The Solar Series: I Background | II: The notch filter | III: The delay | IV: A new solar force? | V: Modeling the escaping heat. | VI: The solar climate model (You are here) | VII — Hindcasting | VIII — Predictions
Open Science live — The story so far: Dr David Evans is building the O-D notch-delay solar model. It’s a much simpler big-picture approach than Global Climate Coupled Models. They use an ambitious bottom-up system where the models add up every small aspect in every small cell of the Earth’s climate atmosphere and oceans and try to predict everything, but the trap is the errors — small errors in 10,000 calculations add up to big-mush. David’s approach is top-down. He looks at the whole system from the outside, and doesn’t try to understand or predict each individual part. It’s a way of starting at the start — to shed light on the big forces and processes that happen as energy arrives on Earth, gets reflected, or blended, and eventually changes the surface temperature. His model won’t tell us what happens to rainfall in Sudan in 2050, but it might do what current models don’t and that is predict the global temperature.
The important development here is to complete the path of the energy flow in the most brutally simple way from Sun –> Earth –> Space. We know the sun provides heat through TSI or Total Solar Irradiance. But this is almost constant — it produces heat for sure, but possibly not much of the variation in temperature on Earth that we are interested in. The discovery of the notch filter means some other force (yet to be specified) from the sun acts with a delay of probably 11 years. This delayed force turns out to cause a lot of the variation in temperature. But Earth is not going to immediately warm or cool with every change. Energy collects in all kinds of pools and buckets before it ends up warming the atmosphere. So the effects of both incoming paths — immediate solar and delayed solar — get combined and run through a “low pass” filter — which blends and smooths the bumps.
Having discovered the pattern in the way TSI is tranformed into temperature, David builds the model with the filters to produce the same “transfer function” as he found in empirical data. Hopefully the model will mimic the overall processes without needing to know the details of all the parts. In a sense all models have to do this at some level. No climate model tracks each molecule or follows each photon. Will it work? It does a good job of hindcasting (and we’ll talk about that soon), but the real test will take a few years. Enjoy the quest to figure it out.
By the way, one of my favourite graphs is below — Figure 4 — some curves are intrinsically beautiful. — Jo
—————————————————–
Building a new solar climate model
Dr David Evans, 21 June 2014, David Evans’ Notch-Delay Solar Theory and Model Home
This is the last of the three posts in which we build the solar model. We assembled a notch filter, a delay filter, and a low pass filter in cascade in part III, in part IV we took a diversion to physically interpret the notch and the delay, and in part V we added the RATS multiplierto model the atmosphere on the yearly timescales of the TSI datasets.
In this post we assemble these four elements in their correct order, and add the immediate path for the TSI changes that obviously warm the Earth directly. This will complete the model. We finish by examining the step response of the model.
The Order of the Filters
The notch-delay solar model so far is simply a computational path from TSI to (surface) temperature that contains a notch filter, a delay filter, a low pass filter, and the RATS multiplier (which is a trivial “filter” whose transfer function is a constant). There are no other filters we can discern from the empirical transfer function, or from elementary physical theory. So with no more to add, let’s put these four in order.
The transfer functions of these four filters, when multiplied together, form the empirical transfer function. The transfer function of two filters in cascade is the products of their two transfer functions, so these four filters must be in cascade (that is, the output of one is the input of the next). But multiplication is commutative, so the empirical transfer function does not indicate their order. For that we turn to physical reasoning.
The filter whose place is most obvious is the low pass filter. It models the Earth as a bucket of heat with unreflected TSI pouring in the top, and its output is the radiating temperature. We can now place the other filters around it.
In the flow of computation the RATS multiplier goes immediately after the low pass filter, because its input is the radiating temperature and its output is the surface temperature. We then have the computational path covered from the unreflected TSI all the way to the output of the entire model.
The notch and delay filters intrinsically go together and are inseparable, and it does not matter if they go notch-delay or delay-notch. The only place left for them to go is between the input to the entire model, namely the TSI, and the input to the low pass filter, which is the unreflected TSI.
Therefore the notch and delay filters are modulating the albedo of the Earth.
Figure 1: The notch and delay filters modulate the Earth’s albedo.
The Immediate Path
The development to date only shows the delayed path from TSI to surface temperature. But obviously any changes in TSI also cause direct and immediate changes in the unreflected TSI, by changing the incoming heat from the Sun, so there is also an immediate path from TSI to the input of the low pass filter. This immediate path must therefore be in parallel with the notch-delay path from TSI to unreflected TSI.
