In the bleak mid-winter

Hang around aurora chasing circles for any length of time and you’ll become aware of the fact that we like our abbreviations. Whether it’s KP or ACE, EPAM or Bt, you’ll hear a lot.

But the abbreviation that’s of most importance is the Bz (the magnetic orientation of the material from the sun). You can have all the KP predictions you like – it’s the Bz that’s the key, and the one you need to keep an eye on if you’re wanting to catch sight of the aurora. If it's southwards (i.e. with negative values) then it's more likely to allow the aurora to extend further south. 

In this blog, I want to take the mid-winter's aurora display from 20th December 2015 as a case study, tracing the time line of the geomagnetic storm and watching what happened to the stats. In it, we’ll see the paramount roll the Bz played – and we’ll see how other factors of the solar wind combined together to produce this beautiful display.

Signs of possible action (Wednesday 16th December)

The origins of the 20th December aurora are found four days earlier, when our sun burst into life.  At 9.03 am the sun launched a partially earth directed CME. Good news, in and of itself. But, soon after, a second CME was launched, also partially earth directed. 

Satellite images showing the two CMEs. The blank disc in the centre is the sun - the contoured areas around it represent the CMEs emerging from the sun.

Satellite images showing the two CMEs. The blank disc in the centre is the sun - the contoured areas around it represent the CMEs emerging from the sun.

The signs were good for some aurora activity in the coming days. But, as ever, there were uncertainties and variable. Both CMEs were relatively slow (CME = 550 km/s and CME = 490 km/s). How would they interact with each other? Would CME 1 outrun the slower CME 2? Or would CME 1 slow down as it ploughed through the ambient solar wind, allowing CME 2 which was travelling in its wake time to catch up?

And what about the Bz? Would it be orientated northwards or southwards when it arrived? 

And the Bt? When two CMEs arrive together, the IMF tends to become more compressed, resulting in higher values of the Bt. If the Bz was southwards, that would tend to produce stronger geomagnetic storming when it arrived.

All these questions remained hanging. For now, it was a waiting game…

Signs of the incoming CME (Friday 19th December)

The first sign of the arrival of the CME was at the ACE satellite, a satellite about an hour’s journey time from the earth for the solar wind. It has various sensors which can tell us about the nature of the solar wind when the CME passes by. The first sign of the possible arrival of the CME is when we begin to see a rise in the low energy protons as shown on the EPAM graph. This graph showed a steady rise on the 18th December. But this still gives us no indication of when the CME will actually arrive and what it will be like when it does arrive.

The CME arrives at ACE (15.28 on Saturday 19th December)

At 15.28 on the 19th, the CME arrived at ACE. The impacts were very noticeable.

The CME arrived - and the values jumped up!

The CME arrived - and the values jumped up!

  • The speed jumped up from ambient levels of around 360 km/s to about 470 km/s;

  • The Bt jumped upwards, as expected from the double CME impact;

  • And the all important Bz? It jumped suddenly down to -10 nT.

The signs were good. It was about an hour until darkness in the UK, giving enough time for the CME to travel from ACE to the earth and stir the geomagnetic field up a bit for some action that evening. 

The Bz shifts north and doesn't spend much time down south at all (late afternoon 19th December until 02.30 20th December)

Notice how the Bz (the red line at the top) stays mostly in positive values from 17.00 hours till 02.30 hours)

Notice how the Bz (the red line at the top) stays mostly in positive values from 17.00 hours till 02.30 hours)

The Bz didn’t stay south for long though, and soon headed north. For the next eight hours or so, it fluctuated north and slight south. This agitated the  Interplanetary Magnetic Field (IMF) and this, combined with the high levels of Bt (the strength of the IMF) and very high density of the solar wind, was enough to spark of some low level action in the far north of Scotland (the only part of the British Isles not blanketed in cloud too!)   But certainly nothing close to the wonderful display we were all hoping for – and with no immediate signs of anything changing, thanks to the fact that the Bz wouldn’t settle southwards for any length of time.

The Bz shifts south - and stays there for nearly 30 hours (02.30 20th until 08.00 21st December)

The Bz shifts suddenly south.

