Direct arrival times and slownesses from wide‐aperture walkaway vertical seismic profile data acquired in a layered anisotropic medium can be processed to give a direct estimate of the phase slowness surface associated with the medium at the depth of the receivers. This slowness surface can, in turn, be fit by an estimated transversely isotropic medium with a vertical symmetry axis (a “TIV” medium). While the method requires that the medium between the receivers and the surface be horizontally stratified, no further measurement or knowledge of that medium is required. When applied to data acquired in a compacting shale sequence (here termed the “Petronas shale”) encountered by a well in the South China Sea, the method yields an estimated TIV medium that fits the data extremely well over 180° of propagation angles sampled by 201 source positions. The medium is strongly anisotropic. The anisotropy is significantly anelliptic and implies that the quasi‐shear mode should be triplicated for off‐axis propagation. Estimated density‐normalized moduli (in units of km2/s2) for the Petronas shale are A11 = 6.99 ± 0.21, A33 = 5.53 ± 0.17, A55 = 0.91 ± 0.05, and A13 = 2.64 ± 0.26. Densities in the logged zone just below the survey lie in the range between 2200 and 2400 kg/m3 with an average value close to 2300 kg/m3.

No abstract

In the seismic exploration domain, the following statements may be said to represent a common understanding. First, a valid earth model with the required properties is crucial for successful pre-stack depth imaging. Second, tomographic inversion is one of the most efficient and appropriate techniques for obtaining such a model. Third, borehole seismic data, because of its acquisition geometry, is very useful for achieving an "all-in-depth" earth model. Finally, integration of different data, such as surface seismic, borehole seismic and well logs, is necessary in order to improve the robustness and reduce the non-uniqueness of the inversion process. However, at the same time, it is known that differences in frequency bandwidth, wave propagation behavior and spatial coverage and resolution mean that such integration is rarely achieved in practice. Integrated tomographic inversion therefore remains a topic of considerable academic and practical interest.In this paper, we present a joint tomographic inversion scheme that brings some hope to resolving the questions above. This technique is demonstrated on a set of different 3D VSP surveys acquired over UNOCAL's Attaka field in Indonesia. The updated 3D model was then used in a 3D pre-stack depth migration. The results were greatly superior to all earlier seismic images obtained over this field.

No abstract

To help design the completion of an Indonesian development well, we have carried outcross-dipole dispersion analyses over a depth interval of approximately 500 ft. Dispersion analyses provide estimates of radial extent of formation damage and indicators of stress- and bedding-induced anisotropies. Most of these sections of this vertical well exhibit sand and shale lithology with shear slowness anisotropy ranging from 5 to 10%. Above the angular unconformity in the deeper section of the well, a clean sand interval (Sand A) shows evidence of cross-dipole dispersion crossovers. Crossing dipole dispersions are indicators of stress-induced anisotropy dominating the sonic data. We have developed a new technique for estimating the maximum horizontal SH, and minimum horizontal Sh stress magnitudes using multi-frequency inversion of wideband cross-dipole dispersions. At the mid-point of sand A, we estimate the overburden stress SV=-2278 psi; the maximum horizontal stress SH=-1843; and minimum horizontal stress Sh=-1698. The fast shear direction is NW5 in this depth interval. Radial profiling of formation shear velocity indicates varying degrees of mechanical alteration extending from one to two borehole diameters in the entire depth. A second clean sand interval B, 300 ft above sand A, shows dipole shear anisotropy on the order of 10%. However, cross-dipole dispersions appear to merge together at high frequencies instead of showing a cross-over. Radial profiling of shear velocity in the two orthogonal directions confirms mechanical damage extending to about 2x the borehole diameter. The near-wellbore region in this interval appears to have deformed (material creep) in an attempt to reduce shear stresses and attain hydrostatic equilibrium. Using the same stress sensitivity coefficients as estimated in the lower sand A, the differential stress (SH-Sh) is estimated to be about 20% larger in the upper interval, than the corresponding value in the lower interval. The fast-shear direction varies abruptly across the angular unconformity, changing from NW5 to NW75. Below the unconformity, this section exhibits beds with dips ranging from 10° to 30°. Cross-dipole dispersions show significant anisotropy and are non-intersecting at higher frequencies. Non-intersecting dispersions indicate bedding-induced anisotropy dominating the cross-dipole data. We have inverted borehole sonic velocities for four combinations of the TI-shale anisotropy, which can be combined with walk away VSPs to obtain all the shale anisotropy constants. These constants are needed in generating synthetic AVO gathers in anisotropic shale formations. Quantitative estimates of the radial extent of near-wellbore damage in this well suggest that perforations should penetrate deeper than twice the borehole diameter to avoid potential permeability impairment caused by near-wellbore mechanical damage. Introduction Optimal completions of wells require both identification and estimation of the radial extent of near-wellbore mechanical alteration that might cause near-wellbore permeability impairment. The near-wellbore mechanical damage characterized by radial profiling of formation shear can be correlated with the skin effect and reservoir productivity index. In addition, a detailed characterization of formation stresses is of help in maintaining wellbore stability and reservoir management. Estimates of formation stresses help us manage reservoirs that are prone to subsidence caused by a significant reduction in pore pressure and an associated increase in the effective stress that exceeds the in-situ rock strength. In addition, the magnitude and orientation of the in-situ stresses in a given field have a significant influence on permeability distribution that help in a proper planning of wellbore trajectories and injection schemes for water or steam flooding.