Jack Norbeck, Colorado School of Mines, Ernesto Fonseca, Shell, D.V.
Griffiths, Colorado School of Mines, Sau-Wai Wong, Shell
An algorithm based on data obtained during underbalanced drilling operations is
described for identifying the location and properties of productive natural
fractures that are intersected by the wellbore. The criteria used to identify
natural fracture locations are (i) total gas concentration measurements from
mud logs and (ii) mud pit volume. The paper also describes a simple approach
for estimating the fracture permeability. The algorithm is applied to data from
three sets of horizontal parallel wells with spacing of roughly 700 ft. In two
of the three well pairs considered, conductive fracture locations identified in
one well aligned with corresponding features in the parallel well. An
attraction of the proposed methodology is that it makes use of data that is
commonly recorded during drilling, reducing the need for expensive image log
tests. The information obtained from this type of analysis can be used to
improve hydraulic fracture treatment designs.
The economic viability of a well drilled in an unconventional gas reservoir is
largely influenced by the level of connectivity between natural fractures,
stimulated fractures, and the wellbore. Engineers have the ability to control
the wellbore path and, to some extent, the hydraulic fracturing process. On the
other hand, natural fracture systems are outside of the engineer’s control.
While knowledge of the geologic conditions and stress history are helpful to
estimate the characteristics of the natural fracture system in a given
reservoir, the true extent of the natural fracture system in any specific
location is typically unknown (e.g., near wellbore). Several well testing
methods are available to the industry to identify natural fractures near the
wellbore, including acoustic and resistivity image logs, but the poor-quality
results of these techniques do not compensate for the expensive costs in most
cases. The desire to learn as much as possible about the natural fracture
systems present in tight gas plays while striving to keep drilling and
completions costs to a minimum has led researchers to explore the use of
drilling data as a means for natural fracture characterization.
Several reports in the literature indicate that it is possible to locate and
characterize the permeability of natural fractures intersected by the drillbit
during conventional, overbalanced drilling operations through the use of mud
loss data (Dyke et al. 1995; Liétard et al. 1999; Lavrov and Tronvoll 2003;
Huang et al. 2010). However, many wells in unconventional gas reservoirs are
drilled underbalanced. For the case of underbalanced drilling, the downhole
pressure condition requires that an alternate approach for natural fracture
characterization be developed.
The investigation discussed within this report is essentially an exploitation
study to determine whether valuable information concerning natural fracture
systems that intersect horizontal wellbores in tight gas reservoirs can be
discovered through the use of underbalanced drilling data. Methodologies are
developed to determine highly probable conductive natural fracture “zones” and,
hence, characterize natural fracture permeability at these locations. The
methods make use only of data that is commonly recorded by drilling engineers
in practice during underbalanced drilling operations. Due to the high level of
uncertainty inherent to this type of analysis, a practical validation analysis
is identified and subsequently performed.
Hydraulic fracture simulators that include the effects of complex natural
fracture systems have already been developed (e.g., Weng et al. 2011). The
quantitative results obtained from the analysis proposed in this study could be
directly integrated into simulators of this type, potentially reducing the
uncertainty in modeling hydraulic fracture growth.