The Notch-Delay Solar Model
Putting it all together, here is the notch-delay solar model. If the recent global warming was associated almost entirely with solar radiation, and if it had no dependence on carbon dioxide, this is how it would work:
Figure 2: Schematic of the notch-delay solar model.
Note the parallel paths:
- The immediate path is for TSI, and has no effect on albedo. This is the direct warming effect of extra TSI.
- The delayed path is for force X, which is the same as TSI but delayed and notched. Force X affects the albedo.
The parameters for the model were found by fitting the model to the observed temperatures since 1610, when yearly TSI data became available, though focused mainly on the last 100 and 200 years. Composite TSI and composite temperature records were created out of the TSI and temperature records analyzed earlier. In forming the composites, the offset of each dataset was adjusted so that the average values for overlapping datasets are the same, datasets were faded in and out of a composite gradually rather than entering the average abruptly, and instrumental data was preferred over proxy data. The fitting process found the model parameters such that the model best reproduced the composite temperature from the composite TSI and best produced a transfer function like the empirical transfer function found earlier.
The most important parameter is the delay parameter, which was found to most likely be 11 years but definitely between 10 and 20 years. The break period of the low pass filter was found to most likely be 5 years, though the possible range is from 4 to 25 years because it might be hiding over to the low frequency side of the notch. (It is very unlikely to be more than about the five years that other researchers have found, but the fitting process held open the possibility.) The most likely set of parameters is called the “P25” set of parameters. The values in P25 were rounded off to form the “P0” set of parameters, which has been used to illustrate the transfer functions and step responses of the filters during this development.
Here is the transfer function of the entire model:
Figure 3: The transfer function of the notch-delay solar model. (Technical note: The phase in lighter blue is not graphed correctly because the sampling of the phase for making the graph does not keep up with the ever faster changes as frequency increases.) Uses the P0 set of parameter values for the solar model, which are the rounded off versions of the parameters determined to best fit the observed temperatures.
It reproduces the amplitude of the empirical transfer function (see Figure 5), in the grayed area.
Here is the step response of the model, in dark blue.
Figure 4: Step response of the notch-delay solar model. It is causal. The response is the sum of the response via the immediate path and the response via the delayed path. Parameter values from P0.
Note that the step response is causal — it is zero before the step stimulus is applied.
The step responses of each of the two paths in the model are also shown.
- The step response of the immediate path, in purple, is due to the direct warming effects of changes in TSI. It is small and has pretty much reached its final value after two years.
- The step response of the delayed path, in light blue, is due to force X. It constitutes the bulk of the overall response. It doesn’t reach its final value until about 15 years after the stimulus. Changes due to force X are not all exactly 11 years after the step change in TSI, but fade in gradually after just a couple of years, reach the crescendo of the “dagger” in year 11, then build to full strength after 15 years.
The final value of the delayed path response in the P0 parameter-set shown is 14 times larger than the final value of the immediate path response. The parameter fitting showed that this was the most likely value, and that the delayed path seems to always be between 10 and 20 times as powerful as the immediate path. Thus, the influence of changes in force X (or TSI via the the delayed path) on temperature is 10 to 20 times as powerful as the changes in TSI (or TSI via the immediate path).
The step response shown in Figure 4 is pristine and clean, with sharp edges, because it is a theoretical model, built of simple components that were inspired by, but do not incorporate, the messy empirical transfer function. If we could measure the step response of the system (TSI in, temperature out) then no doubt it would be lumpy, crinkly, and messy with no sharp edges, because in reality it is far more complicated than the model above. We aim to approximate a messy complicated reality with a simple model.
All of the above and everything in the preceding posts are based on the solar assumption, that all the recent global warming is associated with TSI. Now that the parameter values have been estimated, we can dispense with the solar assumption.
In the next post we will run some climate simulations, to see how well the model does at hindcasting.
Notch-delay solar project home page, including links to all the articles on this blog, with summaries.
*Jo adds that some people find Figure 4 looks unnaturally perfect (I did say it was my favorite). That’s true — it’s the model step response. The actual real one probably looks more complex. But there are no perfect “steps” either.
The most difficult part to believe for me is that the effect of force x would exactly cancel out TSI variation. It seems force x would more likely be greater than or less than the effect of varying TSI due to the 11 year cycle. Having it exactly cancel without explanation is problematic.
41
In Figure 5 of Part II the empirical result seemed to be around an 80% cancellation, not total cancellation.
