Study On Magnetotelluric Constraint Inversion

The magnetotelluric sounding method(MT) is a kind of frequency-domain geophysical exploration methods of study of the structure of the earth using natural electromagnetic field. Compared with the artificial source electromagnetic exploration methods, magnetotelluric sounding method has unique advantages, such as construction is convenient, the great exploration depth, low resistivity body have higher resolution ability and so on. Currently magnetotelluric sounding method is mainly used in the exploration of deep mineral resources, tectonic research, earthquake prediction and so on. Magnetotelluric sounding inversion directly effects the application effect in various fields. Magnetotelluric inversion exists multiple solutions, coupled with electromagnetic rapid decay in the ground, which reduces lower resolution. Study of magnetotelluric constraint inversion can improve accuracy and reduce ambiguity more, which is very important for MT application.In this paper, we review the research situation of magnetotelluric constraint inversion, and then we deduce and summarize the basic theory of magnetotelluric sounding and constraint inversion. Aimed at the problem of difficultly selecting the correction of parameter’s modifier of magnetotelluric inversion model, we propose a constraint inversion method that the parameter’s modifier is updated based on constraints. By least-squares constraint inversion, we calculate and analysis and compare the inversion effect of different parameters inversion, different constraints and different constraint error bounds. The different calculation results of HKHKH type geoelectric section show the following conclusions. The inversion effect of constrained layer is improved. The improved range of constrained layer’s inversion effect and the constraint error bounds is the reverse relation. The more layers of accumulation inversion and the smaller of constraint error bounds that more obviously improve the inversion effect in overall. The average relative error of overall parameters and the constraint error bounds is the positive relation. The improved range of constrained layer’s inversion effect and the constraint error bounds is the reverse relation. The improved range of the next adjacent layer inversion accuracy and the constraint error bounds is the reverse relation. But the inversion effect of the layer that it is better during non-constrained inversion may be reduced. The larger average relative error during non-constrained inversion, accumulation constraint this layer, the inversion effect is more obviously improved in overall, and vice versa. In the situation of accumulation constraint inversion, the average relative error of overall parameters and the constraint error bounds is the reverse relation. The results show that magnetotelluric constraint inversion can improve overall inversion accuracy in some degree and is beneficial to reduce ambiguity more.