Microseismic Monitoring and Geomechanical Modelling of CO2 - bris

More documents

Recommendations

Info

CHAPTER 7. FORWARD MODELLING OF SEISMIC PROPERTIES compute the additional compliance caused by the presence of the discontinuities, using equation 6.8. By rearranging equation 6.7, I can compute the background stiffness from ∆S and the actual stiffness C in . In compliance terms S r = S in − ∆S. (7.5) The background compliance is then rotated into the global coordinate system and inverted to give the background stiffness C r to be used in the stress dependent calculations. The sensitivity of a rock to stress in this model is controlled primarily by the initial crack density tensor. Doug Angus and I have performed extensive analysis of literature data in order to calibrate this parameter (Angus et al., 2009), and have found reasonable consistency. We find that ξ 0 is between 0.05 for stress insensitive rocks and 0.3-0.5 for very stress sensitive rocks (Chapter 6). In contrast, the selection criteria for the background stiffness tensor C r , defined by the behaviour at high stress or petrophysical analysis, is not as well constrained. Therefore I prefer the second initialisation approach described, assigning a fixed initial crack density and then computing the appropriate background stiffness. 7.2.3 Stress dependence Having computed the background stiffness C r , the effects of stress changes on the dynamic stiffness can be computed for subsequent ELFEN timesteps. The initial crack density tensor ξi 0 and the background stiffness tensor C r are rotated into the coordinate frame defined by the principal stress directions. The updated crack density tensor is computed using using equations 6.37 - 6.39, and this additional compliance is added to the rotated background stiffness using equation 6.11. The updated stiffness is then rotated back into the global coordinate system. This provides the full, dynamic, stress dependent, anisotropic stiffness tensor, C, at each timestep. It is assumed that, because the compliant pore space has negligible volume, stress changes have no effect on the rock density. 7.2.4 Fractures Throughout this thesis I have differentiated between cracks, which are small, pervasive features of a similar size to the grains, and fractures, which are larger scale features. As well as the stress dependent cracks, SeisModel c⃝ can include sets of aligned fractures superimposed on top of the stress dependent fabric already calculated above. SeisModel c⃝ uses the method outlined in Chapter 3, adding the additional compliance introduced by the fractures to the inverse of the stress dependent stiffness computed above. The compliance of the fractures can be computed using either the low frequency, high frequency or frequency dependent models (Chapter 3). Any number of aligned fracture sets can be added in this manner. I assume that the fractures are of insignificant volume, so there is no alteration to the rock density. 138
7.3. RESULTS FROM SIMPLE GEOMECHANICAL MODELS 7.2.5 Fluid substitution SeisModel c⃝ also computes the changes in seismic velocity induced by changes in fluid saturation. SeisModel c⃝ provides two options for doing this - using the low frequency anisotropic Gassmann equation (Brown and Korringa, 1975) given in equation 3.9, or the frequency dependent Chapman (2003) model which includes squirt flow and global flow effects. Because the focus of my work is on the effects of stress and not fluid substitution, I will use the low frequency Gassmann approach. The fluid bulk modulus and porosity are provided by the MORE-ELFEN results, but the mineral