DOWNHOLE MEASUREMENTS

Introduction

Downhole logs reveal the physical, chemical, and structural properties of formations penetrated during drilling. A variety of geophysical tools make rapid, closely spaced in situ measurements as a function of depth after the hole has been drilled. Logs can be used to interpret the stratigraphy, lithology, and mineralogy of the penetrated formation. Where core recovery is incomplete or disturbed, log data may be the only way to characterize the borehole section. Where core recovery is good, log and core data complement one another. Geophysical well logs can aid in characterizing lithologic sedimentary sequences and stratal stacking patterns when integrated with core and seismic reflection data.

Individual logging tools were joined together into tool strings (Fig. F12) so that several measurements could be made during each logging run (Table T13). The tool strings were lowered to the bottom of the borehole on a wireline cable, and data were logged as the tool string was pulled back up the hole. Repeat runs were made in some holes to improve coverage and confirm the accuracy of log data. Not all tool strings were run in each hole; refer to individual site chapters for details of logging strings deployed at each site. During logging runs a wireline heave compensator was employed to minimize the effect of the ship's heave, caused by sea swell, on the motion and position of the tool string in the borehole.

During Leg 183, we deployed three different logging strings (Fig. F12; Table T13):

(1) The triple combo (resistivity, density, and porosity) tool string consists of the dual laterolog (DLL), the high temperature lithodensity sonde (HLDS), and the accelerator porosity sonde (APS). The hostile environment natural gamma-ray sonde (HNGS) was included at the top, and the LDEO high resolution temperature/acceleration/pressure tool (TAP) was attached to the base of this tool string. Because of the low resistivities we encountered in the first hole logged (Site 1137), we substituted the dual induction tool (DITE) for the DLL at all subsequent sites. (2) The FMS-Sonic tool string consists of the Formation MicroScanner (FMS), the general purpose inclinometer tool (GPIT), and the dipole shear sonic imager (DSI). The natural gamma-ray tool (NGT) was included at the top of this tool string. The DSI was replaced by the long-spaced sonic tool (LSS) because of tool failure. (3) The well seismic tool (WST). Data from the NGT or HNGS placed at the top of all but the WST tool string provide a common basis for correlation of several logging runs and for depth shifting all logs.

We describe individual logging tools used during Leg 183, including their geological applications and the controls on data quality, below. The properties of the formation logged by each tool, the sample intervals, and the precision of the measurements (including the vertical resolution) are summarized in Table T13. Explanations of tool name acronyms, the acronyms by which the log data generated by the different tools, are referred, and their units of measurement are summarized in Table T14.

More detailed descriptions of individual logging tools and their geological applications can be found in Ellis (1987), Goldberg (1997), Lovell et al. (1998), Rider (1996), Schlumberger (1989, 1994, 1995), Serra (1984, 1986, 1989), and the LDEO-BRG Wireline Logging Services Guide (1994).

The HNGS and the NGT measure the natural gamma radiation from isotopes of potassium, thorium, and uranium in the rocks surrounding the tool. The NGT uses a sodium iodide scintillation detector to measure the natural gamma-ray emission and five-window spectroscopy to determine concentrations of radioactive K (in weight percent), Th (in parts per million), and U (in parts per million). The NGT and HNGS use a similar measurement principle. However, the HNGS uses two bismuth germanate scintillation detectors for gamma-ray detection with full spectral processing, significantly improving tool precision compared to the NGT. The spectral analysis filters out gamma-ray energies <500 keV, eliminating sensitivity to bentonite or KCl in the drilling mud, and improves measurement accuracy. The HNGS generates the same output as the NGT, and it estimates the average borehole potassium contribution to the total potassium signal. Shipboard corrections to the HNGS account for variability in borehole size and borehole potassium concentrations. The NGT and HNGS also measure the total gamma-ray signature (SGR, gAPI [American Petroleum Institute] units) and uranium-free or computed gamma ray (CGR, gAPI units).

The NGT response is influenced by borehole diameter and the weight and concentration of bentonite or KCl present in the drilling mud. KCl may be added to the drilling mud to prevent freshwater clays from swelling and forming obstructions. All of these effects are corrected during processing of NGT data at the LDEO Borehole Research Group (LDEO-BRG).

The HLDS consists of a radioactive cesium (¹³⁷Cs) gamma-ray source (662 keV) and far and near gamma-ray detectors mounted on a shielded skid, which is pressed against the borehole wall by a hydraulic eccentralizing arm. Gamma rays emitted by the source experience both Compton scattering and photoelectric absorption. Compton scattering involves the transfer of energy from gamma rays to electrons in the formation via elastic collision. The number of scattered gamma rays that reach the detectors is directly related to the number of electrons in the formation, which is in turn related to bulk density (RHOM). Porosity may also be derived from this bulk density if the grain- and pore-fluid densities are known.

