[세미나] Prof. Nelson L. Dias

The Hurst Phenomenon in Turbulence: an approach to calculate random errors of turbulence statistics

The Hurst phenomenon is a well-known feature of long-range persistence first observed in hydrological and geophysical time series by E. Hurst in the 1950’s. It has also been found in several cases in turbulence time series measured in wind tunnels and in nature. An investigation is made of the value of the Hurst exponent H in surface-layer atmospheric data, and its impact on the estimation of random errors. Usually, H > 1/2, implying the non-existence of the integral time scale. This has important practical consequences for the estimate of turbulence statistics. Few existing stochastic self-similar models are briefly explored, showing that spectra and autocorrelation functions are not well suited to detect Hurst’s phenomenon. Using data from two micrometeorological campaigns we found that both first- and second-order turbulence data display the Hurst phenomenon. Taking this into account produces larger relative errors than classical approaches based on the existence of the integral scale.