Statistical inverse analysis and stochastic modeling of transition

A computational method is introduced to infer statistical information on boundarylayer perturbations upstream of laminar-turbulent transition in a supersonic boundary layer. The method uses the intermittency function as a basis, which specifies the amount of time a flow is turbulent at a given streamwise location. The methods yields a joint probability density function for amplitude and frequency of boundary-layer perturbations upstream of transition. It relies on linear stability theory to link the probability density function with the intermittency. In order to infer parameters governing the function, we perform a statistical inverse analysis using the Markov chain Monte Carlo method. The approach is applied to a synthetic test case and to experimental data.