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
Leif
‘The spreadsheet is very disorganized and difficult to navigate [simply because of all the stuff crammed into it]. The spreadsheet works with ‘parameter sets’ that may be used to run the model. As we all know ‘with five parameters I can make the elephant wiggle his trunk’ and in Evans’ model there are many more. The important bit is to construct the parameter set from the solar, temperature, and ‘volcanic’ data. That should be the model, but the spreadsheet does not [as far as I can see] do that, so is not really ‘The Model’, just an application of the model. Therefore is no help at all.”
yes,
to explain the folks.
Lets suppose I give you a pile of data
And you come back and say my model is T = f(x,y,z,d,f,t,y,e)
And then he gives you a spreadsheet of the model.
That’s NOT what we want to see. We want to see the SCIENCE BITS.. that is
how did he derive the model.
That is where the real issue is because the model was derived from questionable data and the
thing you want to check is how sensitive is model CONSTRUCTION to the data used.
Without that stuff the spreadsheet is useless.
It looks like the next 6 years should provide a good test.
Any error estimates for the model values?
Anybody knows if the model predicts the pause when given input data preceding it? That would be a fairly good test of its validity. As it is now, it looks as a high-order extrapolation with its usual divergence characteristic. David Evans is an intelligent man so my guess is that this is something he must have done but I would like to see it in order to appreciate the 2015 prediction.
a notch filter attentuates. it does not amplify. so what is proposed is equivalent to a ‘forcing of coolness’.
i do not see any requirement for a lag implicit in a notch filter and would appreciate any explanation of how i may be in error.
also, i can imagine a very simple mechanism that could respond with an 11 yr notch, namely extrasolar cosmic particles that are driven away with more solar activity and that return like the tide when there is less.
anyway- this is an interesting bit where there has been little but reruns for a long time.
Leif Svalgaard doth protest too much … about everything. He’s practically living on these blogs, seemingly trying to shoot down everything in sight. Seems like the “not invented here” syndrome at work here.
Sorry, but his behavior the past week or two sends up red flags for me.
Let the current temperatures in Australia. It does not happen without a reason.
http://earth.nullschool.net/#current/wind/isobaric/1000hPa/overlay=temp/orthographic=136.21,-28.12,1458
I cannot say anything about the model, however Hadsst3 is averaging 0.392 after 5 months. The record is 1998 at 0.416. In the last two months, Hadsst3 was 0.478 and 0.479 respectively. Should these numbers repeat for the next two months, Hadsst3 would at least temporarily rank in first place after 7 months. And warm water takes a long time to cool down. The next few months will be very revealing, with or without an El Nino.
If the climate science thing doesn’t work out, you could make a good living conducting MS Excel seminars.
There is money to be made for your favorite charity if you can find someone who will bet this silly ‘model’ accurately predicts temperatures over the next 6-7 years.
Anthony says:
“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.”
Anthony, Please explain your reasoning for this statement.
Leif says:
“The spreadsheet is very disorganized and difficult to navigate [simply because of all the stuff crammed into it]. The spreadsheet works with ‘parameter sets’ that may be used to run the model. As we all know ‘with five parameters I can make the elephant wiggle his trunk’ and in Evans’ model there are many more. The important bit is to construct the parameter set from the solar, temperature, and ‘volcanic’ data. That should be the model, but the spreadsheet does not [as far as I can see] do that, so is not really ‘The Model’, just an application of the model. Therefore is no help at all.”
And this:
“To me, a useful spreadsheet would be one where you give it 5 inputs: lists of solar, temperature, ‘atomic’, and volcanic data, and a range in years, and the spreadsheet calculates the parameter set, which you can then use in another spreadsheet to calculate the hindcast and the forecast. That would be science.”
Leif, if it was so simple, why didn’t you do it already and solve the problem? Your first post says five parameters and you can make an elephant wiggle its trunk. And then, amazingly, you state it would “USEFUL” to use your “5 parameters” to both hindcast and forecast???? Your lack of logic is unbelievable.
And please explain how you can forecast “volcanic activity” and its impact on changes in the climate. I’d love to see that crystal ball.
Your drive-by sniping on WUWT, in my humble opinion, only reduces your professional credibility.
El Niño will not develop.
http://earth.nullschool.net/#current/wind/isobaric/1000hPa/orthographic=-122.19,-4.98,553
So Werner, this years temps are .478C warmer than an arbitrarily selected anomaly baseline perioid. It can be argued that the anomaly period was at the cold end of an indeterminate cycle.
I have yet to see justification that the anomaly period used to compare todays temperatures to is either 1) normal temperatures for the planet or 2) “the correct” temperature for the planet.
As a result, I find all the tracking and comparisons of month to month temperatures as laughable if not absurd.
@Philip – That is a great explanation and is also my understanding at this point. I would add that since we are at a point where divergence is expected due to CO2 continuing to increase and the TSI decreasing, the path that temperature follows will render one or the other approach rubbish in the near future.
Murray says:
July 8, 2014 at 10:13 am
I have to agree. Even if its forecast proves valid, its hindcast sucks.
The inputs of my table napkin model set to predict atmospheric temperature output would be:
1) Adjustable equatorial oceanic stored energy to 1000 meters to begin the run.
