Automatic inversion of magnetic anomalies of faults

• Published in: Computers & geosciences Vol. 27; no. 3; pp. 315 - 325
• Main Authors: Murthy, I.V.Radhakrishna, Swamy, K.V., Rao, S.Jagannadha
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.04.2001; Elsevier Science
• Subjects: Arbitrarily magnetized faults; Automatic inversion; Earth sciences; Earth, ocean, space; Exact sciences and technology; Geophysics: general, magnetic, electric and thermic methods and properties; Internal geophysics; Magnetic anomalies; Tectonics. Structural geology. Plate tectonics; Australia; Perth Basin; India; Automatic inversion; Arbitrarily magnetized faults; Magnetic anomalies; models; computer programs; aeromagnetic maps; Indian Ocean; magnetization; Australasia; Fortran 77; global; automated analysis; faults; depth; Asia; optimization; magnetic anomalies; direction
• ISSN: 0098-3004; 1873-7803
• DOI: 10.1016/S0098-3004(00)00105-9

An automatic inversion scheme and the related computer program in FORTRAN 77 to invert the magnetic anomalies of faults magnetized in an unknown direction are presented. The inversion scheme finds the initial values of the model parameters and improves them iteratively using Marquardt's optimization technique. The program is generalized in the sense that it can be used for any component of the magnetic anomaly, viz., total, horizontal and vertical. The positions of the dominant minimum and maximum anomaly values are used to estimate the initial values of the depths to top and bottom of the fault model, while the ratio of their anomaly values is used to initialize the value of the direction of magnetization. The position of the fault is located at the point where the anomaly equals to the sum of the maximum and minimum anomaly values. The input to the inversion program consists of the anomalies measured at given distances from an arbitrary reference point on the profile, besides the geological strike of the body and the direction of measurement. The output consists of observed and calculated anomalies, the model parameters, besides the location of the origin of the fault measured from the chosen reference point and the coefficients of linear regional anomaly that may be present in the data.