P - 163 Recent Advances in Boreole Sonic Tecnology Bikas K. Sina, Sclumberger-Doll Researc, Cambridge, Massacusetts 02139-1578, email: sina1@slb.com Summary Mesana tectonic block located in te western Indian state of Gujarat is a fairly well explored, productive block of Nort Cambay Basin. Exploitation in Mesana block as attained a mature stage. Large sized pools are in Kalol Formation occurring at sallower depts in structural plays and operate under water drive wile small sized pools are concentrated at deeper depts in Mesana and Mandali Members of Kadi Formation controlled by strati-structural entrapment operating under depletion drive. Te problems in tese reservoirs are of contrasting nature. Introduction Boreole sonic data plays an important role in seismic exploration, drilling, wellbore stability, well completion, as well as reservoir management. Seismic surveys require boreole sonic data for obtaining accurate compressional slowness (slowness is inverse of velocity) estimates as a function of dept tat are used in te time to dept conversion. Recent studies ave also igligted te importance of using anisotropic velocity models in generating AVO gaters for accurate target locations. Anisotropic velocity models require anisotropic moduli for calculating plane wave velocities as a function of propagation direction. Boreole sonic data from te Sonic Scanner tool in a vertical or deviated well as te potential of providing up to four anisotropy parameters for an assumed transversely-isotropic (TI) or ortorombic formation (Sina et al., 2004). Estimates of formation anisotropy parameters are made using te 3D-anisotropy module. Remaining elastic moduli can be obtained eiter from VSP surveys, core data, or known correlations to enable development of anisotropic velocity model using Kelvin-Cristoffel s equations. Wellbore stability during drilling is planned based on relevant failure models tat require confined compressive and tensile strengts togeter wit te far-field formation stresses to predict safe mud weigt window to avoid tensile fractures or breakouts. Inversion of te Sonic Scanner data can provide estimates of formation stresses and strengts under suitable conditions. Estimates of formation stresses and strengts are obtained by inverting cross-dipole sonic data togeter wit 3D-anisotropy output and radial profiling of sear slownesses (Bratton et al., 2004; Sina 2002; Sina et al., 2006a). Optimal well completion for production requires bot identification and estimation of te radial extent of alteration in reservoir intervals. Formation properties surrounding a boreole can be eterogeneous due to boreole stress concentrations, drilling damage, sale swelling, plastic yielding tat migt cause fluid mobility impairment. We can identify mecanically competent intervals as well as tose exibiting radial alteration in sear slownesses by inverting measured dipole and Stoneley dispersions. Near-wellbore alterations are caracterized by differences between te far-field and nearwellbore slownesses. Optimal depts for fluid sampling are identified by mecanically competent intervals tat exibit nearly uniform sear slownesses away from te boreole surface. In contrast, intervals wit near-wellbore softening caused by mecanical or mobility damage are prone to seal failures or migt lead to tigt pretest. Outputs from te monopole Compressional Velocity Radial Profiling (CRVP), dipole Sear Velocity Radial Profiling (SRVP) of te fast and slow vertical sear velocities, and Stoneley Radial Velocity Profiling (STRVP) of orizontal sear velocity elp in an optimal dept selection for fluid
sampling as well as optimal perforation tunnel design (Sina et al., 2003; Sina et al., 2006a). Fast sear azimut obtained from te processing of cross-dipole sonic data in te presence of dipole dispersion crossovers provides te maximum orizontal stress direction tat is of elp in planning for oriented perforation and stimulation projects. Reservoir management can benefit from an optimal drawdown pressure tat maximizes sand-free production. Accurate estimates of confined rock compressive strengt togeter wit far-field formation stresses enable a more reliable estimate of optimal drawdown pressure tat maximizes production wit reduced risk of sanding (Bratton et al., 2004). Reservoir production can also benefit from te identification of fluid mobility barriers and estimates of te ratio of near-wellbore damaged to far-field mobilities. Radial profiling of fluid mobility in a reservoir interval can provide suc estimates togeter wit mobility logs obtained at an effective radial dept of investigation using eiter MDT pretest drawdown analysis or NMR derived permeability. To provide te aforementioned answer products, we ave developed several data processing algoritms: 1. 