Ball lightning — a free-floating, roughly spherical luminous object, typically tens of centimeters across, that appears in association with thunderstorms and persists for seconds — has been reported for centuries, yet no accepted physical mechanism explains how it forms or sustains itself. The problem, approached in the literature through Boerner's free-electron/negative-corona precursor hypothesis, Nikitin's electrodynamic plasmoid model of creation near a linear-lightning channel, and laboratory pulse-discharge analogue studies, is to identify the plasma-physical pathway by which ordinary lightning or corona activity produces such a structure: the initiation step and its necessary precursors, the energy source, and the stabilization mechanism that maintains luminosity at atmospheric pressure for far longer than an unconstrained air plasma could survive. The single refereed spectrographic capture of a natural event (Cen, Yuan and Xue, 2014) shows emission from vaporized soil throughout the ball's lifetime, constraining composition but not formation or stability. Hypotheses spanning nanoparticle combustion, microwave-trapped plasma, electron-ion plasmoids, and electromagnetic solitons continue to proliferate through 2026 without experimental confirmation, laboratory analogues remain contested as genuine ball lightning, and the mechanism problem is open.

Background

Ball lightning is a luminous, roughly spherical object, most commonly reported to be between about 10 and 50 cm in diameter, that appears during thunderstorms, drifts near the ground or through structures for a few seconds, and then decays silently or explosively. Its existence rests on thousands of eyewitness reports; systematic instrumented captures are extremely rare, and the problem of explaining the phenomenon has been explicitly labeled 'the unsolved problem' in the literature since at least Singer (1963) and Charman (1982). Any explanation must resolve a central paradox: an ordinary electrical discharge in air at atmospheric pressure — a plasma, i.e. an ionized gas with free electrons and ions — recombines and stops glowing within microseconds to milliseconds once the driving electric field is removed, whereas reported ball lightning glows for seconds with no visible connection to a power supply. A mechanism must therefore specify an initiation pathway, a sustained energy source, and a confinement or stabilization principle.

Several pieces of terminology recur in this literature. A corona discharge is a weak, localized ionization discharge at a sharply curved electrode (or any pointed conductor) in a strong electric field; in a negative corona the electrode is negative and free electrons stream away from it into the surrounding air, making free electrons a natural candidate precursor. A cloud-to-ground lightning flash is initiated by a leader — a hot, ionized, propagating channel — and its most powerful current pulse, the return stroke, radiates an intense electromagnetic pulse. A plasmoid is a self-contained configuration of plasma held together by its own magnetic fields and currents. Bead lightning is a related phenomenon in which a lightning channel decays into a chain of lingering luminous balls.

Prior work approaches the formation question from three directions. Boerner (2016, 2023) analyzed the subset of reports in which creation of the ball was actually witnessed and concluded that strong positive cloud-to-ground lightning and very high electric fields are implicated, that in many cases a negative corona preceded the ball, and that free electrons produced by that corona appear required both to create the visible plasma and to form a receiving antenna for the return-stroke electromagnetic pulse. Nikitin (2019) proposed an electrodynamic model in which the ball is a positively charged assembly of rotating electron and proton plasmoids, and discussed a concrete creation mechanism near a loop of linear lightning channel driven by ultraviolet ionization and spatially non-uniform radio-frequency radiation. On the experimental side, Versteegh et al. (2008) produced centimeter-scale, half-second-lived luminous plasmoids from a pulsed atmospheric water discharge, and Théberge et al. (2020) produced confined luminous plasma balls propagating along low-pressure discharge channels as a tabletop thunderstorm analogue — but whether any laboratory object is a genuine instance of natural ball lightning is itself disputed (Shmatov and Stephan, 2023).

