Joint Inversion Of Surface And Borehole Magnetic Data With Compact Constraints

Magnetic data collected on, or above, the earth’s surface can provide a good horizon-tal resolution. Inversion of the surface data can easily recover the horizontal position of ananomalous body. But it cannot resolve the depth distribution of the source because of thelow vertical resolution of the potential field exploration. Although the depth weightingfunction is used to counteract the depth decay of the magnetic kernel, it cannot reduce thenon-uniqueness and produce the depth information of causative body. The measurement inborehole always has high vertical resolution and accuracy, but it cannot be used to inversealone for whole area because the scope of its control is very limited. So the ground-bore-hole joint magnetic inversion which can combine high horizontal resolution of the grounddata and high vertical resolution of well measurement can produce a more three-dimen-sional resolution.The Preconditioned Conjugate Gradient Algorithm (PCG) is used to solve the ground-borehole joint magnetic inversion in this paper by introducing the distance weighting func-tion as a preconditioned matrix to reduce condition number of the magnetic kernel and re-duce iterations of inversion. For the problem of too slowly transition of inversion result tozero-solution space and too small compact distribution center of the properties, the com-pact constraints method proposed by this article is used to overcome them. Three2D and3D synthetic and field data sets are used to illustrate the new inversion algorithm that canget higher resolution which can present a good indication of deep overlapping units by theintroduction of borehole magnetic measurement. The example is also given to prove thatthe compact constrained inversion can help to get a more accurate result with clear boun-dary which is closer to synthetically model. At the end of the paper, the inversion of No.zero profile of Niuling iron mine is given to illustrate the effect of joint inversion of sur-face and borehole magnetic data with compact constraints in practical application.