3D-anisotropy for obtaining a sub-set of transversely-isotropic (TI) elastic moduli from a vertical or deviated well, 2. Compressional Radial Velocity Profiling (CRVP) of compressional velocity, 3. Sear Radial Velocity Profiling (SRVP) of sear velocities in te two ortogonal planes containing te boreole axis, 4. Stoneley Radial Velocity Profiling (STRVP) of sear velocity in te boreole crosssectional plane, 5. Maximum orizontal stress magnitude estimation using te tree sear moduli, and 6. Horizontal stress gradient estimation using differences between velocity dispersions at two depts. Velocity Radial Profile (SRVP) and Stoneley Velocity Radial Profile (STRVP) algoritms (Pistre et al., 2005). Consequently, results from te 3D-anisotropy can be combined wit boreole seismic data (walk-away/walkaround VSPs) to obtain a complete set of transverselyisotropic (TI) elastic moduli. A complete set of formation moduli for a given litology interval can ten be used to obtain an anisotropic velocity model tat would yield plane wave velocities as a function of propagation direction wit respect to te anisotropic axes. Anisotropic velocity models can ten be used to generate AVO syntetic gaters to determine accurate location of reservoir targets. More importantly, te tree sear moduli c 44, c 55, and c 66 as a function of dept can be used to obtain te following formation attributes: 1. Classification of formation anisotropy, 2. Qualitative indicator of fluid mobility in porous sand reservoirs, 3. Far-field formation stresses, and 4. Relative sear rigidity of te tree ortogonal planes. Figure 1a sows scematic diagram of a vertical and orizontal sections of a wellbore trajectory and identifies te sear moduli tat would be determined from te crossdipole sonic data from a well parallel to te X 3 -, X 1 -, and X 2 -axis. Note tat te sear modulus in te cross-sectional plane of te wellbore trajectory is obtained from te Stoneley data. 3D-Anisotropy Te 3D-anisotropy module outputs up to 4 anisotropic moduli of an ortorombic or TI formation using te compressional, fast-sear, slow-sear, and Stoneley slownesses obtained from a single-well data wit known well deviation from te vertical and true stratigrapic dip from an imaging tool. Te 3D-anisotropy algoritm yields verifiable far-field compressional and sear moduli outside any near-wellbore altered annulus caused by stress concentrations or plastic yielding of te formation etc. Te output of tis module is cecked against te far-field values tat are outside any near-wellbore altered annulus from te Compressional Velocity Radial Profile (CRVP), Sear Figure 1a: Scematic diagram of a vertical well parallel to te X 3- axis and a orizontal well parallel to te X 1-axis. Cross-dipole data from a well parallel to te X 3-axis yields te sear moduli C 44 and C 55 in te two ortogonal vertical planes. Similarly, cross-dipole data from a well parallel to te X 1-axis yields te sear modulus C 55 in te vertical X 1-X 3 plane, and C 66 in te orizontal X 1-X 2 planes.
Figure 1b: Scematic diagram of a vertical well parallel to te X 3- axis and a deviated well wit azimut and deviation. Figure 1b displays scematic diagram of a vertical and deviated sections of a wellbore trajectory. Te deviated well azimut is measured from te nort, and well deviation is referred to te vertical. If we assume a layercake model for a given litology interval wit a vertical TIsymmetry axis, te well azimut does not affect te TIanisotropy moduli because of te rotational invariance about its symmetry axis. Stress magnitudes using boreole sonic Formation stresses play an important role in geopysical prospecting and development of oil and gas reservoirs. Bot te direction and magnitude of tese stresses are required in (a) planning for boreole stability during directional drilling, (b) ydraulic fracturing for enanced production, and (3) selective perforation for prevention of sanding during production. Te formation stress state is caracterized by te magnitude and direction of tree principal stresses. Figure 2 sows a scematic diagram of a vertical boreole in a formation subject to te tree principal stresses. Generally, te overburden stress (SV) is reliably obtained by integrating te formation mass density from te surface to te dept of interest. Consequently, estimating te oter two principal stresses (SHmax and Smin) in te orizontal plane is te remaining task necessary to fully caracterize te formation stress state. Figure 2: Scematic of a boreole in te presence of formation principal stresses wit te boreole axis parallel to te overburden stress. Sonic velocities in formations cange as a function of rock litology/mineralogy, porosity, clay content, fluid saturation, stresses, and temperature. To estimate canges in te formation stress magnitudes from measured canges in sonic velocities, it is necessary to select a dept interval wit a reasonably uniform litology, clay content, saturation and temperature so tat te measured canges in velocities can be related to corresponding canges in formation stress magnitudes. Any cange in porosity caused by normal compaction is accounted for by corresponding canges in te formation effective bulk density and stiffnesses. Assuming tat te measured canges in sonic velocities are largely caused by canges in stresses, it is possible to invert boreole sonic velocities for te estimation of canges in formation stress magnitudes (Sina, 1997, Sina et al., 2000a). Te underlying teory beind te estimation of formation stresses using boreole sonic data is based on acoustoelastic effects in rocks. Acoustoelasticity in rocks refers to canges in elastic wave velocities caused by canges in pre-stress in te propagating medium. Elastic wave propagation in a pre-stressed material is described by equations of motion for small dynamic fields superposed on a statically deformed state of te material. Tese equations are derived from te rotationally invariant equations of nonlinear elasticity (Norris, Sina and Kostek, 1994). Te linear equations of motion for isotropic materials contain two independent elastic stiffnesses tat are defined by te dynamic Young s modulus (Y) and Poisson s ratio (), or equivalently, te two Lame parameters, ( and ). Te equations of motion for pre-stressed isotropic materials contain tree additional nonlinear elastic constants (C 111, C 144, and C 155 ) togeter wit te biasing stresses (Sina,
1982; Norris et al., 1999). A forward solution of equations of motion in pre-stressed materials yields plane wave velocities as a function of principal stresses in te propagating medium. Te presence of a boreole in a triaxially stressed formation causes near-wellbore stress distributions tat can be obtained from linear elastic deformation teory. Nearwellbore stress distributions can be mapped into corresponding sonic velocity distributions provided stress coefficients of velocities in terms of te formation nonlinear constants are known (Sina et al., 2006b). Te stress coefficients of velocities are defined in terms of formation nonlinear constants referred to a reference state close to te in-situ stress state of te formation. Canges in cross-dipole dispersions caused by near-wellbore stress distributions can be inverted to estimate far-field stresses and formation nonlinear constants referred to a local reference state. Formation stresses using te tree sear moduli A reservoir sand in te absence of formation stresses and fluid mobility beaves like an isotropic material caracterized by a sear and bulk moduli. However, a complex saly-sand reservoir is caracterized by anisotropic elastic stiffnesses. Anisotropic elastic stiffnesses and te tree sear moduli are affected by (a) structural anisotropy; (b) stress-induced anisotropy; and (c) formation mobility. Structural anisotropy caused by clay microlayerings in sale is described by transverselyisotropic (TI-) symmetry tat exibits te orizontal sear modulus C 66 to be larger tan te vertical sear moduli C 44 =C 55, in te absence of any stress-induced effects. Sales are impermeable and do not constitute part of a producing reservoir. Here we analyze effects of formation stresses on te effective sear moduli tat can be measured wit te axi-symmetric Stoneley and flexural waves in a fluid-filled boreole. Te acoustoelastic teory relates canges in te effective sear moduli to incremental canges in te biasing stresses and strains from a reference state of te material (Sina, 2002). Te tree sear moduli can be estimated from boreole sonic data. Wit te recent introduction of algoritms for te Stoneley radial profiling of orizontal sear slowness and cross-dipole radial profiling of vertical sear slownesses, we can unambiguously estimate te virgin formation sear moduli C 44, C 55, and C 66. Tese algoritms account for te sonic tool bias and possible near-wellbore alteration effects on te measured sonic data. Differences in te effective