Friday, 28 February 2020

Lots more Es in February, or just FT8 keeping us active?

50MHz Sporadic E activity over 12 hours on PSK Reporter 26 February 2020
The accepted wisdom is that Sporadic E is both unpredictable, and predictable.

The unpredictable aspect is that you cannot say if it will occur on any particular day, in any particular direction, or at any time of the day.

The predictable part is that it happens during the Summer and around Christmas. As to what "Summer" means, well that varies with band and latitude. Perhaps April to October in 10m, May to September on 6m, late May to late August on 4m and 3 days around the end of June on 2m. I know nothing about 222Mhz, and 432Mhz has the odd, very rare, event.

You just don't know which days, but you do know generally when to look.

We have discussed the Christmas Es season, or Winter Es as I call it, here before. December and January seem to be the best times, but we don't know when and really only 10m and 6m benefit with rare events on 4m.

Except, the predictable is becoming more unreliable in the sense that we are now finding more event "out of season". And even the unpredictable is becoming more predictable, with more events on more days during the season.

I am no expert but I think that data modes and FT8 are partly responsible for that.

There are many in the amateur world who will go into the technicalities of decibels and sensitivity for an answer to this. I am sceptical about that approach when it comes to unpredictable events. These same people think that using a 30 second meteor scatter program is better than using a 15 second one. In this certainty of theirs they are wrong, but let us leave that thorny topic for a later posting. Amateurs often choose to take the wrong path, and that is a matter for them. I find the 15 second logic easy to grasp, but then I am prefect in every way.

Returning to my thesis, I think that getting all the amateurs in the world onto one frequency, which is more or less what FT8 does, vastly improves the chance of them hearing each other. In fact, I am pretty sure that it will result in far more contacts and receptions being logged than most circuit improvement we could make to our equipment. And, dare I say, more significant than more sensitive data protocols.

The process started with RTTY and its working frequencies. Whereas SSB stations spread out and tend to miss each other, RTTY brought them nearer and allowed reception without tuning or actually paying attention all the time. This got a lot better with PSK, which showed all the stations together on one waterfall. It suddenly got a lot easier to find everyone. As an early RTTY stalwart, PSK so dramatically out performed RTTY that I never went back. I often say how dorky I find RTTY now, and how wasteful it is of resources. However, RTTY opened the door for lots of us.

From PSK I found FSK441 and JT65 great ways to work stations but once again they tended to spread out across the band. I realised what was going on when I understood the distinction between  slow modes and fast modes. WSPR is a desperately slow mode, with 2 minute transmit periods and glacially slow bit rates. But the results were stupendous. I could, and did, work so far on very low power that it would have been the equivalent of 2,000,000 km per watt. But we always knew that the station at the other end would be on the same frequency. Not only did the efficiency of the data mode play its part, but so did having hundreds of others listening for each other, all on the same frequency.

The next step from JT65 was JT9, a simply splendid narrower mode where everybody could share the waterfall. And so eventually to FT8, where the whole world can listen together and you can reasonably expect to find everybody else using the mode. Unlike the fast modes (like MSK144 and fast versions of JT9), you are presented with a waterfall of stations, and they can operate on some sort of shared basis on a single SSB-filter wide part of the spectrum.

FT8 is a slow mode. It is not terrifically efficient. You could make it more efficient by slowing its data rate down some more which would make it more sensitive, but then you are heading towards WSPR territory. FT8 is a compromise which allows us QSOs rather than just beacon operation like WSPR, at the cost of some sensitivity. But never mind the sensitivity, you can work more people because they are all there waiting for you.

Is amateur radio there for chatting about pig farming techniques (yes, I listened to that QSO), or is it here for us to find out something new? Putting everybody on a moderately efficient mode which records reception and does not need to be attended to every moment, all on the same frequency, is likely to maximise the chances of reception. Whatever the sensitivity, it raises the probability of reception taking place, and thus QSOs.

