Predicting auroras

Radio auroras are sporadic events occurring in the E Region in northern and mid-latitudes. They don’t happen often – indeed most radio amateurs would say that they’ve never participated in an auroral event, even though they are equipped for VHF DX operation. Unlike Sporadic E activity which happens annually and is more predictable, radio auroras are much rarer.

And yet radio auroras do have a predictability and cadence too.

I discuss elsewhere that radio auroras occur following the shock on the Earth’s geomagnetic field following arrival of a high energy solar wind. The high energy solar wind typically comes from a solar flare (or other high energy solar event) with plasma departing the Sun some time before.

There are two measures that matter and typically must be exceeded for an aurora to be triggered:

  • Kp-index, the global geomagnetic activity on a scale of 0 to 9 should indicate a Minor Storm (5) through to Extreme Storm (9).
  • Dst, disturbance storm time, the momentary level of geomagnetic disturbance must typically be greater than 200nT.

Dst can be detected on a magnetometer. A typical trace is shown below. This is a trace of the Donbas magnetometer for a 10 hour period. The build-up is obvious. But it takes a big disturbance – a twang of the magnetic field – of more than 200nT to trigger the rotating columns (see the earlier page on the Formation of Radio Auroras).

Linking Dst with alert level

The image below shows the magnetometer Dst reading at Sumburgh Head, UK. 59.8530° N, 1.2760° W just before a radio aurora. The swing in Dst is apparent, peaking at 250nT as the shockwave hits.

Radio aurora follows disturbance peak of 25nT

The global magnetic activity index, Kp, is reported from a network of ground-based magnetometers around the world. A final smoothed figure is issued every three hours by the GFZ organisation in Potsdam, Germany and is available from the Parsec organisation via the SpaceWeatherLive app. The image below shows Kp at 6 during an aurora on 12th May 2021.

Radio aurora occurs while Kp peaks at >5

The immediate disturbance level, Dst, is reported by many aurora watchers using inexpensive magnetometers, and specifically by AuroraWatch UK. Both AuroraWatch and SpaceWeatherLive apps are available from the App Store.

Both figures above are from the AuroraWatchUK web site at https://aurorawatch.lancs.ac.uk/plots/?project=awn&site=sum&date=2021-05-12 accessed on 22nd October 2021.

Solar flares of the size needed to give a Dst of >200nT to trigger an aurora occur rarely. To get a feel for how rarely, and attempt some model for prediction, one needs to look at the bigger picture.

The image below shows the relationship between the Sun’s sunspot activity, as measured by the SSN, and the geomagnetic activity, measured by Kp. SSN is the Smooth Sunspot Number and is described elsewhere on this site. It ranges from zero (at the start and end of cycle) to around 250 (at the peak mid-cycle). Actually, the last two cycles have peaked at about 170.

Assessing aurora likelihood with SSN and Kp

In past solar cycles, there’s been a correlation between peak in Kp and incidence of radio auroras. There’s also been a correlation between Kp and SSN. Speculating the nature of the current Cycle 25, we see that Kp might peak in 2023 as SSN rises.  And Kp might peak again as SSN falls away around 2027 and 2028.

Unsurprisingly, there’s also correlation between occurrence of large solar flares and increased Kp above 5. We can therefore conclude that large solar flares and other major solar events of the type needed to trigger radio auroras will therefore most likely occur in the couple of years before the sunspot maximum and again about three years after the peak.

Regarding month of year, studies of past auroral activity suggest that there’s no best month, though there has been a slight rise in incidence of radio auroras in spring.

That just leaves us interested in the time of day.

Past radio auroras have allowed the following figure to be assembled.

This shows that radio auroras follow the form of the auroral oval descending from the pole with time. Time is on the radial axis from midnight round to mid-day. This is oriented to show auroras available to UK radio amateurs with time in UTC. The circular axis shows circles of latitude in degrees. UK is at about 51˚ through 59˚ north.

Radio aurora incidence by time of day and latitude

Radio auroras start in early afternoon in northern latitudes (65˚N) and the spinning cylinders migrate southward as the afternoon progresses. Only the strongest auroras sink as far south as 45˚N. The peak UK auroral activity is in late afternoon. There’s then a second burst just after midnight, fading after 2:00am.

Anyone wanting to participate in a radio aurora event should therefore monitor space weather and specifically Kp and Dst indices. A report of a major flare could be a good suggestion that something might be about to happen in a day or so. Then you won’t get much warning and so must be ready to switch on as soon as the thresholds are exceeded. Auroral events supporting interesting VHF propagation will last from a few minutes to a few hours.

In this final image, I’ve shown how an aurora builds, appears and then dissipates as a sequence of events.

The progression of an aurora