Jump to content

Solar Cycle 25 Predictions/Forecasts by 1. 'The Panel' & 2. NASA


Recommended Posts

This thread discusses the following two different predictions having different consequences:

  1. In this press releaseNOAA is associated with what I term the 'Panel forecast'.  
  2. I term the 'Kitiashvili forecast' as the 'NASA forecast', since NASA latched on to it in their press release here.

Each of these forecasts is discussed below.

--------------------- --------------------- ---------------------

1. The 'Panel Forecast'

SOLAR CYCLE 25 PREDICTION from the SOLAR CYCLE 25 Prediction Panel https://www.weather.gov/news/190504-sun-activity-in-solar-cycle

  • Solar Minimum (i.e., the Nadir of Cycle 24 which is the "Start" of Cycle 25):   late in 2019 or 2020.
    • (Note, this is a ~1.33yr* Minimum Window; *we'll assume "late in 2019" means sometime in the period Sept-Dec 2019, and "2020" means sometime in 2020.  However, based on clarification from the silso.be article---***linked below, which states, "minimum between the current cycle 24 and cycle 25 is predicted to occur between July 2019 and September 2020"---this gives a 1.17yr Minimum Window.)
  • Solar Maximum (Peak):   between 2023 and 2026.
    • (Note, let's assume they mean between the beginning of 2023 and the beginning of 2026, which is a 3yr Window Peak; however, combined with the 1.17yr Minimum Window, this means they predict Solar Maximum to occur (i.e., the Rise Timesometime within 2.25-to-6.5 years after Solar Minimum.) 
  • Sunspot Peak Range:  95-to-130.
    • (**What does this mean? Are they referring to the peak of the "Monthly Mean", or the peak of the "Smoothed-Monthly"?)

**I wish to nail down the answer to this question.  Later that article states, "Solar Cycle 24 reached its maximum - the period when the Sun is most active - in April 2014 with a peak average of 82 sunspots."   However, from this article on the same subject,http://sidc.be/silso/node/152, we hopefully get some clarification, wherein it states, "This prediction is now given in the scale of sunspot number Version 2. Therefore, solar cycle 25 will be similar to cycle 24, which peaked at 116 in April 2014."


***They also gave clarification on the Minimum Target in that article:  "The next minimum between the current cycle 24 and cycle 25 is predicted to occur between July 2019 and September 2020."

The plot above on the left for Cycle 24 represents the "Monthly Mean" (faint Black Line) and the 13-month "Smoothed-Monthly" (Red Line) Sunspot Number.   We can inspect that plot further, downloading data from here: http://sidc.be/silso/datafiles


Here is the data plotted, after downloading:


Edited by theartist
Link to comment
Share on other sites

The chart below, List of Solar Cycles, lists the past 24 Solar Cycles.  It is annotated to indicate which cycles would have met the Panel's Peak Range prediction, the Rise Time to Solar Maximum, and both criteria.



Seven (7) of the 24 cycles would have met the Panel's Peak Range prediction (after rounding off the Smoothed Maximum ISN to no decimal places).  ALL but one of the 24 cycles would have met the Panel's Rise Time to Solar Maximum prediction. Six (6) of the previous 24 cycles met both the Peak Range and the Rise Time to Solar Maximum prediction.

Clearly, the Rise Time to Solar Maximum prediction was so broad that it would be difficult for Cycle 25 to not meet that window.  Nonetheless, the Panel's prediction on Peak Range is definitely biased toward a weak cycle.  This is specified in the text of their conclusion, "This is well below the average number of sunspots...We expect Solar Cycle 25 will be very similar to Cycle 24: another fairly weak cycle, preceded by a long, deep minimum...Cycle 25 will be comparable in size to Cycle 24."

Before I proceed with the Kitiashvili (NASA) prediction, I came across the following document produced by The Panel, which will help explain the discrepancy above in Peak Range (i.e, 82 sunspots versus 116 sunspots, etc.): https://www.swpc.noaa.gov/sites/default/files/images/u59/10%20Lisa%20Upton%20Official.pdf

Here are a few slides from that document:





--------------------- --------------------- ---------------------

2. The 'Kitiashvili/NASA Forecast'

On 7/10/2019 at 1:02 AM, theartist said:

the Kitiashvili (NASA) prediction

I'm calling this the NASA prediction, since NASA is using it in a news release (cited below) on their website. This prediction is somewhat baffling (to me), because I found two different predictions online.  That may have something to do with the interchange of SSN V1 and SSN V2, over the years, in her model, or possibly it is related to updates of the model as time progresses and more data is input into the model.  I will restrain from further conjecture, and just provide a summation of what I found.

