Data Assimilation of High‐Latitude Electric Fields: Extension of a Multi‐Resolution Gaussian Process Model (Lattice Kriging) to Vector Fields

Abstract We develop a new methodology for the multi‐resolution assimilation of electric fields by extending Gaussian process model (Lattice Kriging) used scalar field originally to vector field. This method takes background empirical as “a priori” knowledge and fuses real observations under framework. The comparison assimilated results two different models three resolutions suggests that (a) significantly reduces fitting errors compared with global spherical harmonic (SHF) because it uses range‐limited basis functions ideal local (b) resolution, determined number functions, is adjustable higher resolution leads smaller errors, indicating more structures in data are captured. also test sensitivity total amount input data: increases, deviate from become impacts can reach remote regions no available. better captures short‐period variations PFISR measurements than SHF maintains coherent pattern surrounding. Lattice Kriging examined via attributing into multiple levels (fine level located region observations). Such has smallest error shortest computation time, making regional high‐resolution modeling efficient. Our be modified achieve other unevenly distributed observations.