Fact-Checked Science

Evidence review · S0 News, 24 October 2026

S0 News, October 24, 2026: a cold vortex, not a weak field, keeps the ozone hole large

On October 24, 2026, S0 News said the southern ozone hole had “beat or tied daily records over a dozen times this season” and that “an excellent explanation for that is the weakening magnetic field of Earth” [1 00:01:47–00:01:53]. Our two paper checks cover the episode’s links. This page checks its own statements against the agencies’ records: the ozone hole [7][6], the Sun [41], an earthquake in Vanuatu [48], Typhoon Champi [57] and a Bay of Bengal storm [65], and the pole-shift frame around the papers [1 00:00:18–00:00:22].

By Fact-Checked Science · Published · Updated

Our analysis marks our own reasoning and calculations, and the verdicts are our assessment. Everything else comes from the numbered sources. For the episode [1], times are positions in the video; for the papers, Abstract or Significance names the part we used and § a section; for Q10 of the Twenty Questions and the India Meteorological Department’s products, p. gives a page of the PDF; the times of NOAA’s, the USGS’s, the warning centres’ and the Japan Meteorological Agency’s products are UTC unless marked.

Code and numbersDownload as PDFSourcesHow this was checkedCorrections

A real quake and a large ozone hole; the magnetic explanation and Champi’s “miss” fail

Our assessment

7 days

Our analysis: dates in 2026 when CAMS’s daily-mean ozone-hole area beat the 1979–2025 record for the date, September 17–21 and October 14–15; NASA’s record has one in the season, October 15; none since [7][6].

−0.92 vs +0.05

Our analysis: correlation, 1990–2026, between year-to-year changes in the hole’s mean area for September 7–October 13 and in the 50 hPa minimum temperature [7], against the same for geomagnetic activity, the Ap index [16].

616 injured

And one dead: Champi’s toll in Taiwan, the island the episode said the typhoon “missed”, from Taiwan’s Central Emergency Operation Center as reported by CNA [62][1 00:01:24–00:01:26].

The episode’s claims

False

Weak field explains ozone hole

“an excellent explanation for that is the weakening magnetic field of Earth. Pollution and volcanic contributions are down. So what gives? Earth is letting in more space particle energy as our shield weakens.” [1 00:01:49–00:02:02]

NASA: “Year-to-year variations in area and depth are caused by year-to-year variations in temperature” [5]; CAMS noted that the hole’s rapid development “coincided with a sudden decrease in the minimum temperature at approximately 20km above the south pole” [9]. Since 1990, year-to-year changes in the hole’s mean area for September 7–October 13 have followed the 50 hPa temperature (r = −0.92), not geomagnetic activity (r = +0.05). The colder vortex accounts for about half of 2026’s growth, 3.3 of 7.0 million km²; the rest does not follow geomagnetic activity, which was lower than in 2024 or 2025 despite three stormy days (our analysis) [7][16]. Particle-driven ozone loss is real but sits at 30–80 km and fades by September [18 Abstract][19 Abstract]; lower down, where the hole forms, particle-made nitrogen oxides have curbed chlorine-driven loss in years with an easterly quasi-biennial oscillation [23 Abstract]. See section 3.

Overstated

Ozone records over a dozen times

“we are at another peak in southern hemisphere ozone hole size. We’ve beat or tied daily records over a dozen times this season.” [1 00:01:43–00:01:49]

The hole is near-record for late October, largest or second-largest for every date from October 10 to 24 in CAMS’s record [7]. But it set records for the date on 7 days in CAMS’s analysis, September 17–21 and October 14–15, and on 1 in NASA’s season, October 15, with none since; on the day of the episode it was shrinking after a small rise on October 22 (our analysis) [7][6]. The only 2026 series with more than a dozen record days is NASA’s 50 hPa minimum temperature: records for cold [6]. See section 3.

Misleading

A pole shift is under way

“We are back at the ongoing magnetic pole shift of Earth.” [1 00:02:45–00:02:50]

The north dip pole drifts toward Siberia and is slowing: 56 km a year in 2002, 36 in 2025–2026, 31 forecast for 2027 [32 p. 3][32 p. 12][33 §3]. The dipole loses about 5% a century [31 Abstract]. That is ordinary secular variation; the field is “probably not reversing” [34], and our review of the forecasts found no excursion likely for centuries. The same words carry the same verdict on our review of the October 20 episode. The line-up sentence’s claim about the jerks paper [1 00:00:18–00:00:22] is rated on the paper’s page. See section 2.

False

Science admits extinction risk

“what mainstream science can’t deny about the extinction risk of magnetic pole ships like the one we’re in now” [1 00:03:31–00:03:35]

Mainstream reviews say the opposite: they find “no currently demonstrable relationship” between extinctions and reversals [35 Abstract] and “no clear picture of a geomagnetic link” [36 Abstract]. The one argued case, the Laschamp excursion, is contested by two formal comments [37][38], and its authors call the events “only temporally associated” [39]; the Blake excursion passed “without apparent serious effects” [37], and the last reversal, about 773,000 years ago [40], is followed by no demonstrable extinction [35 Abstract]. And no excursion is under way: the field “is not in an early stage of a reversal or excursion” [34 Significance], as our review of the October 22 episode also found. See section 2.

True

Vanuatu M7.2 downgraded to 6.9

“Initial 7.2 downgraded to 6.9. … Even with the downgrade, it’s 24-hour uptick blows everything else away. No tsunamis” [1 00:01:02–00:01:17]

The warning centres’ first estimate, a rapid preliminary one, was 7.2, from 10 stations, in the Pacific Tsunami Warning Center’s threat message at 02:37 UTC and the National Tsunami Warning Center’s statement at 02:39 [51][52][48]. The USGS posted Mww 6.9 at 02:45, and the Pacific center’s final message, at 03:28, gave 6.9, “NOTE REVISED MAGNITUDE”, and said the tsunami threat “HAS NOW PASSED” [48][51]. It was the day’s largest quake [49]. The verdict covers these words; section 5 also weighs the next ones, “Vanuatu can handle bigger quakes”.

Misleading

Champi missed Taiwan, hits Japan

“good news. The super typhoon missed Taiwan last night. … That first one will bend back and strike southern Japan tomorrow.” [1 00:01:24–00:01:31]

“Missed” holds only in that Champi made no landfall: the Japan Meteorological Agency’s analyses put its centre 38 km off Taiwan’s easternmost cape at 18:00 UTC on October 23 (our analysis) [57]. It still left one person dead, 616 injured and more than 340,000 households without power [62], and it was no longer a super typhoon [57][58]. It did turn toward Japan, then became an extratropical low over the East China Sea at 12:00 UTC on October 25, 438 km from Kagoshima [56] (our analysis); only its remnant trough was forecast to cross Kyushu, on October 26 [55]. See section 6.

1 · The episode

Two links, a large ozone hole, and a pole shift “we’re in now”

The episode of October 24, 2026, ran four and a half minutes and linked two items: a preprint on geomagnetic activity and world mortality [2] and a peer-reviewed article on the field’s acceleration at the top of Earth’s core [3]. Around them the presenter placed the Sun, an earthquake in Vanuatu, two storms and the southern ozone hole, inside the frame he has used all week, “Earth’s ongoing magnetic pole shift” [1 00:00:20–00:00:22], and closed with a promotion [1 00:03:50–00:04:10].

In the episode’s words

Quoted from the automatically generated transcript. “pole ships” is its rendering of pole shifts, “umbrelles” of umbrae, “corona hole” of coronal hole and “it’s” of its; “2026 into early 2017” almost certainly means 2016; “Ryer forcing” we could not resolve.

“today we’ve got big [clears throat] stories on the ozone, solar storm deaths, and a new data point on the worsening situation of Earth’s ongoing magnetic pole shift. We’ll also hit some weather and seismicity.”

“Folks, we are at another peak in southern hemisphere ozone hole size. We’ve beat or tied daily records over a dozen times this season. And an excellent explanation for that is the weakening magnetic field of Earth. Pollution and volcanic contributions are down. So what gives? Earth is letting in more space particle energy as our shield weakens. And that isn’t going to make this situation any better, is it?”

“From what mainstream science can’t deny about the extinction risk of magnetic pole ships like the one we’re in now, to the evidence explanations that extend to micronova crust unlocking and the turning of the earth, swaying like a drunkard, there is one safe location system that stands high above all others.”

The two links and the promotion, and what the episode made of them

World mortality and the Ap index, 1950–2023

What it is
A preprint posted on SSRN in October 2026 by Vladimir Belkin, not peer reviewed. Following A. L. Chizhevsky, it numbers the years of each solar cycle, groups them, and compares the mean geomagnetic Ap index with the world’s mean death rate per 1,000 people [2 Abstract].
What the author claims
Over the average 12-year cycle “a strong positive correlation, with a correlation coefficient of 0.72”; for years 1–4, “the growth phase”, “a strong negative correlation, with a correlation coefficient of-0.84”; and for years 5–12 “a strong positive correlation, with a correlation coefficient of 0.89” [2 Abstract].
What the episode said
“mortality in solar storms get linked up here for about the 10th time in the science journals. It shows that during the growth phase of the solar cycle, solar storms actually help drop mortality. But once the sunspots are kicking, the big storms increase mortality significantly.” [1 00:02:10–00:02:26]; “the cosmic ray peak at sunspot minimum is actually even worse” [1 00:02:31–00:02:34]

In our check of Belkin (2026): “Mortality tracks Ap, R = 0.72” Misleading; “Link flips with the cycle phase” Unsupported; “Linked again in science journals” Misleading; “Storms cut deaths early in cycle” Unsupported; “Big storms raise deaths at peak” Overstated; “Cosmic-ray peak at minimum worse” Unconfirmed. All three published correlations reproduce to the published two decimals, from UN, GFZ and SILSO data, with the year numbering we reconstructed, and a straight falling line grouped the same way gives nearly the same numbers (0.85 for the 0.89 of cycle years 5–12, which is 0.70 without China’s 1959–61 famine). Daily studies find slightly more deaths on days of high geomagnetic activity, a fraction of 1% per standard deviation, but nothing shows a large rise once the Sun is active. Cosmic rays do peak near solar minimum, and two ecological studies, in São Paulo and Lithuania, tie them to deaths; world death rates do not confirm it, and no study weighs cosmic rays at minimum against storms at maximum, as “even worse” implies.