The main prior theoretical families constrain what a full answer must address. Abrahamson and Dinniss (2000) proposed that a lightning strike vaporizes soil into a silicon-rich nanoparticle network that glows by slow chemical oxidation, a scenario that received circumstantial support when Cen, Yuan and Xue (2014) obtained the first (and so far only refereed) spectrogram of a natural ball lightning event, created by a cloud-to-ground lightning strike, showing emission from soil elements for the entire lifetime of the ball. Lowke (1996) and Morrow (2019) pursued conventional electric-discharge models rooted in the lightning channel itself. Wu (2016) proposed that a relativistic electron bunch at the lightning tip radiates intense microwaves that ionize the air while radiation pressure evacuates a spherical plasma bubble trapping the radiation. Fedosin (2024) calculated an electron-ion model with a positively charged core and a fast-electron shell. Most recently, laser experiments producing millimeter-scale relativistic terahertz solitons — localized, self-sustaining lumps of electromagnetic energy — have been argued to exhibit ball-lightning-like characteristics (Zhou et al., 2026; Bulanov, 2026). None of these mechanisms is experimentally confirmed, and a 2019 review of theoretical, observational and experimental work since 2000 concluded that ball lightning still 'has so far evaded a widely accepted scientific explanation' (Shmatov and Stephan, 2019).

Problem Statement

Identify the plasma-physical pathway by which lightning or corona activity produces natural ball lightning: (1) initiation — which precursors (for example, free electrons from a negative corona, leader-adjacent high-field regions, vaporized soil, or an intense microwave burst at the lightning tip) are necessary and sufficient to convert thunderstorm electrical activity into a localized luminous object; (2) energy source — what powers the observed radiation for seconds without a visible external supply; and (3) stabilization — what confines or sustains the plasma at atmospheric pressure, where free recombination would quench luminosity within milliseconds. The proposed pathway must be quantitative and consistent with the documented observational record, including the association with cloud-to-ground lightning and high ambient fields, reported size and lifetime distributions, the soil-element spectrum of Cen et al. (2014), reported horizontal motion and terminal decay or explosion, and the rarer indoor and aircraft appearances; and it must be testable, either reproducible under laboratory conditions that are realizable in a thunderstorm, or distinguishable from the principal competing mechanisms by a decisive instrumented field observation. The problem is the formation-mechanism core of the long-standing open question of the physical nature of ball lightning.

