IN-SITU STRESSES AND NATURAL FRACTURES OF THE MALLIK GAS HYDRATE RESERVOIR, MACKENZIE DELTA, N.W.T., CANADA

Pat McLellan¹, Kevin Gillen¹, Chris Podetz¹, Scott Dallimore²
¹ Advanced Geotechnology Inc., 1540, 521 – 3rd Ave SW, Calgary, Alberta T2P 3T3, Canada
² Geological Survey of Canada, P.O. Box 6000, Sidney, B.C., V8L 4B2, Canada

The magnitudes and orientations of the three principal in-situ stresses are used for several aspects of oil and gas well design and production optimization, such as: the selection of casing setting depths, designing fracture stimulation treatments, reducing sand production, understanding and controlling casing deformations, and the subsurface disposal of drill cuttings and wastes. In-situ stress data are also used for assessing caprock seal capacity, understanding permeability anisotropy in fractured reservoirs, and determining stress-dependent reservoir properties such as permeability, porosity and compressibility.

Vertical stress (_v) magnitudes in the Mallik area were investigated by compiling and analyzing historical and recently acquired bulk density log and core data. Corrections were applied to these data to account for unlogged portions over the upper permafrost interval and the effects of acute hole enlargement. Minimum horizontal in-situ stress (_Hmin) magnitudes were estimated from historical leak-off test data in the general Mallik area and a profile of micro-fracture stress tests conducted in Mallik 5L-38 with Schlumberger’s Modular Dynamics Tester (MDT).

Data quality problems with leak-off tests and peculiar pressure behaviour in the MDT micro-fracture tests does not permit a straight forward determination of _Hmin in this setting. In general the micro-fracture tests do not have a classic hydraulic fracture breakdown pressure, they have generally flat to slightly inclined propagation pressures on a Nolte plot and there is not a clear fracture closure pressure. Long linear flow periods after shut-in are typical and appear to be related to the presence of natural fractures or previously induced fractures in the same test intervals.

The orientations of the principal horizontal in-situ stresses (_Hmin and _Hmax) have been determined from an analysis of the local structural geology, borehole breakouts in the region, deformed hole ellipticity and shear velocity anisotropy. Evidence for a nearly isotropic in-situ stress regime in the gas hydrate intervals in the Mallik 5L-38 well are reviewed along with its implications for well design and future gas hydrate production.

FMI borehole image log data from the Mallik 5L-38 research well were analyzed for natural and drilling-induced fractures. A total of 47 fractures were identified over the openhole logged interval between 679 mKB and 1169 mKB, of which 34 were interpreted to be natural. The natural fracture development varies with depth in this borehole and is influenced by lithology (Figure 1). Approximately half of the observed natural fractures occur within or proximal to gas hydrate zones. The observed fracture frequency at the borehole is less than the true fracture frequency in-situ due to sampling bias. This effect results in fractures that are sub-parallel to the borehole generally being under-represented in the image log. Observed natural fracture frequencies range from 0 to 3 fractures per metre, with an average of 0.07. Terzaghi-corrected natural fracture frequencies range from 0 to 12.8 fractures per metre, with an average 0.23.

Natural fracture orientation is not random, and appears to be organized into distinct populations. Two conjugate sets are developed in the massive and cross-bedded sands and sandy-silts. The first set has a N-S strike, with both members of the set dipping in opposite directions at approximately 57°. The second set strikes NE-SW, with the two members oppositely directed at dip values close to 65°. The conjugate nature and dip magnitude values of these sets are suggestive of a shear failure origin. A subordinate ESW-WNW trend also appears to be weakly developed. In contrast, natural fractures observed in coals are near-vertical and are more suggestive of cleats. Their strike values are well clustered and average N16°E.

In several instances selected natural fractures observed in the borehole image log can be seen in core from the same interval, although not all fractures observed in the log are present in the core. Peculiar heat distribution patterns observed during the thermal stimulation test in the well may be partially explained by the contribution of the natural fractures observed in the borehole image logs between 916 and 920 mKB (Figure 2). The presence of natural fractures also offers an explanation for long linear flow periods and near-wellbore permeable and impermeable boundaries as observed in the MDT micro-frac stress tests and flow tests.

The presence of what appear to be partially open, permeable, sub-vertical natural fractures in some of the gas hydrate-rich intervals, coupled with relatively low horizontal in-situ stresses in the Mallik setting could have profound implications for gas hydrate production from horizontal wells. For instance, significant permeability anisotropy due to natural fractures will strongly influence the distribution of heat and hence gas production if such wells are thermally stimulated.

Further work is required to characterize the frequency, spacing, connectivity, filling, stress-dependent permeability, and stability of the natural fractures in and adjacent to the gas hydrate zones.

Figure 1. Comparison of natural fracture, core, and selected log data analyzed over the FMI log interval in Mallik 5L-38, from 679 to 1164 m. The stratigraphy of the cored interval is generalized.