The fact that examining Eulerian entities in unsteady velocity fields gives misleading information on Lagrangian coherence is now well-established. In this talk, I will review a range of techniques which have been proposed to extract coherent structures from given velocity data. These include the commonly used finite-time Lyapunov exponents, as well as methods such as curves/surfaces to which there is maximal attraction, transfer (Perron-Frobenius) operator methods for identifying sets which are coherent to transport, clustering methods which group similarly behaving particles, Lagrangian-averaged vorticity for identifying vortices in a frame-independent fashion, and sets which are most susceptible to random perturbations. Each seeks different characteristics, and thus the appropriate method for a given problem needs to be chosen carefully. These methods---not currently well-known in the porous media community---may offer new approaches for extracting coherence in porous flows.
|Acceptance of Terms and Conditions||Click here to agree|