Readers may recall the contentious discussions that occurred on this thread a couple of weeks back. Both Willis Eschenbach and Dr. Leif Svalgaard were quite combative over the fact that the model data had not been released. But that aside, there is good news.
David Archibald writes in to tell us that the model has been released and that we can examine it. Links to the details follow.
While this is a very welcome update, from my viewpoint the timing of this could not be worse, given that a number of people including myself are in the middle of the ICCC9 conference in Las Vegas.
I have not looked at this model, but I’m passing it along for readers to examine themselves. Perhaps I and others will be able to get to it in a few days, but for now I’m passing it along without comment.
Archibald writes:
There is plenty to chew on. Being able to forecast turns in climate a decade in advance will have great commercial utility. To reiterate, the model is predicting a large drop in temperature from right about now:
David Evans has made his climate model available for download here.
The home for all things pertaining to the model is: http://sciencespeak.com/climate-nd-solar.html
UPDATE2:
For fairness and to promote a fuller understanding, here are some replies from Joanne Nova
It can be seen that the strongest ionization occurs at a height of 15 to 30 km.
http://www.temis.nl/profiles/gome2_orbits/daily/images/recent_OzoneProfile_Nadir_IIS_2.png
“the model is predicting a large drop in temperature from right about now:”
Doesn’t look bad:
“Poor man’s polar vortex to make shocking summer return in eastern U.S. next week” By Jason Samenow July 10 at 10:35 am
http://www.washingtonpost.com/blogs/capital-weather-gang/wp/2014/07/10/poor-mans-polar-vortex-to-make-shocking-summer-return-in-eastern-u-s-next-week/
Walter Sobchak look at the distribution of ozone, color brown.
http://www.cpc.ncep.noaa.gov/products/stratosphere/strat_a_f/gif_files/gfs_t100_nh_f00.gif
You can also see cold in Europe. Yours.
JoNova nitpicked and whitewashed: “lubos thought the notch was “on Earth” when we suggest it’s on the Sun, I demonstrated he did not know what he was talking about and his post was irrelevant.” and “Lubos had a bad day.”
Yet it matters not one wit this nitpick to both the headline and the crux of Lubos’ post. Where the physical causality is claimed is moot when you merely call the interaction Force X (“from outer space”). Is it any more preposterous that an 11 year signal is perfectly delayed by an 11 year lag within the Sun itself? If it’s lagged exactly the same duration of the cycle claimed then the lag cancels out to no lag at all except that you can demonstrate peak to peak or valley to valley uniqueness to only the former instead of the current swing. His post was concerned with the algorithms being infinitely tuned and therefore meaningless as a model since by their very nature signal matching algorithms will match nearly any two signals together, which only works well for isolated causally entrained electronic circuits. You are not admitting this clearly at all to explain how your matching system is in any way unique rather than purely arbitrary. Lubos, a former Harvard professor, mentions that he, a string theory level mathematical savant that places him among the top hundred million people most mathematically able, stated that he spent hours delving into the nature of your system, having had access to the details. He acted as a peer reviewer and you failed to pass peer review and now you self publish with magic parameters that create a classic black box. The work appears good as an extended thought experiment at best, and might be used in a textbook to demonstrate the foibles of wanton wiggle matching in the absence of actual physics. The medical field is still chock full of such data mining debauchery as Aussie John Ray pointed out for years on his Food & Health Skeptic blog.
I am perfectly capable of separating the obscure details of your claim with the central force of Motl’s argument against the overall exercise. That WUWT has already had an ongoing spat with career wiggle matchers at the Tallbloke’s Talkshop blog to the extent of moving his link in the blogroll here into the “Transcendent Rant and way out there theory” category means you were likely well aware that your own wiggle matching might get you put on probation here, and at this point your increasingly personal defense of it that focuses on minute details as if that cancels out the major structural and profound criticisms means you are rapidly heading into transcedent rant category yourself. The addition of a massive nuclear testing correction into a pattern matching exercise fails to pass the laugh test since it further obscures the algorithm from internal testing attempts using early/late comparison or arbitrary test data input.
Monckton came out swinging about a mere software setting that created an automatic infill over a gap in Leif’s data, yet now defends your own manual addition of a guesstimated extension so he clearly biased in this affair, but again such quibbles gloss over Motl’s and Willis’ main point that black boxes need be checked internally not turned into highly publicized bets on the future as if a future match proves you right, which it very much does not since you already have 33% odds in your favor towards obtaining a plunge instead of a further lull or a new peak.
You are jumping on techicalities in a way that speaks louder and louder that you in maverick fashion willfully refuse to address the Science 101 concerns about arbitrary wiggle matching which as if such concerns are obscure rather than fundamental.
