Credit: cherishthescientist.net
‘Scientists use an extended, 22-year solar cycle to make the forecast’ is the sub-heading to the article. In other words the Hale cycle. At the end of last year The Talkshop detailed Plenty of predictions from a wide range of research groups, including our own, made in 2013. A possible (?) early indicator is that the ‘smoothed minimum’ of sunspots at the start of solar cycle 25 is given by Wikipedia as 1.8, the lowest recorded since cycle 7 (0.2) in 1823.
– – –
In direct contradiction to the official forecast, a team of scientists led by the National Center for Atmospheric Research (NCAR) is predicting that the Sunspot Cycle that started this fall could be one of the strongest since record-keeping began, says NCAR News.
In a new article published in Solar Physics, the research team predicts that Sunspot Cycle 25 will peak with a maximum sunspot number somewhere between approximately 210 and 260, which would put the new cycle in the company of the top few ever observed.
The cycle that just ended, Sunspot Cycle 24, peaked with a sunspot number of 116, and the consensus forecast from a panel of experts convened by the National Aeronautics and Space Administration (NASA) and the National Oceanic and Atmospheric Administration (NOAA) is predicting that Sunspot Cycle 25 will be similarly weak. The panel predicts a peak sunspot number of 115.
If the new NCAR-led forecast is borne out, it would lend support to the research team’s unorthodox theory – detailed in a series of papers published over the last decade – that the Sun has overlapping 22-year magnetic cycles that interact to produce the well-known, approximately 11-year sunspot cycle as a byproduct.
The 22-year cycles repeat like clockwork and could be a key to finally making accurate predictions of the timing and nature of sunspot cycles, as well as many of the effects they produce, according to the study’s authors.
“Scientists have struggled to predict both the length and the strength of sunspot cycles because we lack a fundamental understanding of the mechanism that drives the cycle,” said NCAR Deputy Director Scott McIntosh, a solar physicist who led the study.
“If our forecast proves correct, we will have evidence that our framework for understanding the Sun’s internal magnetic machine is on the right path.”
Continued here.
– – –
UPDATE: they are using radically different start and end dates for solar cycles 23 and 24 to everybody else, including SILSO and Wikipedia. They say their dates, called ‘terminators’, are ‘the basis for the new study.’
Tallbloke's Talkshop 
the Sun’s internal magnetic machine
Looking forward to the theory of how it got there 😎
– – –
We’ve been here before. In 2006 NCAR predicted SC 24 would be 30-50% stronger than SC 23, but it turned out to be a lot weaker.
Scientists Issue Unprecedented Forecast of Next Sunspot Cycle
https://www.nsf.gov/news/news_summ.jsp?cntn_id=105844
SC23 max: 180
SC24 max: 116
The lowest starting minimum NOT to lead to a low maximum (since records began) was 2.5 in cycle 15, with max 176.
https://en.wikipedia.org/wiki/List_of_solar_cycles
It’s a dynamo, doncha know.
Here’s the set of predictions to Dec30 2019 prepared by Steve Brown.
Of interest in Miles Mathis’ prediction, which is also for a very active cycle.
Click to access solmin.pdf
TB: I had one on my bike, back in the day.
They are so good – why don’t they predict the coronavirus as well?
Here’s a larger version of Steve Brown’s compilation, with Miles Mathis’ prediction added, which is about the same amplitude as this NCAR team are forecasting, though later than their peak in 2025.
Or maybe their climate model uses solar cycle predictions?
Bit of my humour, perhaps, maybe, or not being fully knowledgeable on these things but reading a lot and trying to make sense of all these scientific forecasts ( think they should becalled Polycasts) ….. so will this be a DEAD Rabbit Bounce of activity? I never count my chicks until they’ve layed their first eggs. ( anyone hear about the hatched QUAIL Eggs from the S/Mkt today?)
Data query…
NCAR: Sunspot Cycle 23 began in 1998 and did not end until 2011, 13 years later. Sunspot Cycle 24, which is just ending, was quite weak as well, but it was also quite short — just shy of 10 years long – and that’s the basis for the new study’s bullish prediction that Sunspot Cycle 25 will be strong. [bold added]
– – –
Where are they getting their data from? Wikipedia says SC 23 ended in December 2008, NCAR says 2011. So the ‘basis’ of their new study is way off?
Solar cycle 23 was the 23rd solar cycle since 1755, when extensive recording of solar sunspot activity began. The solar cycle lasted 12.3 years, beginning in August 1996 and ending in December 2008. [bold added]
https://en.wikipedia.org/wiki/Solar_cycle_24
Their SC23 start date also looks seriously wrong, as well as the cycle length – but see the study.
Their study, with 72 references to ‘terminator’, is here:
https://link.springer.com/article/10.1007/s11207-020-01723-y/
Update: https://tallbloke.wordpress.com/2020/12/08/new-sunspot-cycle-could-be-one-of-the-strongest-on-record-says-ncar/comment-page-1/#comment-162907
It will be strong says NCAR with finger up orifice
Is this the ‘official forecast’? [Update – yes]
Solar experts predict the Sun’s activity in Solar Cycle 25 to be below average, similar to Solar Cycle 24
April 5, 2019 – Scientists charged with predicting the Sun’s activity for the next 11-year solar cycle say that it’s likely to be weak, much like the current one. The current solar cycle, Cycle 24, is declining and predicted to reach solar minimum – the period when the Sun is least active – late in 2019 or 2020.
Solar Cycle 25 Prediction Panel experts said Solar Cycle 25 may have a slow start, but is anticipated to peak with solar maximum occurring between 2023 and 2026, and a sunspot range of 95 to 130. This is well below the average number of sunspots, which typically ranges from 140 to 220 sunspots per solar cycle. The panel has high confidence that the coming cycle should break the trend of weakening solar activity seen over the past four cycles.
