I excitedly told my son that the X flare probability was up to 75%. He asked what it usually was at and I said one or maybe 5%. And then he said something interesting, he said so at one percent that means that the sun produces an X flare about once every hundred days of 1%. And at 5% the sun produces an X flare about one in 20 days at 5%. Thinking back, this does not seem to be true. Now we have a probability of 60 to 75% and it is shooting off multiple X flares per day, not one X flare every day and a half or so. This leads me to believe that there is somewhat of a sliding scale, and that 5% really means closer to zero and 75% really means closer to 100.  I would be interested to know what you find folks think about my son’s observation? 

Edited May 10, 20241 yr by WhereingtonEvent
Typos

Postponed my transatlantic flight from Sunday to Thursday next. Not going airborne on a long-haul until that monster is comfortably over the limb...

6 minutes ago, WhereingtonEvent said:

I excitedly told my son that the X flare probability was up to 75%. He asked what it usually was at and I said one or maybe 5%. And then he said something interesting, he said so at one percent that means that the sun produces an ex flare about once every hundred days of 1%. And at 5% the sun produces an X flare about one and 20 days at 5%. Thinking back, this does not seem to be true. Now we have a probability of 60 to 75% and it is shooting off multiple X flares per day, not one X flare every day and a half or so. This leads me to believe that there is somewhat of a sliding scale, and that 5% really means closer to zero and 75% really means closer to 100.  I I would be interested to know what you find folks think about my son’s observation? 

Interesting, shows your son is smart and probably good at math. but i dont think thats how it works.

1 hour ago, WhereingtonEvent said:

I excitedly told my son that the X flare probability was up to 75%. He asked what it usually was at and I said one or maybe 5%. And then he said something interesting, he said so at one percent that means that the sun produces an ex flare about once every hundred days of 1%. And at 5% the sun produces an X flare about one and 20 days at 5%. Thinking back, this does not seem to be true. Now we have a probability of 60 to 75% and it is shooting off multiple X flares per day, not one X flare every day and a half or so. This leads me to believe that there is somewhat of a sliding scale, and that 5% really means closer to zero and 75% really means closer to 100.  I I would be interested to know what you find folks think about my son’s observation? 

I think there's a couple of things to consider there. The first would be what exactly those percentages are based on; I'm not sure what SWPC's process for determining those are, but I don't think it's that systematic, i.e. it probably lacks definite criteria which are updated conditionally based on actual observations, because if it did then you wouldn't just see those nice and round multiples of 5% being used (aside from 1%, of course). So in a sense it's most likely just eyeballing it, although it might be fairly well qualified eyeballing. Would be interesting to know what their process is.

The second thing to consider is however that while it might not seem true to you, intuitions humans have about statistics are often fairly poor, which is part of the motivation of the development of statistics in the first place. I'm not saying you are necessarily wrong, only that it's often the case that percentages reflect the observations more than it might "feel" to someone. For the 5% number I might be inclined to agree myself that it's actually lower when that's shown, but I don't really know if that's true or not. For the 75% number something more to consider is that it's easy to say in hindsight that the region has flared a ton, but it might very well be that there is in fact now roughly a 75% chance of another X-flare occurring. Ideally you'd use a system which marks various empirical features of a flare and updates itself conditionally in hindsight based on whether the region did in fact end up flaring or not.

There does exist systems that use methods like machine learning to try to determine flaring probabilities more accurately, including the infamous DeepFlareNet (DeFN); but as you can probably tell from that site right now some care must be made when devising the algorithms, so that you don't end up with a 100% chance of X-flare and a slightly lower chance of M-flares, heh; or maybe that makes sense for them somehow, i.e. that the algorithm actually thinks it's more likely that the region will X-flare than M-flare, although I'm skeptical about that. But you could always check it out in calmer times and see if it reflects your intuitions better.

But your son's reasoning about the statistics is correct; if there's a daily given percentage of 5%, then you would indeed expect that 1/20 such days would yield the stated flare class, and so on, if those percentages were reflective of reality. But that would be overall, so it would still be a bit more complicated if you were to say, calculate the chance of a flare of a given class happening at least once during the next 20 days where the percentage is 5%; as you can probably understand, even though you'd statistically expect such a flare every 20 such days, it's not guaranteed to happen once in any given sequence of 20 such days, and it could also happen more than once during such a span of time. To calculate those probabilities you'd use a binomial distribution.

Addendum:

DeFN actually also has a network they deem "reliable" (DeFN-R), which seems more reasonable in terms of the percentages, although the X-flare probability there is fairly low now; could very well be a fairly accurate assessment for all I know though.

Edited May 10, 20241 yr by Philalethes
DeFN-R

we're having another flare

And while we wait for the first CME to arrive, another flare in progress 😏

Just now, Ester89 said:

And while we wait for the first CME to arrive, another flare in progress 😏

yees. early squad! how are you?

