Evidence review · Solar storms
Solar eruptions hit Earth in two main ways: a magnetic storm and a shower of particle radiation. Carrington’s magnetic storm is among the largest on record. The isotopes in trees and ice record only the particles, and Carrington’s particle shower was far smaller than the extreme events they record: at most about three times the biggest of the Space Age, or absent [1 Abstract][1 p. 6].
Our analysis marks our own reasoning and calculations, and the verdicts are our assessment. Everything else comes from the numbered sources.
Top 3
Carrington’s magnetic storm ranks among the three largest on record [2 Abstract].
≤ 3×
Its particle shower above 30 MeV was at most about three times August 1972’s, and possibly zero [1 Abstract][1 p. 7].
19–27×
Our calculation: the 5259 BCE shower above 200 MeV, against the median ceiling for any Carrington shower (15–38× across the middle half of its range) [3 p. 8][1 Table 1].
The comment’s claims
“The Carrington event left no isotopes”
A comment quoted, without attribution, on the original infographic
Neither beryllium-10 nor new chlorine-36 measurements in ice show a peak from 1859, and tree rings show no clear, widespread rise; a carbon-14 offset reported in polar trees, statistically significant but unexplained, hasn’t been seen in trees from other regions [1 Abstract][1 p. 2][4 Abstract][1 p. 6]. See section 3.
“It implies the Carrington event was ‘small’ in comparison to others.”
The same comment
Isotopes record only showers of very energetic particles. Carrington’s shower was far smaller than the extreme events they record, or absent, but its magnetic storm ranks among the three largest on record. See section 2.
1 · Two kinds of hit
A big eruption can shake Earth’s magnetic field, a geomagnetic storm, and it can shower Earth with fast particles, a solar energetic particle event. The fiercest storms come from eruptions aimed straight at Earth, near the middle of the Sun’s disc. The particles reach Earth most easily from the Sun’s western side, which the Sun’s magnetic field lines connect to Earth [1 p. 8].
Hits power grids and other long conductors, pipelines, satellite orbits and navigation signals [5 Abstract][6][7]. Measured by the Dst index, from low-latitude observatories, in nanotesla (nT): the more negative, the stronger the storm [8].
Hits satellites, astronauts beyond Earth orbit, and passengers and crew on high-latitude flights [9]; Earth’s magnetic field and thick atmosphere shield people on the ground [10 p. 6][3 p. 2]. The most energetic particles reach the ground as ground-level enhancements (GLEs), measured since the 1940s, and make the isotopes found in tree rings and ice [3 p. 2][10 p. 3].
Judging Carrington by isotopes is like judging a hurricane’s wind with a rain gauge.
A storm can break wind records with ordinary rainfall. The gauge isn’t wrong. It measures something else.
2 · Every event gets two scores
Six events, each scored twice: the magnetic storm on the Dst index, and the particle shower as a rough ranking, because the particle sizes come from different energy ranges and can’t share one scale.
Carrington, 1859the storm in the comment
One of the three largest on record; estimates run from about −625 to −1,760 nT, readings of the Colaba record from −850 to −1,050 [11 §2.2.1][12 Abstract]
At most about 3× August 1972 above 30 MeV, and possibly none at all [1 Abstract][1 p. 7]
March 1989Quebec blackout
−589 nT, the strongest since the index began in 1957 [8]
May 2024auroras far from the poles
August 1972fastest arrival on record
Only −125 nT, yet grids across North America were disrupted [8][6 Abstract]
The largest of the Space Age above 30 MeV [1 p. 5]
774/5 CEfound in tree rings
Extreme: tens of times, up to about 100×, the strongest since the 1950s, above about 200 MeV [17 p. 30][10 p. 4]
5259 BCEconfirmed in ice, 2026
Extreme: about 19–27× Carrington’s median ceiling above 200 MeV (our calculation) [3 p. 8][1 Table 1]
It is often called the greatest storm on record, but its size is uncertain: estimates of its Dst run from about −625 to −1,760 nT, and those read from the Colaba record from −850 to −1,050 nT [11 §2.2.1][18 Abstract][12 Abstract]. Storms in February 1872 and May 1921 were of comparable strength, which puts all three among the largest yet observed [2 Abstract][20 Abstract].
The storm that blacked out Quebec’s grid brought no ground-level particle event [5 Abstract][13]. Satellites still recorded a strong proton storm, peaking on March 13 at 3,500 particle flux units above 10 MeV [14].
3 · Two windows on the past
Magnetometers, telegraph logs and aurora sightings; particle detectors on the ground since the 1940s [3 p. 2].
Sees: magnetic storms, and ground-level particle events in the modern era.