See jetstream.
http://earth.nullschool.net/#current/wind/isobaric/250hPa/orthographic=-69.41,33.80,419
“Rogueelement451 says:
July 10, 2014 at 2:17 am
Willis ,Mosher et al founders of the lost snark..
Most posters at Jo Nova are waiting for expert analysis of “the model” and will continue to do so.
Where you failed ,along with others ,was demanding ,insisting like a spoiled brat ,that you be given data NOW! When it had been stated all along ,that ALL data was going to be released in due course.”
huh.
I dont demand the data or code now.
It is simple. Without data or code it is not worth the time I have already spent
Given the Little Ice Age and various other warm and cold periods in the past, I believe there is evidence that there is a link between temperature and periodic changes in the sun. Therefore, I believe that a reduction in solar activity is going to cause a reduction in temperature. How much of a reduction and how long it takes to happen are the questions.
For the model, I think the notch filter idea is interesting – but rather than thinking in terms of a notch filter, I think it should be a simple delay filter. In other words, between the change in TSI and the change in when a measurable change in temperature are seen will be the delay time.
For the delay time, I think the problem with coming up with a specific delay time is that this should vary depending on various other natural cycles, currents and etc. In other words, conceptually, it makes more sense to me that there is not a simple fixed delay. The 11 year sunspot cycle will complicate it further given TSI varies during solar cycles.
I will be curious to see more information released about the model. The entire discussion hear and on JoNova’s page have been very interesting. I think due to some of the comments, more details will be released about the notch filter model. I also think a number of people will come up with their own new and improved models based on what everyone learns from the discussion.
PROPOSAL: Time for an “Evans-Svalgaard wager” where Anthony puts a count-down clock in upper right corner of WUWT, ditto JoanneNova, ditto ClimateAudit, ditto … counting down to date at which David Evans’ model says XYZ will happen. Call it a beer wager akin to the “Simon–Ehrlich wager”, except call it the “Evans-Svalgaard wager”, maybe rig up a PayPal thing to let the audience participate with monies going to favorite charity of winner. Imagine there would have to be some negotiation of the terms: how measured, by whom, plus-minus predicted measurement, plus-minus time, who measures, exceptions like asteroid strike exceeding certain size, volcano eruption exceeding certain size, nuclear weapons exchange, man-made aerosol discharge by wealthy loony person delusional about saving world, war, etc etc etc)
MY TWO-CENTS:
($0.01) I have seen the TSI decline claimed by David Evans in Leif’s own PNG data graph. You too can see the TSI decline if you focus in on the 2003-2008 time frame of Leif’s own PNG data graph. There is a TSI decline shown both here … http://www.leif.org/research/Monckton-Flaw-3.png … and here … http://www.leif.org/research/Real-Trend-TSI-since-2003.png … NOTE: no claim made by me whether or not this obvious TSI decline is significant, meaningful or otherwise. Only saying there is an obvious TSI decline from 2003-to-2008. Not saying there is a TSI decline from 2003-to-2014
($0.02) I have seen the (never) missing model. Willis Eschenbach simultaneously claims David Evans has “not released … the model, ” then a few paragraphs later claims David Evans “posits an incredibly intricate model,” embedding a PNG schematic of the model block diagram in his own comment here … http://wattsupwiththat.com/2014/06/28/a-cool-question-answered/#comment-1671519 … Although I make no claim that the David Evans’ Electrical Engineering block diagram representation of a model is complete; however, it is very similar to layout & construction to what the Controls Scientists (i.e. Matlab Nth degree warrior junkies) I have previously worked with would call “a model”. Ditto it being in a format that Electrical Design Engineers would use for constructing control systems and filter networks. David Evans obviously looses princess points for using Excel instead of Matlab to construct his model (commence Excel vs Matlab flame war … or should it be an Excel vs R flame war … hmm, maybe this is the real real real issue, a religious war over preferred data analysis software, I don’t want to slog through someone’s spreadsheet do I???, fate worse than death. We are truly fortunate at least all agree EMACS is the world’s absolute best text editor ;-);-);-).
FLAME WAR REVIEW: Flame wars are immensely entertaining to the audience; however, neither side has covered themselves with glory. With printouts of the various threads, both here at WUWT and over at JoanneNova, I can highlight the “tone” that various people have inadvertently used to stoke the fires: David Evans, Leif Svalgaard, Joanne Nova, Willis Eschenbach and others would all suffer hi-lighter marks on their comments. Y’all might have to go to South Park’s “Death Camp of Tolerance” (Season 6, Episode 14) … http://www.southparkstudios.com/full-episodes/s06e14-the-death-camp-of-tolerance
Bob G you are right on. This prolonged solar minimum will be no different then the previous ones in that the temperature trend for the globe will be down.
