Study On Joint Inversion Method Of Gravity And Magnetic Data Based On Model Parameter And Model Gradient Gramian Co-constraints

The joint inversion of gravity and magnetic data utilizes different observation data to invert for obtaining a unified geophysical model,and the obtained three-dimensional(3-D)structures of underground density and magnetism are important bases for the evaluation of mineral resources and the division of metallogenic prospective areas.In order to finely characterize the non-uniform variation of subsurface field sources,I focus on the constraint mode of joint inversion,complex physical correlations and the existence of remanence in the subsurface to establish a high-resolution joint inversion method of gravity and magnetic data.Commonly,different types of geophysical data are linked based on the correlation of physical properties or the structural coupling relationship of underground geological bodies.The Gramian constraint of the model parameters constructed based on the correlation of physical properties and the Gramian constraint of the gradient of the model parameters constructed based on the structural coupling relationship are widely used in joint inversions because the Gramian constraint can be applied to general cases without prior information.In this thesis,I discussed and compared the mechanisms of model parameter Gramian constraint and gradient Gramian constraint in the joint inversion of gravity and magnetic data through model tests,and proposed to combine the two constraints as a co-constraint into the inversion objective function to realize the joint inversion of gravity and magnetic data with co-constraint.The model test results shown that for the combined model with a single physical relationship,the method can take into account the global physical trend of the geological body and the local boundary change characteristics,and effectively improve the accuracy of the joint inversion results.In practice,the correlation between various physical properties is usually difficult to be expressed by a single linear relationship,and physical properties of gravity and magnetic data often have different correlations in different regions.Therefore,in this thesis,I carried out adaptive partitioning based on the correlation of the gravity and magnetic data to achieve zonal joint inversions with co-constraints to further realize the fine inversion.Based on the test results of the combined model with multiple physical property relationships,this method can effectively recover the difference of the physical property correlations,and obtain the inversion results that are more consistent with the real underground physical property distribution.Due to various geological activities,remanence generally exists in the measured data of various regions,resulting in the difference between the magnetization direction of geological bodies and the geomagnetic field direction,which seriously affects the accuracy of the inversion results.Therefore,I considered directly conducting magnetization vector inversion instead of scalar inversion to reduce the influence of remanence.Furthermore,I applied the joint inversion method with Gramian co-constraints to the magnetization vectors,and considered the linear relationship between the three components of the magnetization vectors and the magnetization in the inversion process,and added the self-constrained inversion of magnetic data for the magnetization vectors in the joint inversion process.Model tests showed that this method can effectively restore the magnetization vectors while recovering the correlation between gravity and magnetic data,and the underground magnetization vector field could be obtained directly.Finally,the proposed method was applied to Lu-zong ore concentration area in the middle and lower reaches of the Yangtze River region to obtain a 3-D distribution model of magnetite within 5 km underground.The results showed that the magnetite locations corresponded well with the shallow minerals being mined,and were verified with drilling data to prove the existence of usable deep resources in the ore concentration area,and the extent of deep minerals was circled to delineate favorable mineralization areas.