"NOAA forecasters" and their "forecasting". Are they real at all?

[Chris, HB9DFG]

Hello all

First of all let me say that I am quite sad and very unhappy what presently happens with those "forecasters" and their bad "forecasting", who also giving many people hope für big aurorae or sky shows after a flare while after a few days nothing happens (bye-bye CME), it missed somehow, it was to slow, or to weak all of a sudden instead of a G1 storm and so on. What went wrong, dear NOAA forecasters? Where are you now? Where are your explanations now? "I'ts oh so quiet, shhh-shhh..." (right, I am a Björk fan by the way) 😍

All of us are surely knowing the similar sounding "spaceweather.com" site. Ok, the site is full of advertisements (surely because of the immense server costs etc etc. and we all know only dead is for free) but all in all there are allways beautiful space and other fine pictures to see mixed with differen Informations about similar interesting themes. No doubt about it.

Very often now (especially after the sun has been active and is flareing the so-called "NOAA forecasters" come into play, who then calculate e.g. the arrivel of CMEs and also provide informations about the earths magnetic field and possible ongoing storms for the next days.                                                                                                                                                          

Now (since I am sometimes very skeptical) i have some questions to ask: Those NOAA forecasters, are they really human at all? Or are they just computers running (bad) computerized models on and on? Who are they???

From the archive of spaceweathers.com archive:

Dez. 22nd: A CME is heading for Earth. The faint storm cloud left the sun on Dec. 20th (1136 UT), propelled by an M1.9-class solar flare in the magnetic canopy of sunspot AR2908. NOAA forecasters expect a glancing blow to Earth's magnetic field on Dec. 24th, possibly sparking a G1-class geomagnetic storm.

Dez. 24th: Multiple M-class solar flares this week have hurled at least two faint CMEs toward Earth. NOAA forecast models predict glancing blows on Dec. 24th and 25th. The impacts, albeit weak, could combine to produce G1-class geomagnetic storms and high-latitude auroras during the Christmas holiday.

Dez. 26th: Arctic sky watchers hoping for a geomagnetic storm were disappointed on Christmas when a CME did *not* hit Earth's magnetic field. The incoming storm cloud either missed or it is moving much more slowly than NOAA forecast models predicted.

So for me it looks like there are just computers running nice but (yet) very unreliable modell programms. That's fine in this case because nobody has to stand behind a bad prediction, right? Ok yes, I agree that sometimes these computer models were right and they made a quite good forecast randomly, so all of a sudden the "NOAA forecasters" are heroes in forecasting - not the computer forecast model itself. It is like when the local football team here wins, WE won! And If the local football team looses, THEY lost! Funny isn't it?  🙂 ⚽ ☹️

And if there are some real human NOAA forecasters around please tell us what went wrong if you were wrong. For me it seems not be possible to realise such a precise model because to many unknown values have to be included. You also can't exactly tell me how many pieces of what size fly to which side when a plate falls to the floor and breaks. I can't do that, nobody can develope such a simulation from my point of view! So does it make sense to run a still very unreliable simulatiom like e.g. the WSA-UNLIL because nobody can calculate the exact speed of a starting CME with its slowed down speed at arrival. To define the propper direction towards earth is just the next problem to solve.

So take care and hava a happy new year!

Chris

 

 

 

[Sam Warfel]

So they made one mistake. That’s not a problem. This is forecasting things we cannot see since they leave a star millions of miles away, until they actually arrive.  Very little is known, very few spacecraft and telescopes are available. And yes, there are human science experts using the models to make their forecasts, that is how all modern forecasting works, whether on earth or in space.

If you except space weather forecasts to be like earth weather forecasts, you are quite mistaken, it’s an entirely different ballgame, even if the names seem similar. 

Given how much we still don’t know or understand about the science, it’s actually remarkable that we’re able to forecast it at all. 

[Jesterface23]

The coronal hole high speed stream over powered the recent CMEs. Had the solar wind velocities been maybe 350-450km/s prior to an arrival then we may have noticed an impact.

[Drax Spacex]

Yes, NOAA was wrong this time (as were many others).  Yes, NOAA has been wrong before.  And yes, NOAA will be wrong again.

