DI51B-2632
Forward Modeling Method of Gravity and Magnetic Fields and Their Gradient Tensors Based on 3-D Delaunay Discretization in Cartesian and Spherical Coordinate Systems
Friday, 18 December 2015
Poster Hall (Moscone South)
Yi Zhang1, Chao Chen1, Jinsong Du1,2, Shida Sun1 and Qing Liang1, (1)China University of Geosciences, Hubei Subsurface Multi-Scale Imaging Key Laboratory, Institution of Geophysics and Geomatics, Wuhan, China, (2)Chinese Academy of Sciences, State Key Laboratory of Geodesy and Earth′s Dynamics, Institute of Geodesy and Geophysics, Wuhan, China
Abstract:
In the study of the inversion of gravity and magnetic data, the discretization of underground space is usually achieved by the use of structured grids. For instance, using the regular block as the module unit to divide model space in Cartesian coordinate system and the tesseroid in spherical coordinate system. Structured grids show clear spatial structures and mathematical properties. However, the block can only provide a rough approximation to the given terrain and using the tesseroid to approximate the terrain even seems impracticable. These shape determining errors cause the reduction of forward modeling precision. Moreover, the precision decreases again while using the tesseroid as no analytical algorithm has been acquired. On the other hand, since most terrain data has a limited resolution, unstructured grids, based on the polyhedron or tetrahedron, could fill the space completely, which allows us to reduce errors in shape determination to the minima. In addition, the analytical algorithms for polyhedron have been proposed. In our study, we use the tetrahedron as the module unit to divide the underground space. Moreover, based on the former researches, we supplement new analytical algorithms for tetrahedron to forward modeling gravity and magnetic fields and their gradient tensors in both Cartesian and spherical coordinate systems. The algorithm is testified by comparing the forward gravity and magnetic data of a block with the data obtained using the existed algorithms. The absolute difference between these two data is under 10e-9 mGal. Our approach is suitable for the inversion of gravity and magnetic data in both Cartesian and spherical coordinate systems.This study is supported by Natural Science Fund of Hubei Province (Grant No.: 2015CFB361) and International Cooperation Project in Science and Technology of China (Grant No.: 2010DFA24580).
