DOWNHOLE LOGGING

Four logging runs were made in Hole 899B with only partial success. The total penetration in Hole 899B was 562.5 mbsf. Casing was set to 216 mbsf in an attempt to reduce problems with upper hole stability encountered at earlier sites during Leg 149. Problems with hole stability continued despite the setting of casing and became severe within acoustic basement (Unit IV). The wiper trip made in preparation for logging showed numerous problem zones within the hole from 372 mbsf to total depth. The scientists thus decided to log in two phases. First, the basement was to be logged while holding the end of the pipe at 394 mbsf, just below the sediment/basement contact. The pipe then would be raised inside the casing and the rest of the uncased hole would be logged. This strategy was only partially successful, because obstructions within the acoustic basement prevented the logging tools from descending below 455 mbsf. Three logging runs were made with the pipe held at the 394 mbsf position; these were the geophysical combination, the nuclear-sonic combination, and the geochemical combination tools (see below). The decreasing extent of open hole between runs led to a decision to move the end of the pipe to 193 mbsf (within the casing) and to attempt to log the rest of the hole, beginning with the Formation Microscanner (FMS) string. This tool string was not able to penetrate more than a few meters beyond the end of the casing. Further attempts to log Hole 899B then were abandoned. A summary of the logging tool strings used during Leg 149 and the basis of their measurements is given in the "Explanatory Notes" chapter (this volume). The following is a summary of the logging runs:

1. Run 1, geophysical combination; drill-pipe depth: 394 mbsf. Logged interval: 394-455 mbsf; speed: 600 ft/hr (190 m/hr) Tools: natural gamma-ray/shear sonic/resistivity/temperature.
2. Run 2, nuclear/sonic combination; drill-pipe depth: 394 mbsf. Logged interval: 394-445 mbsf; speed: 1200 ft/hr (380 m/hr). Tools: natural gamma-ray/lithodensity/standard sonic (SDT)/temperature.
3. Run 3, geochemical combination; drill-pipe depth: 394 mbsf. Logged interval: 394-437 mbsf; speed: 600 ft/hr (190 m/hr). Tools: natural gamma-ray/lithodensity/induced gamma-ray spectrometry/aluminum activation/temperature.
4. Run 4, FMS; drill pipe/casing depth: 193 mbsf. Logged interval: Aborted, except for temperature measurement. Tools: natural gamma-ray/FMS/temperature.

The wireline heave compensator was used during all logging runs; sea conditions were calm with minimal swell. Two passes of the first two strings were made to test repeatability of the measurements.

The geophysical combination was run first; the quality of both passes on this run generally were good. No cycle skipping was apparent from the dipole shear imager tool (DSI), and good shear-wave interval traveltimes were recorded from both the dipole and monopole transmitters (Fig. 58). Sonic waveforms through the logged section are presented in Figure 59. Resistivity from the induction phaser tool was noisy and correlation among shallow, medium, and deep resistivities over these intervals is poor. The shallow resistivity was repeatable between the two passes, but the medium and deep resistivities had poor repeatability. The temperature tool malfunctioned and did not capture data.

The nuclear-sonic combination was run second, although an additional 10 m of fill in the hole decreased the logged interval. Bulk density data are of good quality, as verified by laboratory measurements on discrete samples from cores (Fig. 60; see below under "Core-log Integration" section, this chapter). Sonic data from the SDT are of varying quality, with clear cycle-skipping from the long-spaced receivers compromising the quality of the real-time data. These data will require reprocessing to improve the quality. Good quality interval traveltimes (delta T), however, were recorded from the short-spaced receivers with no apparent cycle-skipping. Delta T measurements (µs/ft) from the sonic tools run in the first two strings are compared in Figure 61. Incomplete wall contact of the CNT neutron tool rendered the neutron porosity data useless. Caliper data indicate a reasonably uniform borehole diameter through the section (10.5 in.; 27 mm), except for an increase of 1.5 in. (39 mm) at 405 mbsf. The temperature tool acquired satisfactory data.