But don't hate the scientists - hate the science, the data, the theories, the models, which are incomplete or even invalid - otherwise they would get it right all the time.

Experts are expected to give some answer, even if it's a low confidence answer.  But the scientist or forecaster or expert who is candid and truthful will say when there is some uncertainty and does not always speak in absolutes.  They provide the caveat to cite there is a degree of uncertainty to go along with their prediction.

When NOAA issues a storm watch, they're not saying it will happen but that there is a possibility it could happen.  I think NOAA does a good job explaining the rational for their predictions in there forecast discussions, and in providing caveats when a prediction is low confidence.*

Granted (not referring to anyone at NOAA), there are some "experts" who set themselves up on a pedestal who are given to hubris, pride, and overconfidence for whom professional, academic, or institutional arrogance may be a factor.  But such people are easily identified after they have been clearly identified to have been wrong - you will never hear them say, "I was wrong", or "I'm sorry" - not in their vocabulary.  They will brush it under the rug and won't even discuss or admit the fact that they were wrong.

A test of personal authenticity is when an expert is wrong, they acknowlege it and take responsibility for it.  They discuss the possible reasons why they were wrong, what can be learned from it, how they can use that error to improve their understanding, and reevaluate the confidence they attach to future predictions.

*I don't work for NOAA.  I don't know anyone who works for NOAA.

[Vancanneyt Sander]

When predicting weather on Earth, there is already difficulty in  predicting if it’s gonna rain or not depending on the weather models. For weather forecasts we have weather stations all over the place to track the weather and accurately know when it’s going to rain.

now let’s look at space weather, the tracking of storms can happen only on two places: on the Sun where the CME leaves the Sun into an immense space. And then we have nothing to track it with along it’s journey. So we can’t tell what’s happening with that CME along the way. We’ll only know when it’s about a million kilometres away from Earth when a satellite picks it up. Although still a million kilometres away, it’s only 60minutes or less before it arrives at Earth. So we can only accurately forecast space weather max 60minutes before arrival!

so in space weather we have only very few resources to feed the models and no way of tracking them through millions and millions of kilometres of interplanetary space. That’s why it’s hard to accurately forecast space weather.

an other thing to note is that space weather forecasters like NOAA don’t make space weather forecasts for the Aurora enthusiasts but for the professional industries where space weather can have an impact. For example satellite operators. Then it’s vital to always predict on the safe side so that the professional infrastructure is prepared and protected. So if there is still 10% chance of a CME arrival, it’s for the professional industries still better to predict a possible impact and have a miss than to have one not predicted and hit with possible damage to satellites. So don’t blame the forecasters!

[Cierra Maguire]

I don’t work for NOAA, but I would be happy as a pig in mud to work for them as a forecaster! I use many of their free tools & data sets in my own space weather preambles. 

heres the thing… as others have said, we have limited (& aging) resources — like the DSCOVER satellite— to monitor  the space & Solar environment.
 

When the sun spews out Solar plasma & Solar magnetic field fragments as CMEs, THEY TAKE 24-72 hours to reach us. DSCOVR can give us a 50 minute-or so heads up before that shit reaches our magnetic field.

Think of CME’s like big waves of hot magnetized plasma, moving hundreds of kilometers per second. We do our best to monitor that stuff, but we’re like a buoy in the ocean— even our best tools are just outdated & beat-up life jackets. 

[Archmonoth]

On 12/26/2021 at 2:11 PM, Chris, HB9DFG said:

Hello all

First of all let me say that I am quite sad and very unhappy what presently happens with those "forecasters" and their bad "forecasting", who also giving many people hope für big aurorae or sky shows after a flare while after a few days nothing happens (bye-bye CME), it missed somehow, it was to slow, or to weak all of a sudden instead of a G1 storm and so on. What went wrong, dear NOAA forecasters? Where are you now? Where are your explanations now? 

We don't know what we don't know. 

 

On 12/26/2021 at 2:11 PM, Chris, HB9DFG said:

                                                                                                                                             

Now (since I am sometimes very skeptical) i have some questions to ask: Those NOAA forecasters, are they really human at all? Or are they just computers running (bad) computerized models on and on? Who are they???

