Interpretation of oceanic magnetic anomalies using a linear inverse technique

Abstract

A direct magnetic interpretational technique has been developed and applied to oceanic magnetic anomalies. The method of interpretation computes a distribution of magnetization, within a specified two-dimensional model - given the direction of magnetization, from the observed magnetic anomalies. The technique is based on the numerical solution of a linear integral equation which is approximated by a finite set of linear algebraic equations. These equations relate (n) observed magnetic anomaly field points to (m) unknown magnetization values. Solution of this system of equations is carried out by computer, using matrix operations. The programming procedure allows model elements of irregular cross-section to be incorporated within the magnetic layer and provides a solution to both the completely determined and overdetermined problem (i.e. n≥m). Details of this procedure are presented together with an evaluation of methods of application. Interpretations of magnetic profiles in the North Atlantic Ocean, the Gulf of Aden and the Pacific Ocean are presented in terms of computed distributions of magnetization confined to Layer 2. Results are discussed in terms of the Vine-Matthews hypothesis of sea-floor spreading and certain apparent differences in the bulk magnetization of the oceanic crust. Model studies confirm the feasibility of a thin magnetic layer (0.5 km), situated just below the sea-floor. The approximate shape of this magnetic layer is deduced from known magnetization values obtained from dredged rock samples. Interpretation of magnetic data from the Pacific Ocean indicates that both vertical and inclined source bodies, within Layer 2, represent plausible models, although extensive subhorizontal bodies (dipping at 10 and less) are unlikely.