Reaches back: about 180 years for instruments. Reports of possible auroras go back about 2,700 years, to Assyrian records from about 680–650 BCE [21 Abstract].
Carbon-14 in wood; beryllium-10 and chlorine-36 in ice.
Sees: only big showers of very energetic particles. Carbon-14 and beryllium-10 mainly record particles above about 235 MeV, chlorine-36 those above about 60 MeV [10 p. 3]. Even the strongest ground-level event on record, in February 1956, can’t be reliably seen in beryllium-10 records [3 p. 2]. Completely blind to magnetic storms.
Reaches back: about 15,000 years [10 p. 3].
So the isotopes didn’t miss Carrington’s storm. They were never able to see it.
New chlorine-36 measurements show no significant rise around 1859 in three ice-core records. That rules out an extreme particle event: at most, Carrington’s shower above 30 MeV was about three times August 1972’s, or no shower reached Earth at all [1 Abstract][1 p. 7]. Uusitalo and colleagues report a carbon-14 offset in polar tree rings after 1859, compared with trees at mid-latitudes, with high statistical significance; their own model can’t reproduce it. It hasn’t been seen in trees from other regions, and Mekhaldi and colleagues don’t count it as evidence of a large particle event [4 Abstract][1 p. 6].
4 · Two benchmarks
Magnetic-storm worst case
At risk: power grids and other long conductors, pipelines, satellite orbits and navigation signals [5 Abstract][6][7].
People: no direct harm; the danger is to infrastructure and the services it runs.
How often: rare, and published estimates vary widely; a study of the 1859 storm judged that the probabilities of such extreme storms “cannot be assigned with any reasonable accuracy” [12 Abstract].
Radiation worst case
At risk: satellites, astronauts beyond Earth orbit, and passengers and crew on high-latitude flights [10 p. 6][3 p. 2][9].
People on the ground: well shielded by the air and Earth’s magnetic field [10 p. 6].
How often: six or seven found in about 15,000 years of records, five of them also confirmed in ice [10 p. 3][3 p. 8]. That averages roughly one every one to two thousand years, but they come irregularly: 774/5 and 993/4 CE were about 220 years apart, while two other gaps exceed 4,500 years [10 p. 5][3 p. 2]. Much of the record hasn’t yet been checked year by year [17 p. 34]. The strongest known, in 12,350 BCE, was about 18% stronger than 774/5 CE [10 p. 3].
5 · Do they come together?
Mekhaldi and colleagues took the 50 largest geomagnetic storms since 1957, counted by day, merged those on consecutive days into 31 distinct storms, and found 10 within five days of a ground-level particle event, though not necessarily from the same active region on the Sun [1 p. 8].
Following the paper’s stated rule with Kyoto’s data, we get 34 storms, with the same 10 matches. The paper’s 31 appears only with a looser merge, such as grouping by calendar month. Other reasonable definitions give 20–35%.
In 3 of the 10, the only matching particle event came after the storm. May 2024 shows the two can come together on a modest scale.
That fits the Sun’s geometry. Carrington’s eruption came from near the middle of the disc, so its particles may have missed Earth [1 p. 8].
| Rank | Dates (UT) | Lowest Dst, nT | Ground-level events within five days of the storm |
|---|---|---|---|
| 1 | 13–14 Mar 1989 | −589 | none |
| 2 | 15–16 Jul 1959 | −429 | GLE 7 (2 days after) |
| 3 | 13 Sep 1957 | −427 | none |
| 4 | 11 Feb 1958 | −426 | none |
| 5 | 20–21 Nov 2003 | −422 | none |
| 6 | 10–11 May 2024 | −406 (provisional) | GLE 74 (same day) |
| 7 | 25–26 May 1967 | −387 | none |
| 8 | 31 Mar 2001 | −387 | none |
| 9 | 29–31 Oct 2003 | −383 | GLE 65 (2 days before), GLE 66 (1 day before), GLE 67 (3 days after) |
| 10 | 8 Nov 2004 | −374 | none |
| 11 | 8–9 Nov 1991 | −354 | none |
| 12 | 13 Nov 1960 | −339 | GLE 10 (1 day before), GLE 11 (2 days after) |
| 13 | 10–11 Oct 2024 | −333 (provisional) | none |
| 14 | 8–9 Jul 1958 | −330 | none |
| 15 | 1–2 Apr 1960 | −327 | none |
| 16 | 30 Apr 1960 | −325 | GLE 8 (4 days after) |
| 17 | 14 Jul 1982 | −325 | none |
| 18 | 4–5 Sep 1957 | −324 | none |
| 19 | 13 Apr 1981 | −311 | GLE 34 (3 days before) |
| 20 | 9 Feb 1986 | −307 | none |
| 21 | 23 Sep 1957 | −303 | none |
| 22 | 4–5 Sep 1958 | −302 | none |
| 23 | 15–16 Jul 2000 | −300 | GLE 59 (2 days before) |
| 24 | 24–25 Mar 1991 | −298 | none |
| 25 | 6–7 Apr 2000 | −292 | none |
| 26 | 6 Nov 2001 | −292 | GLE 62 (2 days before) |
| 27 | 6 Sep 1982 | −289 | none |
| 28 | 10 May 1992 | −288 | none |
| 29 | 7 Oct 1960 | −287 | none |
| 30 | 8 Mar 1970 | −284 | none |
| 31 | 10 Apr 1990 | −281 | none |
| 32 | 28 Oct 1961 | −272 | none |
| 33 | 11 Apr 2001 | −271 | GLE 60 (4 days after) |
| 34 | 21 Oct 1989 | −268 | GLE 43 (2 days before), GLE 44 (1 day after), GLE 45 (3 days after) |
Our analysis of hourly Dst from WDC Kyoto, final values to 2020 and provisional after [8], and the GLE list [13]. A storm counts if a GLE falls between five days before its first day and five days after its last. Windows of 4 to 14 days give the same 10.
