“…Analysis of non-stationary turbulent flow dynamics is characterised by highly complex features, exhibiting richer interactions between coherent structures than statistically stationary turbulent flows. Given the large amounts of data available from computational fluid dynamics (CFD) simulations and comprehensive experimental measurement techniques, such as time-resolved particle image velocimetry (TR-PIV), and four-dimensional particle tracking velocimetry (4D-PTV), the emergence of statistically non-stationary turbulence measurements can be expected to become ever more prevalent in the future [Rosi and Rival, 2018, Lacassagne et al, 2020, Steiros, 2022. Given that statistically non-stationarity turbulent flows are more likely to occur in nature and in industrial processes than stationary flows, the present work demonstrates a decomposition method for non-stationary turbulent/chaotic flows based on the proper orthogonal decomposition (POD).…”