Secular acceleration at the core–mantle boundary and geomagnetic jerks

What it is
A peer-reviewed article in Physics of the Earth and Planetary Interiors, open access, by Serrano, Campuzano and Pavón-Carrasco, in the volume for December 2026: “Persistent regional patterns of secular acceleration at the core–mantle boundary and their relation to geomagnetic jerks over the past 125 years” [3].
What the authors claim
As the title puts it, “persistent regional patterns” of secular acceleration at the core–mantle boundary, related to the geomagnetic jerks of the past 125 years [3]; our check of the paper works from its abstract.
What the episode said
“earth has completely lost the plot. The wheels are coming off almost coming completely off the rails in a full equatorial collapse. Concentration points are the South Atlantic anomaly in Bermuda Triangle and then on the West Pacific anomaly and the Dragons Triangle exactly as we expect them to be.” [1 00:03:13–00:03:29]

In our check of Serrano, Campuzano and Pavón-Carrasco (2026): “Data point on a worsening shift” Misread; “Growing power and coverage” Misleading; “A full equatorial collapse” False; “SAA, West Pacific, the Triangles” Misleading; “Acceleration peaks near equator” True. The paper itself could not be read: these verdicts rest on its abstract, from an index copy and paraphrased, and on our re-run of the public field models CHAOS-8.5 and COV-OBS.x2.

The promotion: a “safe location system”

What it is
A paid service that scores locations, “Havenbound — A Safety Report for Any Coordinate on Earth”, sold by Havenbound LLC, a Wyoming limited liability company; not a source, and not rated here [4].
What the episode said
“And in the top two models of the pole shift, Havenbound has every major risk and a whole lot of minor ones, human and natural, across time and whichever model you think is right. Havenbound.co is the number one safe location system on Earth. And that is true for now and for the aftermath.” [1 00:03:50–00:04:08]
What is disclosed
The video’s description opens with “Safe Location Finder: https://havenbound.co” and repeats it under “STEP 4”, beside a promo code for two other firms and a referral code for a third; neither the description nor the presenter says whether S0 is paid by, or has a stake in, Havenbound [1].

Not rated. Havenbound’s own terms say its “Theoretical” layers, which model “hypothetical large-scale geophysical events”, are “speculative by design”, and disclaim “any representation that the modeled events will occur, are likely to occur, or are endorsed as scientific consensus” [4 §3].

“A pole shift is under way” carries the same label and verdict as on our review of the October 20 episode, and “Science admits extinction risk” rests on the sources of our review of the October 22 episode; this page reuses them. The two paper pages rate what the episode said about the papers; this page rates the episode’s own statements.

2 · The frame

No pole shift in the records, and no extinction risk that mainstream science “can’t deny”

The episode called the pole shift “ongoing” twice [1 00:00:20–00:00:22][1 00:02:45–00:02:50] and ended with “the extinction risk of magnetic pole ships like the one we’re in now”, which, it said, mainstream science “can’t deny” [1 00:03:31–00:03:35]. The site has rated both claims before. Here is where today’s field stands and what the reviews say.

Strength: lost since 1840

9%

By about 2015 the axial dipole was “9% weaker than it was in 1840”, a decline of roughly 5% a century [31 Abstract].

Motion: the north dip pole

36 km/yr

Its speed from 2025 to 2026, down from a peak of 56 km a year in 2002 and forecast to slow to 31 in 2027 [32 p. 3][32 p. 12][33 §3].

Direction of travel

Not reversing

“Earth’s magnetic field is probably not reversing” [34]; our review of the forecasts rated “No excursion likely for centuries” True.

Our analysis

“A pole shift is under way”: Misleading, as on October 20

The poles do move and the dipole does weaken, and the episode can call that a shift. But it means more: the frame of the week is an excursion or a flip “we’re in now”, with “the top two models of the pole shift” [1 00:03:35][1 00:03:50–00:03:53]. The records show the north dip pole slowing, from 56 km a year in 2002 to 36 in 2025–2026 [33 §3][32 p. 3][32 p. 12], and the dipole losing about 5% a century [31 Abstract]: ordinary secular variation, in a field that is “probably not reversing” [34]. That an excursion is under way was rated False on our review of October 22 and of October 23. Same words as on October 20, same verdict.

Our analysis

“Science admits extinction risk”: False

What mainstream science says, in its reviews, is that there is “no currently demonstrable relationship” between biological extinctions and geomagnetic reversals [35 Abstract] and “no clear picture of a geomagnetic link” [36 Abstract]. The one case argued for an excursion, the Laschamp of about 42,000 years ago, drew two formal comments [37][38], and its authors replied that the events are “only temporally associated” [39]; the Blake excursion passed “without apparent serious effects” [37]; and the last full reversal, about 773,000 years ago [40], is followed by no demonstrable extinction [35 Abstract]. Our review of the October 22 episode rated “Excursions take out much of life” False on these sources, and no excursion is under way to carry the risk [34 Significance]. “Can’t deny” is the opposite of the record.

The rest of the closing line, rated before

“micronova crust unlocking and the turning of the earth, swaying like a drunkard” [1 00:03:39–00:03:42] lists ideas the site has checked: a solar micronova, False on the video check (micronovae happen on white dwarfs fed by a companion star); the crust turning 90° in a day, False on the video check and on our review of Earth’s energy stores; and ECDO’s turn of the mantle and crust in hours, False on our review of ECDO. The episode’s framing of the jerks paper, “lost the plot”, “a full equatorial collapse” and the two Triangles, is rated on the paper’s page.

The dipole lost about 9% from 1840 to 2015 and the north pole’s drift is slowing; the reviews find no link between field reversals and extinctions.

3 · The ozone hole

A large hole for late October, largely set by a cold vortex; records on 7 days, not over a dozen

The episode’s main claim: the hole is at “another peak”, has “beat or tied daily records over a dozen times this season”, and the weakening field explains it [1 00:01:43–00:02:02]. Our review of the 2026 hole, to October 18, found it unusually wide but not unusually deep. Here is what the two agencies’ data show since then, and what sets a hole’s size.

In the agencies’ words

NASA Ozone Watch measures the hole south of 40°S, CAMS south of 60°S, both as the area where total ozone is below 220 Dobson units [5][8].

“The depth and area of the ozone hole are primarily governed by the amounts of chlorine and bromine in the Antarctic stratosphere. … Year-to-year variations in area and depth are caused by year-to-year variations in temperature. Colder conditions result in a larger area and lower ozone values in the center of the hole.”

NASA Ozone Watch [5]

“Ozone depletion in the Southern Hemisphere stratosphere occurs as cold temperatures allow the formation of polar stratospheric clouds, which play a role in activating the chemical removal of ozone via halogens in the presence of sunlight during the austral Spring (August to November). The rapid development of the 2026 ozone hole coincided with a sudden decrease in the minimum temperature at approximately 20km above the south pole (50 hectopascal / hPa).”

CAMS, September 16, 2026 [9]

“2026年の南極オゾンホールは8月中旬頃に現れ、その面積は、9月上旬にかけて拡大したのち、9月中旬にかけてその時期の最近10年間の最大値程度で推移しています” [the 2026 Antarctic ozone hole appeared around mid-August; its area grew until early September and then, through mid-September, stayed around the largest values for the time of year in the last ten years]

Japan Meteorological Agency, September 29, 2026 [11]

The 2026 hole against the record for each date, September 1 to October 24

Our analysis
  • 2026, CAMS daily mean
  • At or above the record for the date
  • Highest daily mean for the date, 1979–2025
  • 95th percentile, 1979–2023 (dashed)
The 2026 ozone hole against the record for each date, September 1 to October 24 CAMS daily-mean ozone-hole area in 2026, in million square kilometres, with the highest value for each date in 1979–2025 and the 95th percentile of 1979–2023. 2026 peaks at 27.3 on September 20 and again at 26.2 on October 14; it is at or above the record for the date on 7 days: September 17, September 18, September 19, September 20, September 21, October 14, October 15. After October 15 it stays below the record but above the 95th percentile, with a small rise to 24.0 on October 22, and stands at 21.9 on October 24, the day of the episode. 10 15 20 25 30 Sep 1 Sep 10 Sep 20 Oct 1 Oct 10 Oct 20 Oct 24, the episode September 17: 26.42, above the previous record, 25.93 (2006) September 18: 27.17, above the previous record, 25.76 (2006) September 19: 27.12, above the previous record, 25.72 (1998) September 20: 27.30, above the previous record, 26.87 (1998) September 21: 26.40, above the previous record, 25.87 (1998) October 14: 26.18, above the previous record, 25.37 (2015) October 15: 25.98, above the previous record, 25.21 (2015) 2026 peak, 27.3 (Sep 20) 26.2 (Oct 14) small rise, 24.0 (Oct 22) Record for the date, 1979–2025 95th percentile

Area where total ozone is below 220 Dobson units, south of 60°S, million km²

Redrawn by factchecked.science from CAMS’s data files, which run to 18:00 UTC on October 24 [7]; CAMS counts the area south of 60°S [8]; each day is the mean of the four six-hourly analyses, and the record and the 95th percentile are our calculation for each date from the years 1979–2025 and 1979–2023 (the baseline CAMS uses [9]). NASA’s record has the same shape; its one record for the date in the season is October 15, 26.39 million km² against 26.11 in 2015 [6]. Contains modified Copernicus Atmosphere Monitoring Service information 2026 and modified Copernicus Climate Change Service information 2026; neither the European Commission nor ECMWF is responsible for any use that may be made of the Copernicus information or data it contains [10].