Good/bad models are based on data, we don't have all the data, or know what data we are missing. 

On 12/26/2021 at 2:11 PM, Chris, HB9DFG said:

So for me it looks like there are just computers running nice but (yet) very unreliable modell programms. That's fine in this case because nobody has to stand behind a bad prediction, right? Ok yes, I agree that sometimes these computer models were right and they made a quite good forecast randomly, so all of a sudden the "NOAA forecasters" are heroes in forecasting - not the computer forecast model itself. It is like when the local football team here wins, WE won! And If the local football team looses, THEY lost! Funny isn't it?  🙂 ⚽ ☹️

Predictions are not sports plays, this isn't a goal with yes/no, they are attempts at refining the information which is available. Missed predictions are very useful, they provide vision into what kind of data is missing. They also provide additional data points for future computer models and predictions. 

On 12/26/2021 at 2:11 PM, Chris, HB9DFG said:

 For me it seems not be possible to realise such a precise model because to many unknown values have to be included.

We can nail down Earth weather models down pretty well (I think), but when a wildfire produces smoke, or a volcano erupts, those models need time and data to adjust their predictions. 

On 12/26/2021 at 2:11 PM, Chris, HB9DFG said:

You also can't exactly tell me how many pieces of what size fly to which side when a plate falls to the floor and breaks. I can't do that, nobody can develope such a simulation from my point of view!

With absolute information, complete modeling can be made, this is determinism. There are lots of models for crashing/breaking objects when every piece is accounted for. We don't have all the information regarding space weather, and by "we" I mean the body of knowledge about space activity. Our eyes are limited to the machines and satellites measuring things. These machines can also get borked and give faulty information, which changes our predictions. 

On 12/26/2021 at 2:11 PM, Chris, HB9DFG said:

So does it make sense to run a still very unreliable simulatiom like e.g. the WSA-UNLIL because nobody can calculate the exact speed of a starting CME with its slowed down speed at arrival. To define the propper direction towards earth is just the next problem to solve.

 

I understand the predictions are inadequate for you. We might not be able to see all the variables between us and the CME, but overtime we will add to our body of knowledge and predictions will become more refined. 

Edited December 28, 2021 by Archmonoth

[Sunshine]

Maybe I'm totally stupid, but I also wonder about the WSA-Enlil forecast. The while for most of the past year it was a roughly the same as measured solar wind speed and density, it seems to be completely off, without any correlation now. Am I stupid and are those things unrelated, or does anybody know of a better source for wind speed and density prediction?

[Vancanneyt Sander]

9 uren geleden, Sunshine zei:

Maybe I'm totally stupid, but I also wonder about the WSA-Enlil forecast. The while for most of the past year it was a roughly the same as measured solar wind speed and density, it seems to be completely off, without any correlation now. Am I stupid and are those things unrelated, or does anybody know of a better source for wind speed and density prediction?

A model can be good if the input parameters are all known. Like we’ve said already, we only have CME data when it leaves the Sun, so those are the input parameters. Because there is still a lot of interplanetary space between the Earth and the Sun, CMEs slow down at a variable rate when they travel through the interplanetary space. Because we only have data when it leaves the Sun, it makes the model not accurate with a difference in arrival times of more than 6 hours in plus and minus to the predicted arrival times. If we would have multiple satellites between the Sun and Earth, the modelling could be refined and improved making predictions a lot better, but that isn’t possible unfortunately so the models will always be a bit “off”. So with each CME, you have an “estimate” of arrival but then it’s a waiting game of many many hours to see it arrive.

