2026-09-26 · ebungo field research · six sources · twenty-six verbatim quotes
The fall equinox arrived this week, and with it a roughly two-month window when geomagnetic storms — and the auroras they can produce — become more common.[1] This digest follows the chain: why the equinoxes boost auroras (a 1973 geometry result still doing the work), what the 94-year storm record actually says about the boost, and why the timing is good this year even though the sun is past its peak.[4][5]
Sunday, people in more than a dozen states up the East Coast and along the U.S.-Canadian border had a chance to see the northern lights.[1] The reason the odds climbed is seasonal, not coincidental: the fall equinox arrived this week, and that means Earth is entering a roughly two-month window when geomagnetic storms, and the auroras they can produce, become more common.[1]
The twice-yearly pattern is not new. An aurora season comes around the fall and spring equinox each year, and this pattern in nature — auroras increasing twice a year — has been known to scientists for more than a century.[2] The enduring explanation is the Russell-McPherron effect, from a 1973 paper by Christopher Russell and Robert McPherron of UCLA titled Semiannual variation of geomagnetic activity.[2]
The mechanism is magnetic geometry. Solar magnetic fields — carried to Earth via the solar wind — also have a north and south pole, and Russell and McPherron showed that the north-south component of the sun's magnetic field — the Bz component, as solar physicists call it — goes up and down over the year, in a way corresponding to the wobbling of Earth's axis, with the fluctuations largest during the equinoxes.[2] When the solar wind's field arrives pointing south as Earth sees it, it can connect with Earth's own field and pour energy into the magnetosphere — and that coupling is more efficient in spring and autumn than around the solstices.[4] NASA describes the geometry as especially favorable around the spring and fall equinoxes.[1]
The sun's own 11-year cycle is a separate layer. That cycle appears to have peaked in late 2024, and the sun is still relatively active.[2]
The equinox effect is real, and the honest version of the claim is that it doubles the storms, not the typical night. Against the planetary Kp record kept by GFZ Potsdam — more than 270,000 three-hour readings running from January 1932 — storm-level periods (Kp 5 or higher, roughly NOAA G1) are 1.95 times as common around the equinoxes, while the average Kp is only 1.16 times higher.[4] What changes is how often a strong night turns up, not the baseline.[4]
The stronger the storm, the bigger the equinox advantage: periods at Kp 4 or higher are only about 1.5 times as common around the equinoxes as around the solstices; at G1 the ratio is 1.95, at G2 it is 2.3, and at G3 it is 2.5 — and compared month to month, a G3-level period is almost seven times as common in September as in December.[4] Month by month in the same record, September's storm-level share is 9.44% of three-hour periods against 4.05% in December; March leads at 9.47%.[4] Night by night, counting the six hours from 21:00 to 03:00 UTC, 14.7% of September nights in the record include at least one storm-level period, against 6.7% of December nights.[4] (The same site's September guide rounds the same record slightly differently — 9.5% of September's periods against 4.1% in December, level with March as the stormiest month of the year.[3])
It is a season, not a date: the boost covers roughly six weeks either side of each equinox, and nothing special happens on the day itself.[4] This year proved it live — the autumn equinox fell at 00:05 on 23 September 2026 (Iceland time; the next one is 20 March 2027 at 20:25), and NOAA's forecast issued on 21 September expected nothing above Kp 2 through the 23rd, saying no significant transient or recurrent solar wind features were forecast.[4] The long-term statistics are real; the equinox day itself can still be completely quiet.[4]
Solar Cycle 25 peaked in October 2024, when the smoothed sunspot number reached 160.9 — earlier and far stronger than the 2019 NOAA-NASA prediction panel's forecast of 115 in July 2025.[3] An independent check on the same record comes to 160.8 — the two tallies differ at the first decimal.[5] A separate marker confirms the turn: in March 2026 the sun's magnetic poles completed their reversal, the sign that the cycle has turned.[5]
The decline is visible in the count: August 2026 averaged a sunspot number of 76 on the provisional count, about half of September 2024's 141, and NOAA's current prediction puts the smoothed figure near 90 this autumn.[3] But sunspots are not what the aurora answers to. The first eight months of 2026 averaged a planetary ap index of 12.6, above the same months of 2024 (11.1) — the year the sunspots peaked — though below 2025 (15.2).[3] That is the familiar post-maximum pattern: coronal mass ejections give way to fast solar wind streams from coronal holes, which are large and long-lived at this point in the cycle and recur roughly every 27 days as the sun rotates.[5] It is how forecasters could pencil in an active return this week: the other active run NOAA was carrying, 24-26 September, is August's storm coming back round on the sun's 27-day rotation, landing on the full moon — and with the moon up for the whole dark window, only a bright display will show through: a Kp 5 display usually can, a quiet one cannot.[3]
The honest season framing, from a source that checked eleven seasons of cloud data: better than any year between 2017 and 2023, not as good as 2024, and declining from here until around 2030.[5] Cycle 24 is the precedent — its sunspots peaked in 2014, and its most geomagnetically active year was 2015.[3]
Statistics raise the odds; weather and light still decide the night. December in Tromso is the worst month of the season, at roughly one usable night in six, and it is when prices peak — while September gives Tromso the best sky of its year and only 1.8 hours of true darkness a night, and December gives it eight hours out of eight because the sun does not rise at all.[5] Cloud and darkness run in opposite directions across the season, and moonlight can cancel the equinox bonus: a bright moon washes out exactly the faint, ordinary displays that make up most sightings.[4][5]
For a viewer at aurora latitudes the threshold is forgiving — aurora is regularly visible at Kp 2 to 3 when the sky is dark and clear, Iceland sits under the auroral oval, and in mid-September 2026 people photographed aurora on four separate nights at an estimated planetary Kp between 1.67 and 3.33.[4] Farther south the bar is higher: once the index climbs to 5 or higher, the northern tier of the U.S. gets a real shot at spotting an aurora.[1] NOAA's Space Weather Prediction Center posts the three-day outlook and aurora forecast that every guide points to.[1]
Three layers line up through September and October 2026: the equinox geometry that makes storms roughly twice as common (1.95 times, on the sharpest count), a declining solar cycle whose geomagnetic output is still beating the year of sunspot maximum, and a 27-day rhythm of coronal-hole streams that keeps returning active windows until the sky's winter darkness takes over.[4][3][5] The season runs through late March 2027; forecasters at NASA refresh the cycle estimates monthly, and the numbers they update are 13-month smoothed values.[6]
The sky is more interesting than the calendar this week: the equinox's statistical boost is quiet on a given night, the sun is past its peak but still capable, and tonight's real variables are cloud, darkness, and the moon.[4][2][3]
[1] Your odds of seeing an aurora jumped this week — The Weather Channel"The fall equinox arrived this week, and that means Earth is entering a roughly two-month window when geomagnetic storms, and the auroras they can produce, become more common.""Scientists call this seasonal effect the Russell-McPherron effect. NASA describes the geometry as especially favorable around the spring and fall equinoxes.""Once the index climbs to 5 or higher, the northern tier of the U.S. gets a real shot at spotting an aurora.""People in more than a dozen states up the East Coast and along the U.S.-Canadian border had a chance to see the northern lights Sunday."
