A Two-Tiered Method for Accurate and Efficient Prediction of CAT

The goal of the work is to develop a physics-based method for predicting clear air turbulence (CAT). Two levels of prediction will be employed: direct simulation of the intermediate scales of motion based on numerical integration of the Navier-Stokes equations (referred to by the generic term "large eddy simulation," LES), and Reynolds-averaged prediction based on statistical equations of turbulence. The two methods are complementary in terms of cost and accuracy; they are also synergistic, enabling more accurate and efficient prediction when used in combination, than either method when used alone. The work will demonstrate the use of a hierarchy of LES predictions to span a wide range of turbulence scales, enabling the prediction of turbulence intensities as a function of length scale, down to arbitrarily fine scales. The technical feasibility of the components of this approach will be demonstrated in Phase I. Phase II will integrate the methods and encapsulate them in software libraries for easy use in a variety of hosting software programs. Further research, development, and validation of the details which comprise the CAT prediction method will be performed. The software libraries will form the basis of the commercial product.