Quantifying organization of atmospheric turbulent eddy motion using nonlinear time series analysis

Using three methods from nonlinear dynamics, we contrast the level of organization in the vertical wind velocity (ω) time series collected in the atmospheric surface layer (ASL) and the canopy sublayer (CSL) for a wide range of atmospheric stability (ξ) conditions. The nonlinear methods applied include a modified Shannon entropy, wavelet thresholding, and mutual information content. Time series measurements collected over a pine forest, a hardwood forest, a grass-covered forest clearing, and a bare soil, desert surface were used for this purpose. The results from applying all three nonlinear time series measures suggest that ω in the CSL is more organized than that in the ASL, and that as the flows in both layers evolve from near-neutral to near-convective conditions, the level of organization increases. Furthermore, we found that the degree of organization in ω associated with changes in ξ is more significant than the transition from CSL to ASL.

Duke Scholars

Author Gabriel G. Katul Civil and Environmental Engineering