Journal article
Automated detection of pulsating aurora using deep learning methods
Frontiers in astronomy and space sciences, Vol.13, 1852793
08/07/2026
DOI: 10.3389/fspas.2026.1852793
Abstract
We have developed deep learning algorithms to detect and classify pulsating auroras in video data captured by the Time History of Events and Macroscale Interactions during Substorms (THEMIS) All-Sky Imagers (ASIs). We label auroral data into four categories: “bad viewing condition”, “no aurora”, “other aurora” and “pulsating aurora”. In contrast to all prior studies centered on auroral image classification, our primary goal revolves around the classification of pulsating aurora. We introduce two distinct deep learning approaches: first, a convolutional neural network (CNN) model for single-frame classification combined with a smoothing algorithm; second, a hybrid model that combines a CNN with a recurrent neural network (RNN), meaning we are allowing temporal information to inform our classifications. Our models are trained on a large dataset comprised of 100,000 images classified by expert auroral observers. We tested our algorithms on a new dataset with 58 full-night videos and found real world accuracy values of 63.9% for the CNN method and 55.9% for the RNN method. An important outcome of this work is that we used our techniques to classify every one of the more than one billion images that comprise the THEMIS-ASI dataset and thereby produced the largest dataset of automatically classified pulsating aurora to date. Further, this work sets the stage for assimilation of auroral machine-learning outputs into geospace simulations.
Details
- Title: Subtitle
- Automated detection of pulsating aurora using deep learning methods
- Creators
- Fei Wu - University of IowaRiley N. Troyer - Utah State UniversityAllison N. Jaynes - University of IowaYang Gao - University of IowaMary E. Haag - University of IowaJodie McLennan - University of IowaDarren Chaddock - University of CalgaryEric Donovan - University of CalgaryKung-Sik Chan - University of Iowa, Statistics and Actuarial Science
- Resource Type
- Journal article
- Publication Details
- Frontiers in astronomy and space sciences, Vol.13, 1852793
- DOI
- 10.3389/fspas.2026.1852793
- ISSN
- 2296-987X
- eISSN
- 2296-987X
- Publisher
- Frontiers Media SA
- Grant note
- NASA FINESST award: 80NSSC20K1514 NSF: 2045016 University of Calgary ARC HPC cluster
The author(s) declared that financial support was received for this work and/or its publication. RNT was supported by the NASA FINESST award 80NSSC20K1514 to the University of Iowa. AJ, FW, and K-SC were supported by NSF CAREER grant 2045016 to the University of Iowa. The machine learning model classification of the entire THEMIS-ASI data set was carried out with the support of the University of Calgary ARC HPC cluster.
- Language
- English
- Date published
- 08/07/2026
- Academic Unit
- Statistics and Actuarial Science; Physics and Astronomy; University College Courses
- Record Identifier
- 9985218373202771
Metrics
1 Record Views