2 minutes ago, MinYoongi said:

we're having another flare

Just in time for breakfast 😋

2 hours ago, arjemma said:

No need to be scared, but I do understand that you might be. When it comes to the west side of the sun (right side if you are watching SDO images) there is a thing called the Parker Spiral. You can read more about the spiral (more commonly known as "Heliospheric current sheet") here: https://en.wikipedia.org/wiki/Heliospheric_current_sheet

This spiral makes it possible for plasma from the sun to reach earth even if the plasma was ejected from the west side of the sun. A good example of this was the CME launched from a X11 event on September 10 2017 where earth was hit by the solar storm even though it was launched from the west limb.

Here's a video and some picture on that event: https://www.spaceweatherlive.com/sv/arkiv/2017/09/10/xray.html

Here's an article from SWL about the event: https://www.spaceweatherlive.com/en/news/view/306/20170910-x8-2-solar-flare.html

So we have a few days to go until we can't be hit by plasma from this region. We are safe, no need to worry.

Thanks for all of information! Knowing more is always helpful. Still nervous especially with all of the hype/fear mongering out there.

Peaked at M5.97

5 minutes ago, MinYoongi said:

yees. early squad! how are you?

Yeeahh 😎

I'm fine, thanks! And you? Very excited about this whole festival of flares and cmes first time for me

1 hour ago, AlainSki said:

Postponed my transatlantic flight from Sunday to Thursday next. Not going airborne on a long-haul until that monster is comfortably over the limb...

Shame, transatlantic flights are good ways to see the aurora over Greenland. 

This flare looks impulsive, no CME

12 minutes ago, Jeffrey Dikkes said:

AR3664 : 6 CME's due to arrival and GPF warning threshold has been broken.

Still think a G5 event greater than the Halloween solar storm( 2005) is about to ring our doorsteps.

It's rather usual for the 10 MeV protons to rise above the threshold, it happens a lot. What makes you think that this will be greater than the Halloween storms?

24 minutes ago, Jeffrey Dikkes said:

Still think a G5 event greater than the Halloween solar storm( 2005) is about to ring our doorsteps.

The Halloween Solar storms occurred in 2003, not 2005. And no, there's no chance that this is going to be stronger than that.

Not sure why you're posting a picture of the current lowest level of proton event, S1. Perhaps you're trying to make it clear just how much weaker than the Halloween storms this is? During those storms we went into S4 territory, 1000 times the proton flux as we're seeing now:

4 minutes ago, Jeffrey Dikkes said:

@arjemmaI'm not a expert on this just a newbie fresh behind the ears 

I'm from the Netherlands Amsterdam just as I said a newbie but interested in space weather.

Noaa predictions : there is gonna be a teleconference pls go to noaa and read all for urselfs 

The last time that noaa issues this alarm was in 2005

 

We've already experienced 4-5 G4 storms in the last year and a half. This is the first time they've issued a G4 Watch. It's not the same as a warning which is only issued once the threshold is reached. There's nothing all that unusual about the events that are occuring.

14 minutes ago, Jeffrey Dikkes said:

Ah yes @Philalethesthanks for correction 👏 the year was indeed 2003.

That was incorrect from my part. 

Will see @Philalethes nothing to worry about, ur right in all ur outstanding comments debunking every single comment of people who are talking on this form.

Great job! 

The first Cme off AR3664 has just arrived on earth....which is much earlier than expected.

Pls do debunk me @Philalethes

But it hasn't arrived at earth. It arrived at STEREO A. 

17 minutes ago, Jeffrey Dikkes said:

Pls do debunk me @Philalethes

Most of the estimates for the first arrival were between 10T16Z and 11T00Z, with 6 hours of uncertainty on each side, so saying that this is "much earlier than expected" makes little sense. It's on the earlier end of the fully expected range, but still very much as expected.

And yeah, what Parabolic said; currently it's just arrived at SA, not yet at L1, so by that time it will be even more squarely in the expected range. Very much as expected indeed.

My first participation on this fascinating forum.

I've been intrigued by the solar activity that's been happening lately and long story short, I've found this community that provided the necessary information and decided to join.

Hello to each and one of you !

😄

 

CME impact detected. Hopefully this will be a good geomagnetic storm 🙂

IMF just shot up to 45nT. Nice start.

Edited May 10, 20241 yr by Novarupta

2 hours ago, Parabolic said:

Just in time for breakfast 😋

Well, how, where. I will have dinner in a moment.

I would like X 10 for dessert.🍰

Couldn't start worse.

 

Northern Germany still daylight and cloudy as hell.

 

Hope the best for the next few hours. 🙄

2 hours ago, Jeffrey Dikkes said:

AR3664 : 6 CME's due to arrival and GPF warning threshold has been broken.

Still think a G5 event greater than the Halloween solar storm( 2005) is about to ring our doorsteps.

So I would take this option! It's a shame that we won't get to see her, completely different conditions would have to come. Such a pity!

Reminder: please see the message at the top of the page.
Thanks guys!