References

  1. S. Singer, 'The Unsolved Problem of Ball Lightning', Nature 198, 745–747 (1963). DOI: 10.1038/198745b0. https://doi.org/10.1038/198745b0
  2. W. N. Charman, 'Ball lightning: the unsolved problem', Weather 37, 66–75 (1982). DOI: 10.1002/j.1477-8696.1982.tb03556.x. https://doi.org/10.1002/j.1477-8696.1982.tb03556.x
  3. J. J. Lowke, 'A theory of ball lightning as an electric discharge', Journal of Physics D: Applied Physics 29, 1237–1244 (1996). DOI: 10.1088/0022-3727/29/5/018. https://doi.org/10.1088/0022-3727/29/5/018
  4. J. Abrahamson and J. Dinniss, 'Ball lightning caused by oxidation of nanoparticle networks from normal lightning strikes on soil', Nature 403, 519–521 (2000). DOI: 10.1038/35000525. https://doi.org/10.1038/35000525
  5. J. Abrahamson, 'Ball lightning from atmospheric discharges via metal nanosphere oxidation: from soils, wood or metals', Philosophical Transactions of the Royal Society of London Series A 360, 61–88 (2002). DOI: 10.1098/rsta.2001.0919. https://doi.org/10.1098/rsta.2001.0919
  6. A. Versteegh, K. Behringer, U. Fantz, G. Fussmann, B. Jüttner and S. Noack, 'Long-living plasmoids from an atmospheric water discharge', Plasma Sources Science and Technology 17, 024014 (2008). DOI: 10.1088/0963-0252/17/2/024014. https://doi.org/10.1088/0963-0252/17/2/024014
  7. J. Cen, P. Yuan and S. Xue, 'Observation of the optical and spectral characteristics of ball lightning', Physical Review Letters 112, 035001 (2014). DOI: 10.1103/PhysRevLett.112.035001. https://doi.org/10.1103/PhysRevLett.112.035001
  8. H. Boerner, 'Analysis of Conditions favorable for Ball Lightning Creation', arXiv:1606.04421 (2016). https://arxiv.org/abs/1606.04421
  9. H.-C. Wu, 'Relativistic-microwave theory of ball lightning', Scientific Reports 6, 28263 (2016). DOI: 10.1038/srep28263. https://doi.org/10.1038/srep28263
  10. A. I. Nikitin, 'Possible process of ball lightning training in nature', Journal of Atmospheric and Solar-Terrestrial Physics 190, 54–61 (2019). DOI: 10.1016/j.jastp.2019.05.001. https://doi.org/10.1016/j.jastp.2019.05.001
  11. M. L. Shmatov and K. D. Stephan, 'Advances in ball lightning research', Journal of Atmospheric and Solar-Terrestrial Physics 195, 105115 (2019). DOI: 10.1016/j.jastp.2019.105115. https://doi.org/10.1016/j.jastp.2019.105115
  12. R. Morrow, 'The origin of ball and bead lightning from an expanded lightning channel', Journal of Atmospheric and Solar-Terrestrial Physics 195, 105116 (2019). DOI: 10.1016/j.jastp.2019.105116. https://doi.org/10.1016/j.jastp.2019.105116
  13. F. Théberge, G. Nadeau, V. Issa, F. Vidal and J.-C. Kieffer, 'Generation of confined plasma balls propagating along discharge channels: A comparison with ball lightning', Physical Review Research 2, 013266 (2020). DOI: 10.1103/PhysRevResearch.2.013266. https://doi.org/10.1103/PhysRevResearch.2.013266
  14. K. D. Stephan, R. Sonnenfeld and A. G. Keul, 'First comparisons of ball-lightning report website data with lightning-location-network data', Journal of Atmospheric and Solar-Terrestrial Physics 240, 105953 (2022). DOI: 10.1016/j.jastp.2022.105953. https://doi.org/10.1016/j.jastp.2022.105953
  15. M. L. Shmatov and K. D. Stephan, 'Questions regarding alleged laboratory creation of ball lightning', Journal of Atmospheric and Solar-Terrestrial Physics 242, 105995 (2023). DOI: 10.1016/j.jastp.2022.105995. https://doi.org/10.1016/j.jastp.2022.105995
  16. S. G. Fedosin, 'Electron-ion model of ball and bead lightning', Journal of Atmospheric and Solar-Terrestrial Physics 265, 106374 (2024). DOI: 10.1016/j.jastp.2024.106374. https://doi.org/10.1016/j.jastp.2024.106374
  17. H. Boerner, 'The Role of Free Electrons in Ball Lightning Creation', arXiv:2304.01784 (2023). https://arxiv.org/abs/2304.01784
  18. H. Boerner, 'The Link Between Lightning Physics and Ball Lightning', in 'Ball Lightning: A Popular Guide to a Longstanding Mystery in Atmospheric Electricity', Springer Nature Switzerland, pp. 113–129 (2025). DOI: 10.1007/978-3-032-02164-9_9. https://doi.org/10.1007/978-3-032-02164-9_9
  19. Y. Shibuya, M. Maeyama and Y. Inada, 'Massive, long-time generation of OH(X) in ball lightning discharge', Physics of Plasmas 32, 042110 (2025). DOI: 10.1063/5.0251196. https://doi.org/10.1063/5.0251196
  20. C. Zhou, D. Zhang, R. Qi, Y. Bai, Y. Zeng, Z. Zheng, L. Song, Y. Tian and R. Li, 'Ball-lightning-like relativistic terahertz solitons', Nature Photonics 20, 727–733 (2026). DOI: 10.1038/s41566-026-01899-y. https://doi.org/10.1038/s41566-026-01899-y
  21. S. Bulanov, 'Is ball lightning a relativistic soliton?', Nature Photonics 20, 611–612 (2026). DOI: 10.1038/s41566-026-01914-2. https://doi.org/10.1038/s41566-026-01914-2