The most recent prediction was taken from a Kitiashvili presentation found here, starting @ 15:00: https://www.youtube.com/watch?v=W55Zst_ruGA 

The following document appears to match the slides in that video:  https://aas.org/files/resources/kitiashvili.pptx

Kitiashvili Summary (from June 2019 Presentation)

  • Solar Minimum (Period duration):  2019-2021.
    • (Note, this implies Solar Cycle 25 "onset" (i.e., the point at which activity picks up appreciably) is predicted to occur no sooner than early 2022.)
  • Solar Maximum Peak:  in 2024-2025.
  • Sunspot Peak Range: about 50 with an error estimate of ~15-30%. 
    • (Note, per the graph snapshot below, from Slide 10 in her June 2019 Presentation, the Peak Range appears to be ~43-65.)


  • Additional Highlights: The cycle will likely have a double-Maximum extending over 2-2.5 years.

Conclusions from the presentation confirm the above very low cycle prediction, stating, "Solar Cycle 25 (Maximum)...is expected to be significantly weaker than that of the current cycle."

The following article, https://www.nasa.gov/feature/ames/solar-activity-forecast-for-next-decade-favorable-for-explorationpertaining to Kitiashvili's forecast, states: "The forecast for the next solar cycle says it will be the weakest of the last 200 years." 

(I wish to add that a previous prediction, from a Kitiashvili 'poster' dated July 30, 2018, found here https://ntrs.nasa.gov/search.jsp?R=20180007221, stated, "mean sunspot number at the maximum will be about 90 (for the v2.0 sunspot number series) with the error estimate ~15%.)


Edited by theartist
Numbered the two different forecasts, separating them out for clarification.
  • Like 1
Link to comment
Share on other sites

It’s always the same at the end of a solar cycle, always a vast amount of scientist with different outcome on the prediction of the next cycle. The methods they use are always based on good research and data but still, the inner workings of the Sun are still so much unknown that it can’t be predicted in a perfect way. It’s good that there are multiple predictions to see what methods they used and how they think it will evolve. It can improve the predictions of others and improve them. 

With the sc24 predictions we started the cycle with two versions because they couldn’t agree and “the board” chose the stronger and a weaker cycle for sc24. It was untill after the long minimum that they abandoned the stronger cycle. Just to say that there is no good method for a prediction and you better wait and see what time brings...

Link to comment
Share on other sites

Op 10/7/2019 om 21:30, Vancanneyt Sander zei:

but still, the inner workings of the Sun are still so much unknown that it can’t be predicted in a perfect way

Even though we may improve our knowledge of the inner-workings of the sun, that is not going to provide "perfect" predictions for the following two reasons:

  1. We will never have perfect knowledge of the inner workings of the sun (i.e., the solar dynamo physical processes).
  2. The 'solar activity' (e.g., Bright Spot, Active Region, Coronal Hole) manifestation is a function of probabilistic/stochastic processes.

Consider a 'lava-lamp' analogy, as suggested by Dr. Skov, which I will now embellish.  We may know (or at least have a pretty good idea) of the duty-cycle of the lava-lamp's heater.  We may have external sensors measuring temperature and magnetic fields.  We may have a historical record of the lava lamp's visible behavior.  Nonetheless, we are not going to predict with perfect precision the size and number of the lava lamp bubbles that will happen on such-and-such a day next week.  However, as our knowledge and measurement precision of the lava-lamp parameters increase, we can model it pretty well, though it will still be subject to probabilistic/stochastic processes, and thus, (even if we had perfect knowledge, down to quantum mechanics granularity, of all matter-interaction in the universe) any predictions can only be framed within a certain probability.   

And then there are factors that may be affecting those probabilistic/stochastic processes that we hadn't taken into account in our model.  For example, suppose our electrical grid is providing 'clean' repeatable sinusoidal AC energy reliably for years on-end, and after stacking up months of reliable lava-lamp predictions, our AC electricity house-current starts getting 'dirty' due to replacement of generators up-line.  That is when we may find out that any perturbation in the AC house-current has slight influence on the performance of the lava-lamp's heater, enough to measurably influence the stochastic behavior of the lava-lamp.  :P 


Op 11/7/2019 om 20:36, theartist zei:

then there are factors that may be affecting those probabilistic/stochastic processes that we hadn't taken into account in our model

As I learn more and more about this field of study, one of the most confounding things is why the issue of 'planetary clocking' is not already readily understood and incorporated into forecast models.  It would certainly help in understanding some of the probabilistic/stochastic processes, and thus give a much better understanding when combined with increasing understanding of the solar dynamo processes.  Part of the problem is a lack of understanding and confusion on the particular physics by which 'planetary clocking' induces 'solar activity'.  But with our capabilities in computational modeling, I don't see why this is so difficult to understand or accept.  'Planetary clocking' shows up in the data.  