So it is that I think I am finding more Sporadic E in February. Lots more. I cannot remember the figures, but I think that FT8 is less sensitive than JT9. But that does not matter much, as far more stations use FT8, so there seems to be more Es. Ultimate sensitivity does not matter so much for Es, where the signals can be quite strong. WSPR might be more sensitive, but the signals are not steady enough to last 2 minutes. Fast modes might be even better at handling the QSB, but 15 seconds is a good test - it is rather like having an SSB QSO. Fast modes would benefit from scatter enhancement, but with FT8 we can be pretty sure we are witnessing Sporadic E (erm, does this help explain why there are so many modes???)
50MHz sporadic E activity on PSK Reporter over 15 minutes during 28 February 2020 - pleased to reach OY.
OK, I know February has 29 days this year. However, I was away in Grantown on Spey for five of those days, and anyway it did not occur to me that this was happening for the first half of February. My effective study period covered a maximum of 23 days out of a possible 28 (I am not waiting for day 29). During that time I worked DX stations on Sporadic E on 50MHz on 8 days out of a possible 23.

8 days Sporadic E out of 23 in February? That sounds more like April. I would have expected February to produce one or two Es days. I had to check - had this ever happened here in February before?

Now, there are a few issues to resolve before checking the logs. Firstly some days I only worked one station (that happens in April and September too). I have excluded any stations within tropo range though they appear on the maps.

Quite often I worked SM5EPO (in fact on five different days), but it was Per-O who was around. Often the path opened to Sweden, but recently also to the Faeroe Islands where OY1OF heard me repeatedly on several days, but no QSO resulted. I worked mostly into the normal Es distance destinations - France, Germany, Italy, Norway, etc.

I have left out of my figures any one way reports, though there have been lots of them too. I have also left out partial QSOs. These get to exchange of reports but not RRR, so they get a ? in my log and are not counted. Where the callsigns are exchanged but not the reports I do not log them at all.  So I am sure that QSOs on 8 days is a fair statement - there are no partials in that figure. There has been more activity, sometimes just a couple of received messages or a partial QSO, on about 8 more days, but I have ignored them for the figures.

As for the total number of QSOs, well the Es opening on 11 February skews the results. It is not the number of contacts which is the key measure for me, but the number of days it occurs. Still, if you want to see activity on 6m from 1 to 28 February here it is :-
FT8 50MHz contacts at GM4FVM 1 to 28 February 2020
Ploughing through the logs was a tedious business.

The comparable figure for February 2019 is zero. NIL Sporadic E contacts in February 2019. I have not been able to drag up my FT8 listening  log but I bet I was listening on many days in 2019.

Figures for other years:-
2010, 2011, 2012, 2013, All NIL
2014 - 1 QSO (SM5EPO of course)
2015, 2016, 2017, 2018 All NIL.

I was alive and working during all those February dates, and I did work people on meteor scatter or tropo. Apart from that one day in 2014, I never noticed any Es activity. I would, of course, as anybody else on tropo or meteor scatter would, hear any of those stations via Es and reply using whatever mode we were on. There were none in February except one QSO on one day.

So, if it is true (and it might not be true), why is there suddenly Sporadic E propagation now? I doubt it. I think there are more stations listening on FT8 and ready to work someone. In other words, I just noticed it, and so did enough other people to work me.

Sure, there may be more unpredictable Es this month of this year. There was the opening on 11 February 2020, which was associated with increased geomagnetic activity. Likewise, during the short openings on 27 and 28 February 2020 I noticed that the K number as measured at the GM4PMK observatory (see link in sidebar). I certainly believe that on days when Es is not likely to happen, a spike in geomagnetic activity can push it into action. But all these issues arise every year, perhaps even more often at better stages in the sunspot cycle.

I will not bother to comment on the effect of Jet Streams as it appears to be NIL.

Is this Es and not some other means of propagation? It is certainly patchy, there is a lot of one-way propagation and quick QSB. I can see "pings" of signal which looks a bit like meteor scatter but that cannot explain it as there are no major showers now, and when it does get stronger it lacks the characteristic meteor patterns.

