Applying Unsupervised Machine Learning to Solar Wind Classification
MISTGeneral
Téo
Bloch
Date Submitted
2019-03-14 12:23:55
University of Reading
Téo Bloch (University of Reading), Clare Watt (University of Reading), Mathew Owens (University of Reading), Leland McInnes (Tutte Institute for Mathematics and Computing)
We present a new solar wind origin classification scheme developed independently using unsupervised machine learning. The scheme aims to classify solar wind into three types: coronal hole wind, streamer belt wind, and ‘unclassified' which does not fit into either of the previous two categories. The classification scheme is created using non-evolving solar wind parameters, such as ion charge states and composition, measured during Ulysses' three fast-pass latitude-scans. The scheme is subsequently applied to the whole of the Ulysses and ACE datasets. Given the choice of parameter type, the scheme is grounded in the physical properties of the solar source regions. Furthermore, the techniques used are selected specifically to reduce the introduction of subjective biases. We demonstrate significant ‘best case’ disparities (minimum 7%, maximum 18%) with the traditional "fast" and "slow" solar wind determined using speed thresholds.
All attendees are expected to show respect and courtesy to other attendees and staff, and to adhere to the NAM Code of Conduct. To report harassment or violation of the code of conduct please click here.