Two dimensional inversion of marine magnetotellurics data with cross-gradient constraints

Magnetotelluric (MT) method is one of the most famous methods in the investigation of subsurface structures. Because it is easy to implement, it has also become widely used in marine environment surveys. Two dimensional inversion of magnetotellurics data is ill-posed thus can be solved with regularization methods. In this study, an algorithm for two-dimensional magnetotelluric inversion has been developed.The data in MT sounding can be different for marine and land environment. Responses for both cases has been calculated and the reasons for the differences has been discussed. The topographic and coast effects on data has been identified.Conjugate gradient algorithm is used for the non-linear inverse problem. In algorithm, different stabilizers were compared due to their usefulness in 2D inversion of MT data.A joint inversion algorithm has been developed for the inversion of magnetotellurics data. Developed algorithm acts TE and TM mode data as collected from different methods and uses for joint inversion. The joint inversion procedure uses the rule of geometrical similarity between earth models obtained from both TE and TM mode inversion. In order to provide this similarity rule, cross-gradient constraint is used in joint inversion procedure.Developed algorithm has been tested on sythetic models and it is shown that algorithm is usefull for two-dimensional inversion of magnetotelluric data. Results also showed that, joint-inversion process with cross-gradient constraints can provide more similar earth models for TE and TM modes. The results from cross ? gradient inversion could provide usefull information when interpretating the subsurface.