Solar Flare Alert Subscription; Waterfall of HF Black-Out

TL;DR: Waterfall screenshot of HF black-out caused by solar flare. SMS/email alert received six minutes after onset of black-out, potentially useful for SOTA activations.

A few months ago, I subscribed to alerts of various space-weather conditions offered by the NOAA Space Weather Prediction Center. One of the alert triggers I enabled is for “X-ray Events,” which are the most immediate effect of a solar flare and cause short-term HF black-outs on the sun-lit side of Earth.

I receive the alerts by email and also by text message (SMS). The latter is accomplished by subscribing with an email address comprised of my phone carrier’s email-to-text gateway (in my case as a T-Mobile subscriber: <ten_digit_mobile_number>@tmomail.net).

The alerts seem to be timely. I receive them some minutes before the X-ray event/flare/black-out is indicated on the websites where I view current space weather.

Still, I was wondering, were I to be operating / listening on the air at the time of a flare, what the time delta would be between observing the onset of HF black-out conditions (i.e., hearing signals diminish) and receiving the alert. Would it be prompt enough to be useful information during a 30-minute summit activation in which signals disappear for an unknown reason? If signals degraded due to a flare, I might wait it out on summit; on the other hand, if due to the onset of a severe geomag storm, I likely wouldn’t.

To get an idea of the black-out-to-alert latency, I’ve been running my SDR during daytime so as to display the 20 m band on a waterfall with a slow rate of descent, spanning about 25 minutes along the vertical axis.

On receiving the alert for the moderate, M6.7 flare that occurred on 2024 Mar 18 at 1916 UTC, I took a screenshot and captured my first clear, graphical indication of an HF black-out (perhaps “brown-out” would be more fitting for this one):

I would judge the fading to become apparent on the waterfall image at about 3:17 PM local time, or 1917 UTC. Not every signal outright disappears. But there’s a clear drop in signals over on the SSB side of the band, the morass of FT8 signals becomes weaker (more blue and less yellow/red), and an apparent CW contest thins out quite a bit.

The alert for this flare was received at 3:23 PM local time or 1923 UTC, both in my email inbox and on my phone’s SMS app. That’s a latency of six minutes, so not exactly real-time, but still quick enough to be useful information when operating, on summit or otherwise.

The text of the alert says:

Issue Time: 2024 Mar 18 1918 UTC 
ALERT: X-Ray Flux exceeded M5 
Threshold Reached: 2024 Mar 18 1916 UTC 
NOAA Scale: R2 - Moderate

So the time at which the flux reached NOAA’s threshold to qualify as an “event” seems to correspond pretty well, in this case, with when the fade-out first becomes apparent on the waterfall.

I’m going to continue running the slow waterfall as much as practical during daytime. I’d like to see if the alert latency varies. It would also be interesting to capture the black-out from a solid X-class flare.

73,
Matt
W4GO

17 Likes

I have the spaceweather app.
I’m not sure of the latency of the alerts but I have noticed that I get an R1, R2 blackout alert 5-10 minutes before the X-ray flare notification.

If I open the app when I get the radio blackout alert, the chart showing the flare “size” is shown as increasing.

My theory is that the X-ray alert is only sent once the flare has peaked and the scale is known. The radio blackout is one of the early signs of the flare arriving so that alert arrives first.

3 Likes

Thanks, I’ll see if I can evaluate black-out-to-alert latency for that app as well.

The NOAA subscription service also sends a summary of the X-ray event, as you said after the extent of it is known, giving the magnitude and start/max/end times. This seems to arrive generally some 20 to 30 minutes after the initial alert message I referred to in the OP.

Here’s the summary message for the flare of 18 MAR:

SUMMARY: X-ray Event exceeded M5 
Begin Time: 2024 Mar 18 1902 UTC 
Maximum Time: 2024 Mar 18 1919 UTC 
End Time: 2024 Mar 18 1928 UTC 
X-ray Class: M6.7 
Location: S12E62 
NOAA Scale: R2 - Moderate

Interesting that the “begin time” precedes the obvious fading of signals on the waterfall by quite a bit. As I said, it’s the threshold time in the alert that seems to correspond with the start of the fade-out, for this event anyway.

2 Likes

Interesting stuff. I think I will see what is needed to subscribe and handle the warnings so they get displayed on the SOTA cluster.

10 Likes

For comparison purposes, I installed the SpaceWeatherLive app for Android by “Parsec vzw” yesterday and enabled push alerts for X-ray / flare events. (I assumed this was the app you were referring to, Pete. There was another space weather app with fewer downloads and lower ratings.)