The Bz shifts suddenly south.

Then, all of a sudden, the Bz shifted south. At around 02.30 on the 20th, it suddenly dropped down to around -17 nT. And it stayed south.

For many in the UK, we had gone to bed. But the aurora chasers in North America and New Zealand/Australia were treated to great displays. It looked like we in the UK had been unlucky with the timing and that we were about to miss out on the show.

But the Bz stayed south, throughout all the night. And into the next morning. Cue speculation amongst UK aurora chasers – could it hold throughout the day to give us a show later on when it got dark? The Bt and density both remained elevated too. The speed had slowly dropped back to just above ambient levels. But, if things remained like this, it seemed like we might be in for some aurora views from this CME event after all!

As the day progressed, the Bz indeed stayed south. In fact it was to remain south for well over a day in total, producing a geomagnetic storm of unusually long duration – some 30 hours of continuous storming. 

The scale of this graph is days - the 19th to the 21st. Notice how the Bz, once it had shifted south, stayed there for the next 30 hours or so. This is not normal and is one of the reasons for the prolonged nature of this event.

The scale of this graph is days - the 19th to the 21st. Notice how the Bz, once it had shifted south, stayed there for the next 30 hours or so. This is not normal and is one of the reasons for the prolonged nature of this event.

What was the aurora display like to see?

Once it becomes clear that a display is on the cards, I keep an eye on two sets of data.

  • I keep watching ACE data. Most aurora displays produce data that fluctuate much more than this one did. So watching ACE gives you an indication of what might happen within the next hour or so.

  • The magnetometers. I normally use the Scandanavian line magnetometers – these are devices that show the impact the geomagnetic event is having on the earth’s magnetic field. The trace the activity from the north of Scandanvia right down to the UK and are a great real time indication that things may be about to kick off. Basically, the greater the amplitude of the deviation and the further south those deviations occur, the better news it is for us in the UK.

From here on, as the ACE data didn’t change too much over the next number of hours, I’ll be focusing mostly on the story the magnetometers told.

With all the excitement and anticipation rising, we in the UK were just waiting for darkness in the hope that night two would bring more luck than night one had! The only challenge I was facing was cloud. The forecast was for scattered cloud but, as we moved towards darkness, the cloud thickened. Even the light of the nearly full moon was obscured at times. It was clear we were going to have to travel around a bit in search of clear skies. So my fellow aurora chaser, Alistair White, and I headed north, more in hope than expectation. And eventually we found some clear skies, as I will explain later.

In the meantime, for others with clearer skies, the action began just as darkness fell. For much of the duration of the display, there was a noticeable elevated green arc, hanging in the sky towards the north-east. This remained, a reasonably constant feature at more or less the same brightness levels for the entire time I was able to witness it – this corresponds to the prolonged period of negative Bz that allowed the aurora oval to migrate further south and remain here. In fact, it was noticeably elevated. As the cloud began to clear, at first I thought I was seeing the green poking above the clouds, as is often the case in Northern Ireland. However, as the cloud cleared, it became apparent that the darker skies below the arc were indeed cloud free as I could see stars appearing. 

Can you spot the deviations in the magnetometers that corresponded with the three more active periods?

The reason for the arc being elevated was due to just how far south the oval had migrated. Normally, from my latitude, I’m looking hundreds of miles to the north to see the aurora to the far north, which means the green band hugs the horizon. But, in this case, the sustained negative Bz had allowed the oval to move much further south than normal. And, as I was looking at it, I could see it and the clear skies below.

If you look at the magnetometers, you’ll see an interesting pattern that bears this out. The main deviations were towards the centre of the graph – the most northerly magnetometers were not moving as much, indicating that the aurora action was centred more towards the south. This amazing satellite image of the aurora oval from this night, shows just how far south it had indeed moved.

The dnb image from the aurora display - note how the auroral oval has moved south of Icelaand, much closer to the UK. That's one of the reasons for the elevated oval.

In addition to the constant green arc, there were at least three periods of more intense action, complete with pillars of light and movement across the sky.