2) TOA solar insolation calculated for surface incidence on a rotating and tilting sphere minus ocean surface reflectance.
3) Adjustable by amount and length aerosol shading input (clouds, dust, ash, sulfur, etc) as a filter.
4) Adjustable by amount and length of heat loss to the atmosphere via evaporation calculation for rough (IE mixed) versus calm (IE layered) ocean surface.
5) Oceanic fluid dynamics calculations to produce lag.
6) Various calculations to translate energy to joules to watt to atmospheric temperature, etc.
Brad says:
July 8, 2014 at 11:57 am
And please explain how you can forecast “volcanic activity” and its impact on changes in the climate. I’d love to see that crystal ball.
Evans’ ‘model’ uses volcanic activity….and worse ‘atomic bomb tests’
At the very least this is a very serious effort with a lot to digest. Spreadsheet loads up perfectly for me, but is likely well above my paygrade. OK, I do standard C daily but I do not use Excel unless under duress. The fast-draw drive-by shooters here can not POSSIBLY have given due attention in the short time since release, so I shall treat their comments with the disdain deserved. Waiting for those smart enough to be able to digest and at least treat this with some respect and make a useful contribution to the subject on hand.
Two of the biggest unknowns are how will Ocean Heat Content change going forward and how much of an influence might it have on global temperatures. How much of a lag time ? Further if OHC is changing with time the lag time is going to be different then if OHC remained at a steady state..
I think there are circumstances that could drive Ocean Heat Content down to a lower level faster then what most thinking assumes which is, OHC is extremely slow to change. Maybe not so slow if circumstances are right.
What circumstances? Changes in ENSO,VOLCANIC ACTIVITY ,SOLAR , CLOUDS ,ATMOSPHERIC CIRCULATION all of which influence the heat content and temperatures of the oceans.
This is one of the biggest risk in making future temperature prediction in my opinion.
Maybe thresholds come into play that can impact OHC to a great degree if met while OHC changes little until the threshold (x) is reached. Just some thoughts.
Brad and Richard: stop whining about Leif like a bunch of wallflowers. The man is very busy, yet still finds time to mingle with us mere padawans, fow which I am grateful. Leif might point this out himself, but no doubt you would accuse him of bragging if he mentioned he was gearing up for his keynote speech at SCOSTEP’s 13th Quadrennial Solar-Terrestrial Physics Symposium (STP13) in China.
Try not to be jealous of Leif, boys, he actually Walks his Talks.
David Evans/Archibald
Re: Pi/2 (~2.75 year) lag between solar cycle and ocean temperature
Thanks for your innovative efforts.
Now to find out how accurately your model predicts.
See David Stockwell Key evidence for the accumulative model of high solar influence on global temperature 2011
He models the solar cycle on the ocean, calculates a 2.75 year lag (Pi/2 or 90 deg) and shows support for that lag. Further Stockwell links. including the 2.75 year lag in Roy Spencer’s data.
How well does your model show this 2.75 year lag between solar cycle and ocean temperature?
Or does your notch filter avoid looking at this effect?
The base concept of a notch filter is a misinterpretation of the what the spectra represent. HadSST is not “the” output in the sense of TSI proxy convoluted with the transfer function of the system, it is a lot else besides. That makes the basic premise erroneous.
He then brings in a massive, hitherto undocumented, “nuclear cooling” nearly as big as a full 100y of global warming that no one has noticed yet. WTF?
Then he needs another fudge factor to bring in the 11y lag.
I and others have brought up all these major flaws over at JoNova’s and he has not addressed them. I thought he was taking time publishing the details because he was reconsidering some of the implications. It seems not.
While I do think that there will be increased cooling in the coming years, the sudden drop is just a figment of curious data processing and if he’s hanging his hat on that a falsifiable prediction, I think he’s going to be disappointed equally quickly.
I don’t see this going any further.
@Sherry Moore
Aw how sweet. LS has a groupie….
For the “derivation” of David Evan’s model see here (a bit buried in their latest post):
http://jonova.s3.amazonaws.com/cfa/excerpts.pdf
Not at all convincing to me, seems to forget about the existence of noise, and the signal processing arguments are poor/wrong.
Bob Carter has just announced it at the Heartland conference, which makes me worry about handing ammo to the opposition.
Gabriel-hahaha! Maybe, and yes, I have read most of what Leif has published. He has earned the “right” to say whatever he darn well pleases, (being correct an awful lot never hurts either). Let me tell you, Leif would squash me like a bug just as much as he squashes some of the Idiocracy denizens here, if I made an incorrect statement. I would be honored by that fairness, and so should some of these dimwits.
Also, I often illustrate other’s epic Fails, with Leif’s brilliant work, when relevant, so I can’t complain. Maybe Leif will give us all the scoop on the MiniMax 24 findings 😉
[Was that just “an awful lot”, or an “awe-filled lot”? 8<) .mod]
william says:
July 8, 2014 at 12:07 pm
So Werner, this years temps are .478C warmer than an arbitrarily selected anomaly baseline period. It can be argued that the anomaly period was at the cold end of an indeterminate cycle.
That misses the point that this year may well have the warmest sea surface temperature since 1850. And should that be the case, global data sets should not be too different. So 2014 will not likely support the theory. However 2015 may be different.