Rank, October 10–24

Our analysis

2nd

In CAMS’s record for 1979–2026 the 2026 hole was the second-largest, behind 2015, on 13 of the 15 dates from October 10 to 24, and the largest on October 14–15 [7].

Above the 95th percentile

Our analysis

26 of 38 days

From September 17 to October 24 in CAMS’s daily means, every day from October 8; NASA’s values were above its 90th percentile on 26 of 36 days from September 18 to October 23 [7][6].

October 24, the day of the episode

Our analysis

21.9

Million km², CAMS’s daily mean, down from 24.0 on October 22 and 26.2 on October 14 [7]; NASA’s latest value, for October 23, was 23.40 [6].

Records for the date in 2026, counted in each series

SeriesDays at or beyond the 1979–2025 record for the dateWhich days
CAMS, ozone-hole area, daily mean7September 17–21; October 14–15 [7]
CAMS, ozone-hole area, any six-hourly analysis8 days, 30 analysesThe same, plus October 13 [7]
NASA, ozone-hole area, daily value1 in the seasonOctober 15; four more on May 18–20 and June 6, at 0.9–2.7 million km², months before the hole forms [6]
CAMS and NASA, minimum ozone and mass deficit0 in the seasonThe hole was wide, not deep [7][6]
NASA, 50 hPa minimum temperature, record lows16 since August 1From August 1 the file’s values come from a different analysis (GEOS FP) from its 1979–2025 climatology (MERRA-2) [6]
CAMS, 50 hPa minimum temperature, record lows3August 22, September 13 and October 24 [7]

Our analysis, with ozone_checks.py; “at or beyond” means at or above the record area, or at or below the record-low temperature. NASA’s values have two decimals; no 2026 area value ties a record [6].

Our analysis

“Records over a dozen times”: Overstated

The hole is remarkable, and the episode is right about that much: it has been the largest or second-largest for each date from October 10 to 24 in 48 years of CAMS’s record, and above the 95th percentile every day since October 8 [7]. But records for the date fell on 7 days in CAMS’s analysis and on 1 in NASA’s season, all before October 16; counting each of CAMS’s six-hourly analyses gives 30 values, but on 8 days [7][6]. “Another peak” was a small rise on October 22, 2.2 to 2.5 million km² below the mid-October peak, and by the day of the episode both records showed the hole shrinking [7][6]. The only 2026 series with more than a dozen record days is NASA’s 50 hPa minimum temperature, 16 since August 1, in a file that changed analyses that day; CAMS’s own series has 3, the last on October 24 [6][7]. If those are the records the presenter counted, they are records for cold, the agencies’ explanation for a large hole [5][9].

What moves the hole from one year to the next

Year-to-year changes in the hole’s size, against temperature and against geomagnetic activity, 1990–2026

Our analysis
  • Change from the year before, 1991–2025
  • 2026, against 2025
What moves the hole from one year to the next, 1990–2026 Two scatter plots of 36 year-to-year changes, 1991 to 2026. Left: the change in the hole’s mean September–October area against the change in the mean 50 hPa minimum temperature; the points fall along a line from upper left to lower right, correlation −0.92: colder years make larger holes. Right: the same area changes against the change in mean geomagnetic activity (Ap); the points form a shapeless cloud, correlation +0.05. 2026 is marked: the hole grew by 7.0 million square kilometres while the temperature fell 1.8 K and Ap fell 4.1. −15 −10 −5 0 +5 +10 +15 −10 −5 0 +5 +10 Change in the 50 hPa minimum temperature, K r = −0.92 1991: area −0.6 million km², temperature +0.0 1992: area +2.5 million km², temperature +0.7 1993: area −0.9 million km², temperature −2.0 1994: area +0.4 million km², temperature −0.0 1995: area +0.1 million km², temperature +0.1 1996: area +1.1 million km², temperature +0.4 1997: area −0.9 million km², temperature −1.1 1998: area +3.9 million km², temperature −1.9 1999: area −1.6 million km², temperature +0.8 2000: area −0.3 million km², temperature +0.2 2001: area +1.9 million km², temperature +1.1 2002: area −14.5 million km², temperature +6.4 2002 2003: area +13.7 million km², temperature −5.9 2003 2004: area −5.9 million km², temperature +3.0 2005: area +4.0 million km², temperature −3.3 2006: area +2.6 million km², temperature −0.0 2007: area −4.4 million km², temperature +1.0 2008: area +3.7 million km², temperature +0.6 2009: area −2.6 million km², temperature −0.9 2010: area −3.3 million km², temperature +2.2 2011: area +5.1 million km², temperature −3.0 2012: area −6.4 million km², temperature +3.6 2013: area +2.7 million km², temperature −1.8 2014: area +0.1 million km², temperature +0.3 2015: area +5.0 million km², temperature −0.8 2016: area −5.2 million km², temperature +0.7 2017: area −3.0 million km², temperature +3.0 2018: area +5.4 million km², temperature −3.7 2019: area −14.6 million km², temperature +8.0 2019 2020: area +15.6 million km², temperature −9.2 2020 2021: area +0.3 million km², temperature −0.3 2022: area −0.8 million km², temperature +1.7 2023: area +0.2 million km², temperature −0.7 2024: area −3.0 million km², temperature +0.9 2025: area −1.9 million km², temperature +1.9 2026: area +7.0 million km², temperature −1.8 2026 −15 −10 −5 0 +5 +10 +15 −10 −5 0 +5 +10 Change in geomagnetic activity, mean Ap r = +0.05 1991: area −0.6 million km², Ap +9.6 1992: area +2.5 million km², Ap −6.9 1993: area −0.9 million km², Ap −5.1 1994: area +0.4 million km², Ap −0.8 1995: area +0.1 million km², Ap −0.2 1996: area +1.1 million km², Ap −1.3 1997: area −0.9 million km², Ap −2.3 1998: area +3.9 million km², Ap +5.6 1999: area −1.6 million km², Ap +2.0 2000: area −0.3 million km², Ap +1.2 2001: area +1.9 million km², Ap −4.2 2002: area −14.5 million km², Ap +2.6 2003: area +13.7 million km², Ap +2.0 2004: area −5.9 million km², Ap −9.5 2005: area +4.0 million km², Ap +6.6 2006: area +2.6 million km², Ap −7.2 2007: area −4.4 million km², Ap −1.3 2008: area +3.7 million km², Ap −1.2 2009: area −2.6 million km², Ap −2.0 2010: area −3.3 million km², Ap +2.5 2011: area +5.1 million km², Ap +3.0 2012: area −6.4 million km², Ap −0.1 2013: area +2.7 million km², Ap −1.7 2014: area +0.1 million km², Ap +0.9 2015: area +5.0 million km², Ap +8.0 2016: area −5.2 million km², Ap −4.2 2017: area −3.0 million km², Ap +1.5 2018: area +5.4 million km², Ap −4.4 2019: area −14.6 million km², Ap −1.3 2020: area +15.6 million km², Ap −1.4 2021: area +0.3 million km², Ap +0.4 2022: area −0.8 million km², Ap +4.6 2023: area +0.2 million km², Ap −1.5 2024: area −3.0 million km², Ap +8.3 2025: area −1.9 million km², Ap −5.1 2026: area +7.0 million km², Ap −4.1 2026 Change in the hole’s mean area, million km²
Each dot is the change from one year to the next in the hole’s mean area for September 7 to October 13 in CAMS’s analysis, against the change in the mean 50 hPa minimum temperature south of 60°S for September 1 to October 13, also from CAMS [7], and against the change in the mean daily Ap index for August 1 to October 13 [16]; 36 changes, 1991 to 2026. The warm years 2002 and 2019 made the two smallest holes of the period. Contains modified Copernicus Atmosphere Monitoring Service information 2026 and modified Copernicus Climate Change Service information 2026; neither the European Commission nor ECMWF is responsible for any use that may be made of the Copernicus information or data it contains [10]. Ap: GFZ, CC BY 4.0.

Mean Ap, August 1 to October 13

Our analysis

9.2

In 2026, the 12th-lowest of the 37 years since 1990; it was 18.3 in 2024 and 13.2 in 2025 [16], years with smaller holes: NASA’s means for September 7–October 13 were 19.6 and 18.7 million km², against 24.9 in 2026 [6].

Where particles destroy ozone

30–80 km

Four storms of 2024–2026 briefly removed up to 30–80% of the ozone at about 60–80 km [22]; electrons act at 60–80 km [19 Abstract]; a winter loss of 10–15% descends from 45 to 30 km and is gone by September [18 Abstract]. The hole is made lower: on October 10, 2020, ozone over the South Pole was “essentially zero over the altitude region from 14 to 20 km”, and the hole’s low values come from “nearly complete removal of ozone in the lower stratosphere” [14 p. 2].

The field, in a year

Our analysis

≈0.05%

The dipole weakens about 5% a century [31 Abstract]; from 2025 to 2026 the hole’s mean for September 7–October 13 in NASA’s record grew by a third, from 18.7 to 24.9 million km² [6].

