Shake, Rattle and Tilt - Understanding Hydraulic Fractures in Tight Gas Sands

Nancy House¹, Thomas Hewett¹, Julie Shemeta², Stephen Wolfe², Brian Fuller³, and Marc Sterling³
¹ EnCana Oil and Gas (USA) Inc., Denver, CO
² Pinnacle Technologies, Denver, CO
³ Sterling Seismic Services, Ltd., Littleton, CO

Completion techniques in tight hydrocarbon reservoirs often include hydraulic fracturing (“frac’ing”) to increase conductivity and improve deliverability. This study covers the integration of surface seismic, VSP, microseismic and tilt meter data to understand the magnitude and direction of induced hydraulic fractures in a complex, tight gas reservoir. Improved understanding of the fracs leads to more accurately executed treatments, identification of bypassed resources and better overall reservoir management. Many disciplines including completion engineering, geology and geophysics all combined to produce an integrated model of the reservoir.

Microseismic. Microseismic monitoring capitalizes on the induced micro fractures that occur as the pressure increases in the formation during a frac job. The increased pore pressure lowers the effective normal stress along small naturally occurring fractures in the rock, creating a shear failure or a “micro”seismic event, which radiates primary (P) and secondary (S) waves. Using highly sensitive geophones, these P- and S-waves can be recorded and located in time and space (Albright and Pearson, 1982; Warpinski, et al., 2001).

To record the microseismic events, a monitor well close to the proposed treatment well is instrumented by deploying a specially designed geophone tool string attached to a fiber optic wireline. The monitor well must be close enough to the frac well to record the microseism, typically within 1000 ft. The tools are lowered in the monitor wellbore at a depth that straddles the stimulated section of the formation. During the frac treatment, the seismic data is continuously recorded in real-time at the surface and for up to several hours after the treatment ends, until the level of induced seismicity drops.

P- and S-wave arrival times are used to locate each event. The separation time of the P- and S-waves determines the distance; the relative arrival times along the different geophone depths yields the event elevation. Because the particle motion of the P-wave is parallel to its propagation direction, the direction of the event is determined from analysis of the P-wave hodogram, a cross plot of the horizontal components (Figure 1).

Tiltmeter mapping. A frac causes a deformation field in the vicinity of the induced fracture. While the deformation is too small to measure directly, tiltmeters can be used to measure the gradient of the deformation and from this the fracture properties can be inferred. These measurements are almost entirely independent of the reservoir mechanical properties and in-situ stress state.

Surface tilt mapping. An array of tiltmeters are placed on the surface around the well and record the tilt before, during and after the frac. The change in tilt due to the frac is used to calculate the orientation. This far field measurement yields values for the azimuth and dip, as well as volume distribution of the frac fluid.

Downhole tilt mapping. An array of downhole tiltmeters are placed into a nearby offset well. The advantage of downhole tools is the ability to resolve more details than surface tiltmeters. Downhole tilt mapping is more sensitive to the frac dimensions (height, length and width), but is less sensitive to the frac orientation. Information from the two tiltmeter techniques can be used to arrive at a robust solution for frac geometry.

Vertical Seismic Profile (VSP). To fully utilize the resulting fracture geometry, the microseismic and tiltmeter analysis is integrated into both the geologic and geophysical models. Reservoir models derived from the geology data are defined in depth, while the geophysical models are originally defined in two-way seismic travel time, so to get a one-to-one comparison requires an accurate time/depth relationship. To obtain accurate time to depth models, velocity calibration is determined by vertical seismic profile (VSP) data. The time depth relationships are also critical for the placement of microseismic events into the 3D seismic volume, going from depth to time. Once the microseismic data is in the seismic volume, interpretation of the physical constraints to fracture growth, fracture geometry and orientation may be interpreted within a geophysical framework. Because VSP is recorded downhole, it has higher resolution than standard surface seismic data and it can be recorded in 3 dimensions away from the borehole.

3D VSP provides highly detailed images that better illustrate reservoir architecture including stratigraphic, structural, and faulting frameworks around the well bore. Integration of the results into a reservoir model helps to reveal influences of reservoir zones or faults on the direction and magnitude of the hydraulic fractures.

Data acquisition procedures for 3D VSP must be designed to provide detailed subsurface coverage while preserving the high-frequency information for data processing. The vertical axis of the resulting 3D VSP images is in 2-way reflection time with a time scale that is identical to 2-way times in surface seismic data. 3D VSP images are interpreted within the same interpretation framework as surface seismic data without adding the uncertainty of stretching from the depth domain to the time domain.

Summary. Combining 3D VSP recording with microseismic mapping using the specialized tool (designed to capture higher frequency microseismic events) improved the VSP resolution. It also streamlined operations and reduced costs. From the microseismic and tiltmeter data analysis, we have an improved understanding of induced fracture geometry. Mapping this data with the high-resolution 3D VSP images demonstrates the influence from geologic features such as faults, pre-existing fractures and specific geometry of reservoir lithologies on the frac geometries. The frac monitoring data gathered is also a direct measurement of the relative success of the frac design. Combining the geologic, geophysical and engineering disciplines achieved a more comprehensive overall solution of fracture geometry and well stimulation success.

References

Albright, J.N. and Pearson, C.F., Acoustic Emissions as a Tool for Hydraulic Fracture Location: Experience at the Fenton Hill Hot Dry Rock Site, SPEJ 22 (Aug. 1982), 523.

Warpinski, N.R., S.L. Wolhart; C.A. Wright, Analysis and Prediction of Microseismicity Induced by Hydraulic Fracturing; paper number SPE 71648, presented at 2001 SPE Annual Technical Conference and Exhibition, New Orleans, LA, 30 September- 3 October, 2001.

Acknowlegements

We thank EnCana Oil and Gas for permission to publish this work. We are indebted to EnCana’s Dean Dubois, Jeff Johnson, and Mark Turner for their support in implementing this project. We are grateful to Angus Duthie, Charlie Waltman, Sean Machovoe, John Alcott, Scott Malone and Trent Green of Pinnacle Technologies and Bill Schorger of Sterling Seismic Services for their contributions to this project. Thanks also to Jeff Meredith of Veritas DGC for his review of this abstract.

Figure 1. Example of microseismic event recorded on a 3 component geophone with hodogram analysis.