6 · Bottom line
Its magnetic storm is still often used as the benchmark worst case [1 Abstract], and magnetic storms are what threaten power grids [5 Abstract]. The radiation worst case is a different, rarer kind of event, and much bigger than Carrington’s particle shower.
The Dst index starts in 1957. The sizes of the 1859, 1872 and 1921 storms are reconstructed from a few observatories, and their estimates differ [11 Abstract][2 Abstract].
The extreme events are compared above about 200 MeV. Carrington’s firmest limit is above 30 MeV; the 200 MeV ceiling assumes its spectrum looked like a scaled-up modern event [1 p. 6].
Mekhaldi and colleagues listed only the 10 storms that came with a ground-level event (their Table 2), not all 31, so our recount follows their stated rule [1 p. 8].
Bracketed numbers refer to this list; § gives a section, p. a page and Table a table of the source, and Abstract names the part of a work we used. For the 2026 Royal Society papers, p. gives the page of the published PDF, except where a source says we read another version. Links checked October 13, 2026.
Mekhaldi, F., Paleari, C. I., Smith, A. M., Aldahan, A., Beer, J., Christl, M., Vockenhuber, C., Hayakawa, H., Curran, M., Erhardt, T., Plummer, C., Simon, K., Wilcken, K., Zheng, M., and Muscheler, R. (2026). 36Cl concentrations from polar ice cores set new constraints on the Carrington event. Philosophical Transactions of the Royal Society A 384(2329), 20250257. Published September 10, 2026. CC BY 4.0.
Hayakawa, H., Cliver, E. W., Clette, F., Ebihara, Y., Toriumi, S., Ermolli, I., et al. (2023). The Extreme Space Weather Event of 1872 February: Sunspots, Magnetic Disturbance, and Auroral Displays. The Astrophysical Journal 959(1), 23.
Paleari, C. I., Mekhaldi, F., Christl, M., Gautschi, P., Vockenhuber, C., Anderberg, E., Beer, J., Raisbeck, G., Yiou, F., Brehm, N., Wacker, L., Wilhelms, F., and Muscheler, R. (2026). Ice-core cosmogenic radionuclides support the occurrence of an extreme solar particle storm 5259 BCE (7208 years BP). Philosophical Transactions of the Royal Society A 384(2329), 20250347. Published September 10, 2026. CC BY 4.0.
Uusitalo, J., Golubenko, K., Arppe, L., Brehm, N., Hackman, T., Hayakawa, H., et al. (2024). Transient Offset in 14C After the Carrington Event Recorded by Polar Tree Rings. Geophysical Research Letters 51(5), e2023GL106632.
Boteler, D. H. (2019). A 21st Century View of the March 1989 Magnetic Storm. Space Weather 17(10), 1427–1441.
Knipp, D. J., Fraser, B. J., Shea, M. A., and Smart, D. F. (2018). On the Little-Known Consequences of the 4 August 1972 Ultra-Fast Coronal Mass Ejecta: Facts, Commentary, and Call to Action. Space Weather 16(11), 1635–1643.
NOAA Space Weather Prediction Center. Geomagnetic Storms. Accessed October 13, 2026.
Nose, M., Iyemori, T., Sugiura, M., Kamei, T., Matsuoka, A., Imajo, S., and Kotani, T. (2015). Geomagnetic Dst index [data set]. World Data Center for Geomagnetism, Kyoto. Final hourly values to 2020 and provisional values from 2021, read on October 13, 2026.