stiffness is also required to compute the saturated stiffness. In order to estimate this, SeisModel extrapolates from the background stiffness tensor computed in the initialisation step to the limit of zero porosity, using C m = Cr 1 − Φ . (7.6) The fluid saturated density is computed using the porosity and the densities of the fluid and rock, all of which are provided by MORE-ELFEN, using ρ = ρ d + Φρ fl . (7.7) SeisModel c⃝ also provides the option to skip any of these steps if we wish to ignore the contribution from fractures, stress dependence or fluid substitution. 7.2.6 SeisModel c⃝ Output SeisModel c⃝ generates an output file at each timestep. The header of this file contains information about the grid geometry making up the seismic model, and the elevations of any surfaces (such as the top of the reservoir) which can be used by ray-tracing algorithms. For each node, the E, N and Z coordinates of the node are provided, along with the final stiffness tensor computed by the SeisModel c⃝ workflow in 6×6 Voigt notation. These files can then be used to compute seismic observables such as travel-time to the top of the reservoir, reflection coefficients, and shear wave splitting at each timestep. I will now show the seismic results for the geomechanical models constructed in Chapter 5. 7.3 Results from simple geomechanical models 7.3.1 Overburden travel time-shifts The normal incidence travel time for a reflection from the top of the reservoir can quickly and easily be computed by ray-tracing through the elastic model developed, and it is also easily observed on 4-D seismic surveys. This change to this observable has been the most commonly deployed technique used to link geomechanical deformation with geophysical observations (e.g., Hatchell and Bourne, 2005). I use the Christoffel equation (Chapter 3) to compute the changes in velocity of a vertically propagating P-wave through the centre of the 9 simple reservoir models developed in Chapter 5. The user-defined parameters that I use are listed in Table 7.2. Having computed the velocities, I compute the change 139
Page 1:
Microseismic Monitoring and Geomech
Page 5 and 6:
Abstract Capture of CO 2 produced a
Page 7 and 8:
Author’s Declaration I declare th
Page 9 and 10:
Acknowledgments There are many thin
Page 11 and 12:
Table of Contents Abstract Author
Page 13 and 14:
TABLE OF CONTENTS 5.4 Results . . .
Page 15 and 16:
Preface ‘Research is not an end i
Page 17 and 18:
6. D. Angus, J-M. Kendall, J.P. Ver
Page 19 and 20:
1 Introduction A technology push ap
Page 21 and 22:
1.2. CCS OVERVIEW Figure 1.2: CCS s
Page 23 and 24:
1.3. THESIS OVERVIEW for CO 2 to be
Page 25 and 26:
1.3. THESIS OVERVIEW as microseismi
Page 27 and 28:
2 The Weyburn CO 2 injection projec
Page 29 and 30:
2.2. WEYBURN GEOLOGICAL SETTING Fig
Page 31 and 32:
2.2. WEYBURN GEOLOGICAL SETTING N F
Page 33 and 34:
2.4. EVENT TIMING AND LOCATIONS 500
Page 35 and 36:
2.4. EVENT TIMING AND LOCATIONS 500
Page 37 and 38:
2.4. EVENT TIMING AND LOCATIONS Nor
Page 39 and 40:
2.4. EVENT TIMING AND LOCATIONS Fig
Page 41 and 42:
2.5. DISCUSSION 500 1000 1100 North
Page 43 and 44:
2.6. SUMMARY 2.6 Summary • CO 2 s
Page 45 and 46:
3 Inverting shear-wave splitting me
Page 47 and 48:
3.2. INVERSION METHOD the additiona
Page 49 and 50:
3.2. INVERSION METHOD 1. P-wave inc
Page 51 and 52:
3.2. INVERSION METHOD 180 160 140 C
Page 53 and 54:
3.2. INVERSION METHOD 2800 2800 260
Page 55 and 56:
3.2. INVERSION METHOD Loop over γ,
Page 57 and 58:
20 10 5 5 1 20 30 40 80 100 60 40 1
Page 59 and 60:
3.4. SWS MEASUREMENTS AT WEYBURN th
Page 61 and 62:
3.4. SWS MEASUREMENTS AT WEYBURN ei