“We expect Solar Cycle 25 will be very similar to Cycle 24: another fairly weak cycle, preceded by a long, deep minimum,” said panel co-chair Lisa Upton, Ph.D., solar physicist with Space Systems Research Corp. “The expectation that Cycle 25 will be comparable in size to Cycle 24 means that the steady decline in solar cycle amplitude, seen from cycles 21-24, has come to an end and that there is no indication that we are currently approaching a Maunder-type minimum in solar activity.”
https://www.weather.gov/news/190504-sun-activity-in-solar-cycle
“they seem to be using radically different start and end dates for solar cycles 23 and 24 to everybody else […]”
Open your mind to the concept of biased estimation. Consider carefully factors that might bias systematically. Systematic bias needs to be eliminated.
Stay vigilantly aware of the philosophical difference between a parameter and a parameter estimate. Estimates have a distribution of measurement errors. Sometimes the estimates have a systematic bias grounded in methods and assumptions. Keep a wide open mind about possibilities if in fog.
Remember that central limit theorem applies to everything except Cauchy. Even that is no problem as you can just switch from mean to median estimator.
General remark: There’s a lot of sloppy thinking going on about SCL. People have appeared unwilling to learn basics for more than a decade. I think people imagine it to be more complicated than it is (maybe not comfortable with complex numbers even if operations are trivial in that context).
Even more general remark: I advise the west to increase by many orders of magnitude a failsafe commitment to superior math education. It will be glorious.
Date mystery [my comment: December 8, 2020 at 7:07 pm] solved…
See Table 1 (link below). Their dates are all ‘during the rise phase’ rather than at the solar minimum.
– – –
Table 1 List of sunspot cycles, with their peak SSN, the times of their terminators (rounded to the nearest month and expressed as a decimal year), and the difference between subsequent terminators (ΔT; yr). Recall that by definition the terminator of sunspot cycle N occurs during the rise phase of sunspot cycle N+1. [bold added]
From: Overlapping Magnetic Activity Cycles and the Sunspot Number: Forecasting Sunspot Cycle 25 Amplitude
https://link.springer.com/article/10.1007/s11207-020-01723-y/tables/1
– – –
Recall that by definition – ?
Just reading this.
I would advise the researcher to differentiate philosophically between precision and bias, considering the costs of tradeoffs. Some blog commentators were pitching a similarly misleading generalization a decade ago. The widespread inclination towards phase-biased parameter estimates may arise from avoidance of complex numbers.
Precision and accuracy rather — e.g. precisely biased vs. diffusely accurate and unbiased on average …which affords sensible decomposition.
Donner and Thiel are only ones I remember who used a method that was unbiased in a broad sense, but they did not have tunable extent and there was a systematically oscillating phase bias in their measurements (that was simple and easy to correct since its origin was easy to trace to the support span).
There is an immense project I would undertake if I had several hundred thousand dollars in stable, reliable research funding.
‘The 22-year cycles repeat like clockwork’
Except they don’t. They repeat, but the periods vary roughly +/- up to 2 years.
Reblogged this on Climate Collections.
“New sunspot cycle could be one of the strongest on record”
Humm, I wonder what odds the bookies would give?
A small proposal —
All government funded academic institutions that make predictions about the future Must back-up such claims with funds up to the level of 10% of all research grants they have gained in the year when the prediction was made. Failing predictions will see the funds returned back to all taxpayer in the year they come due.
tom0, 2 decades ago I briefly considered working for government (crazy, I know). Just to apply for jobs I had to go for countless exams, including 2 full-day exams. It was ridiculous. I scored perfect on tests for judgement and spatial skills. Then I was told they can’t hire me because I wouldn’t fit in with the people who worked there.
I believe they’re laying the groundwork for faking sunspot activity just as the climate alarmists have faked OHC by fudging the 90% of 1 degree x 1 degree oceanic areas which are not sampled by ARGO floats; just as they did when they ‘adjusted’ ARGO float data (more accurate) to be more in line with ship intake and drag-bucket data (less accurate) because ARGO float data didn’t support their narrative; just as they’ve faked temperature data by ‘adjusting’ it so that the past is cooler, thus making current temperature appear warmer by comparison, establishing a ‘warming trend’ when in actuality much of the raw data shows a cooling trend.
We need some way of cross-checking them, so if they do start faking the sunspot numbers, we can call them on it. They are desperate to maintain their CAGW narrative, and are going to more extreme lengths to do so. They’ve faked or ‘adjusted’ so much of the data that they’re practically forced now to start faking real-time observational measurements to fit their narrative.
Will they apologise if they are wrong?
I don’t see any problem with this site of thing? Failed predictions can tell you more than successful ones – you can “succeed” through blind luck and random chance but you can’t generally fail without being actually wrong. They have made falsifiable predictions so let’s see.
As for money, why not go the Longitude route? Put up prize money to solve a problem in science, winner takes all.
For theorizing on solar activity I tend to the idea that low cycles come in groups. Cycle 24 was the first in a group.
The last group of 3 low cycles was 5,6,7 starting in 1798. Cycle 24 started in 2008, 210 years later, which is very close to the accepted de Vries cycle length (205-210 years).
