This talk addresses the path following behaviors of marine vehicles under the guidance of spatially distributed modeled ocean flow, which is motivated by experimental work on navigating ocean gliders. Experimental data consistently suggests that the navigation error is caused by unknown ocean flow. The method of motion tomography (MT) is proposed to identify the unknown ocean flow from the navigation error along the vehicle trajectories. The method fuses the data collected by multiple marine vehicles along their paths to formulate an “inverse problem” that has been the core problem underlying medical CT imaging. By solving this inverse problem, a high-resolution spatial map of ocean flow in the volume traversed by the vehicles can be reconstructed. Motion tomography provides a “directly measured” spatial map of ocean flow, which can be leveraged by path following controllers on-board marine vehicles to reduce the navigation error. However, due to the limited spatial resolution of the flow map, the tracking performance of the marine vehicles are bounded by a smallest possible error.