There was a recently released paper suggesting tidal effects as the physical mechanism:  https://www.newsweek.com/suns-solar-cycle-governed-alignment-planets-scientists-discover-1441753  A key point in that work was that a little bit of perturbation can induce a great effect in an unstable, and easily-triggered system.  There was a NASA report released several years ago that also suggested tidal effects, but it evidently did not take hold, as there has been, until recently, little mention of the idea in the mainstream.  

I think there obviously will be a tidal component.  However, electromagnetic flux signaling/response, electrostatic-charge-effects, and magnetic-field-connections will also interrelate to constitute the role by which 'planetary clocking' contributes to solar cycle activity.  However, it will not be necessary to have complete understanding of the complex physical interactions before some assumptions of 'planetary clocking' can be incorporated into forecast models.  

This is a brief statement on what is a somewhat complex idea, but I think it will be accepted with further understanding, including communication, on the topic.

Edited by theartist
Link to comment
Share on other sites

  • 1 month later...

This morning I revisited the 'Kitiashvili/NASA Forecast', and something caught my eye.  Below are a couple of slides from her presentation:




If the black line (in the above slides) is her model's prediction, doesn't it significantly undershoot the strength of SC23?

Edited by theartist
  • Like 1
Link to comment
Share on other sites

Guys, thank you for responding, because it may have been the further impetus I needed to look into her forecast further. After reviewing her presentation again, I don't think she mentions the black lines.  Based on her paper from 2008, those black lines (may) represent what they call a "reference solution", which is just one step in the model's production of an eventual forecast, which apparently is both the high fidelity jagged red lines in the top slide, and the smoothed red line in the bottom slide.

She states they predicted SC24 pretty well in 2008, and their model may be able to look-ahead eight years into the future, and possibly longer.

I have a lot more respect for her forecast now, and she may come out looking like a genius by the time SC25 is over.😊 

Edited by theartist
  • Like 1
Link to comment
Share on other sites

On 7/9/2019 at 11:14 PM, theartist said:

I wish to add that a previous prediction, from a Kitiashvili 'poster' dated July 30, 2018, found here https://ntrs.nasa.gov/search.jsp?R=20180007221, stated, "mean sunspot number at the maximum will be about 90 (for the v2.0 sunspot number series) with the error estimate ~15%.

I wish to try and clarify the above statement.  The title of that poster (below), was "Physics-Based Approach to Predict the Solar Activity Cycles"



Additionally, the poster (below) with very similar content [including the statement, "the mean sunspot number at the maximum will be about 90 (for the v2.0 sunspot number series) with the error estimate ~15%"] was published on December 13, 2018 [for American Geophysical Union (AGU) Fall Meeting 2018; December 10, 2018 - December 14, 2018; Washington, DC; United States], titled, "Synergy of Observations and Dynamo Models to Understand and Predict Solar Activity Cycles":


From the figure (repeated below) above the "Conclusions" in both of those posters, it indicates their SC25 prediction was based on data up to 2017.5:



The more recent paper previously cited in this thread titled, Solar Activity Forecast for the Next Decade, given at a presentation on June 09, 2019, [stating, "the mean hemispheric sunspot number at the maximum will be about 50 (for the v2.0 sunspot number series) with an error estimate of ~15%"], from which I took the SC25 forecast parameters, was quite likely using updated data.

That forecast of "about 50" is also found in the following two more recently published posters:

Edited by theartist
  • Like 2
Link to comment
Share on other sites

  • 2 weeks later...

As a hobby, I study the effect the planets have on the sun, that causes sunspots. In the period between 26 October and 5 November we will be able to get some sunspots. This will be the last sunspots in 2019. It should indicate that 2019 will end up with between 278 and 283 days without sunspots. We might even get sunspots in the period belonging to SC25. However, as I said, its just an hobby so I might of course be wrong!


  • Like 1
Link to comment
Share on other sites

  • 5 months later...

Have you seen they have added a forecast to https://www.spaceweatherlive.com/en/solar-activity/solar-cycle that is based on the Panel forecast...

With the continuing low activity, I would suggest the speed of the rise they are predicting to get to their "target" of 115 looks very optimistic, especially when you look at all the forecast models at SILSO http://www.sidc.be/silso/forecasts  which have all been revised down, some more than others...   http://www.sidc.be/silso/predikfcm



Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you also agree to our Terms of Use and our Privacy Policy.