Wikipedia has ‘a comparison of the growth of cycle 25 versus cycle 24, using the 13-month sunspot averages, beginning with the months of the respective minimums.’ Not much has happened to date, but there’s a relevant footnote:
October 2020 had an average sunspot count of 14.4 per day, which was the largest since June 2018. Thus, the 11th month of the cycle was first month with an average of 10 or more, compared with the 13th month for cycle 24. November 2020 had an average of 34 spots per day, the largest since September 2017. Thus the 12th month of the cycle was the first month with an average of 30 or more, compared with the 22nd month for cycle 24. These figures are in early agreement with a new paper (October 2020) by McIntosh et. al.[11] which projects that solar cycle 25 will almost certainly be stronger than SC24 (ISN max 116), and most likely stronger than SC23 (ISN max 180). [bold added]
https://en.wikipedia.org/wiki/List_of_solar_cycles
– – –
That’s the NCAR paper. The plot thickens…and Miles Mathis has jumped on the Wikipedia notes too.
First published December 4, 2020 (by MM)
My Dutch collaborator Steven Oostdijk has just informed me that the Wikipedia page on cycle 25
immediately began changing after our recent paper of a few days ago. Most predictions of cycle 25 are
quickly being revised up, sometimes drastically. The biggest change reported there has come from
Scott McIntosh, who, after earlier predicting cycle 25 would be weaker than cycle 24, published a
paper in mid-October revising that completely. He is now predicting it will be far larger, and might be
the largest in recent times. This after implying that cycle 25 might hardly exist at all, saying odds were
that we were entering a Grand Solar Minimum of extended weakness.
Click to access mcin.pdf
(4 pages)
Wikipedia’s solar cycle 25 page goes into more detail…
As at December 1 2020, solar cycle 25 is showing early signs of being somewhat stronger than solar cycle 24.
• The 13-month average sunspot count for May 2020 was 5.6 spots per day, compared to 3.5 for the corresponding month in the previous cycle.
• November 2020 averaged 34 spots per day, 10 months earlier than the first month to average 30 or more in cycle 24.
• The first single day to have 90 spots has occurred in month 12 of this cycle, compared with month 27 in cycle 24.
• Since June 1 2020, there have been 78 spotless days, compared to 130 in the corresponding period of cycle 24.
https://en.wikipedia.org/wiki/Solar_cycle_25#Early_signs
Miles Mathis has a critique, or demolition, of McIntosh et al on pages 3 and 4 here…
Click to access mcin.pdf
Read Why Tune Blew: a Sleep in 1919
Clyde says:
“I believe they’re laying the groundwork for faking sunspot activity just as the climate alarmists have faked […] We need some way of cross-checking them, so if they do start faking the sunspot numbers, we can call them on it.”
All already done: “ahead by a century“ — Tragically Hip
The red engine of trust revved inverted totalitarianism past blue long ago. IT doesn’t meter now.
Rewriting a past rewrite of a past rewrite of a….
Awake UN’s low lie UN predate Or[well].
the consensus forecast from a panel of experts convened by the National Aeronautics and Space Administration (NASA) and the National Oceanic and Atmospheric Administration (NOAA) is predicting that Sunspot Cycle 25 will be similarly weak.
One of the five authors of the NCAR paper is with NASA Goddard. So whether it’s a high or low solar cycle, NASA can try to claim some kudos either way — aka hedging your bets.
oldbrew says:
December 8, 2020 at 10:12 pm
‘The 22-year cycles repeat like clockwork’
Except they don’t. They repeat, but the periods vary roughly +/- up to 2 years.
.
————————
.
The cycles don’t repeat like clockwork when using the traditional definition of a sunspot cycle (which is minimum to minimum)… but maybe they do when using the the concept of beginning the solar cycle at the ‘terminator’ which occurs during the rise phase, not at the minimum. I’ve only just read this now and haven’t really looked into it yet, but it’s a concept which may have some validity.
It’s important to remember that the start of a solar cycle is at the solar minimum only because we say it is. Is there any justification for this? Nothing really happens at solar minimum and we could just as easily have made each cycle start at solar maximum… or anywhere else in the cycle.
A short cycle and a long cycle aren’t going to be converted to two matching cycles by changing the measurement criteria. Or, they would have to be very strange criteria, but even then it’s likely to be impossible.
I’ve had a chance to look at it now, and I think it’s a pile of crap. They say….
“a relationship between the separation of terminators and sunspot cycle amplitudes: low amplitude sunspot cycles appear to correspond with widely separated terminators while larger amplitude sunspot cycles correspond to more narrowly separated terminators.”
The claim that “The 22-year cycles repeat like clockwork” is just plain wrong. All they’ve done is find some points along the rise phase which give a closer ‘fit’ to connect the level of solar activity with the duration of the cycles. They give it the fancy name ‘terminators’ and hope that nobody notices that they’re actually using variable starting points for cycles. Even with a closer fit, saying that the 22-year cycles repeat like clockwork is nonsense.
Thanks for the comments. Lots of things to discuss. I’m totally open to answering questions about what were thinking/talking about and why my perspective changed between 2014 and 2020 wrt to activity moving out.
Thanks for calling by, Scott. Two things of initial interest for me:
1) has your SC 24 terminator event arrived yet?
2) in Figure 3(b), re. the straight blue line – can you explain ‘local regression trend subtracted’ in layman’s terms?
Hi Scott, and welcome to the talkshop. Thanks for inviting questions here. We really appreciate scientists who are happy to discuss their work.
I just wondered if you are aware of our very different approach to modeling the evolution of solar activity, using planetary orbital resonance patterns rather than directly observed solar indices?
Our group’s 2013 paper by Rick Salvador is here:
Click to access prp-1-117-2013.pdf
After the paper was published, Rick made a 4000 year hindcast using the model, and compared it to Steinhilber et al’s 10 Be based reconstruction.
The model yielded this match to the SSN series from 1749
Finally, here’s how our model is doing since 2013 compared to the data.