sear moduli are related to differences in te principal stress magnitudes troug an acoustoelastic coefficient defined in terms of formation nonlinear constants referred to a cosen local reference state and for a given formation litology. Te following tree equations relate canges in te sear moduli to corresponding canges in te effective principal stresses (Nur and Byerlee, 1971): C C C 44 66 A E V H C ( ), (1) 55 66 A E V C ( ), (2) 55 44 A E H were A E C ( ), (3) C456 2, is te acoustoelastic coefficient, C 55 and C 44 denote te sear moduli for te fast and slow sear waves, respectively; C 456 =(C 155 -C 144 )/2, is a formation nonlinear parameter tat defines te acoustoelastic coefficient; and μ represents te sear modulus in a cosen reference state. Tere are two ways to estimate te acoustoelastic parameter A E for a reasonably uniform litology sand reservoir: 1. Cross-dipole sonic log provides sear moduli C 55 and C 44 in te two ortogonal planes containing te boreole axis. Te acoustoelastic parameter A E is ten given by equation (3) C55 C44 AE, (5) H were C 55 and C 44 denote te fast and slow dipole sear moduli, respectively; and σ H and σ represent te maximum and minimum orizontal effective stress magnitudes, respectively. (4) 2. Wen we ave estimates of te minimum orizontal and overburden stress magnitudes as a function of dept, we can determine te acoustoelastic parameter A E in terms of te farfield sear moduli C 55 and C 66 using te equation (2) C55 C66 AE, (6) V
were we assume tat te effects of permeability on tese sear moduli are essentially similar and small to be negligible. Once we ave determined te acoustoelastic parameter for a given litology interval, we can determine te maximum orizontal stress H magnitude as a function of dept from te following equation ( C55 C44 ) H, (7) AE were C 55 and C 44 denote te fast and slow dipole sear moduli, respectively. Similarly, te minimum orizontal stress magnitude as a function of dept from te following equation track sows te measured compressional (blue), fast-sear (red), and Stoneley (green) slownesses. Te tird track sows te vertical (C 44, red) and orizontal (C 66, green) sear moduli logs for a TI-medium. Te fourt track displays measured dipole sear moduli C 44 and C 55, and an equivalent C 66 * derived from te Stoneley data, all sear moduli in tis track are referred to te boreole axis. Smaller values of C 66 relative to C 44 in te tird track are indicators of larger orizontal mobility tan depts were C 66 is close to or larger tan C 44. Differences between C 66 and C 44 can provide a quick-look indicator of fluid mobility in a sand reservoir. ( C55 C66 ) V. (8) A E Hence, we can estimate formation orizontal stress magnitudes as a function of dept in terms of te tree sear moduli C 44, C 55, and C 66, and te acoustoelastic coefficient A E. Effective stresses are converted into total stresses by adding pore pressure weigted by te Biot coefficient (Nur and Byerlee, 1971). Analysis of Field Data I: 3D-Anisotropy Results Te 3D-Anisotropy algoritm yields up to 4 elastic moduli of an ortorombic formation using boreole sonic data from a vertical or deviated well as sown in Figure 1b. Wen we assume a layer-cake model of te formation caracterized by a transversely-isotropic (TI) anisotropy, te algoritm outputs C 44, C 66, N, and C 33 * as sown in Figure 3. Figure 3 sows a standard output of 3D-anisotropy algoritm tat sows te sear moduli C 44 and C 66 referred to a TI-anisotropy axes. Wen C 66 < C 44, te Tomsen parameter gamma is negative in a sand interval, it is an indicator of iger fluid mobility. In contrast, wen C 66 > C 44, te corresponding interval is caracterized by iger clay content typical of a sale interval. Figure 4 sows a composite log of gamma-ray, caliper, litology, porosity, density, and a sub-set of TI anisotropic moduli in a vertical exploratory well A. Te first track sows gamma-ray (green), neutron porosity (blue), and density (red), wit gray sading denoting saly intervals and yellow sading sowing sandy intervals. Te second Figure 3: 3D-Anisotropy workflow: Compressional, fastsear, slow-sear, and Stoneley slownesses togeter wit well deviation and true stratigrapic dip are te input parameters tat are transformed into te four TI-parameters of a saly formation c44, c66, N, and a combination of c11 and c33, referred to te eart anisotropic axes. Analysis of Field Data II: Estimation of Smax Generally, te overburden stress is obtained by integrating te formation bulk density from te surface to te dept of interest. Te minimum orizontal stress can also be reliably obtained using minifrac or extended leakoff tests at cosen depts. Wen we ave estimates of te overburden stress, minimum orizontal stress, and pore pressure in a sand reservoir, te maximum orizontal stress can be calculated using equations (6) and (7). Note tat wile te sear moduli C 44 and C 55 are readily calculated from te low-frequency cross-dipole sonic data, te sear modulus C 66 in te boreole cross-sectional plane must be