Is FT8 just better than you think it is Jim? Maybe, but I was using it last year too.

For sure, you do get occasional reception over great distances on FT8. Those are random events and almost always just for one reception (I suspect they are caused by either cosmic ray ionisation or random large meteors). This year's events were all long enough to secure a QSO. Looks like Es to me - and everybody else who reports it as such.

So there you are. I might be wrong.  I often am wrong, which you may have noticed.

I think this may simply be a measure of FT8 activity on 6m as much as increased Es.

I cannot lose here. If Es really is getting more common so much the better. Bring it on. If it is not getting more common, thanks to Joe Taylor and FT8.

Actually, thanks to Joe Taylor for FT8 anyway. Nice one Joe.




Wednesday, 12 February 2020

Sporadic E thanks to a kind sun.

Sporadic E on 11 February, who could have predicted that ???
Before I start can I mention that the BBC's excellent "In Our Time" programme featured a talk on Solar Wind recently? The general idea of the programme is that a generalist egg-head who knows little about the specific topic (always played by Melvin Bragg) surrounds himself with three of the best academic experts on the subject, and they all try to tease out the detail for the audience in about 45 minutes. There is also a short round up at the end after the programme itself is finished..

You can find the podcast here:-

Whilst it is not about radio propagation there is a lot of information in it. The science strand of these programmes can be quite interesting, especially for downloading to a phone and keeping for waiting for trains etc.

You can also find it on the BBC Sounds app.

As many of you will know, I am very sceptical about the claims that we can predict Sporadic E using Jet Stream data. The Jet Stream, high speed winds high in the atmosphere may, I concede, play some part in the Sporadic E story. As for them being predictive - I just do not believe it. What evidence I have from here suggests that the predictions are just plain wrong.

For the past few weeks, the RSGB VHF predictions service (and the HF one too) have been predicting the likely path of Sporadic E openings using the Jet Stream predictions and they have been spectacularly inaccurate. No, wrong was the correct word for this. They just did not happen. This week they once again said that there would be lots of Jet Stream energy available, and I thought "so what?". This is interesting, but it does not help me with what day to be ready, where to point my antenna. Anyway, for the past few weeks, nothing has happened anyway.

Much more interesting for me was Solarham's prediction of disturbed geomagnetic conditions on 11 February.
3 day K-index prediction from NOAA as shown on Solarham
As usual, click to enlarge images if necessary.

Solarham (link on side bar) had been using the NOAA K-index predications to issue this warning a couple of days in advance, which meant I was ready.

A K number of 4 might suggest a weak aurora, but it was the prolonged spell of level of 3 during the day of the 11th which interested me.

It is time perhaps for me to return to stating my own theory about unusual Es events. Rather than pinning any credence on the Jet Stream I find that enhanced geomagnetic activity seems to fit in with most out of season Es events. I have a general vague term for this "enhanced geomagnetic activity short of an aurora" to describe when the K level rises. This gets me on aurora watch, but if the level never triggers a full aurora and I find myself listening to Sporadic E instead.

I have been banging on about this for years - here in 2017 ...

Thus the 11 February event looked like just the thing.

I had to go out twice during the day and so missed a lot of contacts. I also got so involved in 6m and even 4m activity that I missed a lot of QSOs on 10m, but here are the day's contacts again with labels this time:-
All contacts (28 to 70MHz) at GM4FVM on 11 February 2020

Over an operating period of 5 hours and 38 minutes I had 53 QSOs (all on FT8) into 12 DXCCs accounting for 38 squares. ODX was UW5ZO in square KO30, a distance of 1984km.

This is pretty much what I had expected. On many days of high Jet Stream energy, I look for Es and find nothing. When there is a day of enhanced geomagnetic activity I looks for Es and find a shed load. I am not saying that the Jet Stream has nothing to do with it, I am saying that Jet Stream predictions are no use in predicting what might happen, whereas solar activity often is.

