21 3
MODIFID PETROPHYSICAL MODEL Non-Clay Matrix Clay Solids Clay Bound Water Organic Matter Adsorbed Gas Free Water Free Matrix Free Gas Component Sub Set Components Response Response Non-shale Matrix -- ~0 matrix Non-y Solids Solids Total Solids: ~0 PETROPHYSICAL CALCULATIONS FREE GAS that can be occupied by free gas in shales is of small to very small volumes. This can be supplemented by additional small volumes of effective porosity (Phi E ) in the non-shale fraction of the rock. In the terminology used in this presentation: Effective + Free Free Available = Phi E + Free This Free Available has both gas and water as contained fluids Free Gas = Free Available x (1-S W ) The challenge is how to calculate the small volumes of free shale porosity. Phi E is available using standard conventional reservoir analysis. Our approach is to segment the reservoir into a number of compartments, and then to determine petrophysical properties for each compartment. Our model currently uses only the density and neutron log combination, but we plan to add Pe responses in the near future. Compartments and their petrophysical characteristics are: Solids: Clay Solids Fluids: Clay Water Fluids:?? 1.0 1.0?? 1.25 0 PhiE 1.0 0 At each level, volumes of the major components can be defined: Component Volume Non Matrix 1 V PhiE V Phi V x PhiE T Phi Where: Phi T = Total (from a density/neutron cross plot) Phi = /neutron cross plot porosity reading in shale 4
Procedures are as follows: 1. Determine if there is gas in the porosity of the clean (non-shale) fraction, if the following criteria are met: a. Effective density porosity (Phi DE ) is greater than density/neutron cross plot (Phi DNE ) b. Effective neutron porosity (Phi NE ) is less than density/neutron cross plot (Phi DNE ) AAPG Annual Convention and Exhibition, Houston Texas, 10-13 April, 2011 2. Run a standard shaley formation analysis to calculate effective porosity, Phi E, and volume of shale, V 3. From previously calculated values of, convert from weight percent to volume percent:, volume % = (RhoB/1.25 x weight% 4. Solve for density and neutron responses in the shale fraction of the formation. Example cross plot of Rho vs. NPhi from the Antrim of Michigan: If differences do exist, adjust density and neutron porosities and recalculate RhoB and NPhi: RhoB Adjusted and NPhi Adjusted Example of the change for a gas reservoir: 3 2.9 2.8 2.7 2.6 2.5 2.4 2.3 2.2 2.1 2 1.9 RHOB_ADJUSTED 3 2.9 2.8 2.7 2.6 2.5 2.4 2.3 2.2 2.1 2 1.9 Salt Sulfer 0 Anhydrite Dolomite Sandstone 20 25 Limestone 45 25 30 30 35 40 5 0 5 10 15-0.05-0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 Salt Sulfer 0 Anhydrite For most shale gas plays, the adjustment is minimal. 20-0.05-0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 NPHI_L_ADJUSTED 25 Dolomite Sandstone 20 25 Limestone 45 25 30 30 35 40 5 0 5 10 15 Original 20 25 30 30 Adjusted 35 35 RHO_WITHOUT_ 3.6 3.4 3.2 3 2.8 2.6 2.4 2.2 2 1.8 1.6 1.4 1.2 1 Clay 1 5. From the Rho vs. NPhi cross plot shown above, choose the following three points to create an envelope encompassing the majority of the data, illustrated by the triangle:, Clay 1, and Clay 2 Clay 2 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 NPHI_WITHOUT_ Then, choose Clay values appropriate to Clay 1 and Clay 2. These Clay Porosities refer to the amount of bound water associated with the clay minerals. For each of the three data points, determine: %, % Clay 1, and % Clay 2 Clay 1 contribution = % Clay 1 x Clay 1 Clay 2 contribution = % Clay 2 x Clay 2 Clay = Clay 1 contribution + Clay 2 contribution 5
