Potential benefit from the application of autoregressive spectral estimators in the analysis of homogeneous and isotropic turbulence

Spectral estimators other than the conventional periodogram technique exist. The autoregressive (AR) method is one of the alternative spectral estimation procedures currently available. The technique is particularly adapted to the detection of narrow-band components in the lower frequency domain for short data records. However, the technique may also be used in the analysis of more “standard” turbulent flow configurations where no specific wave modes are necessarily expected. An example is given here in the case of homogeneous and isotropic turbulence developing freely downstream of a heated grid. The results reported in this paper are in good agreement with earlier findings and would thus tend to confirm the idea that the technique might be used successfully over a wider range of signals in fluid dynamics. They provide us with the opportunity to highlight one particular additional benefit of the AR method with respect to the identification of spectral regions with constant slopes.