Sensitivity Analysis of Rock Physics Parameters for Modeling

Sensitivity Analysis of Rock Physics Parameters for Modeling Time-Lapse Seismic(4 D) response of Norne Field Amit Suman and Tapan Mukerji 25 th SCRF Annual Meeting May 9 – 11, 2012

Joint Inversion Loop Predicted flow and seismic response Generate multiple models Observed flow and seismic response . SCRF Reservoir Model Evaluate misfit 2

Motivation Production data at time t Dynamic modeling Δ Pressure Δ Saturation Optimize mismatch Rock physics modeling Velocity at time t Seismic data at time t Update parameters 3

Previous Work • Last year we investigated parameter sensitivity for modeling time-lapse seismic and flow data of Norne field • One of the investigated parameters was rock physics model • We didn’t investigate sensitivity of varying rock physics parameters on modeling 4 D response SCRF 4

Questions? “Should we investigate sensitive rock physics parameters in modeling 4 D response? ” “What are the sensitive rock physics parameters in modeling 4 D response? ” SCRF 5

Norne Field Segment E F 1 H E 3 H In this study well log data of two wells are used SCRF 6

Data Available • Well logs (Sw, Sonic, Phi) • Horizons • Well data - Oil , gas and water flow rate - BHP (Bottom hole pressure) SCRF 7

Rock Physics Modeling Near the Well Logs Rock Physics K and G (All Brine) Reservoir Vp and Vs (Initial) K and Phi G and Phi Sonic Sw, Phi Gassmann’s Substitution K and G (All Brine) Facies classification K and G (at Reservoir) Calculate Vp and Vs (All Brine) Populate K , G based on Phi K : Bulk Modulus G: Shear Modulus SCRF 8

Facies Classification Shale Vp / Vs Brine Sand Shaly Sand Oil Sand AI Vsh SCRF 9

Rock Physics Modeling Near the Well Logs Rock Physics K and G (All Brine) Reservoir Vp and Vs (Initial) K and Phi G and Phi Sonic Sw, Phi Gassmann’s Substitution K and G (All Brine) Facies classification K and G (at Reservoir) Calculate Vp and Vs (All Brine) Populate K , G based on Phi K : Bulk Modulus G: Shear Modulus SCRF 10

Sensitivity Parameters in fluid substitution • Clay content • Salinity • Gas-oil ratio (GOR) • Pore pressure The sensitivity of varying above parameters to variations in Response: Sum of seismic P-wave velocity after fluid substitution SCRF 11

Experimental Design Clay content (%) 0 20 40 Salinity (ppm) 150000 155000 160000 GOR 175 200 225 Pressure (Mpa) 25 27 30 SCRF 12

Results of fluid substitution Response Sensitivity to clay content 0 20 40 Sensitivity to pore pressure 25 27 30 Sensitivity to salinity 15000 15500 16000 Sensitivity to GOR 175 200 225 Clay content and GOR are the first and second most sensitive parameters in fluid substitution 13

Rock Physics Modeling Near the Well Logs Rock Physics K and G (All Brine) Reservoir Vp and Vs (Initial) K and Phi G and Phi Sonic Sw, Phi Gassmann’s Substitution K and G (All Brine) Facies classification K and G (at Reservoir) Calculate Vp and Vs (All Brine) Populate K , G based on Phi K : Bulk Modulus G: Shear Modulus SCRF 14

Rock physics model Varying clay content and GOR (9 cases) SCRF 15

Constant cement model Clay content Cement fraction Coordination number 16

Fluid mixing • Seismic velocities depend on fluid saturation as well as saturation scale • Reservoirs with gas are very likely to show patchy behavior Sengupta , 2000 SCRF 17

Effective pressure model Two effective pressure models are selected for sensitivity study SCRF 18

Sensitivity Parameters in modeling 4 D response • Clay content • Gas-oil ratio (GOR) • Coordination number • Cement fraction • Effective pressure model • Fluid mixing (Uniform or Patchy) The sensitivity of varying above parameters to variations in Response: L 1 Norm of change in seismic P-wave impedance after 4 years 19

Experimental Design Clay content (%) 0 20 40 GOR 175 200 225 Coordination number 5 7 9 Cement fraction (%) 1 3 5 Effective pressure Model 1 model Fluid mixing Uniform Model 2 Patchy Total number of cases: 324 20

Methodology Dynamic modeling (1997 -2001) Rock physics modeling Δ Pressure Δ Saturation P-wave impedance in 1997 and 2001 Difference in impedance SCRF Compare 21

Results Clay content = 0 % P-wave impedance change in 4 years (m/s. kg/m 3) SCRF Clay content = 20 % 22

Results Sensitivity to clay content Response 0 20 40 Sensitivity to GOR 175 200 225 Sensitivity to effective pressure model Model 1 Model 2 Sensitivity to coordination number 5 7 9 Sensitivity to cement 1 3 5 Sensitivity to fluid mixing Uniform Patchy 23

Conclusions and Future Work • Clay content is the most sensitive parameter in fluid substitution • Salinity and pore pressure have a lesser impact than clay content • Coordination number is the most sensitive parameter in modeling 4 D response of Norne field • The result of this study will be used in joint inversion of time-lapse and production data of Norne field SCRF 24

Acknowledgement • Statoil for data • Norwegian University of Science and Technology (NTNU) SCRF 25

Conclusions and Future Work • Clay content is the most sensitive parameter in fluid substitution • Salinity and pore pressure have a lesser impact than clay content • Coordination number is the most sensitive parameter in modeling the time lapse seismic signature of Norne field • The result of this study will be used in joint inversion of time-lapse seismic and production data of Norne field SCRF 26