Technologies to compute and measure fluid mechanics have grown tremendously in recent years, enabling data of unparalleled accuracy and resolution to be obtained from increasingly complex and important problems. However, methods used to interpret the resulting data have remained relatively unchanged. Fluid mechanics is most commonly visualized and understood by quantities based on instantaneous rate of change information. These quantities become less directly useful, or even meaningful, as increasingly complex problems are analyzed. There is a need to fundamentally change how fluid is visualized and understood. A Lagrangian-based approach that tracks the fluid motion is essential. To date, such methods have been ad hoc and imprecise. The computation of so-called Lagrangian coherent structures (LCS) enables a systematic approach to precisely capture fundamental boundaries, whose coordinated dynamics provide the template for complex fluid motion. This framework has gained considerable popularity in the last few years. A growing body of literature is demonstrating the wide-ranging success of this approach, catalyzing new thinking and paradigms for the study of fluid mechanics. Since there are few publicly-available packages for running LCS computations, the development of robust, usable and sustainable software elements encapsulating this innovative approach are needed.
The aim of this project is to provide software elements that, for users encompass an easy to use toolkit to compute and visualize LCS in a variety of applications, and for developers provide a modular infrastructure amenable to further customization and technological advances. Shawn Shadden (IIT) will lead the detailed design of algorithms and their accurate numerical implementation. John Hart (UIUC) will lead the architectural design of the package and the efficient parallel implementation of its elements. The first year will be dedicated to a complete design of object specifications and communication interfaces. The second year will focus on the detailed implementation of the package and the release of an initial beta distribution. In the third year, this package will be tested and tuned for performance, scalability, precision and robustness.