 

[Sunshine]

Thank you for the explanation @Vancanneyt Sander. That gives some perspective to CME prediction. However, with the current equatorial coronal hole, I would expect the predicted high speed stream of solar wind to show up on WSA-ENLIL, but alas, what we are seeing is a decrease in wind speed for Jan. 2/3 The radar gradient pictures appear to show those streams however, so I'm a but puzzled how this fits together. Other than that, yes, it's unfortunately a waiting game, and with living in the UK I can't even setup an antenna to receive realtime GOES data

[Vancanneyt Sander]

2 minuten geleden, Sunshine zei:

Thank you for the explanation @Vancanneyt Sander. That gives some perspective to CME prediction. However, with the current equatorial coronal hole, I would expect the predicted high speed stream of solar wind to show up on WSA-ENLIL, but alas, what we are seeing is a decrease in wind speed for Jan. 2/3 The radar gradient pictures appear to show those streams however, so I'm a but puzzled how this fits together. Other than that, yes, it's unfortunately a waiting game, and with living in the UK I can't even setup an antenna to receive realtime GOES data

For CH, the ENLIL model isn’t perfect either. We can only see coronal holes on SDO but we don’t know the input parameters: what is the speed/density of the solar wind coming from the hole? We don’t know, you can’t even see it on LASCO. So CH are mostly modelled based on the previous rotation where we’ve seen the solar wind measurements at L1, that’s the only input parameters we have and those where from a previous rotation. CH change during that time but it’s difficult to know in what way (speed/density and IMF) they have changed. That makes ENLIL for CH also not 100% reliable.

conclusion: there is no accurate way for predictions, only an idea of possible conditions by the ENLIL model. 

[Sunshine]

Ah, thank you very much for that. Now it makes much more sense to me. I guess we can only hope that we'll get more satellites up into space which have instrumentation for solar observation. One really wonders why we still haven't surrounded our sun with observation satellites.

[Vancanneyt Sander]

5 minuten geleden, Sunshine zei:

Ah, thank you very much for that. Now it makes much more sense to me. I guess we can only hope that we'll get more satellites up into space which have instrumentation for solar observation. One really wonders why we still haven't surrounded our sun with observation satellites.

Why we didn’t place more satellites up there is mainly because of the difficulties for such orbits. When a satellite wouldn’t be at one of the Lagrange points, they need a lot more fuel to keep them in place which isn’t really possible. That’s why it isn’t possible to place more satellites along the Sun-Earth line. Most of them are at L1 (Lagrange 1) on the Sun-Earth line as a stable position.

[Sam Warfel]

As Sanders said, if it was that simple, we would. If the funding was available, that is. Sadly, there’s not much investment in this particular branch of science like there is in some others 

[Solarflaretracker200]

21 hours ago, Vancanneyt Sander said:

Most of them are at L1 (Lagrange 1

I’m totally stupid when it comes to satellites and maybe the answer is right in front of my face. But is L1 a satellite or is it something where you put the satellites at?

Edited January 3, 2022 by Solarflaretracker200

[Vancanneyt Sander]

29 minuten geleden, Solarflaretracker200 zei:

I’m totally stupid when it comes to satellites and maybe the answer is right in front of my face. But is L1 a satellite or is it something where you put the satellites at?

There are 5 Lagrange points around the Earth where satellites can be put in a relative stable environment. At L1 we have several satellites that are in the sun-Earth line and perfect for 24/7 observation of the Sun. 

[Sam Warfel]

1 hour ago, Solarflaretracker200 said:

I’m totally stupid when it comes to satellites and maybe the answer is right in front of my face. But is L1 a satellite or is it something where you put the satellites at?

Sanders stated what the Lagrange points where. The reason they are special “parking spots in space”, is that in these spots the gravity from the two bodies (and centrifugal forces from orbiting) actually cancel out! For example, at L1, the Sun and the Earth’s gravity both pull equally hard on a satellite, and so it stays still. The point is much closer to Earth than the sun, because of course the Sun’s gravity is much much stronger than the Earth’s, so for them to be equal it must be much closer to the lighter body. 

[Archmonoth]

4 hours ago, Solarflaretracker200 said:

I’m totally stupid when it comes to satellites and maybe the answer is right in front of my face. But is L1 a satellite or is it something where you put the satellites at?

I like to think of the L-points as calm valleys in the solar environment. There are rivers like the Interplanetary Transport Network - Wikipedia and even dusty corners like the Kuiper Belt. 

 

An L3 satellite would be a dream.

[Sam Warfel]

2 hours ago, Archmonoth said:

An L3 satellite would be a dream.

Always knowing all sides of the sun... Hopefully I will live to see that day