Page 238 - Reservoir Geomechanics

219 Determination of S 3 from mini-fracs

and the pressure downhole is determined by adding a pressure corresponding to the
static mud column to the surface. PWD records pressure downhole directly and a
number of comparisons have shown that there can be a significant differences between
downhole pressures calculated from surface measurements and actual downhole LOT
measurements (Ward and Beique 1999). This difference could be caused by suspended
solids, pressure and temperature effects on mud density, or measurement error. There
is an additional error during the pumping and shut-in phases that could be due to the
mud gels, mud compressibility or pressure loss in the surface lines.
PWD also accurately measures the pressures imposed on the formation during a
lost circulation incident (Ward and Beique 1999). There is often some uncertainty
about exactly where the losses are happening in a long open-hole section so the PWD
measurement may need to be referenced to the appropriate depth. Sometimes repeated
resistivity logs can help identify the depths at which the losses occur. Similar to what
happens in a LOT, losses of drilling mud will occur at pressures slightly higher than
the least principal horizontal stress. The accurate determination of such pressures with
PWD data yield reliable estimates of the least principal stress because the fracture must
be propagating into the far field away from the wellbore in order for circulation to be
lost.
Finally, ballooning, sometimes called loss/gain or wellbore breathing, is now gener-
ally thought to be caused by the opening and closing of near wellbore fractures (Ward
and Clark 1998). This phenomenon is especially likely to occur when pore pressure is
significantly above hydrostatic and drilling is occurring with an ECD close to the least
principal stress. In such cases, small mud losses can occur during drilling that, when
the pumps are turned off, bleed back into the wellbore. In Figure 7.6a, we observe
that the pressure drop from the ECD to the static mud weight is quite abrupt when
the pump stops and then increases abruptly when drilling resumes. Note the markedly
different behavior in Figure 7.6b– when the pump is shut off for a connection, the
pressure slowly decays, then slowly builds up when the pump is turned back on. This
behavior is reminiscent of a balloon because it implies the storage of drilling fluid upon
pressurization and the return of this fluid into the wellbore when the pumps are shut off.
Thus, the PWD signature during ballooning has a distinctive curved pressure profile
(Figure 7.6b) as closing fractures bleed fluid back into the wellbore and fractures are
refilled as circulation is resumed. The ECD at which ballooning occurs can be used
as a lower bound for the magnitude of the least principal stress (if S 3 waslower, lost
circulation would have occurred). In fact, it has been argued by Ward and Clark (1998)
that unless the ECD was close to S 3 , ballooning cannot occur. Modeling by Ito, Zoback
et al. (2001) indicates that the most likely reason ballooning occurs is that en echelon
tensile fractures form in the wall of a deviated well (see Chapter 8) that store fluid at
the pressure corresponding to the ECD during drilling. When the pump is shut off and
the pressure drops to the static mud weight, the mud comes out of the fractures and
back into the wellbore.