The HLDS also measures the photoelectric effect factor (PEF) caused by the absorption of low-energy gamma rays. Photoelectric absorption occurs when gamma rays reach <150 keV after being repeatedly scattered by electrons in the formation. Because PEF depends on the atomic number of elements in formation, it is independent of porosity. Thus, PEF varies according to the chemical composition of the sediment. For example, the PEF of pure calcite = 5.08 barn/e^-; illite = 3.03 barn/e^-; quartz = 1.81 barn/e^-; and kaolinite = 1.49 barn/e^-. The PEF values can be used in combination with NGT curves to identify different types of clay minerals. Coupling between the tool and borehole wall is essential for good HLDS logs. Poor contact results in underestimation of density values. Both density correction and caliper measurement of the hole are used to check the contact quality.

The APS consists of a minitron neutron generator, which produces fast neutrons (14.4 MeV), and five neutron detectors (four epithermal and one thermal), positioned at different distances along the tool. The tool is pressed against the borehole wall by an eccentralizing bow spring. Emitted high-energy (fast) neutrons are slowed by collisions. The amount of energy lost per collision depends on the relative mass of the nucleus with which the neutron collides. Much energy is lost when the neutron strikes a nucleus of equal mass such as hydrogen, which is mainly present in the pore water. On degrading to thermal energies (0.025 eV), the neutrons are captured by the nuclei of Si, Cl, B, and other elements, resulting in a gamma-ray emission. The neutron detectors record both the numbers of neutrons arriving at various distances from the source and neutron arrival times, which are a proxy for formation porosity. However, hydrogen bound in minerals such as clays or in hydrocarbons also contributes to the measurement, so the raw porosity value is often an overestimate. The near/array limestone porosity corrected (APLC) log is usually displayed. The pulsing of the neutron source provides the measurement of the thermal neutron cross section () in capture units (cu). It is a useful indicator for the presence of elements of high-thermal neutron capture cross section such as B, Cl, and rare-earth elements.

The dual laterolog provides two resistivity measurements with different depths of investigation: deep (LLD) and shallow (LLS). In both devices, a current beam is forced horizontally into the formation by using focusing (also called bucking) currents. For deep measuring, both measure and focusing currents return to a remote electrode on the surface; thus, the depth of investigation is greatly improved, and the effect of borehole, and of adjacent formations, conductivity is reduced. In the shallow laterolog, the return electrodes that measure the bucking currents are located on the sonde, and, therefore, the current sheet retains focus over a shorter distance than the deep laterolog. The depth of investigation depends on the resistivity of the rock and on the resistivity contrast between the zone invaded by drilling fluid and the virgin (uninvaded) zone. Because of the inverse relationship between resistivity and porosity, the dual laterolog can be used to estimate the porosity of the rock from Archie's equation (Archie, 1942) if the sediments or rocks do not contain any clay. Archie's equation is expressed as FF = a · ø^-m, where ø is the porosity, a is a constant usually with set values of a = 1, and m is known as the cementation factor and it depends on the shape of the particles, which indicates the geometry of the pore channels. Fracture porosity can be estimated from the separation between the deep and shallow measurements based on the observation that the former is sensitive to the presence of horizontal conductive fractures only, while the latter responds to both horizontal and vertical conductive structures. Compared to resistivity induction logging the DLL produces reliable data in highly resistive formations (>1000 m).

The DITE-SFR provides three different measurements of electrical resistivities, each with a different depth of investigation in the formation. Two induction devices (deep and medium resistivity) transmit high-frequency alternating currents through transmitter coils, creating magnetic fields that induce secondary (Foucault) currents in the formation. These ground-loop currents produce new inductive signals, proportional to the conductivity of the formation, which are measured by the receiving coils. The measured conductivities are then converted to resistivity. A third device, a spherically focused resistivity instrument that gives higher vertical resolution, measures the current necessary to maintain a constant voltage drop across a fixed interval.

The TAP is a new "dual application" logging tool (i.e., it can operate as either a wireline tool or as a memory tool using the same sensors and data acquisition electronics depending on the purpose and required precision of logging data) (Table T15). During Leg 183, the LDEO-TAP was deployed as a memory tool in low-resolution mode; the data were downloaded via modem after the logging run was completed. The LDEO-TAP offers greater flexibility in logging operations, and it substantially improves the quality and resolution of data over an extended ambient temperature range (when used in the wireline mode) compared to its predecessor, the LDEO temperature logging tool.

