Authors: Samuel Fordin (University of Delaware), Michael Shay (University of Delaware), Lynn B. Wilson (NASA GSFC)

The Wind spacecraft has yielded several decades of high-resolution magnetic field data, a large fraction of which displays small-scale structures. In particular, the solar wind is full of wavelike fluctuations that appear in both the field magnitude and its components. The nature of these fluctuations can be tied to the properties of other structures in the solar wind, such as shocks, that have implications for the time evolution of the solar wind. As such, having a large collection of wave events would facilitate further study of the effects that these fluctuations have on solar wind evolution. Given the large volume of magnetic field data available, machine learning is the most practical approach to classifying the myriad small-scale structures observed. A multibranch 1D convolutional neural network was used to classify intervals of Wind magnetic field data and determine whether waves were present in each interval. Using this algorithm, fifteen years of Wind magnetic field data from 2005 to 2019 were classified based on their wave content, yielding about 5,000,000 circularly-polarized wave intervals from a total of about 75,000,000. The statistics of the wave dataset is detailed, and it is found that waves occur more often in the fast solar wind, at higher temperatures, and fluctuate primarily perpendicular to the background magnetic field.