Figure 5. Updated Mecanical Eart Model for an exploration well: Te black, red, blue, and green markers denote te overburden stress, maximum orizontal stress, minimum orizontal stress, and pore pressure, respectively. (Sina et al., 2006b) Figure 4: Processed logs sowing gamma-ray, formation bulk density, and porosity in track 1; compressional, sear, and Stoneley slownesses in track 2; C44_TIV and C55_TIV referred to TI-axes in track 3; and C44, C55, C66 * referred to te boreole axes in track 4 (Sina et al., 2006b) obtained from te Stoneley data and requires an accurate estimate of mud compressional slowness and compensation for te sonic tool structure, fluid mobility effects, and any possible near-wellbore alteration. Figure 5 displays an updated Mecanical Eart Model tat includes te maximum orizontal stress magnitude (red circles) obtained using te tree sear moduli togeter wit te overburden (black circles), minimum orizontal (blue circles) stresses, and pore pressure (green circles) as a function of dept. Inverted stress magnitudes ave been cecked for consistency wit wellbore stability predictions using te internal friction angle and UCS from core data and drilling events observed for mud weigts used (Moos and Zoback, 1990). Summary and conclusions We ave presented a brief overview of geopysical, geomecanical, and petropysical applications of new answer products from te Sonic Scanner tool. In particular, we ave described illustrative examples from te 3Danisotropy and te SHmax magnitude estimation algoritms using sonic data from an exploration well in te Nort sea. Yet oter applications include estimation of bot te SHmax and Smin magnitudes and in-situ rock strengt using radial profiles of te tree sear moduli ( Bratton et al., 2004; Sina et al., 2007). References Bratton, T., Bricout, V., Lam, R., Plona, T., Sina, B., Tagbor, T., Venkitraman, and Borbas, T., 2004, Rock strengt parameters from annular pressure wile drilling and dipole sonic dispersions analysis, 45t Annual Logging Symposium, Soc. of Professional Log Analysts, Paper R. Moos, D., and Zoback, M.D., 1990, Utilization of observations of wellbore failure to constrain te orientation and magnitudes of crustal stresses: Application to continental deep sea drilling project and ocean drilling program: J. Geopys. Res., 95, 9305-9325.
Norris, A.N., Sina, B.K., and Kostek, S., 1994, Acoustoelasticity of solid/fluid composite systems, Geopys. J. Internat., 118, 439-446. Nur, A., and Byerlee, J.D., 1971, An exact effective stress law for elastic deformation of rock wit fluids, J. Geopys. Res., 76, 6414-6419. Pistre, V., Kinosita, T., Endo, T., Scilling, K., Pabon, J., Sina, B., Plona, T., Ikegami, T., and Jonson, D., 2005, A modular wireline sonic tool for measurements of 3D (azimutal, radial, and axial) formation properties: Proc. 46 t Annual Logging Symposium, Society of Professional Well Log Analysts, Paper P. Sina, B.K., 2002, Determining stress parameters of formations using multi-mode velocity data, U.S. Patent 6,351,991, Marc 5. Sina, B.K., Sayers, C., and Endo, T., 2004, Determining anisotropic moduli of eart formations, U.S. Patent 6,714,480, Marc 30. Sina, B.K., Burridge, R., and Kane, M.R., 2003, Sonic well logging for radial profiling, U.S. Patent 6,611,761 B2, August 26. Sina, B.K., Vissapragada, B., Renlie, L., and Tysse, S., 2006a, Radial profiling of te tree formation sear moduli and its application to well completions, Geopysics, 71, E65-E77, November-December. Sina, B.K., Vissapragada, B., Renlie, L., and Skomedal, E., 2006b, Horizontal stress magnitude estimation using te tree sear moduli A Norwegian sea case study, 2006 ATCE, San Antonio, TX, SPE 103709. Sina, B.K., Vissapragada, B., Wendt, A., Kongslein, M., Skomedal, E., Renlie, L., and Sandtrov, E., 2007, Estimation of formation stresses using radial variation of tree sear moduli A case study, 2007 ATCE, Anaeim, CA, SPE 109842. Acknowledgments We acknowledge wit tanks close collaboration and elp from many Statoil and Sclumberger staff in te development and testing of tese new applications of boreole sonic data.