I believe there is a strong correlation between sunspot numbers and the mental acuity of our leaders.
/sarc
@oldbrew sorry – my day job intervenes….. two responses for you:
1) has your SC 24 terminator event arrived yet?
We think so, yes – around 11/5/20 – two of the four markers we anticipate to change have shifted but are watching closely for the others – the strong S/N hemispheric asymmetry is adding a bit of complexity. The coronal structure in the equatorial region changed about 5 months ago, waiting on the small-scale stuff to collapse and the cosmic ray flux to take a dive.
2) in Figure 3(b), re. the straight blue line – can you explain ‘local regression trend subtracted’ in layman’s terms?
The Hilbert Transform phase functions can easily get messed up by trend removal in the SSN data, we wanted to analyze the oscillatory part of the signal – so we decided to remove the trend in the data as gently as we could. “rlowess” – is the method we deliberated over to minimally impact the HT and not oversmooth the SSN. The method is covered here: https://www.mathworks.com/help/curvefit/smoothing-data.html
When you undersmooth the background variation, remove it and apply the HT it leads to negative frequencies in the HT – not good and leads to erroneous results.
RLOWESS allows you to have a trend that smoothly varies with time that roughly represents the running mean of the data – in practical terms for this work it avoids issues with the cycles of the Dalton Minimum when the amplitude could skip under the mean from 1750 – 2020, for example. So if you do a blanket timeseries mean trend removal there you can get negative HT frequencies. You can also see a couple of examples of this bad background removal behavior in Fig 2 of https://arxiv.org/abs/1909.06603.
We did a ton of experiments to look at the impacts of any trend subtraction and/or timeseries smoothing on the terminator locations… and hence the correlation. You can see those here – as part of the LENGTHY peer review process – I know many of you are data junkies: https://www.dropbox.com/s/r7722ypop9f1azs/NATASTRON-20033112%20-%20Appendix.pdf?dl=0
@tallbloke – thanks for that and I think I read that paper at the time. I’ve been pretty interested in orbital/tidal work on the SSN. At the very top level, while it might explain some periodicities I’m not sure about amplitude because of the energy density of gravitational interaction vs the massive reservoir of rotational energy in the solar interior.
I am after the Hale Cycle – the 21.8 ish year tighter oscillation. The Hale magnetic cycle is far more cyclic than the sunspot pattern/cycle that it produces. Can I point you here: https://arxiv.org/abs/2010.06048 to see what I mean. The sunspot evolution (and hence the cycle) is almost like an interference pattern established by the interplay and interaction of the global scale field patterns established by the Hale Cycle.
If someone can convince me that gravitational interplay can influence the generation of substantial magnetic fields, their repeated recurrence at about 55 deg latitude on the Sun, their migration to the equator and eventual destruction then I’m all in.
Gravity changes pressure balance and, as a result, could impact magnetic field (sunspot) emergence – that looks very much like a buoyant phenomena. But I am not sure how gravitational interaction can actually generate the field, that looks very much like its dependent on rotation and internal structure of the star.
Look, I believe that what we see in the sunspot production/cycle is ONLY the imbalance in the global scale fields – NOT anything about the strength of those fields. That is most certainly NOT the dynamo theory orthodoxy. Other phenomena, like the terminator itself, can tell us about what lies beneath: https://arxiv.org/abs/1901.09083. The presence of a Hale Cycle termination point is a very telling thing and scares my community since they want phenomena to be very diffusive and slow. So if you are thinking about subtle effects that can tip the scale such that a magnetic structure becomes more buoyant than its surroundings and rises to make a spot then I have to say that it could be consistent with the observations.
I am certainty not mainstream in my work and try to be open-minded in trying to explain what we see.
oldbrew says:
“A short cycle and a long cycle aren’t going to be converted to two matching cycles by changing the measurement criteria. Or, they would have to be very strange criteria, but even then it’s likely to be impossible.”
central limit theorem
extent
not just gain
no one said 2 cycles is a magic extent
Lions everywhere.
type $0[]well: read gain as grain
As soon as I saw a link to springer I assumed paywall. Turns out there’s none. I’ve looked at the various links. If/when I’ve time I can explore the various claims using methods that subsume conventional methods (mainstream locks parameters to reduce public discourse to artificial). A summary of such explorations would fit on 2/3 of a single page. The next time I planned on looking at solar data (if I’m still alive) was between 2036 & 2046 (because things change so slowly). I’m on another project, so there may be a long delay (weeks, months, maybe longer — unknown and in God’s hands).
Watch 25 Years of Solar Cycles in One Incredible Video
UNIVERSE TODAY 13 DECEMBER 2020
Some of the highlights of SOHO’s revelations are:
— First images ever of a star’s convection zone and of the structure of sunspots below the surface.
— The most detailed and precise measurements of the temperature structure, the interior rotation, and gas flows in the solar interior.
— Discovering new dynamic solar phenomena such as coronal waves and solar tornadoes.
— Revolutionizing our ability to forecast space weather, by giving up to three days’ notice of Earth-directed disturbances, and playing a lead role in the early warning system for space weather.
— Monitoring the impact of solar variability on Earth’s climate.
https://www.sciencealert.com/25-years-of-solar-cycles-in-one-incredible-image
– – –
Any results from ‘Monitoring the impact of solar variability on Earth’s climate’?
FYI I’ve boycotted climate and solar data exploration for ~4 years now (was never deliberate or conscious so I’m going on vague recall). The experts in those fields are dead wrong the 4% of the time when it matters most, even if they appear honorable (at least to naive observers) telling the truth 96% of the time. For entertainment value, a random fact I bet no one would ever guess: I’ve never heard Donald Trump’s voice. How many people do you know who can honestly claim that? I may decide to maintain the boycott until Joe Biden sends me several hundred thousand dollars and an assurance there will be a new left wing party assuring: 1. no lockdowns; 2. freedom to believe in natural variations; 3. secure, reliable, equitable income for all. It’s a monstrous disappointment (we ink we yank) that DT never got ERSSTv3b2 reinstated.