A few hours later, there was an X1.1 flare reaching threshold at 9:16 PM my local time (23 MAR 0116 UTC). The newly installed app did indeed push alerts, first for a radio blackout and later for the flare itself. I think the first of the alerts from the app came about five minutes after the SMS/email alerts from NOAA.

I’ll try to note the alert delivery times more definitely for the next flare (before the app retroactively replaces the time the alert appeared with the time of the physical event! [edit: I think actually it was the OS, not the app, that removed the precise alert time at the turn of the calendar day.]).

This flare did not produce any apparent black-out on my waterfall of the 20 m band. Not surprising as it occurred about two hours after my sunset. I suspect most of the signals on that band would have been coming from locations in or near darkness. Had I instead been capturing 15 m or 10 m with the beam turned to the west, I might have seen some indication of a black-out.

2 Likes

Yes, that’s the one I have.
Interesting to note that the alerts come through later than the SMS ones.

2 Likes

Regards to all … interesting… :information_source: 73s

1 Like

@W4GO Dear Matt, you can see X flares developing in real-time and their effect, D-layer absorption, in my free SWX smartphone app.

As soon as Xflux increases from C9.9 to M1.0, you have a minor X flare. So, in a sense, you are even pre-warned. :wink:

2 Likes

Waterfall capturing black-out on the 20 m band caused by this afternoon’s X1.1 flare:

Event notification time-line:

4:42 PM local = 20:42 UTC: fading of signals, increase in noise floor first apparent on waterfall.

4:49 PM = 20:49 UTC: received NOAA alert of X-ray flux >= M5 on phone (via email-to-SMS gateway).

4:51 PM = 20:51 UTC: received NOAA alert of same on PC (via IMAP email client).

5:20 PM = 21:20 UTC: received NOAA summary on PC via email.

The contents say the M5 threshold was reached at 20:41 UTC, maximum flux occurred at 20:56 UTC, and the event ended at 21:01 UTC. One can see signals generally starting to come back at about the latter time on the waterfall.

Interestingly, the SpaceWeatherLive app didn’t produce any alert until I opened the app itself with the intention of confirming alerts were enabled for X-ray events (they were). I have since changed the phone’s OS settings to exclude this app from power saving features. Other apps don’t seem to require this measure to alert properly, though. Next flare, I’ll post an update on whether this worked.

3 Likes

Interesting. I know GOES satellites publish Xflux every minute. Currently, my app probes this data every 5 minutes, but I could easily change that to every minute. That would represent a download of 25.7 kB every minute; so about 1.5 MB an hour.

I never realised this data could be that time critical on a difficult mountain top. I will change my code accordingly and let you know when I did.

1 Like

@W4GO Never believe a programmer; multiply their time estimates by at least a factor 3! :wink: In the end, it took me 3 hours to test and debug, but 1 minute Xflux updates work now and I published the new app version.

However, you will need to reload the web page with CtrlShiftR or clear the browser cache on your smartphone to see the updated version of the app. Otherwise, it takes one week for your browser to realise there is an updated version. You will know you have the latest version when you read the line Updated every minute for Xflux; 5 & 30 minutes for other values.

Please, also note that there is a 3 minute lag between the GOES satellite observation and NOAA publishing the Xflux value in JSON format. To be quicker, one would need to decode the proprietary GOES feed. I am not going to do that, but Python code contributions are always welcome. Anyhow, I assume my app is still quicker than any SMS and e-mail alerts. (Please, test.)

Moreover, their are occasionally small gaps in the satellite observations when the Earth or Moon is obscuring the satellite’s view to the Sun. This tend to happen around noon and midnight and especially around the equinoxes.

Matt, let us know how this works for you.

2 Likes

Thanks for the heads-up - I’ve installed that here too. Looks good!

2 Likes

Thanks Matt. I really like the idea of knowing there is an X Flare before I deploy an antenna at the summit…without adding an extra step to setting up. I looked for aa email - SMS gateway on EE - (a UK carrier) and failed - there isn’t one provided . After a bit more faffing around found iOS has a VIP e-mail that allows an e-mail from a VIP ( Very Important Phenomenon?? ). to appear as a message on a locked phone. It works and will allow iOS users to get an alert that is obvious on a locked phone. Might be useful. 73. Paul

2 Likes

I didn’t see that option on the NOAA site, just email. Is that something you are doing locally or is there a NOAA SMS option?

1 Like

Just email. I’m using my mobile provider’s free email-to-SMS service, per the OP. YMMV.

accomplished by subscribing with an email address comprised of my phone carrier’s email-to-text gateway (in my case as a T-Mobile subscriber: <ten_digit_mobile_number>@tmomail.net).