The tool starts recording automatically after reaching a preset pressure (depth). Temperature, measured by high precision thermistors, and pressure are measured every second. Tool acceleration is recorded four times per second. Data, recorded as a function of time, are correlated to depth based on a synchronized time-wireline cable depth record and pressure recordings. Temperatures determined using the LDEO-TAP do not necessarily represent in situ formation temperatures because water circulation during drilling will have disturbed temperature conditions in the borehole. However, from the spatial temperature gradient it is possible to identify abrupt temperature changes that may represent localized fluid flows into the borehole, indicating fluid pathways and fracturing and/or breaks in the temperature gradient that may correspond to contrasts in permeability at lithologic boundaries.

The DSI employs a combination of monopole and dipole transducers to make accurate measurements of sonic wave propagation in a wide variety of lithologies (Schlumberger, 1995). In addition to a robust and high-quality determination of compressional wave velocity the DSI can determine shear-wave velocity (V_s) in most formations. The configuration of the DSI also allows recording of cross-line dipole waveforms. These modes can be used to estimate shear-wave splitting caused by preferred mineral and/or structural orientations in consolidated formations. A low-frequency source enables Stoneley waveforms to be acquired as well.

We deployed the DSI in two passes of the FMS-Sonic tool string combination, where the DSI replaced the more commonly used array sonic tool. The DSI was run in conventional P and S modes, along with the dipole recording modes.

The DSI tool consists of a transmitter sonde, a receiver sonde, and an acquisition cartridge (see Fig. F13). The transmitter sonde consists of a power amplifier and switching circuitry, which drive one dual-frequency (14 and 1 kHz) monopole transmitter and two pair of dipole (2.2 kHz) transmitters. Separate source functions with appropriate shape and frequency content are used for compressional, Stoneley, and dipole wave modes, respectively.

The receiver sonde houses eight receiver groups spaced every 15 cm, each consisting of four orthogonal elements aligned with the dipole transmitters. During acquisition, the output from these 32 individual elements are differenced or summed appropriately to produce in-line and cross-line dipole signals or monopole-equivalent (compressional and Stoneley) waveforms.

Preliminary processing of DSI data by the Schlumberger Multitask Acquisition and Imaging System (MAXIS 500) estimates monopole and dipole mode velocities using waveform correlation of the digital signals recorded at each receiver. In most instances, the shear-wave data should be reprocessed postcruise to correct for the effects of dispersion, which is caused by the variation of sound velocity with frequency. Processing techniques must be applied to account for a dispersive model without assumptions or to compute a bias correction to minimize any frequency effects on the velocity.

In addition, information such as mode amplitudes, shear-wave polarization, and Poisson's ratio can be extracted postcruise to provide information about lithology, porosity, and anisotropy. Amplitude processing and stacking of Stoneley-wave reflections may also be used to identify fractures, fracture permeability, and aperture in the vicinity of the borehole. The DSI tool is particularly important for determining shear-wave velocities for the upper parts of the basalt flow units. The V_p/V_s ratio in basalts is typically 1.8-2.0. Thus, the part of the lava flow with V_p of <3.0 km/s will have a V_s of <1.5 km/s, which cannot be determined without using the dipole source of the DSI tool.

The LSS measures the compressional wave velocity of the formation. The LSS is configured with two acoustic sources 61 cm apart and two receivers also spaced 61 cm apart. The spacing between the upper receiver pair and the transmitter pair is 2.44 m. The tool measures traveltime in microseconds over a certain distance in the formation. The configuration of the tool allows eight different traveltime measurements that compensate for irregular borehole walls. The velocity data together with the formation density can be used to generate a synthetic seismogram.

The FMS produces high-resolution images of borehole wall microresistivity that can be used for detailed sedimentologic or structural interpretation. This tool has four orthogonally oriented pads, each with 16 button electrodes that are pressed against the borehole walls. Good contact with the borehole wall is necessary for acquiring good quality data. Of a 25-cm borehole, ~30% is imaged during a single pass. Coverage may be increased by a second run. The vertical resolution of FMS images is ~5 mm, allowing features such as burrows, thin beds, fractures, veins, and vesicles to be imaged. The resistivity measurements are converted to color or grayscale images for display. A histogram of image data is used to subdivide the cumulative distribution in classes. During Leg 183, local contrasts in an image of all FMS figures in the site chapters were improved by applying dynamic normalization to the FMS data. A linear transform is applied to the input data to keep a constant mean and standard deviation within a sliding window (1 m was used). When dynamic normalization is used, the values of color indicate relative changes in resistivity. Furthermore, the hole diameter was reduced artificially from 25.1 to 15.2 cm to see more details in the pad tracks. This in effect reduced the blank space between the pad tracks and thereby enlarged the images produced by the FMS data proportionally.