What particles do to ozone, in the research

Energetic particle precipitation does destroy ozone, high up and near the poles. Electron events “strongly affect ozone at 60–80 km, leading to extremely large (up to 90%) short-term ozone depletion” [19 Abstract]. In the Antarctic winter, particle-made low ozone appears “at about 45 km in late June and descend[s] later to 30 km, before disappearing by September”, a stratospheric depletion of “about 10–15% on a monthly basis” [18 Abstract]. The great proton storms of October–November 2003 left “ozone depletions of 5–8% in the southern polar upper stratosphere lasting days beyond the events” [20 Abstract]. Lower down, in the Antarctic spring, the nitrogen oxides that particles make can lock up chlorine: Gordon and colleagues found “consistent ozone increases with elevated EPP during years with an easterly phase of the quasi-biennial oscillation”, and evidence of “EPP-NOx modulating Clx-driven ozone loss” [23 Abstract]. They add that as chlorine declines, “catalytic ozone destruction by EPP-NOx will likely become a major contributor to Antarctic ozone loss” [23 Abstract]. Over the poles the field already lets particles in: the field lines there “extend vertically downwards, allowing the particles to spiral down the field lines” [26].

Pollution and volcanoes, both down

The halogens are falling: chlorine and bromine from ozone-depleting substances “have continued to decline” [13]. NOAA’s Antarctic index has come 28% of the way from its peak back to the 1980 level (ODGI-A 71.6 in 2024), which it expects the halogens to reach “sometime around 2070”; the halogen load itself is about 14% below its 2001–2002 peak, 3,583 against 4,151 parts per trillion of equivalent effective stratospheric chlorine in NOAA’s table (our arithmetic) [15]. The 2022 Hunga Tonga–Hunga Ha’apai eruption’s water vapour was kept out of that year’s vortex [27 Abstract]; in 2023 its direct chemical effect on the hole “was minor (less than 4 DU)”, because the cold vortex dries the air each winter [28 Abstract], or “small and maximized at the vortex edge (10 DU in column)”, with the excess water leaving on a time scale of about four years [29 Abstract]. A model projected more, “enhancing the Antarctic ozone hole by 25–30 DU in 2023”, with responses that “diminish after 2024” [30 Abstract]. In September 2026 the WMO ranked the previous year’s hole “fourth to fifth weakest since severe depletion patterns were observed in 1992” [12], with the same weakening field.

Our analysis

“Weak field explains ozone hole”: False

Three things would have to hold for the explanation to work, and none does. First, the hole’s size would have to follow particle input from one year to the next. It follows the cold instead: across 1990–2026, year-to-year changes in the hole’s mean area for September 7–October 13 track the 50 hPa minimum temperature with r = −0.92, and the change in geomagnetic activity with r = +0.05, no detectable relation [7][16]. The cold accounts for about half of this year’s jump: a drop of 1.8 K predicts about 3.3 of the 7.0 million km². The rest is within the usual year-to-year scatter, though no hole since 1990 has been as large for its temperature, and it does not follow geomagnetic activity; one minimum temperature is only a rough index of the vortex [7][16]. In 2026 the window in which the hole forms was quieter than in 2024 and 2025, mean Ap 9.2 against 18.3 and 13.2, years that made smaller holes [16][6]. It held three days with a daily Ap of 30 or more, August 8 (30), October 4 (38) and October 10 (52), the last two days before the mid-October records; but storms act first at 60–80 km, and their winter effect in the stratosphere descends over months [19 Abstract][18 Abstract]. Second, the particles would have to destroy ozone where the hole is. They act at 30–80 km and the winter loss is gone by September [18 Abstract][19 Abstract][22], while the hole is made lower, by “nearly complete removal of ozone in the lower stratosphere”: at 14–20 km over the South Pole on October 10, 2020, the altitudes where “The lowest winter temperatures and highest abundances of reactive chlorine (ClO) occur” [14 p. 2]. There, in years with an easterly quasi-biennial oscillation, particle-made nitrogen oxides have curbed chlorine-driven loss [23 Abstract]. Third, the field would have to change on the hole’s time scale. It weakens by about 0.05% a year [31 Abstract], while the hole grew by a third from 2025 to 2026 and swung from among the weakest since 1992 to among the largest in a year [12][6]. The mechanism is not imaginary: models of fields far weaker than today’s, such as during a reversal, find particle-driven ozone losses that “significantly increase” and can “exceed the values of ozone hole situations after large SPEs” [24 Abstract][25 Abstract], and particles may matter more as chlorine falls [23 Abstract]. None of it explains a large hole in 2026 under an axial dipole 9% weaker than in 1840 [31 Abstract]. “Pollution and volcanic contributions are down” is right; what the episode left out is the weather.

Cold explains most year-to-year changes in the hole’s size and about half of 2026’s jump; geomagnetic activity shows no detectable link to either.

Code and numbers

Run them

Both need only Python 3. For ozone_checks.py, download the five files listed at the top of the script into one folder and run python3 -I ozone_checks.py FOLDER; the agencies revise recent values, so later downloads can differ slightly from the numbers here, read on October 26 [7][6][16]. python3 -I distances.py needs no download.

4 · The Sun

Quiet, as the episode said; the 2016–2017 lull is in the record, a 2027 uptick is in no forecast

“the sun continues its silent streak. I remember this brief break between peak sunspots and the declining phase uptick last cycle. It was 2026 into early 2017. Thought the sun was done. But the same Ryer forcing comes back next year like it did back then. But for now, there isn’t much doing.” And: “not anything at all happening with the sunspots, especially with the only umbrelles we see heading to the far side of the sun” [1 00:00:28–00:00:48][1 00:00:51–00:00:59].

In the Space Weather Prediction Center’s words

Times are UTC; the episode went up at 11:34 on October 24.

“Solar activity continued at low levels. The largest flare was a C3.3/Sf at 23/0542 UTC from Region 4560 (N13W58, Eai/beta-gamma). Decay was observed in the intermediate spots of AR4560 over the past 24 hours.”

Forecast Discussion, October 24, 0030 UTC, the last before the episode [41]

“Solar activity continued at low levels with the largest flare being a C1.3 at 23/1907 UTC from the only active region with sunspots on disk: Region 4560 (N13W65, Eao/beta-gamma).”

Forecast Discussion, October 24, 1230 UTC [41]

“Solar activity has been at very low levels for the past 24 hours. There are currently 1 numbered sunspot regions on the disk.”

Report and Forecast, October 24, 2200 UTC [42]

Largest flare, 24 hours before the upload

Our analysis

C1.3

At 19:07 UTC on October 23, the largest of three small C-class flares in those 24 hours; October 23 as a whole had eight, the largest a C3.3 at 05:42, and the last M-class flare was an M1.6 on October 19 [43][42].

Sunspot number, October 24

25

The daily value in the Center’s summary, down from 51 on October 23 [42]; SILSO’s estimate was 25, then 19 on October 25 [45].

Spotted regions

1

Region 4560, “the only remaining spotted region on the disk”, “mostly quiet as it rotated towards the W limb” (Forecast Discussion, October 25, 0030 UTC) [41].

The 2016–2017 lull beside 2026, in monthly sunspot numbers

Month201620172026
January57.026.1115.0
February56.426.477.4
March54.117.786.6
April37.932.379.0
May51.518.9101.8
June20.519.293.3
July32.417.878.1
August50.232.676.0
September44.643.760.3
October33.413.2—
November21.45.7—
December18.58.2—

SILSO’s monthly mean total sunspot number, version 2.0 [45]. Cycle 24 peaked in April 2014 at a smoothed 116.4, cycle 25 in October 2024 at 160.9. Spotless days: 27 in 2016, the first on June 3, 96 in 2017, none in 2026 through September 30 [45]. September 2017 is the month of the X9.3 flare, “the largest to date for the nearly concluded solar cycle 24” [46 Abstract].

Our analysis

“The Sun is quiet, spots leaving”: True

In the Center’s vocabulary the Sun went from “low” to “very low” levels over the two days, with one spotted region left on the disk and that one rotating toward the west limb and onto the far side [41][42]. Small C-class flares continued: eight on October 23, the largest a C3.3, and three in the 24 hours before the upload, the largest a C1.3; none was of M class after October 19 [43]. The daily sunspot number fell to 25 [42][45]. The history is right too, once “2026” is read as 2016: monthly numbers sank to 18–26 by early 2017, the first spotless days came in June 2016, and the September 2017 burst produced cycle 24’s largest flare [45][46 Abstract]. Today’s Sun is busier than in 2016–2017, with no spotless day yet in 2026 [45].

Our analysis

“Ryer forcing returns next year”: Unsupported

We could not identify “Ryer forcing”, as we could not identify “a Ryer instigation” on our review of October 23; if it means a Rieger-type periodicity, the 154-day quasi-period in flare occurrence reported in 1984 [47], the episode does not say so. NOAA’s cycle prediction holds the monthly sunspot number near 88 to the end of 2026 and lowers it through 2027 to 61 by December, with no rise [44]. The analogy does not fix a date either: the September 2017 burst came 41 months after cycle 24’s maximum, and the same delay after cycle 25’s maximum in October 2024 points to about March 2028 (our arithmetic) [45]. Large flares do come in declining phases, as in 2005 and 2017 [46 Abstract], but nothing published forecasts one next year.

