Forecasting techniques?

[NEAurora]

As a longtime space weather and aurora nerd, I thought I would join this community and see what others do to decide whether or not to go out and look for aurora.  

Specifically, I’m curious as to how others interpret the data coming in from various sources when making an aurora forecast.

I’m interested because I’ve yet to figure out a one-size-fits-all approach to this. Living at a geomagnetic latitude of about 55 degrees my general rule is a kp 6 or greater with an HPI of 70+ plus and a of at least bz < -8 or so with some persistence.  There are a few other considerations, but these are the main ones. These criteria seem to work well for my latitude, but not always.

Knowing that all of these different pieces of data are related to each other but not identical, I more often than not just kind of basing any decision to go out and look on a “gut feeling” … which sometimes works and sometimes doesn’t. Recently I’ve been looking at magnetic coupling functions (such as the ones described here: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2021JA029946) to try and extract and incorporate more information into my decision making process.
 

Anyone else do anything different?  Any more insights that I should consider? I’d like to refine my process as we head towards solar max.

 

[hamateur 1953]

Respectfully speaking of course. But as murphys law seems to dominate space weather as well as most everything else in our lives, may I suggest a good sense of humor as a healthy thing to preserve sanity?  🤣🤣

16 hours ago, NEAurora said:

As a longtime space weather and aurora nerd, I thought I would join this community and see what others do to decide whether or not to go out and look for aurora.  

Specifically, I’m curious as to how others interpret the data coming in from various sources when making an aurora forecast.

I’m interested because I’ve yet to figure out a one-size-fits-all approach to this. Living at a geomagnetic latitude of about 55 degrees my general rule is a kp 6 or greater with an HPI of 70+ plus and a of at least bz < -8 or so with some persistence.  There are a few other considerations, but these are the main ones. These criteria seem to work well for my latitude, but not always.

Knowing that all of these different pieces of data are related to each other but not identical, I more often than not just kind of basing any decision to go out and look on a “gut feeling” … which sometimes works and sometimes doesn’t. Recently I’ve been looking at magnetic coupling functions (such as the ones described here: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2021JA029946) to try and extract and incorporate more information into my decision making process.
 

Anyone else do anything different?  Any more insights that I should consider? I’d like to refine my process as we head towards solar max.

 

 

[3gMike]

18 hours ago, NEAurora said:

As a longtime space weather and aurora nerd, I thought I would join this community and see what others do to decide whether or not to go out and look for aurora.  

Specifically, I’m curious as to how others interpret the data coming in from various sources when making an aurora forecast.

I’m interested because I’ve yet to figure out a one-size-fits-all approach to this. Living at a geomagnetic latitude of about 55 degrees my general rule is a kp 6 or greater with an HPI of 70+ plus and a of at least bz < -8 or so with some persistence.  There are a few other considerations, but these are the main ones. These criteria seem to work well for my latitude, but not always.

Knowing that all of these different pieces of data are related to each other but not identical, I more often than not just kind of basing any decision to go out and look on a “gut feeling” … which sometimes works and sometimes doesn’t. Recently I’ve been looking at magnetic coupling functions (such as the ones described here: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2021JA029946) to try and extract and incorporate more information into my decision making process.
 

Anyone else do anything different?  Any more insights that I should consider? I’d like to refine my process as we head towards solar max.

 

Hi, Welcome to the forum!

I live at Latitude 52degrees in the UK so I struggle to see Aurora.

As it happens I had just been looking at the Archive https://www.spaceweatherlive.com/en/archive/2022/11/06/xray.html to investigate the temporal distribution of M and X flares this year. As a starting point I used yesterday (6th October) since we had an M flare around midnight. I then stepped back to the previous rotation and examined data for adjacent days. Looking at previous rotations I found that we had M flares on 10th October, 12th September, 16th August, 30th April and 2nd April (and an X on 30th April). There were obviously M flares on other days so this is not a complete solution, but may be useful to add confidence to decisions made on the basis of your other factors. Of course local conditions also contribute and, rather irritatingly, I found that all these dates are within a few days of Full Moon !

Another factor that may be useful is to keep track of the Equatorial Coronal Holes as they tend to survive several rotations.

Edited November 7, 2022 by 3gMike
Added comment re Coronal Holes

[Vancanneyt Sander]

18 uren geleden, NEAurora zei:

my general rule is a kp 6 or greater with an HPI of 70+ plus and a of at least bz < -8 or so with some persistence.  There are a few other considerations, but these are the main ones. These criteria seem to work well for my latitude, but not always

Well… Kp is a global indicator of geomagnetic activity, thus not reliable for nowcasting. Kp is the activity of the past three hours, unless threshold was reached at a magnetometer station. HPI and Bz have a big role in how geomagnetic storms unfold, but as the DSCVR satellite is above the Earth it takes some time before the solarwind reaches the Earth and the magnetic field of the Earth starts to respond. But as a skywatcher you seem to forget to use a nearby magnetometer station as last but not least parameter, for example Kiruna magnetogram. Based on the magnetometer deflection a certain local K-index is reached for that station. So it’s possible to see in the magnetometer data when the deflection is big enough for Aurora at your location. For me as a Belgian I need to have a deflection of -1400 on the Kiruna magnetometer before I even run outside.

[Jesterface23]

The SWPC has a different way of of determining the K indices from what I know as well. There are two different components to watch for fluctuations, but one component needs to be manually calculated now if the public wants to know the value. The SWPC does have both components available only for Boulder on their magnetometer page.

[NEAurora]

Thanks for all of the replies, I hand considered looking at local k-indices instead of the global value.  Oddly enough, the kp index is the factor I generally give the least amount of weight to, if for no other reason than it doesn’t have any predictive value, at least the measurements out at L1 give some lead time.  Again, I appreciate the responses, it’s interesting to see how others interpret the various data sources. 

8 hours ago, hamateur 1953 said:

Respectfully speaking of course. But as murphys law seems to dominate space weather as well as most everything else in our lives, may I suggest a good sense of humor as a healthy thing to preserve sanity?  🤣🤣

 

Do you mean to tell me that the environment doesn’t behave neatly in accordance with my expectations?  And that I’m simply .. just .. to accept this?  I’ll do no such thing. 😀

[hamateur 1953]

Bravo my good man. Bravo!!! 

[Newbie]

Where I live, at just over 37ºS the chances for seeing the aurora are fairly slim (Kp 9), however around 5 years ago the Hobart magnetometer (Australia) indicated sustained Kp of 8.87 over a few hours, the highest reading I've ever seen. The Bz had dropped to -40 and below both indicatve of a strong solar storm. 

This was enough to send me outside and I observed a grey-green naked eye aurora not far above the Southern horizon. I watched it until it disappeared when I came inside the Kp had fallen to around 6 and Bz was around -10.

I found both Kp and Bz best indicators of when you will see an aurora but you need to be able to access current readings requiring mobile data. If you are off the grid you have to wing it!

Newbie

[Scottlarocc E]

Hey @3gMike, I keep track of signifacant flares. I checked all those dates you mentioned and checked my stock charts like SPY and Bitcoin. All those flares created votility. Made the SP500 go down. I love flares because it gives me a heads up for options trading to buy puts. Its a cheat code. lol. 

[3gMike]

5 hours ago, Scottlarocc E said:

Hey @3gMike, I keep track of signifacant flares. I checked all those dates you mentioned and checked my stock charts like SPY and Bitcoin. All those flares created votility. Made the SP500 go down. I love flares because it gives me a heads up for options trading to buy puts. Its a cheat code. lol. 

I think it is more likely due to corporate lunacy - induced by the full moon 😁

[Scottlarocc E]

Full moons usually causes a rally. Like yesterday and today. 

[Archmonoth]

4 hours ago, Scottlarocc E said:

Full moons usually causes a rally. Like yesterday and today. 

Sorry to be contradictory and perhaps this assertion and discussion could be moved to another thread, but the Lunar Effect is a myth, it's been meta studied for many years. The full moon doesn't cause stocks to be more volatile or people to behave differently. 

 

[Scottlarocc E]

There was someone here that listed all the recent M and X flares that happened around a fool moon and I went back in price action on certain stocks. When M or X flares hit, stocks would fall the day or or next day. Its amazing, Especially tech stocks which it really affects. Talk about radio and communication black outs,  It true does cause volatility in the market. There's not many that see this. 

[CalleB]

Hi! 
 

I’ve been doing forecasting for 12 yrs (insane) and for me there a couple of rules : 

- expect the unexpected. 
All the vital signs for a show can still create nothing. Speed, Bz, By and density can be great, double digits in By, Bz and density, and still - nothing. Even some drops on the magnotometers, doesn’t allways create action. 
 

- Learn by your mistakes
Often you’d be wrong in forecasting, of course. Since heliophysics is complex, and space weather too, models and forecasting is wrong. Often. What did you miss? Track it backwards, learn about the magnetic fields and sun-earth connections, and just accept that this field is rather new. Some of the satellites we use has only been up there 20 ish years. Some even less. The field in forecasting is still ongoing a change, and research is being done. 

 

-Use Kp in retrospective 
Take a look at the archive here. Then you can compare all the data (not By if I’m correct) with Kp, and via spaceweather.se you can find magnetometer history via Kiruna.

Potsdam Hp30 index is better for nowcasting, as well as AE index.
 

welcome to a world of frustration an humiliation.  

//Calle www.facebook.com/norrskensverige 
 

 

[Cat Perkinton]

2 hours ago, CalleB said:

Hi! 
 

I’ve been doing forecasting for 12 yrs (insane) and for me there a couple of rules : 

- expect the unexpected. 
All the vital signs for a show can still create nothing. Speed, Bz, By and density can be great, double digits in By, Bz and density, and still - nothing. Even some drops on the magnotometers, doesn’t allways create action. 
 

- Learn by your mistakes
Often you’d be wrong in forecasting, of course. Since heliophysics is complex, and space weather too, models and forecasting is wrong. Often. What did you miss? Track it backwards, learn about the magnetic fields and sun-earth connections, and just accept that this field is rather new. Some of the satellites we use has only been up there 20 ish years. Some even less. The field in forecasting is still ongoing a change, and research is being done. 

 

-Use Kp in retrospective 
Take a look at the archive here. Then you can compare all the data (not By if I’m correct) with Kp, and via spaceweather.se you can find magnetometer history via Kiruna.

Potsdam Hp30 index is better for nowcasting, as well as AE index.
 

welcome to a world of frustration an humiliation.  

//Calle www.facebook.com/norrskensverige 
 

 

Thanks for this, its good to know someone else experiences the same frustrations when all the stats are pointing to a big dollop of activity & I have nothing on camera let alone to the eye. I find the resources and links on this site are excellent.

[Sam Warfel]

On the flip side, sometimes there can be aurora in complete contradiction of everything we know about the data! For example, I took this on Sept 27th my time, at a time when the Bz was 30nT North, when the Hemispheric Power model had dropped all the way down to 13, basically the baseline, meaning there should be no aurora visible even in, say, Fairbanks. 
But here it was, highly visible to the camera from the middle latitudes of Southern WI!

 

[CalleB]

@Orneno exactly! March 13-14 2022 is also a great example of “stunning stuff happening whilst Bz north”. 

Not even going to start mention those sneaky CMEs, this field is so fun to be a part of.

 

 

[Sam Warfel]

44 minutes ago, CalleB said:

exactly! March 13-14 2022 is also a great example of “stunning stuff happening whilst Bz north”. 

Got any pics of what happened then?

44 minutes ago, CalleB said:

Not even going to start mention those sneaky CMEs, this field is so fun to be a part of.

For sure! 😁

[WildWill]

On 11/8/2022 at 10:10 AM, Scottlarocc E said:

Full moons usually causes a rally. Like yesterday and today. 

I might've bought increased volume, risk and volatility - a good time to play the VIX. Or a straddle maybe straddle -spread, but just buying OTM puts is only creating a hedge... or opening a short position.

Aside from last weekend, which you mentioned, would you care to share your (raw)  data for say the last 12 or 18 months? 

I look forward to checking it out.

Thanks Much! 
 
WnA