Abstract: Forward numerical simulation is the basis of inversion imaging. In order to realize fine inversion imaging and quantitative interpretation of magnetic exploration, in this paper, the authors make a full use of the idea of decomposing a large problem into several small parts, and propose an efficient and high-precision numerical simulation method for two-dimensional magnetic anomalies in a space-wavenumber mixed domain based on Poisson equation. In this method, the partial differential equation is transformed into an ordinary differential equation with different wave numbers by one-dimensional Fourier transform, and then the ordinary differential equation is solved by the one-dimensional finite element method with quadratic interpolation, and finally, the magnetic anomaly value in spatial domain is obtained by inverse Fourier transform. In the model example, the two-dimensional models of constant susceptibility and variable susceptibility with rectangular cross sections are designed respectively, and the calculation accuracy and efficiency of the algorithm are verified. The modeling results of synthesis models show that the algorithm has high accuracy and efficiency, the absolute values of relative error are less than 1%, and the simulation time of 2 501×2 501 model nodes is 4.18 s. What’s more, it is suitable for any complex terrain model.

Key words: magnetic anomaly, space-wavenumber mixed domain, 2D model, forward modeling