FMS microresistivity images of borehole walls provide valuable information in exploring the structure and history of oceanic crust. However, the use of these images for structural or core orientation purposes relies on the accuracy of image orientation. In a two-step demonstration involving complementary data sets and a newly developed algorithm, we validated the use of the Schlumberger algorithm, routinely used during ODP cruises, for highly magnetized (hard rock) formations. The basic hypotheses of the two algorithms discussed in this paper are as follows: (1) the dynamics (rotation) of the tool string are intrinsically smooth (of low frequency/wavelength) and (2) the local field fluctuations associated with the oxide-rich layers (thin enough and separated by distance large enough to produce only high-frequency fluctuations in the magnetic records) locally add up a high-frequency component that can be removed by simple filtering operations. Space-scale analysis based on continuous wavelets demonstrated the low-frequency behavior of the tool string rotation in a subvertical borehole. However, at each site, the high-frequency nature of the local anomalous field still needs to be proven. Depending on the geometry (thickness and separating distance), as well as contrasts in magnetic properties (NRM and MS) of the penetrated formations, the local anomalous magnetic field may integrate additional intermediate and low-frequency components along with the referenced geomagnetic field (Parker and Daniell, 1979; Ponomarev and Nechoroshkov, 1983, 1984). An example of such configuration (low-frequency trend) may be visible in the lower portion (870–920 mbsf) of ODP Hole 1203A (Fig. F3D). A similar local low-frequency trend can be expected, for example, at basement contacts where a poorly and uniformly magnetized sedimentary formation is deposited over a thick and highly magnetized basement (Ponomarev and Nechoroshkov, 1983). Using the equations given by Pozzi et al. (1988), computation of the magnetic field in the borehole given the known remanent magnetization (or known induced magnetization) of the formation may help to quantify the effect of the magnetization on the determination of the tool orientation in terms of intensity and wavelength. For example, at Site 735, a magnetization of 1 A/m produces a field of 1,145 nT in the borehole (compared to the main field of 38,400 nT at the site). On the other hand, in addition to the high-frequency effects of oxide-rich layers, other sources such as pipes, casings, basement, and/or magnetic storms can perturb the geomagnetic field in a large spectrum of frequencies, and Bosum and Scott (1988) urge caution, as the proposed filtering techniques will fail.
At present, for ongoing ODP exploration, there are no technical development plans to replace the indirect orientation module (GPIT) of the FMS and UBI tool strings with direct-orientation gyroscopic systems. Hence, in light of both the actual ODP legs and in consideration of (1) the objectives of the future Integrated Ocean Drilling Program (IODP) expeditions, (2) the magnetic properties of penetrated formations, (3) available technical means, as well as (4) time constraints, the following methods are recommended to check tool orientation.
For accurate FMS-based structural investigations or core orientation within highly magnetized formations, a direct and independent log of the total magnetic declination is highly recommended. This log can only be obtained by running an oriented magnetometer associated with a gyroscopic system. Then, this direct record of declination can be compared with the magnetic parameters of the GPIT of the actual FMS-sonic or UBI tool strings. Provided that the magnetometers have similar sensitivity and are logged at similar speeds, one can back-orient the FMS-sonic or UBI tool string using oriented magnetometer logs and double-check the tool rotation computed by filtering techniques.
If, because of lack of time or availability, oriented magnetometers cannot be run, data processing/analysis is necessary. Depending on the frequency nature (high, intermediate, and/or low) of the magnetic anomalies and tool rotation, simple or more sophisticated (nonlinear/adaptative) filtering techniques and complementary analyses must be involved. An example of complementary analysis (quality check) where we take advantage of the use of GPIT logs in combination with different tools is described in Figure F8. In this example we compare the indirect declinations (
dec)—local properties of the formation—deduced from various tool strings with (1) lithological information as well as (2) oxide indicators such as NRM, MS, and/or resistivity logs. Correlation between these logs allows a raw estimation of tool string orientation quality, but in any case does not replace the direct record that oriented magnetometers can provide.
Development of nonlinear/adaptative techniques is in progress. The developed algorithm is based on the multiscale analysis of available magnetic logs. Multiscale analysis using the continuous wavelet transform helps to decipher the multiscale components of magnetic anomalies and tool rotation behavior. However, until we can complete this analysis with geological constraints (lithology), doubts on the estimation of the tool rotation may exist as a consequence of a poorly constrained inversion scheme capable of isolating the rotation component from raw magnetic logs. For now, this detailed study on natural cases encountered during ODP legs is a first step in providing guidelines and error limits for less documented past or future holes.
