Through an interdisciplinary collaboration with the Department of Statistics at the University of California at Davis, we are developing a novel statistical inferential framework for multi-resolution modeling of random vector fields on the sphere. The use of multi-resolution needlet frames on the sphere for analysis of scalar random fields is extended to construct localized and parsimonious representation of vector fields that satisfy natural physical constraints such as being curl-free or divergence-free, thereby enabling a flexible approach to approximating physical processes.
- Funding sources: $50K (2018-2021) from NSF Statistics Program
- PI: Tomoko Matsuo (CU-Boulder)
- Collaborators: University of California at Davis, NOAA NCEI
- Societal relevance: An accurate modeling of the Earth鈥檚 magnetic field is crutial for the navigation, directional drilling, and exploration geophysics.