“Any results from ‘Monitoring the impact of solar variability on Earth’s climate’?”
The standard trick is they do a simple regression based on false assumptions diagnosable from the residuals. If I gave the analysis on a Stat 101 exam a decade ago and called the data “frog leaps” and asked my students to assess the residuals plot they would be able to tell you the model assumptions don’t hold. You cannot believe how much those political-savages irritate me (well actually I know you can and you know why my other boycott has remained deliberate and strict).
Scott, many thanks for your considered response and your open minded approach.
We’ve spent time puzzling over potential mechanisms as well as creating pattern-matching models that correlate planetary orbital resonance and solar variation. As well as tides and gravity, there are other considerations which may bear fruit in the search for a physical basis for a planetary-solar feedback.
One area that may appeal to you in terms of relevance to your work is the Z-axis motion of solar system masses relative to the solar equatorial plane. This work was pioneered by Ray Tomes, who calculated that the combined motion of the gas giants above and below the plane could move the solar core several kilometres up and down, relative to the solar surface over the course of a Hale cycle. This doesn’t sound much, but when you consider the pressures involved, Ray said that this could lead to significant meridional surface flows. He also showed some nice correlations in Fourier transforms of solar activity vs Z-axis periodicities. Unfortunately, Ray never published a paper on this, and the thread where he laid it out on BAUTforum is long gone. I’ll hunt for it on archive.org.
Another plausible Z-Axis effect could arise from gyroscopic action applying a torque to the Sun’s spin axis.
In addition to Z-Axis considerations, barycentric motion in the X-Y plane could be affecting rates of energy release from the solar interior. The physical basis for this was laid out by Wolff and Patrone in 2010.
http://www.springerlink.com/content/47023810v8100nk0/
The next few years are going to be exciting. There are a large number of predictions from various groups covering a very large range of cycle amplitudes. Your prediction and ours are at the extremes of the envelope. One or the other of us is going to be seriously wrong. Worse still, if we’re equally wrong, it means Leif gets the last laugh. 😉
That last paragraph literally made me laugh out loud. Not sure either of us could tolerate that! 😉
He and I have something fun cooking on the Hale Cycle. If I can find time to hack it together.
I’ll dig in on the various orbital matters – defo not an expert there!
Did a quick exploration. The suggested relationship gets considerably tighter for about 2/3 to 3/4 of a Schwabe cycle but with phase residual left for hierarchical study (buy pays D-Comm. pose IT $yen).
Comm. reds: This is amenable to rig “Or.” US$ algebraIC treat mint$ because r, theta, and function sov. theta (in pole “Or.” CO[]air done IT$) are all function sov. x & y.
Edge effects and extrapolation: deep exploration needed before commentary (maybe in the next life…)
Out liars: O-Bay general lies? bei. own-D the term innate “Or.” 2 sea[]awei from pays-buy-US D-est time IT$.
most import ant a11y “leave” a11IT tall mist tory
Scott: Let me know when you are able to exactly reproduce the key figure in Donner and Thiel and then I will give you a tip.
This one? https://www.researchgate.net/publication/248852424_Scale-resolved_phase_coherence_analysis_of_hemispheric_sunspot_activity_A_new_look_at_the_north-south_asymmetry
1) I’m a scientist, not a magician…
2) that’s a fairly rudimentary analysis of something for which there’s a non-trivial mapping to field strength. Why do I need to reproduce it?
3) let’s look at a concluding statement: “We suggest that the temporally varying phase relationship between the northern and southern hemispheric sunspot areas is a main contribution to the north-south asymmetry of solar activity” the N-S asymmetry of the areas is a symptom of the physical process underlying the production of spots. I cannot see this the other way around. I have to ask fairly significant questions about the review process for that – IF it has been accepted to publication.
What is the tip?
I’m pointing to the methods, not the subject-matter.
The methods used to produce figure 4 are the second-best thing out there.
If you can get to figure 4 then I can concisely guide you to a generalization of their methods that subsumes conventional methodology as a special-case with arbitrarily locked parameters that artificially limit both exploration and discourse.
You’re well-positioned for a fruitful collaboration with those guys using the more generalized next-level method.
Their calipers take 12 (for monthly resolution) or 365.25 (for daily resolution) measurements of both amplitude and phase (and hence cycle length & deceleration) per year.
For anyone who can replicate their figure 4 I am able to quickly explain how to eliminate their span-bias (the high-frequency oscillation you see in their figure 4).
The next level after that is to generalize their method for variable extent. 4 sharp red “knock, knock” read “Or.” hears the dog whistle: “What curry US IT?” An ongoing source of acute suspicion and deep cynicism (ware IT = in vert-D too tell IT eerie UN nos. ’em!) : I’ve yet too D-tact anyone comfortable publicizing their competent awareness that far.
Talk shh!ops miss story? There may be silent knights ahead. Sea my last comment on this[s]thread. Hear IT is!
NOAA predicts a sunspot peak around mid-2025 — ‘Predicted values are based on the consensus of the Solar Cycle Prediction Panel’.
Long-Term Prediction of SSN and F10 for Solar Cycle 25
https://spawx.nwra.com/spawx/listpredict.html
They list monthly prediction values, and a high/low range.
