Abstract
Forecasting of the dynamically relativistic electrons is crucial to alert of the flux enhancements that pose a significant threat to satellite operation at geostationary orbit (GEO). This work investigates contribution of solar wind and geomagnetic activities in the hourly electron flux (E > 2 MeV) at GEO measured by GOES satellites during the solar cycles 23–24. Time series forecasting models based on three Base learners: linear regression, Sequential Minimal Optimization for Regression (SMOreg), and feed forward neural network (NN) scheme with back-propagation learning in Waikato Environment for Knowledge Analysis (WEKA) are adopted to predict the GEO electron fluxes (f). It is found that the computation time of SMOreg method is much longer than other methods. For 48-hour lagged times of short testing sets of 5 days, the prediction capability of all Base learners is fairly good and comparable, but the SMOreg model based on the auroral (electrojet) lower (AL) overlay on logarithmic flux (log-f) is more suitable for the forecasting particularly during years 2008, 2011, and 2015. Sometimes the ultra-low frequency (ULF) wave indices as inputs provide better prediction accuracy than the solar wind speed as found in 2008. The significance of the AL suggests the role of substorm activities on the relativistic electron enhancements. The results suggest to nonlinearity in short-term variations in the relativistic electron flux in relation to the solar wind and magnetospheric conditions. For a larger testing sets (4 months) with 1-day lagged time, the values of the mean absolute error (MAE) and root mean square error (RMSE) increase. All Base learners provide similar good prediction accuracy results for each year with lower values of MAE and RMSE during 2008 and 2011. The prediction accuracy obtained from the input of selected parameters to overlay on the log-f is similar to the input of the log-f only. The model results indicate a linear dependence of relativistic electron fluxes on the solar wind and magnetospheric conditions in a longer term.
Similar content being viewed by others
REFERENCES
Baker, D.N., Erickson, P.J., Fennell, J.F., Foster, J.C., Jaynes, A.N., and Verronen, P.T., Space weather effects in the Earth’s radiation belts, Space Sci. Rev., 2018, vol. 214, no. 117. https://doi.org/10.1007/s11214-017-0452-7
Boberg, F., Wintoft, P., and Lundstedt, H., Real-time Kp predictions from solar wind data using neural networks, Phys. Chem. Earth C, 2000, vol. 25, no. 4, pp. 275–280.
Chen, Y., Reeves, G.D., Fu, X., and Henderson, M., PreMevE: New predictive model for megaelectron-volt electrons inside Earth’s outer radiation belt, Space Weather, 2019, vol. 17, no. 3, pp. 438–454. https://doi.org/10.1029/2018SW002095
Chu, X., Ma, D., Bortnik, J., Tobiska, W.K., Cruz, A., Bouwer, S.D., Zhao, H., Ma, Q., Zhang, K., Baker, D.N., Li, X., Spence, H., and Reeves, G., Relativistic electron model in the outer radiation belt using a neural network approach, Space Weather, 2021, vol. 19, p. e2021SW002808. https://doi.org/10.1029/2021SW002808
Degtyarev, V.I., Kharchenko, I.P., Potapov, A.S., Tsegmed, B., and Chudnenko, S.E., The relation between geomagnetic pulsations and an increase in the fluxes of geosynchronous relativistic electrons during geomagnetic storms, Geomagn. Aeron. (Engl. Transl.), 2010, vol. 50, no. 7, pp. 885–893.
Elkington, S.R., M. K. Hudson, and A. A., Acceleration of relativistic electrons via drift-resonant interaction with toroidal-mode Pc5 ULF oscillations, Geophys. Res. Lett., 1999, vol. 26, no. 21, pp. 3273–3276.
Fan, R.E., Chen, P.H., and Lin, C.J., Working set selection using second order information for training support vector machines, J. Mach. Learn. Res., 2005, vol. 6, pp. 1871–1918.
Fan, R.E., Chen, P.H., and Lin, C.J., A study on SMO-type decomposition methods for support vector machines, IEEE Trans. Neural Networks, 2006, vol. 17, pp. 893–908.
Freeman, J.W., O’Brien, T.P., Chan, A.A., and Wolf, R.A., Energetic electrons at geostationary orbit during the November 3–4, 1993 storm: Spatial/temporal morphology, characterization by a power law spectrum and, representation by an artificial neural network, J. Geophys. Res., 1998, vol. 103, no. A11, pp. 26 251–26 260.
Fukata, M., Taguchi, S., Okuzawa, T., and Obara, T., Neural network prediction of relativistic electrons at geosynchronous orbit during the storm recovery phase: Effects of recurring substorms, Ann. Geophys., 2002, vol. 20, pp. 947–951.
Hall, M.A., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., and Witten, I.H., The WEKA data mining software: An update, SIGKDD Explor., 2009, vol. 11, no. 1, pp. 10–18.
Haykin, S., Neural Networks: A Comprehensive Foundation, Prentice Hall, 1999.
Lazutin, L.L., Injection of relativistic electrons into the internal magnetosphere during magnetic storms: Connection with substorms, Geomagn. Aeron. (Engl. Transl.), 2013, vol. 53, no. 6, pp. 716–732.
Mathie, R.A. and Mann, I.R., A correlation between extended intervals of ULF wave power and storm-time geosynchronous relativistic electron flux enhancements, Geophys. Res. Lett., 2000, vol. 27, pp. 3261–3264.