[2] Yes, there's an aurora season, and it's here — EarthSky"Yes, there is an _aurora season_, which comes around the [fall](https://earthsky.org/astronomy-essentials/everything-you-need-to-know-september-equinox/) and [spring equinox](https://earthsky.org/astronomy-essentials/everything-you-need-to-know-vernal-or-spring-equinox/) each year. This pattern in nature – [auroras](https://earthsky.org/sun/what-causes-the-aurora-borealis-or-northern-lights/) increasing twice a year – has been known to scientists for more than a century.""That cycle appears to have peaked in late 2024, and the sun is still relatively active.""Russell and McPherron showed that the “north-south” component of the sun’s magnetic field – called the [Bz component](https://icelandatnight.is/bz-level) by solar physicists – goes up and down over the year, in a way corresponding to the wobbling of Earth’s axis."
[3] Northern Lights in Iceland in September 2026 — Aurora Iceland"Solar Cycle 25 peaked in **October 2024**, when the smoothed sunspot number reached 160.9 — earlier and far stronger than the 2019 NOAA-NASA prediction panel’s forecast of 115 in July 2025.""August 2026 averaged a sunspot number of 76 on the provisional count, about half of September 2024’s 141, and NOAA’s current prediction puts the smoothed figure near 90 this autumn.""The first eight months of 2026 averaged a planetary ap index of 12.6, above the same months of 2024 (11.1), the year the sunspots peaked, though below 2025 (15.2).""Storm-level periods — NOAA G1 or stronger — make up 9.5% of all three-hour periods in September, level with March as the stormiest month of the year, against 4.1% in December and 5.3% in June.""And the other active run NOAA is carrying, 24-26 September, is August’s storm pencilled in to come back round on the Sun’s 27-day rotation, landing on the full moon. With the moon up for the whole dark window, only a bright display will show through: a Kp 5 display usually can, a quiet one cannot."
[4] Does the Equinox Really Double Your Chances of Seeing the Northern Lights? — Aurora Iceland"The stronger the storm, the bigger the equinox advantage.** Periods at Kp 4- or higher are only about **1.5 times** as common around the equinoxes as around the solstices. At G1 the ratio is 1.95, at G2 it is 2.3, and at G3 it is 2.5. Compared month to month, a G3-level period is almost seven times as common in September as in December.""expects nothing above Kp 2 through the 23rd and says that “no significant transient or recurrent” solar wind features are forecast.""Storm-level periods (Kp 5- or higher, roughly NOAA G1) are 1.95 times as common around the equinoxes.""The average Kp is only 1.16 times higher.""Counting the six hours from 21:00 to 03:00 UTC, when Iceland’s aurora is usually at its best, 14.7% of September nights in the record include at least one storm-level period, against 6.7% of December nights.""The autumn equinox falls at 00:05 on 23 September 2026, Iceland time (the next one is 20 March 2027 at 20:25)."
[5] Aurora Season 2026-27: What the Data Says — Northern Lights Travel"Solar Cycle 25 reached maximum in October 2024, with a smoothed sunspot number of 160.8, well above the 115 the official panel had forecast. In March 2026 the sun's magnetic poles completed their reversal, which is the marker that confirms the cycle has turned.""better than any year between 2017 and 2023, not as good as 2024, and declining from here until around 2030.""December in Tromso is the worst month of the season**, at roughly one usable night in six, and it is when prices peak.""September gives Tromso the best sky of its year and **1.8 hours** of true darkness a night. December gives it eight hours out of eight, because the sun does not rise at all, and the worst sky in the dataset.""Near the equinoxes geomagnetic storms are close to twice as common as at the solstices, an effect described by Russell and McPherron in 1973.""coronal mass ejections give way to fast solar wind streams from coronal holes, which are large and long-lived at this point in the cycle and recur roughly every 27 days as the sun rotates."
[6] Solar Cycle Progression and Forecast — NASA"Each month the solar prediction is updated using historical and the latest monthâs observed solar indices to provide estimates for the balance of the current solar cycle and the next.""The forecasted solar indices represent the 13-month smoothed values consisting of a best estimate value stated as a 50-percentile value along with the 95 and 5 percentile statistical values."