Previous   |    Next

doi:10.2204/iodp.proc.314315316.112.2009

Structural geology and geomechanics

Structural analysis was performed primarily on geoVISION resistivity images using GMI Imager (Geomechanics International Inc.), GeoLog/​Geomage (Paradigm Geotechnology B.V.), and GeoFrame (Schlumberger) software. These software packages present resistivity image data of the borehole wall as a planar “unwrapped” 360° image with depth. The software also allows visualization of the data in a three-dimensional (3-D) borehole view.

The geoVISION downhole tool provides resistivity images of shallow, medium, and deep depths of investigation, imaging the formation at ~34, ~43, and ~55 cm diameters of investigation, respectively. Resistivity image data were imported into the software packages and displayed as both statically and dynamically normalized images. Static normalization displays the image with a color range covering all resistivity values for the entire logged interval. Static normalization is preferred for comparing relative changes in resistivity throughout the borehole and is therefore ideal for correlating lithologic or facies changes and comparing the resistivity of particular fault zones. Dynamic normalization scales the color range for resistivity values over a specified interval. Dynamic normalization is commonly used for detailed identification of fractures and sedimentary structures and often allows more subtle changes in the image to be identified. Further filtering of the data was also possible with the software. Image data were compared with borehole caliper data to investigate the changing average diameter of the borehole.

Methods of interpreting structure and bedding differ considerably between cores, wireline logs, and LWD data sets. Horizontal and vertical resolution of resistivity images is considerably lower than comparable data from cores and wireline image logs (e.g., Fullbore Formation MicroImager [FMI]). Vertical resolution for LWD resistivity images is ~5–7.5 cm if ROP is maintained at ~20–30 m/h. Horizontal resolution across the image is a function of several factors. Some of these factors cannot be precisely constrained; therefore, this resolution has some uncertainty. These factors include the diameter of investigation (i.e., shallow, medium, or deep), which is also influenced by borehole elongation; the difference between formation resistivity and that of the borehole fluid; and the number of measurements made around the hole (56 for the geoVISION tool). The ratio between R_t (true resistivity of formation) and R_xo (resistivity of zone invaded by drilling fluid) also influences the diameter of investigation, but we assumed the ratio to be 1.0, owing to minimal invasion because resistivity was measured soon after drilling. For the geoVISION tool and the likely range of formation resistivities encountered during this expedition, the approximate horizontal resolutions range from ~2 to 3 cm. The ability to image a feature (feature detection) is also a function of the resistivity contrast and resolution. It may therefore be possible to resolve features smaller than the expected vertical or horizontal resolution if they contrast strongly with the background resistivity. The vertical resolution also controls whether the thickness of a layer can be determined. These resolutions should be compared with cores (millimeters) and FMI wireline resistivity images (~0.5 cm); therefore, smaller features are not resolvable within the LWD images. For example, individual microfaults (“small faults” <1 mm width) and shear bands (1–2 mm to 1 cm in width) identified in cores (e.g., Site 808) should not be resolvable in LWD resistivity image data. This should be considered when directly comparing reports from previous and future data sets.

In the unwrapped geoVISION resistivity images, sinusoidal lines are planar surfaces inclined to the borehole axis. Curved lines differing from sinusoids are nonplanar surfaces. To pick planar features (bedding planes, beds, fractures, faults, etc.), sinusoids were interactively fitted to determine dip and azimuth. Features were further classified according to type of fracture, width/​aperture, shape, and relative resistivity (conductive versus resistive). Further analysis (fracture frequency, azimuth distribution, etc.) was performed in GMI Imager, GeoLog/​Geomage, GeoFrame, and other software packages. We also compared resistivity images directly with other logging data for interpretation of bedding planes and for correlation of deformation style, lithology, and physical properties.

