Multi-scaling of extreme values, intermittence detection using waves and evidences of determinative chaos in the wind dynamics in the Central Amazonian

Different characteristics from the high temporary variability of turbulence in High Resolution (60 Hz) series of wind and temperature are studied. Data was obtained in the Amazonian forest by project LBA (large-scale atmosphere-biosphere experiment in Amazonian) on March 1999. The scaling properties of the extreme events distribution are studied, and analysis of the properties of scale for extreme values is performed, for both, fixed and different time scales. The relations of scaling between the tails of the probability density function are calculated, and upon these relations, predictions are done for the functions of density for a particular scale. The wavelet analysis is used in order to detect intermittency in the energy dissipation rates. Such a phenomenon, associated sometimes with processes of momentum and heat transfer, is responsible for the insufficiency of Kolmogorov's theory in some scales. It is used the method of the false neighbors and the Shannon mutual information to reconstruct the attractor of the wind dynamics, moreover, estimation of the Lyapunov exponents is made to detect low dimensionality chaos in the Amazonian turbulence