TSI which was off by.15% during the recent solar lull( 2008-2010 time period) will be off by at least that much if not more going forward into this decade.
It is not just the primary solar effects but the many secondary effects from solar variability that will impact the climate ranging from more clouds due to an increase in galactic cosmic rays, to a more meridional atmospheric circulation due to changes in ozone distribution/concentrations to an increase in volcanic activity due to extended periods of very quite solar activity.
I posted earlier and many times in the past a set of solar parameters and duration of time these parameters need to have in order for the sun to achieve a meaningful effect upon the climate.
Willis Eschenbach says:
July 9, 2014 at 10:14 am
joannenova says:
July 9, 2014 at 12:34 am
Thanks to Anthony for updating the post
It’s getting a little rambunctious here so I don’t want to get more involved than to say the output on your graph in the indicated post is exactly what I would expect to see from a low pass filter — and yes indeed it does “clip the fast peaks” and “fill in the valleys — but the actual content (energy?) remains constant — mostly. There are other issues with boxcars for low pass filters as the frequency response is not uniform — but since the peaks and valleys are relatively uniform it’s not likely an issue here…
The signal out indicates where to find the peaks and valleys and depending on the time constant will “hold” at a peak or a valley as long as the next pulse occurs withing a set time.
You may want to request an electrical engineer in the group to comment on the validity of the technique — for various purposes. On the face of the issue I see no problems with that output being used in certain circumstances.
…and yes padding at the end or beginning of a series of data points is (can be) a valid technique — otherwise you can get bogus output for the adjacent data. The only thing I would want to see is the justification for the padding… I would likely have run it through GNU Octave — and got similar results with the built in filter algorithms based on what your graph shows.
Since I don’t want to participate in the abuse, and I have enough to keep me busy for the next few weeks, may I again suggest an electrical engineer with a thick hide and a strong capability in filter design comment on this issue… and if I’m wrong I’ll need an even thicker hide.
Cheers!
Maybe I used some bad language as I have a comment in moderation — or maybe you are (wisely) moderating this thread.
Cheers all! This thread looks like a winner!
lukemullen indulges in the betting game: “PROPOSAL: Time for an “Evans-Svalgaard wager” where Anthony puts a count-down clock in upper right corner of WUWT, ditto JoanneNova, ditto ClimateAudit, ditto … counting down to date at which David Evans’ model says XYZ will happen.”
But if temperature plunges in what possible way does a wiggle matching algorithm become confirmed by it being that it’s a theory only in being called one rather than actually being one? Or if temperature soars again why not just tweak the parameters again and claim victory? That’s why internal testing is so important, for if the link to solar activity is real it should robustly survive all manner of testing, with future testing being rather irrelevant in comparison due to the 33% random chaotic chance of yet another plunge in temperature. And a real solar link certainly wouldn’t require such complex mathematics but be robust to different and possible much simpler formulations. If such a serious solar influence exists then there is no delay except in the inappropriate mathematical formulation, unless an actual *delay* is demonstrated in variation shape of solar output eleven years ago and temperature today. For now it seems the 11 year delay is but a mathematical kludge to escape falsification by non-causality in time, all to escape utter lack of solar correlation in the first place. The laugh test has been applied and this “model” that willfully lacks internal testing in favor of black box betting failed it. If the Evans really want to place a bet, let them publicly short sell green energy companies in the real stock market, eh, for indeed only a steep plunge will scare off their existing investors.
WillR said in part July 10, 2014 at 3:06 pm:
“ …and yes padding at the end or beginning of a series of data points is (can be) a valid technique — otherwise you can get bogus output for the adjacent data. The only thing I would want to see is the justification for the padding… “
Good question WillR,
But before asking how to justify padding, can you justify the smoothing in the first place! It depends on how much you admire hiding actual data and fooling yourself and other people! (Here we are talking specifically about looking for patterns in relatively clean data – not the removal of true noise.)
Under what circumstances can you justify padding the ends. Well – NEVER! Unless you prefer a guarantee of a “bogus output” to the mere possibility of one.
Svalgaard still goes on about “900 days of fabricated data”, despite clear and repeated variations on this quote from Joanne (emphasis mine):
Svalgaard:
Not spin – statement of fact = NOT USED IN THE MODEL (read my lips!).
Are you being deliberately obtuse, or does it come naturally?
Leif, that is funny as hell! I have felt that level of frustration.