No way it makes it peak in 2025 – we think mid-to-late 2023. That would be ~21.5 years after the maxima of the same Hale configuration.
binomial smooth (mods see concise comment in filter) :
B2=average(A2,average(A1,A3))
different measures of SCL to answer different questions
Let’s try this in pieces since the filter has a prejudice.
2020 Leamon, McIntosh, Chapman, Watkins
Timing Terminators: Forecasting Sunspot Cycle 25 Onset
=
We acknowledge that since the Hilbert phase wrap is mathematically consistent with the time of maximum rate of change of the underlying quantity, one could construct other methods of terminator proxy calculation, based on derivatives of the sunspot number time series. However, as Fig. 2(d) shows, the Hilbert phase wrap is a more robust indicator, especially when the data is noisy,” with a first or second derivative equal to zero every second or third data point.
=
After digging out that money quote I instantly understood that all the papers linked can be summarized in a 1/2-page for those with no. time too waste.
1 minute spreadsheet calculation says ITa11!
Repeat binomial smooth of first differences (with result equivalent to a Gaussian smooth) gives dates predicting Scott’s dates with r = 1.0.
My “figure 2D” is a scatterplot of Scott’s “terminator dates” vs. “the 1 minute spreadsheet date”. Guess what feature appears to be D-C[11ew:RO]BUST!line (strait!T!wan).
Concise D-missed stuff IC shh! “O-enough!” Term.in[NAT]OR:we11-done.
_
Upfront filter for caste, “a side” didn’t like “the titles key” words:
co[i]ns[n]icely TERM.in at OR.[$0]WELL
Two papers suggesting a long-ish cycle 25, in theory.
A phenomenological study of the long‐term cosmic ray modulation, 850–1958 AD
K. G. McCracken F. B. McDonald J. Beer G. Raisbeck F. Yiou
First published: 10 December 2004
(2) the heliomagnetic cycle (∼25 years at times of low solar activity) continued throughout the minimum
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2004JA010685
Citations:109
– – –
The change of the solar cyclicity mode☆(2015)
A.G.Tlatov
Abstract
Our analysis of groups of sunspots since the year 1610 till indicates that the Gnevyshev–Ohl rule (GO) displays cycles of inversion with the period of 200 years. The latest inversion occurred in the Hale double cycle 22–23. Due to that, in several subsequent double cycles the odd cycles should be weaker than their preceding even cycles.
https://www.sciencedirect.com/science/article/abs/pii/S027311771400386X
mods: demystification is in the filter — looks like we may be heading for another sharp change of course like last time that happened (last XR thread — & the time before that was early Jan. 2020 right before the sense soar……)
FYI: Readers likely have no idea what a monstrous can of worms this thread has opened up.
Readers will recall that solar cycle deceleration (SCD) and sunspot integral account for 83% of global average SST. The leftover is almost all ENSO with a small amount attributable to BDO (bidecadal oscillation).
The wavelet tachometer I refined (starting from Donner & Thiel) to measure SCL & SCD can be systematically biased by prefiltering second-order-central differences with gaussian smoothing (at 1/2 Jupiter-Earth-Venus period) to exactly measure Scott’s T-T SCL (terminator-to-terminator solar cycle length).
If you understand how different methods relate to one another mathematically (and how some methods — including systematically phase-biased methods — are subsumed by more generalized methods, including unbiased methods) these are the kinds of things you can intuitively KNOW before verification (and after with r=1.0 measured).
This is all very interesting. There’$no. doubt I could help sharpen methods. Logistics problem: I was already swamped on another project that is many orders of magnitude more important.
mods: “O say Can. EU sea” acute reason for left-write cynicism negotiating Joe’s “unity” filter?
Hi Paul. There are no comments pending or in trash or in spam folders.
TB: 2 comments completely vanished without giving a #comment in the link. A comment restructured into 2 out-of-order pieces now appears above after 1 passed and a moderator released the other.
Let’s move on…
Triggers
This is a cautionary note about data interpretation.
Readers will recal that USGS has recorded thousands of geological events coincident with lunisolar events even though lunisolar cycles do not set all of the underlying background variations of geological susceptibility to trigger.
If I repeatedly hammer a brittle substance it doesn’t break on every hammer blow, but it’s a safe bet that if it’s going to break it will be on a hammer strike (compared with say well-timed attempts to break with assertive discourse).
I have a more clear example from firsthand observation. I’ve shared this before.
Where I grew up there was a large bay about 5km wide on an estuarine inlet from the sea where 2 large rivers merged.
Many years the ice would go out in a single day — a WINDY day.
Structures that were securely frozen in place for months were SUDDENLY ripped apart.
You don’t have to be the sharpest reader on the net to realize what was different on that spring day than a day in the middle of winter with strong (maybe stronger) wind.
Defining an ice-presence variable as 1 for presence of ice or 0 for absence of ice there were many years with a QUICK -1 change on a windy SPRING day after the ice had become what locals called “rotten” from seasonal warming (including from rain that would have been snow a month earlier) in the annual cycle.
Let us take care interpreting the timing of trigger events.
We have a versatile toolset with flexible calipers adjustable to measure “trigger-blasted bias” into phase deviations from central limit.
Sensible phase decomposition (in the temporal sense) is feasible to explore across a range of scales the triggers ordering superficial global partitioning through local deviation (in both temporal and material senses) from hierarchical central limits (remains true with “temporal” extended to “spatiotemporal” in this case).
Poll Lore CO[]air Don IT
From my comments quantitatively alert readers will realize how to increase the resolution of Scott’s study to monthly (or daily) while maintaining a perfect correlation with his decadal estimates.
Buy extension a few will realize what curry US ITees $implied.
SILSO: Provisional international and normalized hemispheric daily sunspot numbers for November 2020
Click to access monthlybull202011.pdf
South getting active in late November, last 4 days: 69, 74, 83, 74.
