This study has shown that the same properties of the gravity gradient tensor are valid for the pseudogravity gradient tensor derived from magnetic field data, assuming that the magnetization direction is known. Eigenvectors of the pseudogravity gradient tensor are used to estimate depth to the center of mass of geologic bodies. The strike directions of 2D geological structures are estimated from the eigenvectors corresponding to the smallest eigenvalues. For a set of data points enclosed by a square window, a robust least-squares procedure is used to estimate the source point which has the smallest sum of squared distances to the lines passing through the measurement points and parallel to the eigenvectors corresponding to the maximum eigenvalues. The dimensionality of the pseudogravity field is defined from the dimensionality indicator I, derived from the tensor components. In the case of quasi-2D sources, a rectangular window is used in the robust least-squares procedure to reduce the uncertainty of estimations.
Based on synthetic data sets, the method was tested on synthetic models and found to be robust to random noise in magnetic field data. The application of the method was also tested on a pseudogravity gradient tensor derived from total magnetic field data over the Särna area in west-central Sweden. Combined with Euler deconvolution, the method provides useful complementary information for interpretation of aeromagnetic data.