1 Like

Thanks, will take a look.

I never realised this data could be that time critical on a difficult mountain top. I will change my code accordingly and let you know when I did.

I wouldn’t say it’s critical, but potentially helpful in some situations. Experienced operators are used to dealing with degraded HF propagation without the benefit of mobile internet service, e.g., be prepared with multiple bands/modes on HF, handheld for VHF. As my waterfall captures show, the term “black-out” is something of an exaggeration for ordinary flares.

I think an app/page that one has to actively check would be of most use when one already has reason to suspect a propagation event might be occurring. Signals disappear while operating; is it my gear, X-ray event, geomag event, or something else? Check current space weather to find out.

On the other hand, a push alert is useful in situations that are not conducive to interacting with ones phone. For example, you’ve just reached the summit and are about to start setting up the HF station. You hear an alert on your phone indicating that an X-ray event is in progress, so you decide to do VHF first, have a snack, etc. to let the black-out run its course.

3 Likes

I’m late posting an update, but this seems to have gotten the alerts working (and I haven’t noticed any effect on my phone’s power usage):

An M5.4 flare occurred on 11 April, reaching maximum flux at 17:06 UTC.

The SpaceWeatherLive app raised an alert for this event correctly – that is, without me needing to interact with my device to “wake up” the app – at 17:09 UTC.

The email and SMS-via-email-gateway alerts from NOAA for the same event were received at 17:11 UTC.

The NOAA alert text indicated the flux threshold was reached at 17:05 UTC.

My SDR’s slow waterfall capture of the event, below, shows visible signal fade-out beginning at about 17:04 UTC.

(I don’t have an explanation for the distinctive noise signature appearing on the waterfall at about the same time; it doesn’t look like the increase in noise floor that I would expect from a flare, but I haven’t been able to find a local source for it either.)

To recap a rambling thread:

The SpaceWeatherLive app for android appears to provide timely push alerts of solar flares and the consequential HF “black-out.” In order to get the alerts, you may need to delve into your phone’s settings to exclude the app from power-saving measures.

If you want alerts of solar flares by email, or by SMS provided you have access to an email-to-SMS gateway service, you can subscribe to the NOAA service linked up-thread.

Based on my observations so far, the latency between discernible HF fade-out and alert arrival appears to be about five minutes for the app and about six to seven minutes for the NOAA service. This is probably subject to network-related variability.

2 Likes

I didn’t know NOAA offered this. Thanks for sharing!

For anyone using other carriers, the 2 major carriers in the USA can receive texts like mentioned in original post.

Just use the email address ##########@vtext.com if you have Verizon or ##########@txt.att.net if you have AT&T.

This site has a list of what email address to use if using many other carriers.

2 Likes

Caught the big one. X8.8 flare today near local noon. Noise burst and dramatic black-out from 9.5 MHz to 14.5 MHz captured on the SDR:

It’s interesting to attempt to correlate observed onset with the GOES X-Ray flux plot from NOAA:

16:46 UTC: X-ray flux (GOES -16/18 long) crosses M-class threshold
16:47 UTC: black-out begins on waterfall (host time is set accurately via NTP)
16:48 UTC: flux crosses X-class threshold
16:51 UTC: flux reaches peak magnitude

Reliability of the alerts from NOAA and the SpaceWeatherLive app has been generally good over the last few weeks. Occasionally, alerts from one or the other service arrive late or not at all. Having both has been useful redundancy.

Today, I needed triple redundancy. Luckily I happened to glance at the waterfall about 10 minutes into the black-out. Otherwise I likely would have missed getting the screen grab.

SpaceWeatherLive only raised the alert for the flare when I woke my phone up – a recurrence of the earlier problem which excluding the app from power saving in the phone settings has mitigated, though not totally resolved.

Worse, the NOAA alert via email-to-SMS gateway arrived about two hours after the flare and the alert destined for my normal email came about four hours late. I wonder if the sheer volume of space weather events recently has exposed scaling deficiencies in NOAA’s alert service.

It’d be interesting to hear others observations on the reliability of the alerts from either service.

2 Likes

I now use the noaa alert - but it does seem to arrive after rather than during the event - but as my phone goes ping and it stays o the screen I at leas notice it - even with summit brain. (Only used 2m on the last summit…)

Q). What are the thoughts on the progression of Cycle 25? I know that the last couple of weeks excitement has not yet fed into the data. I think it already looks bigger than Cycle 24 - will we go back to the late 50’s with a SSN of over 300 and how will all the space hardware cope?

3 Likes