Cross-over error and reference station location for a marine magnetic survey

Abstract

A marine magnetic survey was conducted over the Dellwood Knolls area of the northeast Pacific in September 1977 using a proton precession magnetometer and Loran-C positioning. Geomagnetic field reference stations were located at Victoria and Port Hardy. In addition, a recently developed ocean bottom magnetometer was deployed in the survey area. The cross-over error, a measure of survey accuracy, and the factors contributing to it are discussed. In high magnetic gradient regions, the part of the error related to the uncertainty in ship position implies that the repeatability of the Loran-C navigation system is of the order of 30 metres. Part of the error is also due to time variations of the geomagnetic field. This latter effect can be significantly reduced by correcting to a well located reference station. Because of the areal distribution of the geomagnetic coast effect, a land station close to the survey area may not be as effective as a more distant site. Referencing to an ocean bottom magnetometer, at least to moderate levels of disturbance, is comparable to correcting to a well located land station. Using a combination of land and ocean bottom reference data produced a reduction in the cross-over error from 25 to 11 nT for this survey, effectively removing the geomagnetic variation term. This technique should prove even more useful for surveys conducted during periods of higher geomagnetic activity.