Dike swarms are igneous structures of continental expression accounting for major episodes of magmatism in igneous provinces, mantle plume heads, and continental breakup. In regional magnetic maps, dike swarms are recognized by high-amplitude lineaments indicative of lengthy and juxtaposed magnetized bodies. High-anomaly amplitudes from such tabular (2D) bodies tend to obscure lower amplitude contributions from localized 3D sources, representative of magmatic structures that once served as magma plumbing and storage. The recognition of such subtle signals with conventional filtering techniques is prevented due to spectral overlapping of individual contributions. We have developed a processing scheme to remove contributions from elongated, homogeneous sources to make clear contributions from 3D sources located below, in the middle of, or above a framework of elongated homogeneous sources. The canceling of 2D fields is accomplished by evaluating the horizontal component of the magnetic anomaly along the lineament strike, which for true elongated and homogeneous sources gives a null response. The gradient intensity of the transformed field is then evaluated to enhance residual fields over 3D sources. Lineaments thus removed identify tabular bodies with homogeneous magnetization, interpreted as being indicative of the uniform distribution (mineral type, concentration, and grain-size distribution) of magnetic carrier content in the rock. We evaluated our technique with synthetic data from multiple 2D-3D interfering sources and then applied it to interpret airborne data from the Ponta Grossa Dike Swarm of the Paraná-Etendeka Magmatic Province in Southeastern Brazil.

Many approaches to magnetic data inversion use assumptions that the source has homogeneous magnetization in terms of direction of intensity. Such a premise can not always be verified, but it allows the use of quantities that, for 2D sources, are independent of the magnetization slope (Amplitude of the Analytical Signal, the Intensity of the Anomalous Vector Field, among others) to recover the shape and intensity of the magnetization. This work presents a method to test the homogeneity of the distribution of magnetization to two-dimensional sources. This test can be done using a binary link when inverting magnitudes resulting from the ratio between field strengths, ICVA and IGA. The procedure uses the inversion in steps to facilitate the corvergence in the iterative process. The first stage recovers a magnetization distribution by reversing the ICVA and IGA with smooth bonds; the second uses this solution as input, also adjusting the ratio between fields with the compactness link, applying in this solution the uniformity bond that prescribes a binary distribution (zeros and ones). The identification of homogeneous sources is done when the binary link solution provides adjustment to the data at the same threshold as that observed in the previous step. Simulation tests and real data suggest that if there are two or more homogeneous domains, the ratio between the magnetization intensity between each domain (at least two) can be determined. The proposed procedure is tested with synthetic data and applied to reverse a terrestrial magnetometry profile on a dike of the Paraná State Etendeka Magmatic Province in the Paraná Basin, Brazil.

No abstract