Using data assimilation techniques to validate fluid flow models

This project involves the use of statistical techniques such as data assimilation for validation of the construction of models of fluid flows. It is often not possible when constructing a model for a fluid flow to include all physical processes; this is particularly true for turbulent flow in geophysics, and in astrophysical and laboratory plasmas. Usually modelling these flows involves making assumptions. In this project we shall investigate how statistical techniques such as parameter estimation via data assimilation can be used to validate models of fluid flows (for example for magnetised and unmagnetised Taylor-Couette flows). Simple models may be constructed via truncations or symmetry arguments and we shall test how well these simple models perform against full solutions of the Navier-Stokes equations. Extension to flows in plasmas is also of interest.