-
Y = nT-
, (2)
The FMS tool is a four-pad microelectrical resistivity device that allows detailed investigation of vertical and lateral variations of formation resistivity (Bourke et al., 1989; Goodall et al., 1998). Schlumberger introduced the FMS tool in 1986. A slim version of this tool was designed in collaboration with the ODP Borehole Research Group at Lamont-Doherty Earth Observatory and was first used during Leg 126 (Pezard et al., 1992). The four pads produce an image that covers ~25% of the 25-cm-diameter borehole wall. Proper pad contact with the formation is ensured only where the borehole diameter is <37 cm. Formation resistivity is measured in 64 buttons: two rows of eight on each pad. Electrical current flows from the buttons into the formation and returns to the tool body. Current intensity variations in each button are proportional to the formation resistivity close to the borehole wall. Resistivity measurements are recorded every 2.5 mm, and the vertical resolution of the tool is of the order of 2.5 cm, although beds thinner than 2.5 cm can be detected if the resistivity contrast is high. The individual resistivity values are binned into color or grayscale classes to produce an image of the borehole wall.
Formation resistivity depends primarily on the resistivity of the formation fluid, the connectivity of the pore space, and mineralogy. The high cation-exchange capacity of clay minerals generally leads to lower formation resistivity in clay-rich beds. In the largely unconsolidated sediments encountered at Woodlark Basin sites, the pixel tone of the images tends to be correlated with grain size, so that silt and sand beds and laminae show lighter tones and muddy sediments show darker tones. Where muds have a higher than average carbonate content, the resistivity increases and image tones are lighter (Célérier et al., this volume). The FMS images from Holes 1118A, 1109D, and 1115C are generally of fair to good quality with data from at least one to two of the four pads on the tool (sufficient for mapping of bed boundaries). Locally poor contact of some of the pads with the borehole wall was the result of borehole rugosity caused by factors such as poor consolidation and caving.
The FMS images from the Pliocene turbidite succession at Sites 1118, 1109, and 1115 were printed at a scale of ~1:5 and interpreted side by side with conventional well logs and the detailed bed-by-bed descriptions and core photographs acquired during ODP Leg 180 (Taylor, Huchon, Klaus, et al., 1999). Bed-by-bed sedimentary sections (Fig. F4) were produced directly from the FMS images by picking the bases and tops of beds in the images. Gradational bed tops were positioned at the upper limit of the apparent grading, based on grayscale gradients from light (more resistive sand and coarse silt) to dark (more conductive claystone). Sharp tops of nongraded beds were recognized in some cases. Sand- and silt-bed thicknesses were then measured and tabulated in a spreadsheet file for each hole.
Three models have been advocated for the distribution of bed thicknesses in turbidite successions: a lognormal model (Ricci Lucchi, 1969), an exponential model (Drummond and Wilkinson, 1996), and a power-law model (Hiscott et al., 1992, 1993; Malinverno, 1997). Following a logarithmic transformation of bed thickness, data that are lognormally distributed should plot as a normal distribution with a "bell"-shaped frequency curve. Lognormal bed-thickness distributions characterize many sedimentary successions (Hinnov and Goldhammer, 1991; Drummond and Wilkinson, 1996).
If the data instead fit an exponential distribution, then the number of thin beds is much greater than the number of thick beds, there is no modal bed thickness, and the distribution is consistent with the following relationship:
where
If a power-law distribution with a negative exponent characterizes a turbidite succession, then the number of thin beds is much greater than the number of thick beds. The power-law relationship is given by the following:
, (2)
where n and T are defined above and is a constant of order 1.0 given by the slope of a plot of Y vs. T in log-log space.
As a final step in the analysis of the Pliocene turbidite succession, the base of each turbidite bed at the three drill sites was assigned an absolute age using the age model of Takahashi et al. (this volume) and linear interpolation between picks on their accumulation-rate curves (Table T1). Then, bed number (starting with the deepest bed in each hole) was plotted against age to assess the frequency through time of turbidity currents reaching each of the three sites in the rift basin. The goal of this analysis was to discover (1) whether turbidity currents were focused into different parts of the basin at different times and (2) whether the number of sand and silt beds deposited per unit time was dramatically different from the basin axis to its northern margin.
