Evaluation of large-scale turbulent/non-turbulent interface detection methods for wall-bounded flows

The experimental analysis of the spatial organization of turbulent/non-turbulent interfaces (TNTI) is an important task in many fields of fluid dynamics, especially to characterize mixing processes. Mixing processes are often associated with the macroscopic motion of coherent flow structures. A classical example for illustration is the turbulent mixing layer visualization by Brown and Roshko (J Fluid Mech 64:775–816, 1974). Today, a question of actual research is if coherent large-scale motions observed in turbulent boundary layer flows have an impact on the structure of the TNTI. As the length of these motions extends over many boundary layer thicknesses and their turbulent energy, and thus significance or impact, raises with Reynolds number, the TNTI detection technique must be accurate at large Reynolds numbers. Furthermore, the technique must be able to resolve the TNTI locally with microscopic spatial resolution and, at the same time, globally over a large macroscopic spatial domain. As the last two points require techniques with a large dynamic spatial range (ratio between largest and smallest scales that can be resolved), only tracer particle-based imaging techniques are suited, as the spatial resolution and field of view (FOV) can both be tuned by adjusting the magnification of the lens and the size and number of camera sensors. In this work, three suited techniques are compared to assess the sensitivity of the TNTI measurement of the method applied. The techniques considered are based on the turbulent kinetic energy, the homogeneity of the non-turbulent flow region, and the particle image density. The effect of bias errors on the TNTI measurement is particularly considered, but the implication of the results for the working range of the various techniques is also outlined. The analysis illustrates exemplary the sensitivity of the intermittency factor and the length of the TNTI with respect to the method applied.