Diagnosis of magnetic structures and intermittency in space-plasma turbulence using the technique of surrogate data

Intermittency is usually identified in turbulent flows as non-Gaussian tails of the probability density functions (PDFs) of the turbulent field derivatives. Here we investigate the role of phase coherence among the Fourier modes in creating intermittency in magnetized space plasmas using the technique of surrogate data. We apply the technique to two examples: (i) synthetic data and (ii) magnetic field fluctuations recorded in the terrestrial magnetosheath by the Cluster spacecraft. We use a set of four series of data, one observed and three surrogate, and their PDFs and moments (q < or = 4) as discriminating statistics. We show that the technique allows for detecting coherent structures and estimating their scales. We show furthermore that the phases, but not the amplitudes, create the non-Gaussian tails of the PDFs. We show also that the surrogate data used cannot account for asymmetries of the PDFs of the observed data. This enables us to confirm a scenario of turbulent cascade of mirror structures proposed in previous publications, by showing the existence of an approximately constant energy flux in the inertial range.