I’ve assumed that since the model filter is built on that Notch frequency analysis that the modelled cancellation is less than 100%. You can see in Figure 4 of the current post (above) that the 11-year notch only goes down to 0.2 for a 1.0 degree step input, so it again looks like it’s not an exact cancellation.
30
jim2’s penetrating kickstart tempered by Andrew McRae’s careful response is a reassuring beginning.
The 170 page document is almost certainly destined to become a ski racing course (a place where people skim & skip at high speed).
I remain ready to carefully consider the previously promised (but not yet delivered) 2 page overview before attempting constructive commentary.
Meanwhile, there’s no reason to delay in applauding David Evans & Joanne Nova for not succumbing to relentless hateful harassment intended to coerce people into being terrified to state the truth:
The sun drive’s Earth’s climate.
290
I also applaud “David Evans & Joanne Nova for not succumbing to relentless hateful harassment intended to coerce people into being terrified to state the truth:
The sun drive’s Earth’s climate.”
211
I made a mistake when I stated the following:
“The sun drive’s Earth’s climate.”
The correction can be stated even more succinctly:
The sun drives Earth’s climate.
easy litmus test:
Stating this fact on corrupt climate blogs triggers hatefully coercive response.
150
Excellent point. The lack of cancellation at 5 yrs and less (higher frequency) is also information.
00
Is there also evidence of cancellation at frequencies lower than 1/11years? Perhaps this one is a harmonic of something much slower than the 22 year Hale cycle! Think Fibonacci.
00
It not difficult at all to imagine this situation. Example.
Solar TSI, causes warming of 1%, at the same time increased UV by 10% increases the thickeness of the ozone layer resulting in 1% more energy being reflected, nett result is zero. the two effects are opposite, which is the same as delaying by 11 years, UV is in phase with TSI if you delay the UV by 11 years. Both reinforcement or destructive interference of the many solar effects are just as likely, what David shows is that destructive interference is what happens at 11 years. I’d draw an analogy to the two slit diffraction experiment that everyone does in high school (assuming you did physics) but that would probably not reasonate with many.
Remember this is an example I’ve pulled out of my hat to illustrate, I am not asserting that UV is force X, although it might be. The challenge is to show how the various solar effects combine to zero. That’s decades of research away, for now we must be content that it merely does sum to (near) zero.
41
bobl.
I don’t think the ozone layer reflects.
However, if affected differently between equator and poles it can affect tropopause heights leading to global circulation changes that affect cloudiness which does reflect.
31
I would guess that ‘Force X’ is likely to be a combination of things, UV, magnetic, cosmic rays, etc.
00
See Post II, the paragraph starting “(To put some numbers on it:”. The temperature peaks expected from the TSI peaks are not a huge amount above the detectablity limit. Presumably after they are mainly cancelled by the notching outlined in Post IV, they fall below the detectability threshold.
20
Roger Tattersall’s most recent article outlines the many frequencies that can play a role in the function of the planet’s thermostat. In vibration analysis of machinery, the multitude of frequencies generated in a large machine (such as an aircraft) often produces unexpected responses, depending on the resonance of the components. Therefore it is not difficult to imagine that David’s model represents only a basic portrayal and that the actual system is immeasurably more complex. As with all things in nature, one can only marvel at the infinite precision of this thermostat.
100
Rod, this model does not incorporate other basic climate ‘frequencies’ such as the Wyatt/Curry stadium wave. That would be an interesting future development. This is just the simplest possible model for solar inputs, with each element having an explicit ‘net sum’ physical explanation. The Earth ( stadium wave, etc) can be superimposed later.
10
I apologise for a sloppy explanation.
It was for the most part in reference to this passage in Dr. Evans’work: “The step response shown in Figure 4 is pristine and clean, with sharp edges, because it is a theoretical model, built of simple components that were inspired by, but do not incorporate, the messy empirical transfer function. If we could measure the step response of the system (TSI in, temperature out) then no doubt it would be lumpy, crinkly, and messy with no sharp edges, because in reality it is far more complicated than the model above. We aim to approximate a messy complicated reality with a simple model.”
After reading Tallbloke’s most recent article, I was reflecting on my own experience in trying to deduce what the heck is going on in a machine studied using vibration analysis (VA). VA is often more art than science. Given the stock standard set of causes for a particular VA anomaly, the truth is often discovered when one uses imagination and experience to understand what the data really means.
20
The model uses TSI as input and global temperature as output. The oscillations you talk about are masked in the global average temperature record as while some places warm, others cool. That is fortunate because these types of oscillations are chaotic and inherently difficult to model.
