Research on Modeling Regional Geomagnetic Field Using Poisson Wavelets
Geophysical information systems (GIS) have long had a need to represent observation data on the sphere. In the recent years, the observation data of global geomagnetic field are classically represented using spherical harmonics. However, without the localization of space domain, this decomposition leads to numerical difficulties when computing high order model or regional model. A new representation approach which based on Poisson wavelets can overcome this drawback. We first analysis the causation of bringing these drawbacks, then introduce the basic theorem of Poisson wavelets on the sphere, and expatiate how to modeling the regional geomagnetic field using Poisson wavelets. In order to build appropriate frame, we use both regular distribution meshes and data-driven distribution points to select wavelets center positions, use generalized cross validation (GCV) to choose the scales of wavelets, and use least-squares(LS) method to obtain the model coefficients. Finally we build the Poisson wavelets model of the target regional geomagnetic field.
geophysical information systems spherical harmonics geomagnetic field model Poisson wavelets generalized cross validation least-squares method
Hui Zhang Lei Zhao Longwei Chen
College of Mechatronics Engineering and Automation, National University of Defense Technology, Changsha, China
国际会议
北京
英文
117-121
2010-11-27（万方平台首次上网日期，不代表论文的发表时间）