Study On Joint And Laterally Constrained Inversion Of MRS And TEM Data

The resistivity information of subsurface has an important impact on the results of MRS inversion. Among all the geophysical prospecting methods, the transient electromagnetic method can be jointed with MRS method to improve the accuracy of the inversion results because it has many advantages such as high resolution, high efficiency and high sensitivity to low-resistivity structure, and has a relatively accurate explanation of the resistivity distribution within several hundred meters underground. However, when simulating complicated structures of the subsurface by one-dimensional (single point) MRS-TEM joint inversion method, the poor-continuity detecting results of adjacent points, result from the inherent non-uniqueness of the inversion, may deviate the results from the actual geological structure. Based on the continuity of the electrical structure underground, this paper introduced laterally constrained inversion (LCI) and applied it to the joint inversion to solve the problem of lateral discontinuity in the inversion, therefore, improving the accuracy of joint inversion.Firstly, this paper described the two conventional forward algorithms of MRS and TEM, and made the appropriate improvements to the insufficiency of the existing algorithms such as low efficiency and accuracy of the results,therefore, laying a solid ground for joint inversion. In addition, this paper introduced the basic principle of joint and laterally constrained inversion, and then elaborated on the concrete process and the constrained conditions of this kind of inversion. This paper also adopted the traditional one-dimension method without LCI, moderate-constraint LCI and excessive-constraint LCI method to respectively simulate the three-layer and four-layer models designed by the author. After comparing the results of the three methods, the advantages of the LCI method and the significance of constraint to LCI were verified. Meanwhile, to prove the ability of the joint inversion method to resist the noise disturbance, this paper added white noise to the theoretical models, which were 5% and 10% respectively. The results showed that under the interference of the noise, the joint inversion still enjoyed high accuracy. Finally, through the joint constraint inversion of the data obtained from Anhui Huangshan landslide research, the practicality and accuracy of joint and laterally constrained inversion of MRS and TEM data were verified, after comparing the results with local geological data. The joint and laterally constrained inversion provides a new method for efficiently obtaining the hydrogeology distribution underground.