Distribution of magnetic field components in the solar wind plasma

Abstract

We examine the probability distribution functions (PDFs) of fluctuations of magnetic field components using the Ulysses and Omnitape data sets to evaluate departures from Gaussian distributions. Functional fits as well as moment comparisons (kurtoses) are used in drawing conclusions concerning the degree of non‐Gaussianity. Short‐timescale fluctuations are separated, to the extent possible, from the longer‐timescale variation of the mean field, and attention is paid to data selection issues such as stationarity. At the present level of comparison it seems that departures from Gaussianity of the distributions of the fluctuations are not severe. The analysis is carried out in the mean field coordinates in which the fluctuations of components are close to being uncorrelated, and it is observed that the perpendicular components are closer to being Gaussian than the parallel one. It is shown that the kurtosis is highly exaggerated when the variation of the mean field is not taken into account. We further examine the distributions of hourly magnetic field fluctuations in fast and slow solar wind at different phases in the solar cycle in order to quantitatively describe departures from the Gaussian distribution. It is found that the kurtosis lies between 2.8 and 4.8 for all components in the mean field coordinates, in both fast and slow wind and near solar maximum and near solar minimum. It appears that the distributions of fluctuations are rather more similar with regard to their degree of non‐Gaussianity than might have been expected based upon the well‐documented differences in the characteristics of fast and slow solar wind intervals. This suggests a robust form of statistical similarity that may be associated with either in situ or source region nonlinear effects. In addition to the distributions of the fluctuations, we present the distributions of the mean field and of the variances of the components which are more likely to be influenced by solar sources than by interplanetary dynamical processes. The PDFs of variances of the magnetic field components over different subintervals approximate lognormal curves, and provide the motivation to compute the PDFs of the fluctuations using the approach of Castaing et al. [1990], which is a superposition of Gaussian distributions with variances that are distributed lognormally. The resulting PDFs of the fluctuations provide a good model to describe the small departures from Gaussian distributions seen in the observed PDFs. An additional step is taken to compute the expected PDF of the magnitude of the fluctuations from the Castaing PDFs of the components and is also seen to be in good agreement with the observed PDF.