A Characteristic Mapping Method for Tracer Transport on the Sphere

Abstract

A semi-Lagrangian method for the solution of the transport equation on a sphere is presented. The method evolves the inverse flow-map using the Characteristic Mapping (CM) [1] and Gradient-Augmented Level Set (GALS) [2] frameworks. Transport of the advected quantity is then computed by composition with the numerically approximated inverse flow-map. This framework allows for the advection of complicated sets and multiple quantities with arbitrarily fine-features using a coarse computational grid. We discuss the CM method for linear transport on the sphere and its computational implementation. Standard test cases for solid body rotation, deformational and divergent flows, and numerical mixing are presented. The unique features of the method are demonstrated by the transport of a multi-scale function and a fractal set in a complex flow environment.