Preliminary check done — a few highlights:
The relationship measured anywhere on the wavelet phase dial (doesn’t matter whether min-min, max-max, T-T, decline-decline, or whatever in between) is tighter than the Hilbert T-T (terminator-to-terminator) measure reported in the paper.
TB will love this news:
11% increase in r^2 by moving 11.07 grain calipers -Φ cycles from T-T with 11.86 extent. This combination was the optimal setting — delightful insight discovered with pure objectivity.
Compare and Contrast
Today’s new notes focus on aggregation at decadal grain and extent.
MULTIDECADAL SCL has the same cycle grain but the gaussian extent is ~66 years, measuring how the partitioning between cycles shifts to balance out over a number of cycles — i.e. central limit theorem applied to phase …meaning we can’t take a measurement for 2020 until 2053.
–
Back to exploring Scott’s work, an informal look with wavelet calipers shifted -Φ cycles (royal blue) from T-T — alongside T-T Hilbert (light blue) :
The bad news and the good news mixed together:
The model residuals have systematic features.
Forecasters: These demand careful study.
In particular, note clusters with backward slants on the scatterplot above.
In good talkshop fashion: Leave a little mystery.
The question both amateurs and pros can explore:
What accounts for the groupings?
If Phi Halve a CO[2]Will…
a11 IT a11 mist[]0R ea. sea$We11….
=
….win the still see con spires an arm “Or”
and her sol UN and AB “Or” tied cure rants BRI-D tie NY monsters
truce ail in’s D_D
hawk ware dawn scent — and the Fearstan numb a11 is J(ET tie)SUN
lags fear US sly pump in they’re stiff “green gall” ops
and heads bob up:
ploys Delhi caught pause comm. sent-in-mute nos. dr. ill ag UN NY
cure fail eerie find
and sea L[ ]D [O] vert
=
— the Do[]0R $heirs 11 attitudes
“this is the strange est…
…lie phi half ‘eve Or.’ 11eave-D” 00R$ “weight in 4 EU too ‘tally’ what went wrong”
classy Φ
centering “terminator” window at 0 the highlight is not ±1/2 (which is ~T-T but in gaussian sense)
rather the highlights are
from +Φ/2 to +Φ/2+Φ/4 = +3Φ/4
best correlation is with current cycle
peak correlation at +Φ/2 which is ~min-min (but in gaussian sense)
everywhere else best correlation is with next cycle
peak correlation at -Φ/2 which is ~max-max (but in gaussian sense)
this is elegant, beautiful nature illuminated by beautiful instrument
mods: filter
alert readers will realize another interpretation of these measurements:
min to max is 1-Φ cycles
max to min is Φ cycles
T & ±1/2 are midpoints
the right answer to the riddle left above:
Φπ radians = Φ/2 cycles
Optimal Morlet Wavelet Settings
1/12 = resolution (i.e. monthly data)
11.0696157491919 = grain (Bollinger 1952 with Seidelmann 1992 orbital elements)
11.8626151546089 = extent (Seidelmann 1992 orbital elements)
44.2784629967674 = span (Bollinger 1952 with Seidelmann 1992 orbital elements)
-0.618033988749895 = phase
-0.849725938710943 = r
0.722034170918193 = r^2 — i.e. ~72% of variance explained
Grain scales period of sine & cosine waves.
Extent scales width of Gaussian envelope (local bell scaling sin & cos).
Span/resolution sets how many points are in the wavelet.
No sunspots…
To summarize what we just learned about long-run solar cycle shape deviation from uniform (in Gaussian central limit), here’s a spreadsheet recipe.
Make column A increase in equal steps of something like 1/100 — e.g.:
0.00, 0.01, 0.02, etc.
Then put this formula in cell B1 and copy/paste down column B:
phase in π radians
=2*(MOD((ATAN2(COS(2*PI()*A1),SIN(2*PI()*A1))/PI()+(-COS(2*PI()*A1)+1)/2/AVERAGE(1,5^(1/2))^2),1)-MOD((ATAN2(COS(2*PI()*A1),SIN(2*PI()*A1))/PI()+(-COS(2*PI()*A1)+1)/2/AVERAGE(1,5^(1/2))^2)/2,1))
For column C:
amplitude
=-COS((ATAN2(COS(2*PI()*A1),SIN(2*PI()*A1))/PI()+(-COS(2*PI()*A1)+1)/2/AVERAGE(1,5^(1/2))^2)*PI())
Beyond amplitude and cycle length, shape is just another parameter to modulate (held constant in the above equations and graph).
Note well a course for next level exploration.
I’m not pointing out the things I’m refining as I go. There simply isn’t time.
Suggested exercise: Make column A step sizes a function of instantaneous frequency (before shaping B & C).
OB: Remember that Scott’s forecast is based on a forecast of a measurement that won’t be finalized for a few years. I’m sure you’ve also considered (given the review links you posted above) how the model might change given detailed records of a long deep minimum (which we don’t have) for careful study.
…but I suspect you are correct to assume discourse trouble-mmmaking (misunderstanding, misinterpretation, and misrepresentation) will be a popular response to Scott’s model.
I’m not sure if RJ Salvador is still modeling, but we have a lot more insight available now than when he developed his last model. I’m not finished telling the “perfect” so-called “60 years” s(496)tory. (This thread was a diversion — worthwhile, but at a cost.)
For anyone (at least anyone sensible) trying to use 1 of my graphs above to check Scott’s work: The Gaussian window with 72% r^2 is centered roughly on the midpoint between max & max, meaning there’s still edge effect for years IF you try to extrapolate …but not if you’re reviewing the record with the benefit of precise, sharp hindsight — such is the nature of central limit theorem ….at least for those willing and able to understand it (…and that leaves out what few eXpeRts might curry US sly be $0[]well incline-D to mislead).