Page 63 and 64:
3.4. SWS MEASUREMENTS AT WEYBURN ξ
Page 65 and 66:
3.4. SWS MEASUREMENTS AT WEYBURN da
Page 67 and 68:
4 2 2 3.4. SWS MEASUREMENTS AT WEYB
Page 69 and 70:
1 1 1 1 1 1 1 1 1 1 3.5. DISCUSSION
Page 71 and 72:
3.6. SUMMARY 3.6 Summary • I have
Page 73 and 74:
4 A comparison of microseismic moni
Page 75 and 76:
4.2. EVENT LOCATIONS 2400 Velocity
Page 77 and 78:
4.2. EVENT LOCATIONS 200 150 Northi
Page 79 and 80:
4.3. EVENT MAGNITUDES Pressure (MPa
Page 81 and 82:
4.3. EVENT MAGNITUDES 160 140 120 N
Page 83 and 84:
4.4. SHEAR WAVE SPLITTING 50 −3 E
Page 85 and 86:
4.5. INITIAL S-WAVE POLARISATION In
Page 87 and 88:
4.5. INITIAL S-WAVE POLARISATION 6
Page 89 and 90:
4.5. INITIAL S-WAVE POLARISATION of
Page 91 and 92:
4.6. INTERPRETATION OF SHEAR WAVE S
Page 93 and 94:
1 1 5 5 30 4.6. INTERPRETATION OF S
Page 95 and 96:
80 4.6. INTERPRETATION OF SHEAR WAV
Page 97 and 98:
2 3 3.5 4 5 2 1.8 1.4 4 4.6. INTERP
Page 99 and 100:
4.7. DISCUSSION 4.7 Discussion The
Page 101 and 102:
pressure, P fl σ ′ ij = σ ij
Page 103 and 104:
5.2. EFFECTIVE STRESS AND STRESS PA
Page 105 and 106: 5.3. NUMERICAL MODELLING 5.3.1 Flui
Page 107 and 108: 5.3. NUMERICAL MODELLING MORE FLUID
Page 109 and 110: 5.3. NUMERICAL MODELLING (a) (b) Fi
Page 111 and 112: 5.4. RESULTS 0 500 1000 Depth (m) 1
Page 113 and 114: 5.4. RESULTS 1 0.9 0.8 Soft Med Sti
Page 115 and 116: 5.4. RESULTS Overburden Extension i
Page 117 and 118: 5.4. RESULTS 1z:100x:100y 1z:100x:5
Page 119 and 120: 5.5. SURFACE UPLIFT 1 0.9 0.8 Soft
Page 121 and 122: 5.5. SURFACE UPLIFT Figure 5.17: Ma
Page 123 and 124: 6 Generating anisotropic seismic mo
Page 125 and 126: 6.2. STRESS-SENSITIVE ROCK PHYSICS
Page 127 and 128: 6.3. A MICRO-STRUCTURAL MODEL FOR N
Page 145 and 146: 6.4. CALIBRATION WITH LITERATURE DA
Page 147 and 148: 6.4. CALIBRATION WITH LITERATURE DA
Page 149 and 150: 6.5. COMPARISON OF ROCK PHYSICS MOD
Page 151 and 152: 6.5. COMPARISON OF ROCK PHYSICS MOD
Page 153 and 154: 7 Forward modelling of seismic prop
Page 155: 7.2. SEISMODEL c⃝ WORKFLOW time s
Page 159 and 160: 7.3. RESULTS FROM SIMPLE GEOMECHANI
Page 161 and 162: 7.3. RESULTS FROM SIMPLE GEOMECHANI
Page 163 and 164: 7.4. SUMMARY 7.4 Summary • I have
Page 165 and 166: 8 Linking geomechanical modelling a
Page 167 and 168: 8.2. MODEL DESCRIPTION Layer Thickn
Page 169 and 170: 8.2. MODEL DESCRIPTION Unit E (GPa)
Page 171 and 172: 8.3. RESULTS Marl Vugg 0.8 0.8 Norm
Page 173 and 174: 8.3. RESULTS 15 10 Injector Produce
Page 175 and 176: 8.3. RESULTS 2000 15 2000 15 1500 1
Page 177 and 178: 8.4. A SOFTER RESERVOIR 8.4 A softe
Page 179 and 180: 8.4. A SOFTER RESERVOIR 2000 15 200
Page 181 and 182: 8.4. A SOFTER RESERVOIR 2000 Max SW
Page 183 and 184: 8.5. DISCUSSION pore-pressure being
Page 185 and 186: 8.6. SUMMARY 8.6 Summary • I appl
Page 187 and 188: 9 Conclusions Geological storage wi
Page 189 and 190: calibrate. This model has been cali
Page 191 and 192: 9.1. NOVEL CONTRIBUTIONS techniques
Page 193 and 194: 9.2. FUTURE WORK Finally, the compa
Page 195 and 196: Bibliography Abt, D. L., Fischer, K
Page 197 and 198: BIBLIOGRAPHY Gassmann, F., 1951. Ub
Page 199 and 200: BIBLIOGRAPHY Kendall, J.-M., Pilido
Page 201 and 202: BIBLIOGRAPHY Sayers, C. M., 2002. S
Page 203 and 204: BIBLIOGRAPHY White, D., 2009. Monit
Page 205 and 206: A List of Symbols The table below l
Page 207 and 208:
B In Support of Carbon Capture and
Page 209 and 210:
Both China and India are aggressive
Page 211 and 212:
Utsira rock properties vary signifi
show all

Microseismic Monitoring and Geomechanical Modelling of CO2 - bris

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?