FMS images are oriented to magnetic north using the GPIT (see below). This allows the dip and strike of geological features intersecting the hole to be measured from processed FMS images. FMS images can be used to visually compare logs with the core to ascertain the orientations of bedding, fracture patterns, sedimentary structures, and to identify stacking patterns. FMS images have proved particularly valuable in interpreting sedimentary structures, and they have been used to identify cyclical stacking patterns in carbonates (Eberli, Swart, Malone, et al., 1997), turbidite deposits (Lovell et al., 1998), cross-beds (Hiscott et al., 1992), facies changes (Serra, 1989), and volcanic sequences (Demant et al., 1995). Detailed interpretation of FMS images in combination with other log and core data in the sense of core-log integration will be carried out postcruise.

The GPIT is included in the FMS-Sonic tool string to calculate tool acceleration and orientation during logging. The GPIT contains a triple-axis accelerometer and a triple-axis magnetometer. The GPIT records the orientation of the FMS images, and it allows more precise determination of log depths than can be determined from cable length, which may experience stretching and/or be affected by ship heave.

The WST is used to produce a zero-offset vertical seismic profile and/or check shots in the borehole. The WST consists of a single geophone used to record the full waveform of acoustic waves generated by a seismic source positioned just below the sea surface. During Leg 183, a 300-in³ air gun positioned at a water depth of 3-7 m, depending on wave height, and offset from the borehole by 50 m on the port side of the JOIDES Resolution, was used as the seismic source. The WST was clamped against the borehole wall at 30- to 50-m intervals, and the air gun was typically fired between five and 15 times at each station. The recorded waveforms were stacked and a one-way traveltime was determined from the median of the first breaks for each station, thus providing check shots for calibration of the integrated transit time calculated from sonic logs. Check-shot calibration is required for the borehole to seismic tie because compressional wave velocity derived from the sonic log may differ significantly from seismic stacking velocities and velocities obtained via well-seismic surveys. This is caused by (1) differential frequency dispersion (the sonic tool operates at 10-20 kHz; seismic data is in the 50- to 100-Hz range), (2) difference in travel paths between well-seismic and surface-seismic surveys, and (3) borehole effects caused by formation alterations (Schlumberger, 1989). In addition, sonic logs cannot be measured through pipe, and the traveltime to the uppermost logging point has to be estimated by other means.

Log data quality is largely determined by the state of the borehole wall. It may be seriously degraded by rapid changes in the hole diameter and in sections where the borehole diameter greatly decreases or is washed out. Deep-investigation measurements, such as the resistivity and sonic compressional wave velocity, are least sensitive to borehole conditions. Nuclear measurements (density and neutron porosity) are more sensitive because of their shallower depth of investigation and the effect of increased drill-fluid volume on neutron and gamma-ray attenuation. Corrections can be applied to the original data to reduce these effects. We cannot, however, correct for very large washout. By use of the HNGS and the NGT on the strings, data can be depth correlated between logging runs. Logs from different tool strings may still, however, have depth mismatches caused by either cable stretch or ship heave during recording. Ship heave is minimized by a hydraulic wireline heave compensator designed to adjust for rig motion during logging operations.

Data for each logging run were recorded and stored digitally and monitored in real time using the Schlumberger MAXIS 500 system. After logging a hole, data were transferred to the shipboard Downhole Measurements Laboratory for preliminary processing and interpretation. FMS image data were interpreted using Schlumberger's Geoframe 3.1.4 software package. Well-seismic, sonic, and density data were interpreted using GeoQuest's IESX software package to establish the seismic-to-borehole tie. We plotted logs from the shipboard-processed data as depth-related curves, or images, representing the physical and chemical properties of the strata penetrated.

Log data were also transmitted to LDEO-BRG for processing using a FFASTEST satellite high-speed data link, soon after each hole was logged. Data processing at LDEO-BRG includes (1) depth-shifting all logs relative to a common datum (i.e., mbsf), (2) corrections specific to individual tools, and (3) quality control and rejection of unrealistic or spurious values. Once processed at LDEO-BRG, log data were transmitted back to the ship providing near real-time data processing. Log curves of LDEO-BRG processed data were then replotted on board for refining interpretations (see "Downhole Measurements" in each site chapter). Further postcruise processing of the log data from the FMS is performed at the Laboratoire de Mésures en Forage (LMF) in Aix-en-Provence, France. Mismatches of data sets can be related to ship heave, which caused irregular tool motion.

Postcruise-processed data (acoustic, caliper, density, gamma-ray, magnetic, neutron porosity, resistivity, and temperature) are available in ASCII format (see the "Related Leg Data" contents list). Access to the log data is free. A summary of "logging highlights" is also posted on the LDEO-BRG website at the end of each leg.

Downhole logging on board the JOIDES Resolution is provided by LDEO-BRG in conjunction with Leicester University Borehole Research, the LMF, University of Aachen, University of Tokyo, and Schlumberger Well Logging Services.