Geophysical electromagnetic modeling and evaluation: A review

Electromagnetic forward modeling is the cornerstone of geophysical electromagnetic inversion. During the last 50 years, numerical simulation methods have been rapidly developed and widely used in geophysical area as the computational capacity continued to increase, such as from single-core to the most modern multi-core processing cards. This paper reviews the literature of electromagnetic fields simulation, particularly focusing on the forward modeling methods include finite difference method, finite element method, integral equation method, and several hybrid methods. We also discuss the possibility of deep learning methods for EM modeling. By sorting out the work done by the predecessors, this review briefly introduces the basic principles and traces back the development of these methods. We propose a Qualitative Evaluation Model named STAMP Model and some criterias of qualitative evaluation on these methods will be discussed in this model.