Calculate from the Rho vs. NPhi cross plot Calculate Free : Free = Clay On a vs. Clay cross plot, check that is mostly equal to or greater than Clay : CLAY_POROSITY_FINAL 0 0.05 0.1 0.15 0.2 0.25 0 0.05 0.1 0.15 0.2 0.25 ALE_POROSITY_FINAL If it is not, adjust choice of either Clay 1 or Clay 2. 6. Add Free to Phi E, to yield Free Available. This is the porosity that is available for free gas. Compare with core porosity measurements if they exist. In the event of significant mismatch, edit and repeat item 5. 7. Calculate water saturation by empirical observation; the best calculation has been found by using a standard water resistivity approach: S W = (R W /R Wa ) 0.5 R W = Water Resistivity R Wa = Apparent Water Resistivity (Phi T m x R t ) Phi T = standard calculation of total porosity m = cementation exponent R t = formation resistivity 8. Calculate free gas porosity Free Gas = Free Available x (1 S W ) 9. Using appropriate gas formation volume factor, determine free gas-in-place MECHANICAL PROPERTIES If acoustic data are available, mechanical properties can be calculated using: Young s Modulus Bulk Modulus Shear Modulus Poisson s Ratio Curves required for these calculations are: Compressional Acoustic Shear Acoustic If no shear data are available, reasonable estimates can be made through Rock Physics modeling. Using density and neutron as input, reconstruction of both pseudo compressional and pseudo shear data are possible. If there is convergence on measured compressional data, then it is reasonable to presume the shear pseudo curve is reliable. 6
Comparisons of Young s Modulus with Poisson s Ratio can be used to distinguish brittle from ductile rock. This distinction is valuable when completion intervals are chosen. 0 GAPI 400 V 0 V/V 1 0 1-0.01 wt 100-0.01 wt - Sonic 100 0.01 wt 100 1:100 MD in ft B/D 0 21 B D Young's Modulus 10 8 6 4 2 0 Ductile Brittle 0 0.1 0.2 0.3 0.4 0.5 Poisson's Ratio 0.01 wt 100 1420 1440 1460 1480 1500 1520 1540 1560 1580 7
ADSORBED VS. FREE GAS The distribution of Adsorbed vs. Free Gas can be shown by comparing cumulative values of the two entities. Since well productivity is influenced by both types of gas free gas will tend to be produced more quickly than adsorbed gas it is important to understand their relative abundance in the reservoir. An example from the Antrim, Michigan: Cumulative Adsorbed Gas - MMCF 0 50 100 150 200 250 300 EXAMPLES Data from two gas shale reservoirs are presented: 8 0 10 20 30 40 50 60 70 80 90 100 Cumulative Free Gas - MMCF a. Antrim, Michigan b. Devonian, Western Canada Both wells have core pyrolysis measurements of, and both core measurements include porosity, water saturation, grain density, and permeability. Data interpretation includes a. Standard analysis to calculate total and effective porosity, shale volume, and water saturation the starting point for shale evaluation b. Petrophysical evaluation of using the Passey et al technique c. analysis, to compute free shale porosity and associated water saturation 0 GAPI 200 200 400 Zone01 Zone02 Zone03 Zone04 1350 1400 1450 1500 1550 1600 1.95 g/cc 2.95 0.45 v/v 0.15 Sonic 140 us/ft 40 Resistivities RT 0.2 ohmm2000 RXO 0.2 ohmm2000 d. Component analysis, showing volumes of: Non- matrix Dry clay Clay bound water Free water in shale porosity Gas in shale porosity expressed as a volume Example from the Antrim, Michigan note comparisons of core measurements (illustrated by symbols) with petrophysical calculations: Logs RhoB 2 g/cc 3 0.45 V/V 0 Caliper 5 IN 25 Bulk 2 g/cc 3 Resistivity Logs Deep and Total 2 OHMM 20000 1 V/V 0 Effective 1 V/V 0 0 GAPI 400 1:100 MD in ft 0 1 Water Saturation 1500 1520 1540 1560 1580 1600 1620 shows fair to good comparison to petrophysics Saturations Sw 1 v/v 0 100 % 0 Matrix Properties 2.5 g/cc 3 Rhoma Wet 2.5 g/cc 3 2.5 g/cc 3 S = Sandstone L = Limestone D = Dolomite A = Anhydrite S L D A Bulk Volumes Free Gas 0.2 v/v 0 20 % 0 Hydrocarbons Water 0 21 0.1 unkn 100 0.1 unkn 100 0.1 wt 100 Pay GS NS Pay Gas Content 0 scf/ton 200 0 scf/ton 200 Perm 0.001 md 10 Stacked Data Clean Matrix 1 2 3 4 5 6 7 8 9 10 11 12 Track descriptions are as follows: compared with petrophysical is always very good 1: Raw Data and logs 2: Raw Data Resistivity 3: Bulk Volumes shale, porosity, matrix 4: Water Saturation 5: Grain wet and dry 6: Fluids in Free Available 7: Level of hot colors indicate high 8: from Passey et al 9: Net Pay 10: Adsorbed gas content 11: Permeability 12: Component Analysis Clay 1 Clay 2 PhiE B/D porosity shows fair to good comparison to petrophysics B D
