A Dynamical Model of Equatorial Magnetosonic Waves in the Inner Magnetosphere: A Machine Learning Approach. Issue 6 (21st June 2021)

• Record Type: Journal Article
• Authors: Boynton, R. J.; Walker, S. N.; Aryan, H.; Hobara, Y.; Balikhin, M. A.
• Is Part Of: Journal of geophysical research. Volume 126:Issue 6(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-06-21
• Subjects: magnetosonic waves -- machine learning -- NARMAX; Magnetospheric physics -- Periodicals; Space environment -- Periodicals; Cosmic physics -- Periodicals; Planets -- Atmospheres -- Periodicals; Heliosphere (Astrophysics) -- Periodicals; Geophysics -- Periodicals; 523.01
• DOI: 10.1029/2020JA028439
• Languages: English
• ISSNs: 2169-9380

Abstract

Abstract: Equatorial magnetosonic waves (EMS), together with chorus and plasmaspheric hiss, play key roles in the dynamics of energetic electron fluxes in the magnetosphere. Numerical models, developed following a first principles approach, that are used to study the evolution of high energy electron fluxes are mainly based on quasilinear diffusion. The application of such numerical codes requires statistical models for the distribution of key magnetospheric wave modes to estimate the appropriate diffusion coefficients. These waves are generally statistically modeled as a function of spatial location and geomagnetic indices (e.g., AE, Kp, or Dst). This study presents a novel dynamic spatiotemporal model for EMS wave amplitude, developed using the Nonlinear AutoRegressive Moving Average eXogenous machine learning approach. The EMS wave amplitude, measured by the Van Allen Probes, are modeled using the time lags of the solar wind and geomagnetic indices as inputs as well as the location at which the measurement is made. The resulting model performance is assessed on a separate Van Allen Probes data set, where the prediction efficiency was found to be 34.0% and the correlation coefficient was 56.9%. With more training and validation data the performance metrics could potentially be improved, however, it is also possible that the EMS wave distribution is affected by stochastic factors and the performance metrics obtained for this model are close to the potential maximum. Plain Language Summary: The Nonlinear AutoRegressive Moving Average eXogenous machine learning methodology is used to deduce a model of the equatorial magnetosonic (EMS) waves in the inner radiation belt. Machine learning algorithms automatically deduce a model from input‐output data. Here, the output data is the measurements of the EMS wave amplitude measured by the Van Allen Probes, while the inputs are the solar wind conditions, measured by the Advanced Composition Explorer spacecraft at L1, and geomagnetic conditions, measured by ground based magnetometers. Equatorial magnetosonic waves, together with chorus and plasmaspheric hiss, play key roles in the dynamics of the radiation belts, where high electron radiation can lead to spacecraft damage. Thus, developing a model of the EMS waves is an important step in the development of forecasts that can predict the amount of radiation a spacecraft will experience as it transits the radiation belts. We go on to assess the performance of the developed model and then the model is mapped out to the inner magnetosphere to inspect the response of the magnetosonic waves to changing solar wind and geomagnetic conditions. Key Points: Machine learning is used to develop a novel dynamical model of the magnetosonic waves using solar wind, geomagnetic indices and location This model is important because it can potentially lead to an improvement to the forecasting capabilities of numerical radiation belt models The performance metrics of the model could potentially be improved with more data or the system could be affected by stochastic factors