We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Machine‐Learning Based Identification of the Critical Driving Factors Controlling Storm‐Time Outer Radiation Belt Electron Flux Dropouts.
- Authors
Hua, Man; Bortnik, Jacob; Ma, Donglai
- Abstract
Understanding and forecasting outer radiation belt electron flux dropouts is one of the top concerns in space physics. By constructing Support Vector Machine (SVM) models to predict storm‐time dropouts for both relativistic and ultra‐relativistic electrons over L = 4.0–6.0, we investigate the nonlinear correlations between various driving factors (model inputs) and dropouts (model output) and rank their relative importance. Only time series of geomagnetic indices and solar wind parameters are adopted as model inputs. A comparison of the performance of the SVM models that uses only one driving factor at a time enables us to identify the most informative parameter and its optimal length of time history. Its accuracy and the ability to correctly predict dropouts identifies the SYM‐H index as the governing factor at L = 4.0–4.5, while solar wind parameters dominate the dropouts at higher L‐shells (L = 6.0). Our SVM model also gives good prediction of dropouts during completely out‐of‐sample storms. Plain Language Summary: The outer belt relativistic and ultra‐relativistic electrons, also known as "killer" electrons due to their deleterious effects on satellites, can exhibit fast and significant losses (also called dropouts), which can result from the combined effects of various physical processes. This study aims to identify the critical driving factors controlling dropouts using a machine‐learning approach, which enables us to extract physical insights by isolating different drivers, and ranking their importance by comparing the model performance. Our study adopts a unique way to relate the inputs to dropouts in a nonlinear way compared to the traditional statistical method. We construct Support Vector Machine models using a time series of geomagnetic indices and solar wind parameters as inputs to predict storm‐time dropouts based on 5‐year Van Allen Probes observations. Our results demonstrate that the SYM‐H index is the most informative input at L = 4.0–4.5, suggesting the dominant effects of the ring current in the inner magnetosphere. Solar wind pressure and density are regarded as the governing factor at L = 6.0, indicating the important impacts of solar wind drivers at higher L‐shells. Our SVM models give good predictions of dropouts during completely unseen storms, which are crucial for the understanding and forecasting of outer belt electron flux dropouts. Key Points: We investigate the critical driving factors controlling dropouts by constructing dropout prediction models using Support Vector Machines (SVMs)The most informative (critical) inputs controlling dropouts are SYM‐H at L ≤ 4.5 and solar wind drivers at L = 6.0 with mixed impact in betweenOur ultimate best SVM models can capture the observed relativistic and ultra‐relativistic dropouts during completely unseen storm events
- Subjects
RADIATION belts; MACHINE learning; SOLAR wind; SUPPORT vector machines; WIND pressure; ELECTRONS
- Publication
Geophysical Research Letters, 2024, Vol 51, Issue 10, p1
- ISSN
0094-8276
- Publication type
Article
- DOI
10.1029/2024GL108268