5 · Vanuatu

A 7.2 that the Pacific warning center revised to 6.9 within the hour, far offshore and without a tsunami

“Vanuatu stole the seismic show last night. Initial 7.2 downgraded to 6.9.” Then, after a joke about dating-app photos: “Even with the downgrade, it’s 24-hour uptick blows everything else away. No tsunamis and Vanuatu can handle bigger quakes.” [1 00:00:59–00:01:05][1 00:01:13–00:01:21]

USGS, event us6000u4j2

M 6.9

Moment magnitude, reviewed, at 02:31:28 UTC on October 24, 13:31 in Vanuatu, 197 km west-north-west of Port-Olry on Espiritu Santo, 10 km deep (a depth the operator assigned), from 57 stations [48].

From the threat message to “passed”

Our analysis

51 min

The Pacific Tsunami Warning Center’s first message, at 02:37, warned of “HAZARDOUS TSUNAMI WAVES” within 300 km; its final message, at 03:28, said the threat “HAS NOW PASSED” [51].

PAGER

Green

“There is a low likelihood of casualties and damage”: 53,173 people exposed to intensity IV shaking and 125 to V, the strongest where people live [48].

The first hour, in the warning centres’ messages

Time, UTCSourceWhat it said
02:31:28The earthquakeOrigin time, 14.06°S 165.55°E [48]
02:37Pacific Tsunami Warning Center, threat message 1“MAGNITUDE 7.2”; “HAZARDOUS TSUNAMI WAVES ARE POSSIBLE FOR COASTS LOCATED WITHIN 300 KM OF THE EARTHQUAKE EPICENTER”, along Vanuatu’s coasts; first waves forecast at Espiritu Santo at 03:01 [51]
02:39National Tsunami Warning Center, statement 1“Magnitude 7.2”; for Alaska, British Columbia and the US West Coast, “Based on earthquake information and historic tsunami records, the earthquake is not expected to generate a tsunami” [52]
02:39Pacific center, statement for Hawaii“A TSUNAMI THREAT EXISTS FOR PARTS OF THE PACIFIC LOCATED CLOSER TO THE EARTHQUAKE”; none for Hawaii; magnitude “7.2 MOMENT” [51]
02:40The National center’s rapid preliminary origin, at the USGSMwp 7.2 from 10 stations, 33 km from the final epicentre (our analysis) [48]
02:45USGS, reviewed originMww 6.9, from 57 stations [48]
03:28Pacific center, threat message 2, final“MAGNITUDE 6.9”; “NOTE REVISED MAGNITUDE”; “BASED ON ALL AVAILABLE DATA... THE TSUNAMI THREAT FROM THIS EARTHQUAKE HAS NOW PASSED”; “MINOR SEA LEVEL FLUCTUATIONS MAY OCCUR IN SOME COASTAL AREAS NEAR THE EARTHQUAKE OVER THE NEXT FEW HOURS” [51]
11:34The episode goes up9.1 hours after the quake (our analysis) [1]

The two warning centres’ messages are quoted from the Iowa Environmental Mesonet’s archive [51] and from tsunami.gov [52]; they report no observed waves. Vanuatu’s Meteorology and Geo-Hazards Department lists only its latest automatic solutions, which no longer reached back to October 24 when we read it [54].

The USGS on Vanuatu’s bigger earthquakes

On the M 7.3 of December 17, 2024, 24 km west-north-west of Port Vila and 54 km deep: “At least 14 people killed, including several by a landslide, 265 injured and 1,092 displaced; 731 homes damaged or destroyed”, with intensity IX at most and a tsunami of 25 cm at Port Vila; and in the century before it, “24 earthquakes of magnitude 7 or larger within 250 km of this event” [50]. For this one, PAGER notes that “the population in this region resides in structures that are highly vulnerable to earthquake shaking, though some resistant structures exist” [48]. In the USGS catalogue we count 76 earthquakes of magnitude 7 or more named for Vanuatu since 1900, 23 since 2000 and 5 since 2022, the largest an M 8.1 in 1920 (our analysis) [49].

Our analysis

“Vanuatu M7.2 downgraded to 6.9”: True

“Initial 7.2 downgraded to 6.9” is exactly what happened: the 7.2 was the warning centres’ rapid preliminary estimate from ten stations, sent six to eight minutes after the shock, and the reviewed moment magnitude, 6.9, replaced it 14 minutes after the shock [51][52][48]; such revisions are routine, which the dating-app joke [1 00:01:05–00:01:13] leaves out. “Its 24-hour uptick blows everything else away” holds: of 21 earthquakes of magnitude 5 or more worldwide in the 24 hours before the upload, 17 belonged to this sequence, and an 18th, 195 km deep and 199 km from the mainshock, was a separate event near Espiritu Santo; the M 6.9 alone released about four-fifths of the 21 quakes’ energy (our estimate) [49]. “No tsunamis” matches the final message [51]. The next clause, “Vanuatu can handle bigger quakes”, which the panel leaves out, is partly true: big ones are routine there, and this one, about 200 km offshore with light shaking on land, harmed no one in the records we read [48]; but the M 7.3 of 2024 killed at least 14 [50].

A catalogue trap: the Pacific center’s statements on a deep earthquake in the Bonin Islands, issued at 02:24–02:26 UTC on October 25, gave its origin as “0209 UTC OCT 24”, so the USGS catalogue lists two “M 7.0” origins on October 24 [51][53]; the USGS (M 6.5, 492 km deep) and the National center put the quake at 02:09 UTC on October 25, after the episode [53][52].

6 · The storms

Champi missed Taiwan by about 40 km and hurt it anyway, then faded before Japan; the Bay of Bengal storm landed as forecast

“Weather is up next and good news. The super typhoon missed Taiwan last night. Indian Ocean storm is cranking. That first one will bend back and strike southern Japan tomorrow. While it looks like the Indian Ocean storm is going to pick a brutal location to make landfall on the Indian subcontinent. Eyes open. That is going to be bad if this forecast plays out.” [1 00:01:21–00:01:40]

Champi, October 23 to 26, in UTC

Our analysis
  • The storm in the records
  • Taiwan’s land warning
  • The episode goes up
Positions, pressures and ten-minute winds from the Japan Meteorological Agency’s analyses as listed by Digital Typhoon [57] and its bulletin of 21:45 JST on October 25 [56]; one-minute winds and the forecast from the Joint Typhoon Warning Center [55]; Taiwan’s warnings and toll from CNA [61][62]. Distances are ours, by the haversine formula [57].

Taiwan, in the emergency centre’s figures

CNA on the evening of October 24: “Tropical Storm Champi left one person dead, injured 616 and caused some damage to infrastructure before moving away from Taiwan on Saturday afternoon”; “more than 340,000 households lost power during the storm”; of 9,027 damage reports, 3,287 were fallen trees; “Yilan County recorded 650.0 millimeters of accumulated rainfall from Thursday through 4 p.m. Saturday, the highest in the country”, and 4,960 people had been evacuated [62]. The death came on Thursday, October 22, “in Liuqiu Township, Pingtung County, where a man was swept away by storm-driven waves on a beach” [60]. “Late Friday night, Sandiao Cape in New Taipei, Taiwan’s easternmost point, recorded a gust exceeding 221 kph” [61]. On Friday morning Taiwan’s weather agency gave the storm maximum sustained winds of 155 km/h [60]. In the Hong Kong Observatory’s classification, by ten-minute mean winds, a super typhoon has winds of “185 km/h or above” [58]; the Japan Meteorological Agency’s ten-minute winds for Champi were 115 knots, about 213 km/h, on October 20–21, but 80–85 knots, about 148–157 km/h, as it passed Taiwan [57].

What the forecasters said about Japan

Weathernews, which took the storm’s position and strength from the Japan Meteorological Agency’s bulletin of 09:45 JST on October 23, wrote that morning: “The typhoon is forecast to gradually weaken as it tracks west of Iriomote Island through the night of October 23. It is then expected to turn eastward over the East China Sea, weakening further before transitioning into an extratropical cyclone” [59]. CNA, on October 24: “With the storm currently on a course toward southern Japan” [62]. The agency’s analysis at 21:00 JST on October 25 classed it “温帯低気圧” [extratropical low], at 31.0°N 126.0°E [56]. The Joint Typhoon Warning Center’s final warning, at 21:00 UTC on October 25: “THE 251144Z ASCAT DATA REVEALED THE SYSTEM HAS OPENED INTO A TROUGH”; “ELEVATED SURFACE WINDS ARE POSSIBLE AS THE TROUGH PASSES OVER KYUSHU” [55].

Our analysis

“Champi missed Taiwan, hits Japan”: Misleading

Each half has something true in it. Champi made no landfall in Taiwan; its centre passed about 38 km off Sandiao Cape late on October 23 [57], as the forecasters had expected [59]. And it did bend back toward Japan [62]. But “good news … the super typhoon missed Taiwan” leaves the impression that the island was spared, and it was not: one dead, 616 injured, a third of a million households without power and 650 mm of rain in Yilan [62][60], from a storm that had stopped being a super typhoon a day and a half before [57][58]. Its centre had already passed 30 km from Yonaguni, in Japan’s southernmost islands, on October 23, before the episode [57]. The strike forecast for “tomorrow”, October 25, did not come: that day Champi was a weakening tropical storm, then an extratropical low, 300–650 km from Kagoshima [57][56][55], and only its remnant trough was forecast to reach Kyushu, on October 26, with winds of 35 knots [55].

