Schlumberger introduced the FMS in 1986 to measure the electrical resistivity of boreholes with an array of sensors sufficiently dense to produce high-resolution resistivity images of the borehole wall (Ekstrom et al., 1987). A slim version of this tool was designed to pass through the drill pipe and was introduced during Leg 126 (Shipboard Scientific Party, 1990; Pezard et al., 1990). The image resolution is on the order of a few millimeters. However, because of the slim design, the four pads produce an image that covers only about 25% of the usual 25-cm-diameter ODP boreholes and good pad contact with the formation is ensured only in holes with a diameter <37 cm.
The FMS data are processed separately from the conventional logs. In particular, the depth computation includes a sticking detection procedure that uses the tool accelerometer data to correct the winch cable depth data when the tool gets stuck and the two measurements conflict. This correction is applied from the bottom of the hole, which is set to be consistent with that of the other logs, and can accumulate significant depth differences upward in some instances. It is therefore necessary to correlate FMS and classical logs. This is possible because the FMS tool string records natural gamma ray and caliper data that are processed in two different manners: a first processing is identical to that of classical logs and produces ASCII data given in the Initial Reports volume; a second processing that is made together with the electrical images, and therefore includes the same depth correction as these images, is included in the input DLIS file used in the Schlumberger Geoframe interpretation software. Correlation can then be made between these two different outputs for the same measurements. For Leg 180, caliper data provide a good reference for correlation and are therefore systematically displayed alongside the FMS images in this paper. The "Appendix" lists a few offsets at selected tie points for each hole. In the case of Holes 1118A and 1115C, this offset from the conventional logs can reach a few meters.
Two types of images are constructed: a dynamic image, where the color equalization is done in a 2-m sliding window, and a static image, where the color equalization is done for the whole borehole. The static FMS image is a qualitative measure of resistivity, whereas the dynamic image, by enhancing local contrast, reveals detailed sedimentary or tectonic features.
These FMS electrical images are then analyzed with the Schlumberger Geoframe software. A key feature of this software lies in its ability to interactively renormalize the color scale within any depth interval to help identify features of various sizes. The image variations were noted to build a catalog of typical facies that are naturally related to the core-defined lithologic units and to the logging units defined by the Shipboard Scientific Party (1999c, 1999d, 1999e).
The sedimentary fill of the northern margin is mainly composed of alternating clay, silt, carbonates, and sands mixed in various proportions, as summarized in the lithologic units (numbered I-XII; Table T2). Holes 1118A and 1109D bottom in a prerift dolerite overlain by a conglomerate, whereas Hole 1115C bottoms in prerift forearc sediments.
The logging units (numbered L1-L12) were defined after a quick-look analysis of mainly neutron porosity (APLC), lithodensity (RHOM), photoelectric effect (PEFL), and gamma ray (NGR) (Shipboard Scientific Party, 1999a). The photoelectric effect is a direct indication of carbonate content. Clay content is usually estimated from gamma ray, but the presence of radioactive sands made this impractical in the Woodlark Basin. The difference between neutron and lithodensity porosity was therefore used instead as an indicator of clay content. The density porosity (DPHI) is computed from the density measurement (RHOM) by
where,
The porosity difference (DPORO) is then computed by
DPORO ranges from -1.8 in pure sand to 0 in pure limestone to 0.4 in clay or dolomite. The analysis uses the separation of the APLC and DPHI curves, which represents DPORO. Further details can be found in Shipboard Scientific Party (1999a). It is useful to remember that the terminology clay, silt, and sand used in core descriptions is based on grain size, whereas that used in log interpretation, clay, carbonate, and sand, is related to mineralogy and porosity. As a consequence, an interval with low clay and carbonate content may be described as sandy from logs while being described as silty from core.
The structural analysis determines planar structure orientations by fitting sinusoids of observed features on 1/10 to 1/20 vertical scale dynamic images. Most of these structures are either sedimentary bedding or fractures and were labeled as bedding 1 and 2 or fracture 1 and 2 according to a decreasing degree of confidence. Bedding confidence is limited by the lack of resistivity contrast within some formation and/or by sediment reworking (synsedimentary deformation or bioturbation), which attenuates that contrast. Fractures are identified indirectly by bed truncation or abnormal contacts. These abnormal contacts have to be traced from pad to pad, which is made difficult by the low coverage of slim FMS data. Ambiguous planar features that could not be properly identified were also recorded and labeled as unknown. Erosional surfaces or unconformities were also attributed to this category to separate sedimentary structures such as channels from deformational structures. These measurements are discussed and related to the core-derived structural domains (numbered I-III) below.
The intervals where bedding orientation varies are then isolated to determine whether these orientations can be accounted for by simple folds, and, in the cases where they can, whether folding is related to nearby faulting. This is done by using the stereographic projection where the poles of cylindrical or conical structures fall within a great or small circle, respectively.
FMS images and structural interpretation are presented at three different scales: