g).Overall, the LWD density (RHOB) log shows a good fit to the core bulk density (Fig. F21). The LWD density log generally underestimates the core values between 0 and 550 mbsf, and overpredicts the core values at 550-970 mbsf, by up to 0.1 g/cm3. LWD density (Fig. F21) mostly records the density of seawater between 0 and 156 mbsf where the log was taken inside the 20-in borehole casing. Below 156 mbsf the LWD density log shows a steady increase from ~1.7 to ~1.95 g/cm3 at 389 mbsf (log Unit 1 and Subunit 2a). At the frontal thrust zone (log Subunit 2b) density exhibits large variations associated with fluctuations in the differential caliper, corresponding to the thrust identified between 389 and 415 mbsf in RAB images at Hole 808I, which was also identified in Hole 808C from cores at 357-395 mbsf. Below the frontal thrust zone density decreases sharply, then remains fairly constant at ~1.85 g/cm3 to a depth of ~530 mbsf (log Subunit 2c), where density once more increases steadily to ~2.1 g/cm3 at 725 mbsf. Between 725 and 776 mbsf density drops sharply to ~1.75 g/cm3, corresponding to a period of borehole wiper trips; thus the data may be of questionable quality. Density increases more rapidly from ~1.95 g/cm3 at 776 mbsf to 2.25 g/cm3 at 930 mbsf. From 930 to 965 mbsf it decreases steadily to ~2.15 g/cm3, where it steps down to 1.4 g/cm3. This corresponds to the décollement zone, identified in cores at 945-964 mbsf. Below the décollement zone density increases steadily from ~1.7 g/cm3 at 975 mbsf to ~2.0 g/cm3 at 1034.79 mbsf.
Grain density data from
Site 808 core measurements (Fig. F22)
were used to calibrate a density to porosity transform for the LWD density logs
below 156 mbsf. Core grain density measurements indicate changes with depth
associated with lithologic variation. Although small-scale trends are observed,
the data follow two general trends below 156 mbsf, divided by an abrupt change
in grain and bulk density corresponding to the transition from upper to lower
Shikoku Basin facies at ~823.7 mbsf. Least-squares regression (Fig. F22),
after manual removal of low and high density spikes, was used to evaluate these
trends in grain density (g).
For 156 to 823.7 mbsf,
g
= 2.6353 + (1.2065 x
10-5 x
z),
and for 823.7 to 1223.79 mbsf,
g
= 2.5805 + (1.0 x
10-4 x
z),
where z is the
depth in meters. Porosity (
)
was calculated from the LWD density log values (b)
using
= (b
- g)/(w
- g),
assuming a water density (w)
of 1.035 g/cm3. The resulting porosity (Fig. F23)
shows an overall slightly steeper gradient with depth than that calculated from
cores. Initially, at shallow depths, core data underestimate LWD values, but
this trend is gradually reversed around 535 mbsf. Laboratory measurements tend
to overestimate porosity (Brown and Ransom, 1996) due to clay-bound water and
the rebound of samples from in situ temperature and pressure conditions
(Hamilton, 1971). Rebound effects increase with depth. Between 156 and 270 mbsf,
where porosity (calculated from LWD density data) shows large fluctuations from
40% to 90%, the LWD-calculated porosity shows mostly higher values than those
calculated from cores. Between 270 and 535 mbsf core and LWD porosity are in
close agreement at ~45%, except at the frontal thrust zone at ~390-415 mbsf,
where LWD porosity fluctuates from 30% to 80%. From 535 to 965 mbsf LWD porosity
is increasingly offset from core values, reaching a maximum misfit of ~7% at
~930 mbsf, except from 725 to 750 mbsf, where porosity is artificially high due
to the effects of the enlarged borehole caused by the wiper trip (see "Operations").
At 965 mbsf LWD-calculated porosity increases sharply and is higher than core
values by up to 40%, corresponding to the base of the décollement zone. From
970 to 1223.79 mbsf LWD porosity decreases more abruptly than in the cores. Most
of the increase in porosity and decrease in density in and below the
décollement zone is due to borehole effects (see "Quality
of LWD Logs").
Lithologic units sampled by LWD at Sites 1173 and 808 are of similar composition but at different stages of diagenesis and distances from the influence of tectonic deformation. Comparison of density and porosity results from these sites may provide information on the processes associated with subduction, as primary changes in physical properties are the result of the tectonic forces associated with subduction and formation of the accretionary prism at the Nankai Trough. Below the nearly 400 m of lower slope-apron and trench-wedge sediments at Site 808 (lithologic Units I and II), the trench-basin transition facies (Unit III), although having a greater thickness than its contemporary Subunit IB at Site 1173, exhibits very similar LWD density (Fig. F24) and porosity (Fig. F25) characters, although with a shift in absolute values. Density shows quite large fluctuations (0.4 g/cm3 at Site 808 and 0.2 g/cm3 at Site 1173) with fairly broad peaks and troughs but no significant average variation with depth (2.0 g/cm3 at Site 808 and 1.7 g/cm3 at Site 1173). The higher density at Site 808 is reflected in the lower porosity (37% at Site 808 and 60% at Site 1173) (Fig. F25).