The Bay of Bengal storm

In the India Meteorological Department’s words

Two and a half hours before the episode, at 14:30 IST on October 24, the storm was a “Well Marked Low Pressure Area over Eastcentral and adjoining Southeast Bay of Bengal”, with a “Pre-Cyclone Watch for Andhra Pradesh and adjoining South Odisha coasts”; it was expected to become a cyclonic storm “by 26th October morning” and to “reach over Westcentral Bay of Bengal near Andhra Pradesh coast around morning of 28th October, 2026” [63 p. 1], with “Major damage to thatched houses/ huts” among the impacts expected on that coast on October 27–28 [63 p. 3]. It named no stronger category and no crossing point. By 03:00 UTC on October 25 it was a depression, and that day’s forecast, issued at 12:00 UTC, is the first we found with a very severe cyclonic storm crossing the coast between Bapatla and Visakhapatnam, around 06:00 UTC on October 28 [64 p. 1]. At 02:30 IST on October 26, a deep depression, BOB 08/2026, under a “Cyclone Alert … Orange Message” [65 p. 1], it was “very likely to cross Andhra Pradesh & Yanam coasts between Bapatla and Visakhapatnam as a very severe cyclonic storm with the maximum sustained wind speed of 115-125 kmph gusting to 140 kmph by forenoon of 28th October, 2026” [65 p. 2].

Our analysis

A forecast that held

“Cranking” was early: on October 24 the Joint Typhoon Warning Center rated the disturbance at 25 knots, and it issued its first warning, on Tropical Cyclone 02B, at 03:00 UTC on October 26, with landfall expected “SOUTHEAST OF VISAKHAPATNAM AROUND TAU 48” and rapid-intensification aids ranging up to 105 knots [66]. Before the episode, the India Meteorological Department expected a cyclonic storm by October 26, still strengthening as it neared the Andhra Pradesh coast around the morning of October 28 [63 p. 1]; from October 25, after the episode, it expected a very severe cyclonic storm to cross the coast between Bapatla and Visakhapatnam by the forenoon of October 28 [64 p. 1][65 p. 2]. The episode does not show its source. The forecast held: the storm, named Arnab, crossed the Andhra Pradesh & Yanam coasts south-west of Yanam, about 20 km east-northeast of Narsapur, between 02:00 and 08:00 IST on October 28, 20:30 UTC on the 27th to 02:30 UTC, as a very severe cyclonic storm with sustained winds of 125–135 km/h gusting to 150 [67 p. 1][68 p. 1], and by 11:30 IST on October 29 it had weakened to a depression over central Chhattisgarh, 110 km south of Raipur [69 p. 1]. That outcome was added on October 29; it is not rated here.

7 · Limits

Where this check could be wrong

What would change these verdicts

  1. 01

    A study attributing part of the 2026 hole’s area to particle precipitation or to the field’s decline, or an ozone loss in the lower stratosphere in September–October 2026 tied to a solar proton event [23 Abstract][24 Abstract].

  2. 02

    Revised CAMS or NASA files with more record days for the date, or a typhoon-strength landfall in Japan on October 25 in the agencies’ best tracks [7][57].

  3. 03

    Reports of a tsunami or of damage in Vanuatu from the October 24 earthquake, or a NOAA cycle update that puts a rise in sunspot number in 2027 [48][44].

The Bay of Bengal storm’s landfall, on October 28, is recorded in section 6 and the change log; it does not change the verdicts above.

Sources

Bracketed numbers refer to this list. For the episode [1], times are positions in the video (hours:minutes:seconds). For the papers, Abstract or Significance names the part we used, § a section and p. a page; for Q10 [14] and the India Meteorological Department’s products [63]–[65] and [67]–[69], p. gives a page of the PDF. The times of NOAA’s, the USGS’s, the warning centres’ and the Japan Meteorological Agency’s products are UTC unless marked; Taiwan is UTC+8, Japan UTC+9, Vanuatu UTC+11 and India UTC+5:30. We read the sources on the evening of October 25, 2026, US time, which was October 26 in UTC; “read October 26” in the list means UTC. The three sources on the storm’s landfall [67]–[69] were read on October 29, 2026, in UTC. Quotations from Japanese sources are in the original, with our translation in brackets. Links checked October 25, 2026.

  1. 1

    SpaceWeatherNews (S0) (October 24, 2026). Earth Field Chaos, Solar Storm Deaths | S0 News Oct.24.2026. YouTube video, https://www.youtube.com/watch?v=e526hqAnTwg, published 11:34 UTC, with its description as read on October 26, 2026. Quoted from the automatically generated transcript; times are positions in the video.

  2. 2

    Belkin, V. (2026). Heliodemography: Global Mortality Rate and Geomagnetic Ap Index (1950–2023): R = 0.72. SSRN preprint, not peer reviewed; the episode’s link. Crossref lists a CC BY 4.0 licence for it. Abstract only, as for our check of the preprint.

  3. 3

    Serrano, M., Campuzano, S. A., and Pavón-Carrasco, F. J. (2026). Persistent regional patterns of secular acceleration at the core–mantle boundary and their relation to geomagnetic jerks over the past 125 years. Physics of the Earth and Planetary Interiors 381, 107615. CC BY 4.0; the episode’s link. Checked in our check of the paper, which could not read the paper and rests on its abstract, from an index copy and paraphrased, and on a re-run of the public field models CHAOS-8.5 and COV-OBS.x2; cited here for its title and record.

  4. 4

    Havenbound LLC. Havenbound — A Safety Report for Any Coordinate on Earth (home page) and Terms of Service (effective September 13, 2026), §§2–3. Read October 26, 2026, for the promotion only; the product is not rated.

  5. 5

    NASA Ozone Watch. 2026 Antarctic OMPS and MERRA-2 Ozone: data description, page last updated October 25, 2026, 20:59 UTC. Read October 26, 2026.

  6. 6

    NASA Ozone Watch. Daily files for 2026 with their 1979–2025 climatology: ozone-hole area (data to October 23), minimum ozone, mass deficit, and 50 hPa minimum temperature, 50–90°S (“Data starting 2026-08-01 are from GEOS FP”); and the mean area for September 7 to October 13 of each year since 1979. Read October 26, 2026. Credit: NASA Ozone Watch.

  7. 7

    Copernicus Atmosphere Monitoring Service. Ozone monitoring data: Southern Hemisphere ozone-hole area, ozone column minimum, ozone mass deficit and minimum 50 hPa temperature south of 60°S, every six hours from January 1, 1979, to October 24, 2026, 18:00 UTC (data CAMS/C3S). Read October 26, 2026. Records, ranks, percentiles and correlations are our own calculations from these files.

  8. 8

    Copernicus Atmosphere Monitoring Service. Monitoring of the ozone layer, its monitoring page: “The ozone hole area is calculated as the area with ozone values below 220 DU south of 60S.” Read October 26, 2026.

  9. 9

    Copernicus Atmosphere Monitoring Service (2026). Copernicus: Ozone hole reaches notable threshold early again as area increases above average in 2026. Press release, Bonn, September 16, 2026; CAMS’s news and press pages list nothing newer on the ozone hole as of October 26.

  10. 10

    ECMWF. Licence to use Copernicus Atmosphere Monitoring Service products, as read for our review of the 2026 ozone hole.

  11. 11

    Japan Meteorological Agency. 南極オゾンホールの状況（2026年） [The state of the Antarctic ozone hole, 2026], updated September 29, 2026. In Japanese; translations in brackets are ours.

  12. 12

    World Meteorological Organization (2026). WMO Ozone and UV Bulletin No. 4 – September 2026, publication page (key messages, on the 2025 season); the full report could not be reached.

  13. 13

    World Meteorological Organization (2022). Executive Summary. Scientific Assessment of Ozone Depletion: 2022, GAW Report No. 278, WMO, Geneva. As read for our review of the 2026 ozone hole.

  14. 14

    NOAA Chemical Sciences Laboratory (2022). Q10: How severe is the depletion of the Antarctic ozone layer? Twenty Questions and Answers About the Ozone Layer: 2022 Update, part of the WMO Scientific Assessment of Ozone Depletion: 2022; p. gives a page of the PDF. Read October 26, 2026.

  15. 15

    NOAA Global Monitoring Laboratory (2024). The NOAA Ozone Depleting Gas Index: Guiding Recovery of the Ozone Layer, fall 2024 update. Read October 26, 2026.

  16. 16

    GFZ Helmholtz Centre for Geosciences. Geomagnetic Kp index: daily Ap values since 1932, in the file Kp_ap_Ap_SN_F107_since_1932.txt, to October 25, 2026. CC BY 4.0. Read October 26, 2026.

  17. 17

    NOAA Space Weather Prediction Center. Solar Proton Events Affecting the Earth Environment, 1976–Present, hosted by NOAA’s National Centers for Environmental Information; the file was last modified on August 3, 2026, by the directory listing, and its last event began on July 30, 2026. The text version in the Center’s archive (ftp.swpc.noaa.gov, pub/indices/SPE.txt) ends in November 2025. Read October 26, 2026.

  18. 18

    Damiani, A., Funke, B., López Puertas, M., Santee, M. L., Cordero, R. R., and Watanabe, S. (2016). Energetic particle precipitation: A major driver of the ozone budget in the Antarctic upper stratosphere. Geophysical Research Letters 43(7), 3554–3562. Abstract only.

  19. 19

    Andersson, M. E., Verronen, P. T., Rodger, C. J., Clilverd, M. A., and Seppälä, A. (2014). Missing driver in the Sun–Earth connection from energetic electron precipitation impacts mesospheric ozone. Nature Communications 5, 5197. CC BY 4.0. Abstract only.

  20. 20

    Jackman, C. H., DeLand, M. T., Labow, G. J., Fleming, E. L., Weisenstein, D. K., Ko, M. K. W., Sinnhuber, M., and Russell, J. M. (2005). Neutral atmospheric influences of the solar proton events in October–November 2003. Journal of Geophysical Research: Space Physics 110(A9), A09S27. Abstract only.

