Turbulent Dissipation Rate Estimation Based on Spectrum Extrapolation

When estimating the turbulent dissipation rate from the energy spectrum, the high-frequency range predominantly governs the dissipation. However, experimentally acquired data are often contaminated by noise in the high-frequency region, making it difficult to directly trust the information in the dissipation range, which results in an overestimation of the dissipation rate. In this study, we develop a method that first identifies a reliable range within the measured spectrum, then reconstructs the unmeasured high-frequency portion using a theoretical spectral model. This reconstruction is based on the expected physical behavior in the dissipation range, where the spectrum decays due to viscous effects. The proposed approach yields consistent turbulent dissipation rate estimates for data contaminated by high-frequency noise. The validity of the proposed method is verified by reconstructing the dissipation spectrum using experimental data acquired in the Modane S1MA wind tunnel with a sufficiently resolved high-frequency time series extending to the dissipation range. The methodology is then applied to three-dimensional particle image velocimetry (3D-PIV) data contaminated by high-frequency noise.