Quantifying Global and Random Uncertainties in High Resolution Global Geomagnetic Field Models Used for Directional Drilling

Abstract

Summary Total field strength, declination, and dip angle of the Earth's magnetic field, in conjunction with gravity, are used by magnetic-survey tools to determine a wellbore's location. Magnetic field values may be obtained from global models that, depending on the model, have a wide range of spatial resolution at the Earth's surface from large scale (3000 km) to small scale (28 km). The magnetic field varies continuously in both time and space, so no model can fully capture the complexity of all sources; hence, there are uncertainties associated with the values provided. The SPE Wellbore Positioning Technical Section/Industry Steering Committee on Wellbore Surveying Accuracy (ISCWSA) published their original measurement-while-drilling (MWD) error model in 2000. Such models and uncertainties define positional error ellipsoids along the wellbore, which assist the driller in achieving their geological target, in addition to aiding collision avoidance. With the recent update to Revision 5 of the ISCWSA error model, we have reassessed the uncertainties associated with our latest high-resolution global magnetic field model. We describe the derivation of location-specific global and random uncertainties for use with predicted geomagnetic values from high-resolution models within magnetic MWD survey-tool-error models. We propose a sophisticated approach to provide realistic values at different locations around the globe; for example, we determine separate errors for regions where the models have high spatial resolution from aeromagnetic data compared to regions where only satellite data are available. The combined uncertainties are freely available via a web service with which the user can also see how they vary with time. The use of the revised uncertainty values in the MWD-error model, in most cases, reduces the positional error ellipsoids and allows better use of the increased accuracy from recent improvements in geomagnetic modeling. This is demonstrated using the new uncertainty values in the MWD-error model for three standard ISCWSA well profiles. A fourth theoretical well offshore Brazil where the vertical magnetic field is weak shows that with drillstring interference correction relying on the more uncertain magnetic dip, the positional error ellipsoids can increase. This is clearly of concern for attaining geological targets and collision avoidance.