Solar Flare Prediction with Time Series Analysis of HMI Data
- The rapid emergence of magnetic flux and free magnetic energy on the solar photosphere often indicates an increased probability of future flaring activity. For example, two active regions with the same total unsigned flux but different flux emergence rates may have different flaring activity. However, the nature of this change remains poorly understood. What patterns surround flux emergence and the temporal evolution of other magnetic variables governing eruptive activity? And how can we use these patterns to improve our predictions of solar flares? To answer this question, we took a three-part approach by creating time series data, featurizing these data, and using the resultant features in a machine learning model to determine which features govern eruptive activity. We found that more lag time (time series data length) increases model accuracy, indicating that important flare prediction data is encoded in the time series. We also found that certain individual variables, such as total unsigned flux and total free energy perform well on their own. This study represents a first step in time series analysis of flaring active regions.
|Type of resource
|June 1, 2019 - September 1, 2019
- Use and reproduction
- User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
- This work is licensed under a Creative Commons Attribution 3.0 Unported license (CC BY).
- Preferred Citation
- Pauker, Lucas and Bobra, Monica and Jonas, Eric. (2019). Solar Flare Prediction with Time Series Analysis of HMI Data. Stanford Digital Repository. Available at: https://purl.stanford.edu/yv269gc6873
Stanford Research DataView other items in this collection in SearchWorks
Also listed in
Loading usage metrics...