Frontside halo coronal mass ejections (CMEs) are the main drivers of severe geomagnetic storms, yet their storm impact cannot be reliably predicted before the CME arrives at Earth. Michalek, Gopalswamy, Lara and Yashiro (2006) showed, from cone-model deprojection of SOHO/LASCO halo CMEs, that space speed and source longitude give a useful first-order geoeffectiveness forecast, while the strongest predictors of storm intensity -- the total strength BB and southward component BZB_Z of the interplanetary CME (ICME) magnetic field -- are measured in situ and hence unavailable before arrival. The unresolved problem is to predict, from observations made before the ICME reaches Earth (remote sensing of the Sun and CME, or in-situ measurements of the ambient solar wind upstream of the ICME), the orientation and strength of the ICME magnetic field at the near-Earth point -- in particular the time profile and extrema of BZB_Z -- together with arrival time, and thereby the storm severity. Helicity-rule heuristics, semi-empirical flux rope models, heliospheric MHD simulations, and machine learning have each produced partial results, but as of 2026 none provides a validated, general pre-arrival prediction; this is the open problem known in the field as the Bz problem.

Background

A coronal mass ejection (CME) is an eruption of magnetized plasma from the solar corona; when launched toward Earth it is seen near the Sun as a 'halo' CME, a brightening surrounding the occulting disk of a coronagraph (an instrument that images the faint corona by blocking the solar disk), such as the Solar and Heliospheric Observatory (SOHO) and its Large Angle and Spectrometric Coronagraph (LASCO). Halo CMEs appear haloed only because they propagate approximately along the Sun-Earth line, so their measured sky-plane speeds suffer a projection effect. The 'cone model' used by Michalek et al. (2006) approximates the CME as a radially propagating cone to deproject the observed halo and recover the space (true) speed VV, angular width, and source longitude γ\gamma; symmetric and asymmetric versions of this inversion were developed by Michalek and collaborators in 2003-2007, and were later superseded by multi-view reconstructions using the two Solar Terrestrial Relations Observatory (STEREO) spacecraft and the graduated cylindrical shell (GCS) forward model of Thernisien (2011). Geoeffectiveness is the capacity of a solar or interplanetary structure to disturb Earth's magnetosphere; it is conventionally scored by geomagnetic storm indices such as DstDst (an hourly index of the ring current, more negative during stronger storms) or KpKp. During the passage of an interplanetary CME (ICME; the heliospheric counterpart of a CME), the dominant storm driver is magnetic reconnection between the ICME field and Earth's magnetopause, which is most efficient when the interplanetary field points southward, i.e. when its north-south component BZB_Z (measured in geocentric solar magnetospheric, GSM, coordinates) is large and negative; storm intensity correlates strongly with the peak southward BZB_Z and total field BB inside the ICME. Many ICMEs contain a magnetic flux rope -- a twisted, rope-like magnetic structure whose axis orientation, chirality (handedness of the twist), and axial and azimuthal field strengths determine the observed BZB_Z profile, which also depends on the observing spacecraft's path through the rope. Propagation from the Sun to 1 AU (one astronomical unit, the mean Sun-Earth distance) takes one to four days and modifies the structure: flux ropes deflect, rotate, expand, and erode as they interact with the ambient solar wind and with other CMEs, and ambient-wind preconditioning changes the resulting storm. Heliospheric propagation models used in forecasting include the magnetohydrodynamic WSA-ENLIL+Cone model (the empirical Wang-Sheeley-Arge solar-wind model coupled to the ENLIL three-dimensional magnetohydrodynamic heliospheric simulation, with the CME injected as a cone), drag-based models (which treat the CME as a body dragged by the solar wind), and data-driven flux rope emulations such as 3DCORE and INFROS (semi-empirical flux rope models fit to observations) and the spheromak inserts (a spheromak being a self-contained, roughly spherical twisted magnetic configuration) of EUHFORIA (EUropean Heliospheric FORecasting Information Asset, a three-dimensional magnetohydrodynamic heliospheric prediction model). Because no routine measurement of coronal or CME magnetic fields exists (there is no operational 'coronal magnetograph'), the field orientation must be inferred indirectly from source-region proxies -- the hemispheric helicity rule (the statistical preference for one sign of magnetic twist in each solar hemisphere), filament channel morphology (the structure of the sheared-field lanes that host filaments, curtains of cool dense plasma suspended in the corona), extreme-ultraviolet (EUV) loops, flare ribbon morphology (the pattern of the bright ribbons that mark the footpoints of newly reconnected loops during solar flares), and 'magnetic tongues' (elongated extensions of a bipolar source region, produced by the flux rope's azimuthal field) in magnetograms (maps of the Sun's surface magnetic field).