Example from Devonian, Western Canada: Logs RhoB 2 KG/M3 3 2 ohm.m 20000 1 V/V 0 Effective 0.45 unkn 0 DT 110 USEC/M 45 Caliper 5 cm 25 Resistivity Logs Deep and Total 1 V/V 0 0 gapi 400 0 1 1:200 MD in m 2330 2340 2350 2360 2370 2380 2390 2400 2410 2420 2430 porosity shows fair to good comparison to petrophysics 1 2 3 4 5 6 7 8 9 10 11 12 See previous example for track descriptions. CONCLUSIONS Saturations Matrix Properties Bulk Volumes Pay Gas Content Perm Sw Free Gas 0 21 GS 0.001 md 10 1 v/v 0 2.5 G/C3 3 0.2 v/v 0 0.1 unkn 100 0 scm/ton 200 Rhoma Wet NS 0.001 unkn 10 2.5 unkn 3 0.1 unkn 100 0 scm/ton 200 20 unkn 0 DT Pay Hydrocarbons 0.1 unkn 100 S = Sandstone Water L = Limestone D = Dolomite permeability 0.1 unkn 100 data is A = Anhydrite S L D A The methodology requires a standard suite of open hole logs: Resistivity Acoustic desirable but not essential Pe desirable but not essential Adsorbed gas volumes are available from the straight forward technique of Passey et al. Free shale gas porosity i.e. porosity available to contain free (not adsorbed) gas is determined by logical distinction of reservoir components built into the system and checks for impossible results due to unrealistic input. AAPG Annual Convention and Exhibition, Houston Texas, 10-13 April, 2011 minimal, but shows fair correlation with petrophysics compared with petrophysical is always very good Stacked Data Clean Matrix Clay 1 Clay 2 PhiE If core data are available, additional fine-tuning of procedures is possible. However, the system does not require core data to run. Clearly, it is helpful to have knowledge of the reservoir sequence. Probably the most important piece of information is the likely range of. Knowing the clay mineral species assemblage helps in the interpretation of the shale log responses. Comparison of free gas volumes with adsorbed gas volumes will help in deciding which intervals to complete. If acoustic data are available, or can be estimated from Rock Physics modeling, mechanical properties comparisons can be used to distinguish ductile from brittle rocks. This data is again helpful in deciding when to complete. REFERENCES 1. Passey, Q. R., S. Creaney, J. B. Kulla, F. J. Moretti, and J. D. Stroud, 1990, A Practical Model for Organic Richness from and Resistivity Logs, AAPG Bulletin V. 74, No. 12, p. 1777-1794. 2. Denver Well Logging Society, 2008 Spring Workshop, Special Analyses. 3. Denver Well Logging Society, 2010 Fall Workshop, Petrophysics 4. Passey, Q.R., K.M. Bohacs, W.L. Esch, R. Klimentidis, and S. Sinha, ExxonMobil Upstream Research Co, 2010, From Oil-Prone Source Rock to Gas-Producing Reservoir Geologic and Petrophysical Characterization of Unconventional -Gas Reservoirs, SPE 131350 ABOUT THE AUTHORS Michael Holmes has a Ph.D. from the University of London in geology and a MSc. from the Colorado School of Mines in Petroleum Engineering. His professional career has involved employment with British Petroleum, Shell Canada, Marathon Oil Company and H.K. van Poollen and Associates. For the past 20 years he has worked on petrophysical analyses for reservoirs worldwide under the auspices of Digital Formation, Inc. 9
Dominic I. Holmes has a BS in Chemistry and a BS in Mathematics from the Colorado School of Mines. He has been involved with the development of petrophysical software for 20 years with Digital Formation, Inc., particularly with regards to the presentation of petrophysical information in a graphical format. Antony M. Holmes has a BS in Computer Science from the University of Colorado. He has been involved with the development of petrophysical software for 20 years with Digital Formation, Inc., particularly with regards to the implementation of petrophysical analyses. 10