Page 275 - Reservoir Geomechanics

256 Reservoir geomechanics

∼1–5kHz.Adispersiveflexuralwavepropagatesalongtheboreholewallwithavelocity
that is a function of the formation shear modulus. The dispersive nature of the flexural
wave is used to filter out the high frequencies corresponding to short wavelengths that
sampletherockssubjectedtothestressconcentrationaroundtheborehole(Sinha,Norris
et al. 1994). In fact, at low frequencies, the flexural wave velocity approximates the
shear velocity of the formation and has a depth of investigation of approximately 1.5 m
into the formation such that it should be insensitive to the stress concentration around
the wellbore. When the sondes are oriented (either geographically or with respect to the
top or bottom of the well) the polarization direction of the fast and slow shear velocity
directions can be obtained as measured in a plane perpendicular to the wellbore. The
observations of interest are thus shear velocities that correlate with low frequencies that
penetrate deeper into the formation beyond the altered zone around the wellbore. In
addition, borehole ovality is known to bias the results of a shear wave splitting analysis
with dipole sonic logs (Leslie and Randall 1990; Sinha and Kostek 1996), and care must
be taken not to mistake shear polarization in the formation for the effects of ovality.
An example of cross-dipole data in vertical wells is shown in Figure 8.14 (after
Yale 2003). Using cross-dipole data in vertical wells, they showed the direction of
maximum horizontal stress in the Scott field of the North Sea was equally well deter-
mined from shear velocity anisotropy (solid arrows) and wellbore breakouts (dashed
lines). Note that while the stress orientations obtained from the breakout data compare
extremely well with that implied from the cross-dipole analysis, overall the stress ori-
entations in the field seem to be quite heterogeneous with no overall trend apparent. In
fact, the stress orientations seem to follow the trend of faults in the region. We return
to this case study in Chapter 12 where we offer a model to explain these varied stress
orientation observations.
A second example is shown in Figure 8.15 from an oil field in Southeast Asia.
Fifteen vertical wells with dipole sonic logs were used to determine the direction
of maximum horizontal compression (Figure 8.15a). This analysis delineates subtle
differences in stress orientation in the northwest, southwest and southeastern parts of
the field. Ten vertical wells with electrical image data that detected wellbore breakouts
(Figure 8.15b) show the same stress directions in the northwest and southwest parts
of the field. Although no breakout data are available in the southeast part of the field,
the direction implied by the dipole sonic log can be used with confidence to determine
the stress orientation. It seems clear that with appropriate quality control (see below),
shear velocity anisotropy obtained from dipole sonic data in vertical wells can provide
useful information about stress orientation if structural sources of velocity anisotropy
can be ruled out.
Following closely Boness and Zoback (2006), I briefly address the topic of stress
orientation determination from shear velocity anisotropy in this chapter to look at the
more complex problem of determining stress orientation from cross-dipole sonic data
when wells are highly deviated. In this case, the potential influence of bedding planes