Alert: Above there is 1 major error in my comments — where I contrast models for “next” and “current” cycles. The boundaries are correct, but when time permits I’ll need to clarify how the definition of “current” and “next” becomes a philosophical consideration when comparing methods based on different aggregation criteria. It’s possible to generalize to a next-level to get around this obstacle which is caused for example by some wanting to do things like number cycles. Like I said: You can’t believe what a can of worms this thread opened. A LOT of exploratory advances are feasible with the insight that’s sharpening.
If Joe Trump or Don Biden give me several hundred thousand dollars of secure, reliable, stable research funding I’ll develop a visualization tool like nothing anyone has ever seen.
Jupiter Earth & Venus (Didn’t) Meet with 22 Taxis to Review the Numbers
Many stars of freedom MET (not) too D-Clear:
“Right: Now’s the Beat Period.”
1.59868955949705 = (1.00001743371442)*(0.615197263396975) / (1.00001743371442 – 0.615197263396975)
CRies Toll Bell [L]ove ID “Free” Dem.
Santa be ware familly sleigh. “Terre Or.” “well” lost WHO gather UN D-air curve “Or” sov. Joe‘s Am. UN IT shh! UN = fear “Jan. YOU weary” luck D-own $0[]well.
Harmonic Mean = 2*(Axial Period)
0.761766209372163 = (1.00001743371442)*(0.615197263396975)/AVERAGE(1.00001743371442,0.615197263396975)
One-Sided Split Harmonic Mean Orientation (Bollinger 1952 “knew more” IC a11y)
0.814040387734912 = (11.8626151546089)*(0.761766209372164) / (11.8626151546089 – 0.761766209372164)
0.407020193867456 = 0.814040387734912 / 2
0.203510096933728 = 0.814040387734912 / 4
44.2784629967674 = slip(1.59868955949705,0.814040387734912)
22.1392314983837 = slip(1.59868955949705,0.407020193867456)
11.0696157491919 = slip(1.59868955949705,0.203510096933728)
Broke UN Full Lock the Fierce to shut the CO[2]IDoor[well]
Manage con tact when it matters: DON’T meet with anyone for the next 2 weeks. Weave human rights to beat WHO’s at UN. Centuries later “The January Excuse” links his story to Joe1984 WHO “won” “free” dem[ise].
I may have missed something in the above discussion so apologies if I have. Just a thought for everyone to ponder. Given that the earths magnetic field is weaker now than at any point in modern history, is it not possible that even an ordinary solar cycle could be a greater danger to us simply because our shield is weaker, making the discussion over how strong the solar cycle is rather academic and missing the point? Just a thought. The New World Order will rely totally on their ability to control via digital tech that may have a huge problem staying up.
Another thought is our closest star, and some of the outer planets have seen changes to their own behavior/weather patterns. If like me you believe that electromagnetic connections control “everything” then we could be in for something life changing within the next 100 to 200 years that may make the Carrington even look like a day at the beach. Again just a thought.
Given all the work that has been done that shows radio dating is flawed, it follows that our timelines cant be relied on and could be way out. The Younger Dryas was about 14,000y ago, and all the established theories look at a mechanism on earth as its cause. What if the cause is a galactic electro-magnet mechanism that can effect everything, our magnetic field, the sun, all our weather on all the planets. What if it can create an extinction event, perhaps one that leads to an ice event. What if it can cause instant ice, like that that must have hit the mammoths. There is a pattern of these events every 12 to 14k years in the geology, even if our timelines are out a bit.
TB:
“modeling the evolution of solar activity, using planetary orbital resonance patterns”
Quadrupole heliocentric planetary geometry orders the actual timing of each sunspot maximum, this shows cycle 25 maximum occurring in 2025, and with a likely second peak in 2027.
Researchers forecast ‘peak sunspot number for solar cycle 25 is likely to be 62 (±12)’.
– – –
A 20 year decline in solar photospheric magnetic fields: Inner-heliospheric signatures and possible implications [2015]
We report observations of a steady 20 year decline of solar photospheric fields at latitudes ≥45° starting from ∼1995. This prolonged and continuing decline, combined with the fact that cycle 24 is already past its peak, implies that magnetic fields are likely to continue to decline until ∼2020, the expected minimum of the ongoing solar cycle 24. In addition, interplanetary scintillation observations of the inner heliosphere for the period 1983-2013 and in the distance range 0.2-0.8 AU have also shown a similar and steady decline in solar wind microturbulence levels, in sync with the declining photospheric fields. Using the correlation between the polar field and heliospheric magnetic field (HMF) at solar minimum, we have estimated the value of the HMF in 2020 to be 3.9 (±0.6) nT and a floor value of the HMF of ∼3.2 (±0.4) nT. Given this floor value for the HMF, our analysis suggests that the estimated peak sunspot number for solar cycle 25 is likely to be 62 (±12).
Abstract: ‘https://ui.adsabs.harvard.edu/abs/2015JGRA..120.5306J/abstract’
PDF: ‘https://arxiv.org/pdf/1506.03589.pdf’
Caption: ‘Observed decline in solar magnetic field strength, starting roughly in the mid-1990s. The grey dots are the data points while the filled blue circles are annual means.’ Source: Author provided
From this article:
Studies of the Sun Suggest Our Star’s Magnetic Field Is Becoming Weaker
04/11/2020
https://science.thewire.in/the-sciences/solar-magnetic-field-sunspots-solar-cycle-25-maunder-minimum/