  21. 21

    Mironova, I., Doronin, G., and Mironov, A. (2026). Ozone depletion in January 2026: impact of energetic particle precipitation. Frontiers in Astronomy and Space Sciences 13, 1924724. CC BY 4.0. Abstract only, from its Crossref record, read October 26, 2026.

  22. 22

    Jia, J., Orsolini, Y., Szeląg, M. E., et al. (2026). Comparative analysis of middle atmospheric ozone response to four large geomagnetic storms of Solar Cycle 25. EGUsphere preprint, under review for Atmospheric Chemistry and Physics. CC BY 4.0. As read for our check of the preprint.

  23. 23

    Gordon, E. M., Seppälä, A., Funke, B., Tamminen, J., and Walker, K. A. (2021). Observational evidence of energetic particle precipitation NOx (EPP-NOx) interaction with chlorine curbing Antarctic ozone loss. Atmospheric Chemistry and Physics 21(4), 2819–2836. CC BY 4.0. Abstract only.

  24. 24

    Winkler, H., Sinnhuber, M., Notholt, J., Kallenrode, M.-B., Steinhilber, F., Vogt, J., Zieger, B., Glassmeier, K.-H., and Stadelmann, A. (2008). Modeling impacts of geomagnetic field variations on middle atmospheric ozone responses to solar proton events on long timescales. Journal of Geophysical Research: Atmospheres 113(D2). Abstract only.

  25. 25

    Sinnhuber, M., Burrows, J. P., Chipperfield, M. P., Jackman, C. H., Kallenrode, M.-B., Künzi, K. F., and Quack, M. (2003). A model study of the impact of magnetic field structure on atmospheric composition during solar proton events. Geophysical Research Letters 30(15). Abstract only.

  26. 26

    NOAA National Weather Service. Types of space weather storms, section on solar radiation storms. As read for our check of storm ozone loss.

  27. 27

    Manney, G. L., Santee, M. L., Lambert, A., et al. (2023). Siege in the Southern Stratosphere: Hunga Tonga-Hunga Ha’apai Water Vapor Excluded From the 2022 Antarctic Polar Vortex. Geophysical Research Letters 50(14), e2023GL103855. CC BY-NC 4.0. Abstract only.

  28. 28

    Wohltmann, I., Santee, M. L., Manney, G. L., and Millán, L. F. (2024). The Chemical Effect of Increased Water Vapor From the Hunga Tonga-Hunga Ha’apai Eruption on the Antarctic Ozone Hole. Geophysical Research Letters 51(4), e2023GL106980. CC BY-NC-ND 4.0. Abstract only.

  29. 29

    Zhou, X., Dhomse, S. S., Feng, W., et al. (2024). Antarctic Vortex Dehydration in 2023 as a Substantial Removal Pathway for Hunga Tonga-Hunga Ha’apai Water Vapor. Geophysical Research Letters 51(8), e2023GL107630. CC BY-NC-ND 4.0. Abstract only.

  30. 30

    Fleming, E. L., Newman, P. A., Liang, Q., and Oman, L. D. (2024). Stratospheric Temperature and Ozone Impacts of the Hunga Tonga-Hunga Ha’apai Water Vapor Injection. Journal of Geophysical Research: Atmospheres 129(1), e2023JD039298. Abstract only.

  31. 31

    Finlay, C. C., Aubert, J., and Gillet, N. (2016). Gyre-driven decay of the Earth’s magnetic dipole. Nature Communications 7, 10422. CC BY 4.0. Abstract, as read in its Crossref record on October 26, 2026, and for our review of the October 22 episode.

  32. 32

    NOAA National Centers for Environmental Information and British Geological Survey (2025). State of the Geomagnetic Field, December 2025. As read for our review of the October 20 episode; p. gives a page of the PDF.

  33. 33

    Beggan, C. D., Kloss, C., Amblard, P., et al. (2026). International geomagnetic reference field: the fourteenth generation. Earth, Planets and Space 78, 127. CC BY 4.0. As read for our review of the October 20 episode.

  34. 34

    Brown, M., Korte, M., Holme, R., Wardinski, I., and Gunnarson, S. (2018). Earth’s magnetic field is probably not reversing. Proceedings of the National Academy of Sciences 115(20), 5111–5116. Title, and the Significance statement as read in its Crossref record on October 26, 2026.

  35. 35

    Plotnick, R. E. (1980). Relationship between biological extinctions and geomagnetic reversals. Geology 8(12), 578–581. Abstract only, as read for our review of the October 22 episode.

  36. 36

    Glassmeier, K.-H., and Vogt, J. (2010). Magnetic Polarity Transitions and Biospheric Effects. Space Science Reviews 155(1–4), 387–410. Abstract only, as read for our review of the October 22 episode.

  37. 37

    Picin, A., Benazzi, S., Blasco, R., et al. (2021). Comment on “A global environmental crisis 42,000 years ago”. Science 374(6570), eabi8330. Author manuscript. As read for our timeline review.

  38. 38

    Hawks, J. (2021). Comment on “A global environmental crisis 42,000 years ago”. Science 374(6570), eabh1878. As read for our timeline review.

  39. 39

    Cooper, A., Turney, C. S. M., Palmer, J., et al. (2021). Response to Comment on “A global environmental crisis 42,000 years ago”, a reply to Hawks. Science 374(6570), eabh3655.

  40. 40

    Singer, B. S., Jicha, B. R., Mochizuki, N., and Coe, R. S. (2019). Synchronizing volcanic, sedimentary, and ice core records of Earth’s last magnetic polarity reversal. Science Advances 5(8), eaaw4621. CC BY-NC 4.0.

  41. 41

    NOAA Space Weather Prediction Center (October 23–25, 2026). Forecast Discussion, the issues of October 23 (1230 UTC), October 24 (0030 and 1230 UTC) and October 25 (0030 and 1230 UTC), from the Center’s archive (ftp.swpc.noaa.gov, pub/forecasts/discussion/). Read October 26, 2026.

  42. 42

    NOAA Space Weather Prediction Center and US Air Force (October 23–25, 2026). Solar and Geophysical Activity Summary nos. 297–298 (data for October 23 and 24), Report and Forecast of Solar and Geophysical Activity nos. 296–297 (October 23 and 24, 2200 UTC), Solar Region Summaries nos. 296–298 and the Edited Events for October 23–24, from the archive (pub/forecasts/SGAS/, RSGA/, SRS/ and pub/indices/events/).

  43. 43

    NOAA Space Weather Prediction Center. GOES-18 X-ray flare list, last 7 days, October 19–25, 2026. Read October 26, 2026, 03:41 UTC.

  44. 44

    NOAA Space Weather Prediction Center. Solar Cycle Progression: predicted sunspot number, unchanged since our reading of October 24. Read October 26, 2026.

  45. 45

    SILSO, Royal Observatory of Belgium. Sunspot number, version 2.0: monthly means, 13-month smoothed means and daily values (provisional for 2026), and the estimated international sunspot number for October 2026. Read October 26, 2026.

  46. 46

    Redmon, R. J., Seaton, D. B., Steenburgh, R., He, J., and Rodriguez, J. V. (2018). September 2017’s Geoeffective Space Weather and Impacts to Caribbean Radio Communications During Hurricane Response. Space Weather 16(9), 1190–1201. CC BY-NC-ND 4.0. Abstract only.

  47. 47

    Rieger, E., Share, G. H., Forrest, D. J., Kanbach, G., Reppin, C., and Chupp, E. L. (1984). A 154-day periodicity in the occurrence of hard solar flares? Nature 312(5995), 623–625. Cited for the term; not read.

  48. 48

    US Geological Survey, Earthquake Hazards Program. M 6.9 - 197 km WNW of Port-Olry, Vanuatu, event us6000u4j2, October 24, 2026, 02:31:28 UTC: the reviewed origin and the National Tsunami Warning Center’s rapid preliminary origin it superseded, PAGER (alerts, comments, exposures, cities and historical earthquakes) and ShakeMap. Read October 26, 2026.

  49. 49

    US Geological Survey. ComCat searches through the FDSN event service: earthquakes of magnitude 5 and above worldwide, October 23, 11:34 UTC, to October 24, 11:35 UTC; magnitude 5.5 and above near Vanuatu, October 1–26, 2026; and magnitude 7 and above in 10–22°S, 164–172°E since 1900. Read October 26, 2026; counts are our own.

  50. 50

    US Geological Survey, Earthquake Hazards Program. M 7.3 - 24 km WNW of Port-Vila, Vanuatu, event us7000nzf3, December 17, 2024: impact summary and tectonic summary. Read October 26, 2026.

  51. 51

    NWS Pacific Tsunami Warning Center (October 24 and 25, 2026). Tsunami Threat Messages no. 1 (WEPA40 PHEB, 0237 UTC) and no. 2, final (0328 UTC), for Vanuatu; Tsunami Information Statements for Hawaii (0239 UTC October 24; 0226 UTC October 25) and the Pacific statement on the Bonin Islands quake (0224 UTC October 25). From the Iowa Environmental Mesonet’s text-product archive, because tsunami.gov’s copies refused scripted reading. Read October 26, 2026.

  52. 52

    NWS National Tsunami Warning Center (October 23 and 24, 2026, Pacific time). Tsunami Information Statement Number 1 (WEAK53 PAAQ 240239, October 23, 7:39 PM PDT), for Alaska, British Columbia, Washington, Oregon and California, on the Vanuatu earthquake; and Tsunami Information Statement Number 1 on the Bonin Islands earthquake (WEAK53 PAAQ 250226, October 24, 7:26 PM PDT), from the Iowa Environmental Mesonet’s archive. Read October 26, 2026.