Mike Jowsey says:
July 10, 2014 at 5:24 pm
“The largely irrelevant 3 dots are clearly listed on the graphs as an “assumed average” extension. They have no impact on the model (they are not used in it).
The important ‘spin words’ are ‘largely irrelevant’. If not used, why are they there? And they are used to calculate the smooth. And whether the fabricated data are used or not is actually irrelevant. They serve to fool the lay reader into accepting the graph and the [non-existing] sharp drop. Typical tricks in the PR-business.
Mike Jowsey says:
July 10, 2014 at 5:24 pm
“They have no impact on the model “
Furthermore, what has been ‘revealed’ [like tablets from the gods] is no model, just a means to run a model. The ‘model’ would be have to derive [and justify their existence] the 11 parameters that make up the ‘parameter set’. That is where the potential science would be. What we have now is no science, just black-box’ curve fitting. But, of course, the goal seems to be to run a PR-stunt rather than science, and in that respect they seem to have whipped their adherents into the required mouth-foaming frenzy.
NicFromNYC asked: “But if temperature plunges in what possible way does a wiggle matching algorithm become confirmed by it being that it’s a theory only in being called one rather than actually being one?”
Luke Mullen replies: In the same possible way that the commodity price plunge of the the famous Simon–Ehrlich wager confirmed Simon’s “cornucopian theory”. Note that I use the word “confirmed” loosely, aware of the ongoing debate about the significance of the Simon-Ehrlich wager since its 1990 payoff in favor of Simon.
NicFromNYC wrote: “If the Evans really want to place a bet, let them publicly short sell green energy companies in the real stock market, eh, for indeed only a steep plunge will scare off their existing investors.”
Luke Mullen replies: But we the audience would miss out on the open-source collective negotiation of the terms, seeing a counter counting down, watching a Skype Beer drinking event, seeing a virtual check being sent to [fill in name of favorite Evans charity or Svalgaard charity]. OBVIOUSLY Evans could still short appropriate industries; however, then we loose the benefit of the count down clock, the community spirit of opposing camps, watching the beer consumed on Skype, etc
Examples of wager escape clauses:
(a) any asteroid impact with Torino Scale 8,9,10 occurs
(b) Any one volcanic eruption with Volcanic Explosivity Index of 6 or higher (Mount Pinatubo)
(c) More than XYZ volcanic eruptions with Volcanic Explosivity Index of 5 (Mount St. Helens)
(d) Carrington Event
(e) others TBD by the community …
LSvalgaard says: The important ‘spin words’ are ‘largely irrelevant’. If not used, why are they there? “…to make the fall slightly less dramatic by indicating it is ending”, as Joanne said.
They have no impact on the data, it is merely a rounding by assumption, as indicated in the text box and by using a different colour and line type. Why does it cause you to foam at the mouth so much?
Mike Jowsey says:
July 10, 2014 at 7:01 pm
Why does it cause you to foam at the mouth so much?
Scientific fraud can have that effect on me. A PR-stunt with all attendant tricks and fabrications not so much.
The model does not forecast how the suns activity will behave in the future, because there has been no scientific mechanism put forward for this.
The premise that solar activity effects earth is especially interesting because of the fact that the sun drives our climate on various time scales, if I can put the understanding of the suns output in an easily understood way, it will be this;
What is the maximum potential of solar activity output?
What is the minimum potential of solar activity output?
You see? there’s a very straight forward question being asked!
As an Engineer David Evans has put forward his honest (for the lack of words) analysis of the sun-earth relationship, there is no issue with his engagement in understanding this area of science, I believe in his sincerity of his exploration of it..
I also agree with Leif, ideas can be “right for the wrong reason”, even with the best intentions. It should be acknowledged that David Evans may in fact be developing a tool for correctly analyzing solar activity, and further development of scientific tools by engineers for this area of science should be appreciated!
Sparks says:
July 10, 2014 at 8:12 pm
What is the maximum potential of solar activity output?
http://www.leif.org/research/Report-on-Extreme-Space-Weather-Events-2014.pdf
What is the minimum potential of solar activity output?
nothing at all.
cold fusion deja vu.
Moral of the story: Skeptics are very rare. Being against global warming does not make you a skeptic any more than it makes you a scientist.
lsvalgaard says:
July 10, 2014 at 8:17 pm
Big laugh there Leif, because you don’t know yet…
There is a particular area of your “blood sport” I’ve been looking into recently I’ve found very interesting.
😉
The steady state druids have not a hint of an explanation:
http://www.zerohedge.com/news/2014-07-10/polar-vortex-back-middle-july
Just put out the fires of disbelief and blind them with BS. That’s the game of the National Mascot, so who cares?