Problem Statement

Given remote-sensing observations of a frontside halo (Earth-directed) CME -- white-light coronagraph images, solar disk magnetograms, and EUV imaging obtained around the eruption time -- predict, before the interplanetary CME arrives at the near-Earth measurement point (an L1 monitor or Earth's magnetosphere), the magnetic field that will be observed there: the orientation of the embedded magnetic flux rope and the strength and time profile of its components, in particular the geocentric solar magnetospheric north-south component BZB_Z and total field BB, together with the arrival time and speed, and hence the geomagnetic storm severity (e.g., peak DstDst or KpKp). The prediction may use any observations made before the ICME reaches this near-Earth measurement point -- remote sensing of the Sun and CME, or in-situ measurements of the ambient solar wind upstream of the ICME -- but may not use in-situ measurements of the ICME itself, whether at the near-Earth measurement point or at an upstream monitor that the ICME passes before reaching Earth; a nowcast based on the ICME already impinging on such an upstream monitor affords only tens of minutes to hours of lead time and does not meet the objective, which is a prediction made over the one-to-four-day Sun-to-Earth transit window. The central scientific question is which pre-eruption and eruptive solar observables (flux rope chirality, axis orientation and axial field from source-region magnetograms, filament channels, EUV loops, flare and shock signatures, deprojected kinematics) determine the ICME field at 1 AU, and how deflection, rotation, expansion, erosion, and interaction with the ambient solar wind and other CMEs modify it en route. Predicting this magnetic structure -- the flux-rope orientation and the strength and time profile of BZB_Z and BB -- before arrival is the mandatory deliverable of the task and the open core of this problem; the arrival time and speed, and the storm severity they imply, are derived consequences of that prediction rather than separately required goals, and resolving the problem requires a validated, general pre-arrival prediction -- one that demonstrably applies to CMEs beyond those used to build or tune it -- not merely a proposed method, a single-event hindcast, or a sign-of-BZB_Z classification. The deprojection of the CME's space parameters is now routinely performed and serves as an available input rather than an open component of the objective, while kinematic and source-location criteria alone demonstrably cannot separate geoeffective from non-geoeffective halo CMEs -- the missing discriminant is the ICME magnetic field, especially BZB_Z.

References

  1. Michalek, G.; Gopalswamy, N.; Lara, A.; Yashiro, S. (2006). Properties and geoeffectiveness of halo coronal mass ejections. Space Weather 4, S10011. DOI: 10.1029/2005SW000218; arXiv:0710.4526. https://arxiv.org/abs/0710.4526
  2. Michalek, G.; Gopalswamy, N.; Yashiro, S. (2007). Prediction of Space Weather Using an Asymmetric Cone Model for Halo CMEs. Solar Physics 246, 399-408. DOI: 10.1007/s11207-007-9081-8; arXiv:0710.4372. https://arxiv.org/abs/0710.4372
  3. Gopalswamy, N.; Yashiro, S.; Akiyama, S. (2007). Geoeffectiveness of halo coronal mass ejections. Journal of Geophysical Research: Space Physics 112, A06112. DOI: 10.1029/2006JA012149. https://doi.org/10.1029/2006JA012149