  53. 53

    US Geological Survey, Earthquake Hazards Program. M 6.5 - Bonin Islands, Japan region, event us6000u4r9, October 25, 2026, 02:09:23 UTC, 492 km deep, and the two Pacific Tsunami Warning Center origins in the catalogue dated October 24 (pt26297050, pt26298050). Read October 26, 2026.

  54. 54

    Vanuatu Meteorology and Geo-Hazards Department. Earthquakes: the last 25 automatic solutions in the region, as listed on October 26, 2026, which no longer reached back to October 24.

  55. 55

    Joint Typhoon Warning Center. Tropical Storm 29W (Champi) Warning Nr 046, the final warning, 252100Z October 2026, with its track file of six-hourly positions and intensities since October 11 (wp2926.tcw), and the Significant Tropical Weather Advisory for the western Pacific of 252100Z. Read October 26, 2026.

  56. 56

    Japan Meteorological Agency. Typhoon information for Typhoon No. 30 (Champi), analysis of 21:00 JST, October 25, 2026 (issued 21:45 JST): category “温帯低気圧” [extratropical low], 31.0°N 126.0°E, East China Sea, 1000 hPa. Read October 26, 2026.

  57. 57

    Kitamoto, A., National Institute of Informatics. Digital Typhoon: Typhoon 202630 (CHAMPI), detailed track information, the Japan Meteorological Agency’s three-hourly analyses of position, pressure, wind and class, October 14–25, 2026, and the general information page. Read October 26, 2026.

  58. 58

    Hong Kong Observatory. Classification of Tropical Cyclones, by the maximum ten-minute mean wind near the centre, following the World Meteorological Organization’s recommendation. Read October 26, 2026.

  59. 59

    Weathernews (October 23, 2026, 10:30 JST). Typhoon CHAMPI: Strong Wind, Crosswind, and Rain Risks at Taiwan Airports, which cites the Japan Meteorological Agency’s bulletin of 09:45 JST for the storm’s position and strength; the forecast sentences quoted are Weathernews’s own.

  60. 60

    Focus Taiwan (CNA) (October 23, 2026, 12:32 PM). Typhoon Champi causes 1 death, injures 94.

  61. 61

    Focus Taiwan (CNA) (October 24, 2026, 11:29 AM). Land warning for Typhoon Champi expected to be lifted Saturday evening: CWA.

  62. 62

    Focus Taiwan (CNA) (October 24, 2026, 8:17 PM). Champi leaves 1 dead, 616 injured as storm moves away from Taiwan, citing Taiwan’s Central Emergency Operation Center.

  63. 63

    India Meteorological Department (October 24, 2026, 1430 hours IST). Press Release-2: Well Marked Low Pressure Area over Eastcentral and adjoining Southeast Bay of Bengal [Pre-Cyclone Watch for Andhra Pradesh and adjoining South Odisha coasts]. p. gives a page of the PDF.

  64. 64

    India Meteorological Department, Cyclone Warning Division (October 25, 2026, 1200 UTC). Tropical Cyclone Forecast Programme Report.

  65. 65

    India Meteorological Department (October 26, 2026, 0230 hours IST). National Bulletin No. 7 (BOB/08/2026): Deep Depression over Eastcentral and adjoining Southeast Bay of Bengal [Cyclone Alert for Andhra Pradesh and adjoining South Odisha coasts—Orange Message].

  66. 66

    Joint Typhoon Warning Center. Tropical Cyclone 02B (Two) Warning Nr 001, 260300Z October 2026, its prognostic reasoning and track file, and the Significant Tropical Weather Advisory for the Indian Ocean of 260300Z. Read October 26, 2026.

  67. 67

    India Meteorological Department (October 28, 2026, 1430 hours IST). Press Release-6: Severe Cyclonic Storm “Arnab” over Coastal Andhra Pradesh & Yanam and adjoining Westcentral Bay of Bengal [Red Message], which reports the landfall. Read October 29, 2026.

  68. 68

    India Meteorological Department, RSMC New Delhi (October 28, 2026). Tropical Cyclone Advisory No. 18 for the WMO/ESCAP Panel’s member countries, issued 0700 UTC, based on 0300 UTC, which gives the landfall in UTC. Read October 29, 2026.

  69. 69

    India Meteorological Department (October 29, 2026, 1430 hours IST). National Bulletin No. 33 (BOB/08/2026): (A) Depression (Remnant of Very Severe Cyclonic Storm “Arnab”) over Central parts of Chhattisgarh and adjoining Odisha and (B) Low Pressure Area over South Andaman Sea and neighborhood, based on 1130 hours IST; the latest national bulletin when read, at 15:43 UTC on October 29, 2026.

Method

How this was checked

On October 25, 2026 (October 26 in UTC), we checked every statement, number and quotation on this page against its source: the episode’s automatically generated transcript, quoted verbatim with its timestamps, and the video’s description [1]; the two links’ records and the preprint’s abstract, with our checks of them, and the promotion’s pages [2]–[4]; for the ozone hole, NASA Ozone Watch’s description and daily and seasonal files, CAMS’s six-hourly analyses since 1979, its monitoring page, September release and licence, the Japan Meteorological Agency’s and the WMO’s 2026 pages, the 2022 Assessment and its Twenty Questions, NOAA’s halogen index, GFZ’s Ap index, the Space Weather Prediction Center’s list of solar proton events, and thirteen papers and pages on particles, the Hunga Tonga water vapour and the polar field lines [5]–[30]; for the frame, ten sources on today’s field and the extinction record, reused from our reviews of the October 20 and 22 episodes, with Finlay’s and Brown’s records re-read [31]–[40]; for the Sun, the Space Weather Prediction Center’s discussions, summaries, reports, region lists, events, flare list and cycle prediction for October 23–25, SILSO’s sunspot numbers and two papers [41]–[47]; for Vanuatu, the USGS’s records and searches, the two tsunami warning centres’ messages and Vanuatu’s geohazards page [48]–[54]; and for the storms, the Joint Typhoon Warning Center’s and the Japan Meteorological Agency’s products, Digital Typhoon’s track list, the Hong Kong Observatory’s classification, Weathernews, three CNA reports and the India Meteorological Department’s bulletins [55]–[66]. About 150 items in all. The two papers are checked on their own pages; nothing here re-rates them.

The verdicts and the items tagged “Our analysis” are our own assessment, built on the cited sources: the counts of record days, ranks and percentiles in CAMS’s and NASA’s files; the correlations between year-to-year changes in the hole’s area and in the 50 hPa temperature or the Ap index, the share of 2026’s jump that the temperature accounts for, and the Ap means and storm days; the halogen load’s fall from its peak, from NOAA’s table; the field’s change in a year beside the hole’s; the distances from Champi’s track to Taiwan, Yonaguni and Kyushu and from the Vanuatu epicentre to the islands, by the haversine formula; the time gaps between the quake, the warning centres’ messages and the episode; the count of earthquakes in the 24 hours before the upload, with the sequence picked by distance from the mainshock, and their shares of the energy; the count of Vanuatu’s magnitude-7 earthquakes since 1900; and the 41-month arithmetic of the solar-cycle analogy. Both charts are drawn from the agencies’ files, with the Copernicus notice and GFZ’s licence.

Re-run here: every ozone count and correlation, with the hosted ozone_checks.py, and every distance and time gap, with distances.py. Not read in full here: the two papers, quoted from their records and our checks of them (the article’s own page rests on its abstract, from an index copy, and on a re-run of the field models CHAOS-8.5 and COV-OBS.x2); the papers on particles, on Hunga Tonga and on the extinction record, read in their abstracts or as read for our earlier pages; the WMO’s 2026 Bulletin, whose full report could not be reached; and the Japan Meteorological Agency’s forecast at the moment of the episode, which it no longer serves. Not checked: solar proton events in August–October 2026, which the Space Weather Prediction Center’s list does not yet cover, and conditions on Kyushu on October 26. The Taiwan toll and the agencies’ latest ozone values are preliminary; the Bay of Bengal storm’s landfall, on October 28, was added on October 29 from the India Meteorological Department’s press release, advisory and latest bulletin [67]–[69].

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.

The verdicts on this page

False
The evidence contradicts the claim.
Misleading
Contains something true but leaves a false impression, usually by leaving out key context.
Overstated
Has some basis but claims much more than the evidence supports.
True
The evidence supports the claim.

All ten verdicts, and how we check

Corrections

Corrections and changes

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.

Change log

  1. Bay of Bengal storm outcome added. The India Meteorological Department reported that the storm, named Arnab, crossed the Andhra Pradesh and Yanam coasts near Narsapur, south-west of Yanam, between 20:30 UTC on October 27 and 02:30 UTC on October 28 (02:00–08:00 IST) as a very severe cyclonic storm, with winds of 125–135 km/h gusting to 150, and was a depression over central Chhattisgarh by October 29; section 6, its heading and the note in section 7 now say so (sources 67–69), and no verdict changes.
  2. Published on factchecked.science after a full check against the sources. Built from the episode’s transcript and description, the two links’ records and our checks of them, CAMS’s and NASA’s ozone files and statements, GFZ’s Ap index, the Space Weather Prediction Center’s and SILSO’s products, the USGS’s records and the tsunami warning centres’ messages, the Joint Typhoon Warning Center’s and the Japan Meteorological Agency’s products, CNA’s reports from Taiwan and the India Meteorological Department’s bulletins, all read on October 25, 2026 (October 26 in UTC).

Suggest a correction

Spotted an error, or know a better source? Tell us here. Submissions aren’t published.