Dealing with the bands, as usual with Es, action tends to start here on 10m and work upwards:-
10m (28MHz) contacts at GM4FVM on 11 February 2020
Early in the morning I was finding that the openings on 6m were very short and moving around geographically, so many failed to complete. I could get callsigns and reports, but not RRR so they failed in terms of duration being less than a minute long before moving somewhere else. On 10m they were long enough to complete. Later on in the day, when 6m and 4m became more stable I mostly sat and watched 10m as the activity rolled along. I could have made twice the number of contacts on 10m that I actually attempted, if my attention was not diverted elsewhere. It was on 10m that I worked my ODX for the day, to UW5ZO. This was early on - Es is best when the ionisation is weakest.

I devoted more attention to 6m, hoping to test out my recently installed antenna. Whilst signals were strong at times, up to +17dB, they tended to fade quickly, so more gain might extend the operating period long enough to complete the contact. So far so good. ODX was 9A2DI in JN95 (as it often is) at 1870km. I had 32 QSOs on 6m in just under 4 hours,  working 11 DXCC and 25 squares.
6m (50MHz) contacts at GM4FVM on 11 Feburary 2020
32 QSOs producing 25 squares shows how variable this opening was. Most of the time there were only two or three stations heard at once. As it moved around without much of a pattern, the stations changed location. Some of the French stations were there for a long time, as were some of the Balkan stations, whereas the Germans and Czechs stayed only briefly.

Only rarely do these events stray onto 4m. I think this is partly due to the low activity on that band, but also the trend for small pockets of territory to be covered as you get higher in frequency (which gets much more pronounced on 2m). Both these factors mean that your chances of picking out an active amateur are lower. However, I still did have some success:-
4m (70MHz) contacts at GM4FVM on 11 February 2020
Of course there are a lot fewer stations active on 70MHz. I was heard in Spain, which surprised me. I was pleased with that performance from the new antenna. 3 QSOs on 3 DXCC and 3 squares, of course. ODX was HA7WFN in JN97 at 1733km. Just to prove me wrong, he stayed on my waterfall for half an hour. Maybe the variation is not so much in which area is covered, but how wide the area of coverage. Perhaps as the area of ionisation varies this changes the stations we hear. So a station at the centre would stay on the propagation zone for longer than one at the edge. I have no firm evidence for this, I just wonder.

During Summer Es, the common type, sometimes you can just about see the effect of the Sun moving in relation to any point on the Earth (well, the Earth rotating really). The Es tends to move, as you might expect, East to West as the Sun does. People talk about "Es clouds drifting", but there are no Es clouds in that sense, and whatever you call them they don't drift anyway. The very way Es clouds work is that they are trapped layers of ionised particles, so by any definition they do not move much. During this event there was not much sign of any pattern. Probably the ionisation was popping up wherever the geomagnetic effect was strongest.

This was a good Es opening worthy of a Summer's day, when you might have expected it. That it happened in mid-February is interesting. It happened during a period of raised, but not very high, geomagnetic activity. Higher geomagnetic activity might have triggered an aurora and this posting would be different. Lower geomagnetic activity, well, that would just be another day when the weathermen would be spouting about Jet Streams, and nothing would have happened at all.

That is my theory, and I only have my blog to put it forward. Others, with the might of a national society and weekly prediction broadcast on amateur radio frequencies can put theirs forward in their way. Quite possibly we are both wrong. There could be 101 reasons why this event happened, but I know I could be wrong, I state it is just a theory, whereas others claim to have the answer.

Science moves on by rejecting old theories. It is bad science if you only look at events when they happen and ignore what is going on when they don't happen. It is bad science when you only look at what happens near you and not at what happens elsewhere. And it is bad science when you take your own speciality, however good you may be at it, and apply it's solutions to problems which have many and varied causes.

And finally...

As the Jet Stream theory enthusiasts are talking about a weather phenomenon in the North Atlantic, you would not expect to see Es activity enhanced in different parts of the world by such events.
Activity on 6m on 11 February 2020 on PSK reporter

It seems this event produced effects in the other parts of the world which use 6m, even during the night when VHF activity is usually very low. Is the Jet Stream around our UK shores responsible for all this?