The 10-11 May 2024 geomagnetic storm was the most severe storm in over 2 decades. It also had a very long duration. According to the final Kp values published by GFZ Potsdam, the Kp index was at 8- or higher, i.e. at severe or extremely severe levels, for no less than 33 consecutive hours or 11 3-hour intervals (see graph underneath). This duration is very comparable with the 13-14 March 1989 and 18-19 September 1941 storms, the trio exceeding most of the other long and severe storm durations (since 1932) by 3 to 5 3-hour intervals. On some occasions, a succession of passing coronal mass ejections (CMEs) resulted in a near-continuous level of geomagnetic storming. An example are the Halloween storms in 2003, when two CMEs distinctly separated in time by nearly a half day passed the geospace resulting in extreme geomagnetic storms on 29-30 and 30-31 October. Observers in Belgium saw the aurora on both nights, getting the impression this was one continuous storm - which it wasn't. Of course, when the CMEs are launched in quick succession and with increasing speed, the more recent CMEs may catch up with the preceding ones ("CME cannibalism") or follow each other in close time intervals (hours or less) when they pass the earth environment. Then the result is one strong, long-lasting geomagnetic storm, such as the May 2024 event when "... the fast and dense solar wind structures from (up to 6) CMEs interacted with one another, forming a composite interplanetary CME (ICME) with complex solar wind density and magnetic field structures embedded. ... " (Tulasi Ram et al. - 2024).
It can be worse... The famous Carrington storm of 2-3 September 1859 seems to have been the result of a single CME. Yet, based on worldwide aurora observations, polar lights were visible from locations below 50 degrees latitude for at least 42 hours (Green and Boardsen - 2006). From this, one can reasonably assume that Kp was at severe and extremely severe levels for at least that duration, so at least 14 3-hour intervals! Mind-boggling.
It can get even worse... A few years ago, Hisashi Hayakawa and co-workers (2017) as well as Kataoka and Iwahashi (2017) reported from historical documents in East Asia that low-latitude, i.e. below 30 degrees magnetic latitude (MLAT), auroral displays appeared in succession for almost nine nights from 10 to 19 September 1770, except for the 12th. The most equatorward sightings were from 18.8 MLAT from Dòngtínghú (Hunan, China), and from near Timor Island in the southern hemisphere (-20.6 MLAT) as observed by Joseph Banks and Sydney Parkinson on board HMS Endeavour as specialist members of Captain Cook’s crew. Hence, these observations are now known to be the earliest record of simultaneous auroral observations in both hemispheres. The latitudinal extents of the 1770 auroral events were at least comparable with those of the Carrington event.
The aurorae were extremely bright, especially on 17 September when they were as bright as the full moon according to local reports. The color was deep red, and many suspected fires in distant cities. As the aurora grew higher and brighter in the sky, people protectively started pouring water on the roofs of their houses because they thought the fire from the sky might rain onto their roofs. Others started divine dances and prayed to the Buddha, fearing the end of the world had arrived. Still others commented also on the luminous stripes within the red vapour, comparable to solar rays shining through the clouds.
There are not many solar drawings from that time (Clette et al. - 2015), with September 1770 being just one year after the maximum of moderate solar cycle 2. Though the accuracy and level of detail of Johann Caspar Staudacher’s drawings are not of the best quality, his sunspot drawings revealed an extremely large and complex sunspot group that rotated over the solar disk from 12 to 22 September 1770 (see Hayakawa et al. - 2017). This large active region looked like a (large) bunch of grapes and was even observed by naked eye in Japan around 17 September 1770. Measurements on the drawing from the day before indicates a sunspot area up to 6000 millionths of the Sun’s visible hemisphere (MH), more than twice the size of the sunspot group during the Carrington event in 1859 and comparable to the largest known sunspot group that appeared in April 1947 (see the STCE newsitems here and here).
The low-latitude aurorae were almost continuously observed for nine nights, but most likely only the aurora from 15 to 19 September were due to eruptions (and associated CMEs) from the large sunspot region. Most of the storm highlights date from 16 to 18 September. But even then, a duration of 3 to 5 days of severe to extremely severe geomagnetic storming is exceptionally long. In many geomagnetic storms, the duration is only one or two nights. The long duration of the auroral observations in 1770 indicates long-lasting geomagnetic storm activities resulting from continuous solar activity, such as multiple, consecutive CMEs most likely -in this case- originating from the same sunspot group.
