“…We use a similar methodology to previous work (Beavan & Haines, ; Haines & Holt, ; Holt, Shen‐Tu, et al, , Holt, Chamot‐Rooke, et al, ; Holt & Shcherbenko, ), where the horizontal displacement gradient tensor field (i.e., strains and rotations) is modeled as a continuous field on the surface of a sphere. The forward model of the continuous field is parameterized using a continuous time‐varying 3‐D rotation vector function W ( r , t), where r is a directional cosine vector that passes through the Earth's center, t is time, and both W ( r , t) and r are perpendicular to the Earth's surface.…”