  4. Yermolaev, Yu. I. (2008). Comment on "Geoeffectiveness of Halo Coronal Mass Ejections" by N. Gopalswamy, S. Yashiro, and S. Akiyama (J. Geophys. Res. 2007, 112, doi:10.1029/2006JA012149). Cosmic Research 46, 540-541. DOI: 10.1134/S0010952508060099; arXiv:0801.2709. https://arxiv.org/abs/0801.2709
  5. Thernisien, A. (2011). Implementation of the Graduated Cylindrical Shell Model for the Three-Dimensional Reconstruction of Coronal Mass Ejections. The Astrophysical Journal Supplement Series 194, 33. DOI: 10.1088/0067-0049/194/2/33. https://doi.org/10.1088/0067-0049/194/2/33
  6. Shen, Chenglong; Wang, Yuming; Pan, Zonghao; Zhang, Min; Ye, Pinzhong; Wang, S. (2013). Full halo coronal mass ejections: Do we need to correct the projection effect in terms of velocity? Journal of Geophysical Research: Space Physics. DOI: 10.1002/2013JA018872; arXiv:1303.4059. https://arxiv.org/abs/1303.4059
  7. Mays, M. L.; Taktakishvili, A.; Pulkkinen, A.; MacNeice, P. J.; Rastätter, L.; Odstrcil, D.; Jian, L. K.; Richardson, I. G.; LaSota, J. A.; Zheng, Y.; Kuznetsova, M. M. (2015). Ensemble Modeling of CMEs Using the WSA-ENLIL+Cone Model. Solar Physics 290, 1775-1814. DOI: 10.1007/s11207-015-0692-1. https://doi.org/10.1007/s11207-015-0692-1
  8. Savani, N. P.; Vourlidas, A.; Szabo, A.; Mays, M. L.; Richardson, I. G.; Thompson, B. J.; Pulkkinen, A.; Evans, R.; Nieves-Chinchilla, T. (2015). Predicting the magnetic vectors within coronal mass ejections arriving at Earth: 1. Initial architecture. Space Weather. DOI: 10.1002/2015SW001171. https://doi.org/10.1002/2015SW001171
  9. Savani, N. P.; Vourlidas, A.; Richardson, I. G.; Szabo, A.; Thompson, B. J.; Pulkkinen, A.; Mays, M. L.; Nieves-Chinchilla, T.; Bothmer, V. (2017). Predicting the magnetic vectors within coronal mass ejections arriving at Earth: 2. Geomagnetic response. Space Weather 15, 441-461. DOI: 10.1002/2016SW001458. https://doi.org/10.1002/2016SW001458
  10. Möstl, Christian; Amerstorfer, Tanja; Palmerio, Erika; Isavnin, Alexey; Farrugia, Charles J.; Lowder, Chris; Winslow, Reka M.; Donnerer, Julia M.; Kilpua, Emilia K. J.; Boakes, Peter D. (2018). Forward Modeling of Coronal Mass Ejection Flux Ropes in the Inner Heliosphere with 3DCORE. Space Weather 16, 216-229. DOI: 10.1002/2017SW001735; arXiv:1710.00587. https://arxiv.org/abs/1710.00587
  11. Palmerio, Erika; Kilpua, Emilia K. J.; James, Alexander W.; Green, Lucie M.; Pomoell, Jens; Isavnin, Alexey; Valori, Gherardo (2017). Determining the Intrinsic CME Flux Rope Type Using Remote-sensing Solar Disk Observations. Solar Physics. DOI: 10.1007/s11207-017-1063-x; arXiv:1701.08595. https://arxiv.org/abs/1701.08595
  12. Ameri, Dheyaa; Valtonen, Eino (2017). Investigation of the Geoeffectiveness of Disk-Centre Full-Halo Coronal Mass Ejections. Solar Physics. DOI: 10.1007/s11207-017-1102-7. https://doi.org/10.1007/s11207-017-1102-7
  13. Scolini, Camilla; Messerotti, Mauro; Poedts, Stefaan; Rodriguez, Luciano (2017). Halo coronal mass ejections during Solar Cycle 24: reconstruction of the global scenario and geoeffectiveness. Journal of Space Weather and Space Climate 7, A36. DOI: 10.1051/swsc/2017046; arXiv:1712.05847. https://arxiv.org/abs/1712.05847
  14. Kilpua, E. K. J.; Balogh, A.; von Steiger, R.; Liu, Y. D. (2017). Geoeffective Properties of Solar Transients and Stream Interaction Regions. Space Science Reviews 212, 1271. DOI: 10.1007/s11214-017-0411-3. https://doi.org/10.1007/s11214-017-0411-3
