A novel multi-spacecraft technique, applied to measurements made by the four Cluster
spacecraft in the solar wind, is used to determine the field-aligned anisotropy of magnetohydrodynamic
inertial range turbulence. Time-lagged two-point correlation functions
are used to construct spatial auto-correlation functions. These are used in conjunction
with a phenomenological elliptical scaling model to estimate the ratio of field-parallel to
total correlation lengths, obtaining a mean value of 0.61 ± 0.02. A simulated turbulent
field model is also fitted to the spatial auto-correlation functions in order to estimate
the power in the 2D fluctuations, obtaining a mean value of (79 ± 3)%. The results obtained
from both models, assuming a slab-2D paradigm, are consistent with solar wind
fluctuations being anisotropic with energy mainly in wave vectors perpendicular to the
mean magnetic field.
Eight intervals of multi-point magnetic field data are analysed, and the degree of
variation in the estimates of anisotropy about the mean is larger than expected for both
models. This variation is not correlated with the solar wind velocity or the plasma beta.
However, these anisotropy estimates are correlated between different field components,
suggesting that the assumption of axisymmetry is valid.
An alternative multi-spacecraft approach, where time-lagged second order structure
functions are used to estimate the field-aligned anisotropy of the power levels and the
spectral index, is also presented. The mean value obtained for the power in the 2D
fluctuations, (93 ± 15)%, is consistent with the simulated turbulent field results, while
the observed anisotropy in the spectral index suggests that multiple energy transfer
mechanisms are present in solar wind turbulence.
Results from all three analysis techniques are discussed and found to be correlated
with each other. This suggests that the observed variation in the anisotropy
estimates could be a physical effect that different data intervals exhibit different degrees
of anisotropy.