Myagkova, I.N., Dolenko, S.A., Efitorov, A.O., Shirokii, V.R., and Sentemova, N.S., Prediction of relativistic electron flux in the Earth’s outer radiation belt at geostationary orbit by adaptive methods, Geomagn. Aeron. (Engl. Transl.), 2017, vol. 57, no. 1, pp. 8–15.
Perry, K.L., Ginet, G.P., Ling, A.G., and Hilmer, R.V., Comparing geosynchronous relativistic electron prediction models, Space Weather, 2010, vol. 8, p. S12002. https://doi.org/10.1029/2010SW000581
Pesnell, W.D., Lessons learned from predictions of Solar Cycle 24, J. Space Weather Space Clim., 2020, vol. 60. https://doi.org/10.1051/swsc/2020060
Pires de Lima, R., Chen, Y., and Lin, Y., Forecasting megaelectron-volt electrons inside Earth’s outer radiation belt: PreMevE 2.0 based on supervised machine learning algorithms, Space Weather, 2020, vol. 18. https://doi.org/10.1029/2019SW002399
Platt, J., Sequential Minimal Optimization: A Fast Algorithm for Training Support Vector Machines, Technical Report MSR-TR-98-14, 1998.
Potapov, A.S., Relativistic electrons of the outer radiation belt and methods of their forecast (review), Sol.-Terr. Phys., 2017, vol. 3, pp. 46–58.
Qian, Y., Yang, J., Zhang, H., Shen, C., and Wu, Y., An hourly prediction model of relativistic electrons based on empirical mode decomposition, Space Weather, 2020, vol. 18, p. e2018SW0022078. https://doi.org/10.1029/2018SW002078
Ripoll, J.-F., Claudepierre, S. G., Ukhorskiy, A. Y., Colpitts, C., Li, X., Fennell, J., and Crabtree, C., Particle dynamics in the Earth’s radiation belts: Review of current research and open questions, J. Geophys. Res., 2020, vol. 125, p. e2019JA026735. https://doi.org/10.1029/2019JA026735
Sakaguchi, K., Nagatsuma, T., Reeves, G.D., and Spence, H.E., Prediction of MeV electron fluxes throughout the outer radiation belt using multivariate autoregressive models, Space Weather, 2015, vol. 13, no. 12, pp. 853–867.
Shevade, S.K., Keerthi, S.S., Bhattacharyya, C., and Murthy, K.K., Improvements to the SMO algorithm for SVM regression, IEEE Trans Neural Networks, 2000, vol. 11, no. 5, pp. 1188–1193. https://pubmed.ncbi. nlm.nih.gov/18249845/.https://doi.org/10.1109/72.870050
Simms, L.E., Pilipenko, V., Engebretson, M.J., Reeves, G.D., Smith, A.J., and Clilverd, M., Prediction of relativistic electron flux at geostationary orbit following storms: Multiple regression analysis, J. Geophys. Res., 2014, vol. 119, no. 9, pp. 7297–7318.
Stringer, G. A., Heuten, I., Salazar, C., and Stokes, B., Artificial neural network (ANN) forecasting of energetic electrons at geosynchronous orbit, in Radiation Belts: Models and Standards, Lemaire, J.F., Heynderich, D., and Baker, D.N., Eds., Washington, DC: AGU, 1996, vol. 97, pp. 291–295.
Thorne, R.M., Radiation belt dynamics: The importance of wave-particle interactions, Geophys. Res. Lett., 2010, vol. 37, no. 22, p. L22107. https://doi.org/10.1029/2010GL044990
Wang, R. and Shi, L., Study on the forecasting method of relativistic electron flux at geostationary orbit based on support vector machine, Chin. J. Space Sci., 2012, vol. 32, no. 3, pp. 354–361.
Watari, S., Tokumitsu, M., Kitamura, K., Ishida, Y., Forecast of high-energy electron flux at geostationary orbit using neural network, Trans. JSASS Space Tech. Jpn., 2009, vol. 7, no. 26, pp. Tr_2_47–Tr_2_51.
Witten, I.H., Frank, E., Hall, M.A., and Pal, C.J., Data Mining: Practical Machine Learning Tools and Techniques, Morgan Kaufmann, 2016.
Wrenn, G.L., Rodgers, D.J., and Ryden, K.A., A solar cycle of spacecraft anomalies due to internal charging, Ann. Geophys., 2002, vol. 20, no. 7, pp. 953–956. https://doi.org/10.5194/angeo-20-953-2002
Yeeram, T., A neural network forecasting of relativistic electron flux at geostationary orbit: Solar activity phase dependence, J. Phys.: Conf. Ser., 2020a, vol. 1593, p. 012026. https://doi.org/10.1088/1742-6596/1593/1/012026
Yeeram, T., Solar activity phase dependence of the magnetospheric processes and relativistic electron flux at geostationary orbit, Astrophys. Space Sci., 2020b, vol. 365, no. 5. https://doi.org/10.1007/s10509-020-03794-6
ACKNOWLEDGMENTS
This research was financially supported by Mahasarakham University 2020. The author is grateful to the GOES mission. The OMNI data were obtained from (http://omniweb.gsfc.nasa.gov/). The author also wishes to thank the reviewers for constructive comments that have led to notable improvement in the manuscript.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
The authors declare that they have no conflicts of interest.
Rights and permissions
About this article
Cite this article
Thana Yeeram Time Series Forecasting of Hourly Relativistic Electrons at Geostationary Orbit during Solar Cycles 23–24. Geomagn. Aeron. 62, 278–287 (2022). https://doi.org/10.1134/S0016793222030185
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0016793222030185