  15. Vourlidas, A.; Patsourakos, S.; Savani, N. P. (2019). Predicting the geoeffective properties of coronal mass ejections: current status, open issues and path forward. Philosophical Transactions of the Royal Society A 377, 20180096. DOI: 10.1098/rsta.2018.0096. https://doi.org/10.1098/rsta.2018.0096
  16. Kilpua, E. K. J.; Lugaz, N.; Mays, M. L.; Temmer, M. (2019). Forecasting the Structure and Orientation of Earthbound Coronal Mass Ejections. Space Weather. DOI: 10.1029/2018SW001944. https://doi.org/10.1029/2018SW001944
  17. Sarkar, Ranadeep; Gopalswamy, Nat; Srivastava, Nandita (2020). An Observationally Constrained Analytical Model for Predicting the Magnetic Field Vectors of Interplanetary Coronal Mass Ejections at 1 au. The Astrophysical Journal 888, 121. DOI: 10.3847/1538-4357/ab5fd7; arXiv:1912.03494. https://arxiv.org/abs/1912.03494
  18. Temmer, Manuela (2021). Space weather: the solar perspective. Living Reviews in Solar Physics 18, 4. DOI: 10.1007/s41116-021-00030-3. https://doi.org/10.1007/s41116-021-00030-3
  19. Reiss, Martin A.; Möstl, Christian; Bailey, Rachel L.; Rüdisser, Hannah T.; Amerstorfer, Ute V.; Amerstorfer, Tanja; Weiss, Andreas J.; Hinterreiter, Jürgen; Windisch, Andreas (2021). Machine Learning for Predicting the Bz Magnetic Field Component From Upstream In Situ Observations of Solar Coronal Mass Ejections. Space Weather 19, e2021SW002859. DOI: 10.1029/2021SW002859; arXiv:2108.04067. https://arxiv.org/abs/2108.04067
  20. Narock, Thomas; Narock, Ayris; Dos Santos, Luiz F. G.; Nieves-Chinchilla, Teresa (2022). Identification of Flux Rope Orientation via Neural Networks. Frontiers in Astronomy and Space Sciences 9, 838442. DOI: 10.3389/fspas.2022.838442; arXiv:2202.05901. https://arxiv.org/abs/2202.05901
  21. Ledvina, Vincent E.; Palmerio, Erika; Kay, Christina; Al-Haddad, Nada; Riley, Pete (2023). Modeling CME encounters at Parker Solar Probe with OSPREI: Dependence on photospheric and coronal conditions. Astronomy & Astrophysics. DOI: 10.1051/0004-6361/202245445; arXiv:2303.10793. https://arxiv.org/abs/2303.10793
  22. Sarkar, Ranadeep; Pomoell, Jens; Kilpua, Emilia; Asvestari, Eleanna; Wijsen, Nicolas; Maharana, Anwesha; Poedts, Stefaan (2024). Studying the Spheromak Rotation in Data-constrained Coronal Mass Ejection Modeling with EUHFORIA and Assessing Its Effect on the Bz Prediction. The Astrophysical Journal Supplement Series 270, 18. DOI: 10.3847/1538-4365/ad0df4; arXiv:2311.15616. https://arxiv.org/abs/2311.15616
  23. Sarkar, Ranadeep; Srivastava, Nandita; Gopalswamy, Nat; Kilpua, Emilia (2024). Modeling the Magnetic Vectors of Interplanetary Coronal Mass Ejections at Different Heliocentric Distances with INFROS. The Astrophysical Journal Supplement Series 273, 36. DOI: 10.3847/1538-4365/ad5835; arXiv:2406.09247. https://arxiv.org/abs/2406.09247
  24. Alobaid, Khalid A.; Wang, Jason T. L.; Wang, Haimin; Jing, Ju; Abduallah, Yasser; Wang, Zhenduo; Farooki, Hameedullah; Cavus, Huseyin; Yurchyshyn, Vasyl (2024). Prediction of Geoeffective CMEs Using SOHO Images and Deep Learning. Solar Physics 299, 159. DOI: 10.1007/s11207-024-02385-w; arXiv:2501.01011. https://arxiv.org/abs/2501.01011
  25. Rüdisser, Hannah T.; Davies, Emma E.; Amerstorfer, Ute V.; Möstl, Christian; Weiler, Eva; Weiss, Andreas J.; Le Louëdec, Justin; Reiss, Martin A.; Nguyen, Gautier (2026). Towards a Fully Automated Pipeline for Short-term Forecasting of In Situ Coronal Mass Ejection Magnetic Field Structure. The Astrophysical Journal 1003, 202. DOI: 10.3847/1538-4357/ae5e64; arXiv:2602.06926. https://arxiv.org/abs/2602.06926