The first of these was just after dark. If you look at the magnetometers, you’ll see distinct fluctuations around and after 16.00. At that stage, reports were coming in of dancing skies and pillars of light (and I, frustratingly, was still under cloud then!)

The second one was around 20.00, when the magnetometers dove dramatically. At this stage, Alistair and I were driving near Ballymoney and the skies were just beginning to clear. As I looked around, now more with expectation than hope, in one of the clear patches to east I could see clear pillars of light stretching high up into the sky – and that even with all the light pollution from the dash board and other cars all around us.

“Quick!” I shouted, “Pull over – it’s happening!” We found a side road and somewhere safe to stop and just about managed to catch the tail end of the pillars, before the display settled down into the green arc. But we had found clear skies – it was game on! 

You can just about make out a couple of pillars in the centre of the image - the tail end of the middle outburst.

Given where we had stopped, I suggested going to the Dark Hedges to see what we might catch there. We arrived at this most mystical of places and got set up. There were hints of green between gaps in the cloud. But the cloud was still an issue. There were gaps – but then they’d be covered in again. So it was back to hope over expectation. We settled down to play the waiting game.

Eventually, we were rewarded. The clouds got thinner and thinner before clearing more or less completely. The bright waxing moon shone through the trees, casting wonderful shadows and lighting the intertwining branches. But even this was not enough to wash out that glorious arc to the north east.

And so we spent the next hour or so shooting away. I’d never photographed the aurora here before, so I tried out various positions with the trees, drinking in the beauty of the scene around us.

Eventually, with a 06.00 start the next morning looming for my friend, we headed home. It’s always risky leaving an active aurora in mid-display. At any point, the aurora pillars of light may make a re-appearance. The only indication that it’s about to happen is the magnetometers. So I kept an eye on them as we travelled south. And indeed, around 23.30, it all kicked off once more as the magnetometers took another dive south. And I just about had time to get into location to capture this image from a location I’d scoped out previously.

After this, the cloud closed in for me again, and that was it for the night for me.

Why no mention of the KP?

For many budding aurora chasers, the first abbreviation they hear of is the KP. The reason for this is that most of the automated aurora alert websites and apps use it as the basis of the warnings they give. This is often combined with maps showing you with precision how far south a KP6 aurora will be visible from The only problem is that this information is not much use. Here’s why:

  • If it’s a KP forecast, then it’s purely speculative. You are much better to keep an eye on the actual data coming from ACE, and how the magnetometers reacting in real time.

  • If it’s the actual KP value for the earth you are reading, then you need to be aware of two things:

    • First, this is a planetary average, and specific lines of longitude can be below – or above – average.

    • Secondly, the actual KP values are summaries for the past three hours. So, in a typical shorted duration aurora display, by the time you realise it’s KP7, the aurora may have already passed.

That’s why this blog has been focused on the ACE data (especially Bz) and the magnetometers.

That said, we can look back in retrospect and look at how the planetary KP values reflected this long duration storm as it progressed.

The global KP values for the three days over which this geomagetic storm was active.

The global KP values for the three days over which this geomagetic storm was active.

So what do we learn from this chase?

  1. The Bz is the key. Even though the CME had hit b midy afternoon on the 19th, there was no significant aurora action for most of the UK as the Bz hadn't shifted south. It wasn't until 02.30 on the 21st that the Bz turned south - then the action kicked off.

  2. Other factors are not unimportant, however. The high levels of the Bt helped maintain the potential for a good show once the Bz shifted south.

  3. The low speed of the CME was a good thing. Although it stopped the KP values reaching the 7 or 8 some had predicted, the slow speed was one of the reasons for the long duration of the show. Also, as Tamitha Skov observed, if the speed had been higher and the KP values reached those dizzying heights, there was the potential for some negative impacts on the electricity grids (it's looking out for these possible impacts that is one of the main reasons why we even get space weather forecasts!)


Read about the Mid-Summer aurora I observed here

Find out more background info on some of the abbreviations used throughout this blog here


Thanks to James Rowley-Hill for some very helpful advice on parts of this blog and as ever the good folks at AUK for being a wonderful group of people to share the chase with!