The upper Shikoku Basin facies (lithologic Subunit IVA at Site 808 and Unit II at Site 1173) exhibits a similar trend at both sites, although the wiper trip at Site 808 has somewhat degraded the density data at around 750 mbsf (Fig. F24). At Site 1173 the density oscillates by ~0.1 g/cm3 but follows a near constant trend of ~1.66 g/cm3 to a depth of ~200 mbsf, where there is a gradual reduction in density, before returning to a near constant trend of ~1.64 g/cm3. Site 808 conversely has a greater oscillation of ~0.15 g/cm3 and shows a greater offset between the upper and lower near-constant trend, dropping from ~2.11 to 2.06 g/cm3.
The upper part of the lower Shikoku Basin facies (lithologic Subunit IVB at Site 808 and Unit III at Site 1173) shows a very similar density trend at Sites 1173 and 808. At Site 808 there is a gradual increase from ~2.17 to 2.26 g/cm3 just above the décollement zone (930-965 mbsf). Although no décollement zone is present at Site 1173, density increases from ~1.91 to ~2.0 g/cm3 in the same lithologic interval. Below the décollement zone at Site 808, LWD density recovers rapidly following reduction at the base of the accretionary wedge, with a trend much steeper than the contemporary sediments at Site 1173. However, the LWD results below the décollement zone at Site 808 are not consistent with core results (Fig. F24), which show a similar trend to LWD and core density at Site 1173.
Figure F26
shows a comparative plot of the five resistivity logs run at this site (see "Logging
while Drilling" in the "Explanatory
Notes" chapter for technical details). Identical downhole
trends in resistivity occur in all five resistivity logs and log unit boundaries
are clearly identified. As is the case for other parameters from the upper 156
m, logs from this interval did not give meaningful readings, as they were
acquired in the 20-in borehole casing. The bit resistivity signal is generally
smoother with a smaller degree of variation than other resistivity measurements,
but the resolution is lower. Following is a description of each log unit as
observed on the ring resistivity log. Log Unit 1 has an average resistivity of
~0.8-0.9 
m.
After a sharp decrease in resistivity from 1.3 to 0.5 m
between 156 and 168 mbsf, the signal shows an increasing trend downward to the
log Unit 1/2 boundary. Log Unit 2 is characterized by an overall decreasing
trend in resistivity from ~0.9 to ~0.6 m.
However, this trend is sharply offset at 389-415 mbsf (log Subunit 2b), where
the resistivity values are ~ 0.3 m
higher. This zone corresponds to the frontal thrust zone. Log Unit 3 is
characterized by a higher degree in variability of the resistivity signal. The
variation in the resistivity signal is lower again in log Unit 4. The average
value of resistivity in this interval is ~0.6 m.
At about the top of the décollement zone (~925 mbsf) in log Unit 4 the
resistivity trend changes from gradually increasing to gradually decreasing
(Fig. F26).
Figure F27
presents a superposition of the different resistivity measurements to emphasize
the differences between bit and ring resistivity on one hand and shallow-,
medium-, and deep-focused resistivity on the other hand. There is a noticeable
difference in the resistivity values obtained at the bit and at the ring, with
the latter being on average 0.05 m
higher. The correlation diagram (Fig. F28A)
emphasizes this observation. It also shows that the correlation between the two
measurement types stays very good throughout the whole section. A superposition
of the deep, medium, and shallow button resistivity measurements (Fig. F27)
shows good agreement between them. As was observed at Site 1173, shallow
resistivity values are systematically higher than both medium and deep
resistivity values. This feature is probably related to the measurement
conditions and applied corrections. The general shape of the shallow and deep
button resistivity correlation diagram (Fig. F28B)
confirms this trend.
As at Site 1173, the preliminary ISONIC P-wave velocity values from Hole 808I do not closely correspond to core measurements and previous wireline data (Fig. F29). The preliminary data show a strong tendency for a near-constant value of ~2.2 km/s from ~500 to 1000 mbsf, with excursions to lower values, raising suspicion about log quality. Depth trends also correlate poorly with the density log, except between ~700 and 800 mbsf. All of these observations suggest systematic error in the preliminary coherency analysis and traveltime picks from which the velocity was derived. Further postcruise waveform analysis will be required to derive a reliable velocity log from the ISONIC data.