NOAA Space Weather Prediction Center. Solar Radiation Storm. Accessed October 13, 2026.
Vasilyev, V., Usoskin, I., and Krivova, N. (2026). Linking solar magnetism, extreme solar particle events and stellar superflares. Philosophical Transactions of the Royal Society A 384(2329), 20250261. CC BY 4.0. We read the accepted version, arXiv:2602.10243; p. gives its page.
Cliver, E. W., and Dietrich, W. F. (2013). The 1859 space weather event revisited: limits of extreme activity. Journal of Space Weather and Space Climate 3, A31.
Tsurutani, B. T., Gonzalez, W. D., Lakhina, G. S., and Alex, S. (2003). The extreme magnetic storm of 1–2 September 1859. Journal of Geophysical Research: Space Physics 108(A7), 1268.
International GLE Database. List of ground-level enhancements, GLE 5 to GLE 77. Sodankylä Geophysical Observatory and University of Oulu. Accessed October 13, 2026.
NOAA Space Environment Services Center. Solar Proton Events Affecting the Earth Environment, 1976–present. Preliminary listing, hosted by NASA’s Solar Data Analysis Center; its entries run to 2017. Accessed October 13, 2026.
Hayakawa, H., Ebihara, Y., Mishev, A., Koldobskiy, S., Kusano, K., Bechet, S., et al. (2025). The Solar and Geomagnetic Storms in 2024 May: A Flash Data Report. The Astrophysical Journal 979(1), 49. We read the accepted version, arXiv:2407.07665; § gives its section.
Papaioannou, A., Mishev, A., Usoskin, I., Heber, B., Vainio, R., Larsen, N., et al. (2025). The High-Energy Protons of the Ground Level Enhancement (GLE74) Event on 11 May 2024. Solar Physics 300, 73. CC BY 4.0. We read the arXiv version, arXiv:2505.09180.
Usoskin, I., Miyake, F., Baroni, M., Brehm, N., Dalla, S., Hayakawa, H., et al. (2023). Extreme Solar Events: Setting up a Paradigm. Space Science Reviews 219, 73. CC BY 4.0.
Hayakawa, H., Nevanlinna, H., Blake, S. P., Ebihara, Y., Bhaskar, A. T., and Miyoshi, Y. (2022). Temporal Variations of the Three Geomagnetic Field Components at Colaba Observatory around the Carrington Storm in 1859. The Astrophysical Journal 928(1), 32.
Brehm, N., Christl, M., Knowles, T. D. J., Casanova, E., Evershed, R. P., Adolphi, F., et al. (2022). Tree-rings reveal two strong solar proton events in 7176 and 5259 BCE. Nature Communications 13, 1196. CC BY 4.0.
Love, J. J., Hayakawa, H., and Cliver, E. W. (2019). Intensity and Impact of the New York Railroad Superstorm of May 1921. Space Weather 17(8), 1281–1292.
Hayakawa, H., Mitsuma, Y., Ebihara, Y., and Miyake, F. (2019). The Earliest Candidates of Auroral Observations in Assyrian Astrological Reports: Insights on Solar Activity around 660 BCE. The Astrophysical Journal Letters 884(1), L18.
Method
On October 13, 2026, we checked every statement, number and source on the original infographic, 62 items in all, against three 2026 papers on extreme solar events (two in their published versions, one in its accepted version), Kyoto’s Dst index and the international list of ground-level enhancements. We looked up each storm’s Dst ourselves and recounted the storms that came with particle events. The check corrected several statements; the change log lists them.
After the page was built, a second reviewer who hadn’t seen the drafting redid the storm count from Kyoto’s hourly Dst and the list of ground-level events, which matched ours exactly, and checked the sources. The changes that followed: −850 nT is the low end of the estimates read from the Colaba record, not of all published estimates, which start at about −625; Mekhaldi and colleagues did list the 10 storms that came with a ground-level event; the extreme particle events come about once every one to two thousand years on average, with gaps from about 220 to over 5,000 years; the carbon-14 offset in polar trees after 1859 is now described as its authors report it; Carrington’s particle shower is called far smaller than the extreme events, not small; and a dozen smaller fixes.
The verdicts and the items tagged “Our analysis” are our own assessment, built on the cited sources: the analogy, the 19–27× comparison, which divides fluences from two papers, and the recount of storms and particle events. The three 2026 papers are open access under CC BY 4.0; the charts are our own, drawn from the numbered sources.
This page was researched and written with Claude, an AI assistant made by Anthropic, and then re-checked against its sources in a separate verification pass.
Corrections
If we get something wrong, we fix it and record the change below with the date. If a correction changes a verdict, the log says what the old verdict was.