Medium Energy Electron Flux in Earth's Outer Radiation Belt (MERLIN): A Machine Learning Model

Abstract

The radiation belts of the Earth, filled with energetic electrons, comprise complex and dynamic systems that pose a significant threat to satellite operation. While various models of electron flux both for low and relativistic energies have been developed, the behavior of medium energy (120-600 keV) electrons, especially in the MEO region, remains poorly quantified. At these energies, electrons are driven by both convective and diffusive transport, and their prediction usually requires sophisticated 4D modeling codes. In this paper, we present an alternative approach using the Light Gradient Boosting (LightGBM) machine learning algorithm. The Medium Energy electRon fLux In Earth's outer radiatioN belt (MERLIN) model takes as input the satellite position, a combination of geomagnetic indices and solar wind parameters including the time history of velocity, and does not use persistence. MERLIN is trained on >15 years of the GPS electron flux data and tested on more than 1.5 years of measurements. Tenfold cross validation yields that the model predicts the MEO radiation environment well, both in terms of dynamics and amplitudes o f flux. Evaluation on the test set shows high correlation between the predicted and observed electron flux (0.8) and low values of absolute error. The MERLIN model can have wide space weather applications, providing information for the scientific community in the form of radiation belts reconstructions, as well as industry for satellite mission design, nowcast of the MEO environment, and surface charging analysis. Plain Language SummaryThe radiation belts of the Earth, which are the zones of charged energetic particles trapped by the geomagnetic field, comprise complex and dynamic systems posing a significant threat to a variety of commercial and military satellites. While the inner belt is relatively stable, the outer belt is highly variable and depends substantially on solar activity; therefore, accurate and improved models of electron flux in the outer radiation belt are essential to understand the underlying physical processes. Although many models have been developed for the geostationary orbit and relativistic energies, prediction of electron flux in the 120-600 keV energy range still remains challenging. We present a data-driven model of the medium energies (120-600 keV) differentialelectron flux in the outer radiation belt based on machine learning. We use 17 years of electron observations by Global Positioning System (GPS) satellites. We set up a 3D model for flux prediction in terms of L-values, MLT, and magnetic latitude. The model gives reliable predictions of the radiation environment in the outer radiation belt and has wide space weather applications.

Energetic Charged Particles in the Terrestrial Magnetosphere: Cluster/RAPID Results

“…It can be useful for the long‐term statistical analysis of electron flux behavior in the outer radiation belt, and can be incorporated into the empirical models of the medium energy electron flux, for example, by Smirnov, Berrendorf, et al. (2020). Moreover, the electron fluxes converted to PSD can be used to set up boundary conditions for radiation belts modeling, as well as for data assimilation purposes.…”

The Short‐Time Prediction of the Energetic Electron Flux in the Planetary Radiation Belt Based on Stacking Ensemble‐Learning Algorithm

“…The linear regression model tends to have better performance when L > 4.5. Smirnov et al (2020) predicted the electron flux for the medium energy (120-600 keV) based on satellite position, geomagnetic index and solar wind parameters (including velocity time history). Chu et al (2021) used a neural network (ANN) to build a relativistic electron model for the prediction of the Earth's outer radiation belt.…”

Outer Radiation Belt Flux and Phase Space Density Response to Sheath Regions: Van Allen Probes and GPS Observations

“…Due to intermittent noise, the fits from ns60 can be unreliable, so we have excluded it from our study. Data from other satellites were filtered based on the goodness of fit following Smirnov et al (2020): the measured counts were compared to the modeled counts from the fit in the five lowest energy channels, and the data was discarded if the discrepancy was too high. This approach is conservative and tends to exclude particularly low flux data.…”