We identified fractures within geoVISION resistivity images by their contrasting resistivity or conductivity, from contrasting dip relative to surrounding bedding trends, or by truncation of other features. Resistivity of fractures is defined relative to the full range of resistivity values within the hole and is nonquantitative. If relative resistivity is unclear, the fracture resistivity is undefined. In some cases, conductive fractures may be easier to identify relative to the background resistivity, biasing the results slightly. Differentiation between fractures and bedding planes is complex and less accurate where bedding is inclined. Care was taken to avoid misinterpreting borehole artifacts as natural geological features. We compared initial structural interpretations from the logging data with seismic reflection data. Comparisons with cores will be possible following later expeditions. Our interpretations are based on the above criteria, but we acknowledge that some fractures and bedding planes may have been misinterpreted.

Borehole wall failure analysis

Breakouts and/or tensile fractures, two types of drilling-induced borehole wall failure, form when the state of local stress field at the borehole wall exceeds rock/​sediment strength. Breakouts form parallel to the minimum principal horizontal stress (S_hmin) and perpendicular to the maximum horizontal principal stress (S_Hmax), resulting in elongation of the borehole. Breakouts are recorded in resistivity images as two parallel conductive vertical features 180° apart. Drilling-induced tensile fractures may form in conjunction with breakouts or independently. The tensile fractures form perpendicular to S_hmin, 90° from the azimuth of the breakouts.

We recorded the orientation, downhole extent, and width of breakouts with the image analysis software considering all three borehole images (shallow, medium, and deep). Caliper data allow visualization of the changing average borehole diameter with depth, but azimuthal caliper data were not referenced to geographic coordinates and so cannot be used to assess borehole elongation. We compared breakout distribution and width with lithology (from image resistivity and defined lithologic units derived from all logging data) and drilling parameters. Further breakout analysis was conducted with GMI and other software.

Constraining stress from drilling-induced tensile fractures

Drilling-induced tensile fractures occur when the hoop stress at the borehole wall exceeds rock tensile strength. Where the tensile strength of sediments is negligible, the occurrence of drilling-induced tensile fractures is an indicator of tensile hoop stress at the borehole wall. We attempted to estimate in situ stress magnitudes by constraining possible stress ranges that allow failure in the borehole, specifically the formation of drilling-induced tensile fractures. Most of the following discussion is drawn from Zoback et al. (2003).

Because the stress in sediment is limited by the strength of frictional sliding on faults, it is possible to constrain the range of possible stress states at any depth and pore pressure. If the ratio between the two extreme effective principal stresses goes beyond certain values defined by the coefficient of friction, sliding occurs along critically oriented faults (Byerlee, 1978), which in turn releases the excess stresses. More explicitly, the limiting condition for failure can be expressed by Coulomb friction law:

where µ² is the coefficient of friction, P_p is pore pressure, and S₁ and S₃ are the maximum and minimum in situ principal stresses, respectively. This equation can be plotted as three lines (1, 2, and 3 in Fig. F10) representing the conditions of failure and defining the “stress polygon” in the S_Hmax versus S_hmin or horizontal stress domain. This stress polygon encompasses the possible states of stress at a given depth. Above and to the left of these three lines the sediment/​rock would be at failure in its natural state. The interiors of the stress polygons define allowable values for horizontal principal stresses for conditions favoring but not at the threshold of normal, strike-slip, and thrust faulting. The size of the stress polygon depends on the coefficient of friction, depth, and pore pressure.

IODP holes provide information on the vertical stress from the overburden, borehole pressure by direct downhole measurements with MWD tools, failure through the presence of breakouts and drilling-induced tensile fractures, and potentially, style of deformation from offset features. These observations are used with the theoretical framework above to provide estimates of the stresses in the borehole and whether the stress ratios favor normal, thrust, or strike-slip faulting.

Estimation of stress magnitude is possible when information on rock strength is available. Since no core was recovered from this LWD expedition, indirect estimations of rock strength parameters (unconfined compressive strength and internal friction coefficient) were made using a set of empirical equations that relate rock strength to other physical properties measured from geophysical logging (Chang et al., 2006). The two strength parameters are sufficient to construct any well-known rock strength criteria (Colmenares and Zoback, 2002). We used a strength criterion that is suitable for describing the general strength characteristics of the sedimentary rock. Ranges of possible in situ stress magnitudes were constrained by comparing rock strength with the state of the local stress field at the borehole wall where we observed borehole breakouts.

Top of page   |    Previous   |    Next