The initial run of the geochemical combination was aborted when the geochemical tool neutron source failed during testing at 160 mbrf; the string then was recovered, and the geochemical tool replaced. A failure of the neutron source in the second tool at total depth, however, precluded any data acquisition from the induced gamma-ray spectrometry tool (GST). Natural gamma-ray and aluminum measurements were obtained from a depth of 437 mbsf (representing a further 8 m of hole fill) to the base of the drill pipe. Only a single pass was made with this tool string. During this run, the temperature tool (the same unit as was used during the first run) again failed to acquire data.

Measurements by the temperature tool were the only data acquired in the fourth logging run (FMS), which penetrated only a few meters below the base of the drill pipe. These data are of good quality, but began at a shallower depth (195 mbsf) than the previous temperature run.

Natural gamma-ray activity is below the reliable levels of detection for the spectroscopy mode (K, U, Th) and the total (K + U + Th) data (a mean of 8-10 API units) must also be regarded with caution. For this reason total NGT data were not used for depth shifting. Instead, five strong peaks recorded by the sonic, resistivity, and density tools provided much more reliable reference points for depth shifting. In addition, all nuclear data were subjected to a linear five-point (0.5 m) moving-average filter.

Borehole temperatures measured by the Lament-Doherty temperature tool reflect the temperature of the seawater in the borehole, rather than the true formation temperature. The formation cools during drilling operations because of drill-fluid circulation and temperatures begin to rebound only after drilling has ceased. Data were obtained from the second and fourth runs, deployed 14 and 20 hr, respectively, after the wiper trip. The second run yielded a temperature of 13.8°C at a depth of 455 mbsf, while the fourth run measured 8.6°C at the deepest point, 193 mbsf. Measured bottom-water temperature was 2.4°C.

One characteristic log unit was determined from tool responses over the short interval logged. It is characterized by a general lack of change in log curves and correlates with the serpentinite breccia unit seen in the top of lithostratigraphic Unit IV. The main logs are summarized in Figure 58.

This unit is characterized by very low total natural gamma-ray values (of the order of 5-10 API units) and low aluminum content (1-1.5 wt%), which is indicative of an ultramafic lithology; it also exhibits a high resistivity from the short-focus log (100-120 m), a low porosity (0%-10%), and a compressional-wave interval transit time of 70 to 80 µs/ft (corresponding to velocities of 4.3-3.8 km/s). A slightly higher interval transit time than that of pure serpentine (about 53 µs/ft or 5.75 km/s velocity) can be ascribed to the brecciated nature of the "matrix." The DSI sonic tool employed in the first string indicated a good shear-wave response from the monopole source that correlates very well with that from the dipole source and is indicative of a coherent, unfractured ("fast") crystalline formation (approximately 125-135 µs/ft; shear-wave velocity of 2.4-2.3 km/s).

Within the logged section, a local variation toward the base is shown by five distinct peaks that can be correlated between short- focus resistivity, bulk density, and sonic traveltime. These probably correspond to individual blocks of serpentinite within the brecciated subunit that systematically cause an increase in resistivity (up to 1000 m), a decrease in traveltime (to 50 µs/ft, equivalent to 6.1 km/s), and a relative increase in bulk density of 0.25 g/cm³ (resulting in a decrease in porosity to 0%-1%). High-resolution (2 in.) processing of the density data indicates that the vertical thickness of these blocks is on the order of 20 to 40 cm. An indication that a relatively permeable zone exists between 413 and 417 mbsf is illustrated by DSI Stoneley waveform data (Fig. 59B) and a corresponding decrease in focus resistivity and shear-wave velocity (Fig. 58). The Stoneley waveform also indicates, very clearly, the borehole caliper deviation at 404.5 mbsf (Fig. 59B).

Bulk-densities derived from core measurements over the logged section correlate closely with the log-derived values (see Fig. 60). The core-derived bulk density (measured in discrete core samples in the laboratory) at 423.3 mbsf corresponds to one of the large serpentinite blocks (Interval 149-899B-21R-2, 78-81 cm) and probably correlates with one of the two distinct peaks (at 421 or 424 mbsf) in the log-derived bulk-density data. The difference between core and log measurements here is the result of a slight depth mismatch. In Figure 62, we compare laboratory-measured velocities with velocities derived from both sonic tools. In Figure 63, we compare laboratory-measured resistivity with log-derived resistivity (see the "Physical Properties" section, this chapter). In general, excellent agreement is seen between log-derived and discrete laboratory measurements in cores.