We present numerical evidence that in strong Alfvenic turbulence, the critical balance principle---equality of the nonlinear decorrelation and linear propagation times---is scale invariant, in the sense that the probability distribution of the ratio
of these times is independent of scale. This result only holds if the local alignment of the Elsasser fields is taken into account in calculating the nonlinear time. At any given scale, the degree of alignment is found to increase with fluctuation amplitude, supporting the idea that the cause of alignment is mutual dynamical shearing of Elsasser fields. The scale-invariance of critical balance (while all other quantities of interest are strongly intermittent, i.e., have scale-dependent distributions) suggests that it is the most robust of the scaling principles used to describe Alfvenic turbulence. The quality afforded by situ fluctuation measurements in the solar wind allows for direct verification of this fundamental principle.
We investigate the anisotropy of Alfvenic turbulence in the inertial range of slow solar wind and in both driven and decaying reduced magnetohydrodynamic simulations. A direct comparison is made by measuring the anisotropic second-order structure fun
ctions in both data sets. In the solar wind, the perpendicular spectral index of the magnetic field is close to -5/3. In the forced simulation, it is close to -5/3 for the velocity and -3/2 for the magnetic field. In the decaying simulation, it is -5/3 for both fields. The spectral index becomes steeper at small angles to the local magnetic field direction in all cases. We also show that when using the global rather than local mean field, the anisotropic scaling of the simulations cannot always be properly measured.