00
This is very exciting and I can’t wait to see the results in the next installment. As you state, it does not matter in what order the filter stages are cascaded. In a linear system,the output will be the same for a given input, no matter what the order. The only difference it makes is that the signals between the stages match the signals in the physical system, which is theoretically constructed in a particular order.
However, in a non-linear system, it matters very much the order of the filter stages and the non-linear functions. This brings me to my question. You state that in the caption of Figure 1: “The notch and delay filters modulate the Earth’s albedo.” By this, do you mean that the ratio of 30/70 changes by the resulting surface temperature? In this case, the system is most definitely not linear. The amount of modulation would then be a very important parameter. And if this is the case, your block diagram should show a path from the output temperature of the system back up to the albedo box. Then the albedo box would show the change in percentage of energy reflected to space as a function of the surface temperature. Or, am I not understanding something?
10
I think David ment albedo as only “something” between the total solar output and the surface temperature. Could be anything,or many things. Consider if the Suns Magnetic field were any time function symetric about the time origin. Consider F(m) = cos(PIt) with t being 11 years. a 22 year even function, with no harmonics. Any other even function would have only even harmonics 11 year, 5.5 year 2.75 year etc. We can observe the 11 year cycle in TSI and in the atmosphere but not on the surface. This leads to somthing in the atmosphere is responding, could be albedo could be total water vapor content. Any other ideas? Now those that can, can start to figure it out. Those that cannot, should go back to teaching.
12
Forgot 3.16666 years sorry!
00
The to most important feedbacks are water vapor (specific humidity) and clouds, which produce more than just albedo. Water vapor forms clouds, and precipitation from clouds removes water vapor while leaving latent heat in a place up in the atmosphere where it may more favorably radiate away, especially from Tropical T storms -the Lindzen adaptive iris/Eschenbach thermoregulation hypothesis.
You are correct that albedo in the usual sense is not the best choice of words.
I suggest Dave and Jo might consider something like the following formulation of their basic proposition:
Using a new Fourier transform tool, OFT, we find in the observational data evidence for a solar originated ‘force X’ that affects net climate response with a delay. The most likely physical interpretation is the Svensmark hypothesis. X is the solar magnetosphere, the mediating agent is cosmic ray influence on cloud nuclear ion, and the feedback affected is clouds including both their albedo and humidity influences. The IPCC AR5 said clouds are the most uncertain thing in climate models. Here we have a simple rigorous theoretical alternative,parameterized from observational data, which makes falsifiable near term (scale > 1 year, < 11 year) predictions.
21
Rud, why would you choose specific humidity (and at what altitude or pressure?) over total column water vapor.
When water vapor is releases at the surface, typically from ocean but also can come from land, it will take some period of time to mix into the bulk of the troposphere. Total column water vapor is a better measure of driving force, IMO, since it averages out the response to surface heating.
00
Specific humidity is the measure of actual water vapor per mass of dry air, that is the proportion of that GHG present in that mass. It is most important In the upper troposphere where it determines the altitude (depth) of the optical window. (top of fog analogy). Experts like Glenn Paltridge use 400 hPa. There is reason to think 300 would be as good. By 250, one is frequently above the window. GCMs have UTrH roughly constant at 400, which means a substantial increase in UTsH with warming. Both corrected (for changed insrument generations dry cold bias) radiosonde and most direct MSU and indirect (GPS reflection) observations show decreasing UTrH despite a slight increase in UTsH. And they agree well with each other in specific locations like Singapore near the equator. So model water vapor feedback is overstated by something like half, model dependent. This is why even CMIP5 produces a tropical troposphere hotspot when none exists.
00
Rud Istvan June 22, 2014 at 10:25 pm
Konrad,June 20, 2014 at 2:24 pm
“Specific humidity is the measure of actual water vapor per mass of dry air, that is the proportion of that GHG present in that mass.
It is most important In the upper troposphere where it determines the altitude (depth) of the optical window. (top of fog analogy).
Rud this is most egregious nonsense. It is all a “fantasy proposed by those that think they know.”
00
Arrgh,
Albedo is misleading.
What the transfer function is saying is that some thing at or between the suns total emission, and the thermalisation of that at the earth is reducing the effect of TSI at 11 years with respect to the Phase of TSI. This may be simply that the earth merely responds less to the frequency distribution of TSI at peak sunspot than the frequency distribution at min sunspot. That is it may be the frequency selectiveness of the earth and the changing spectrum of TSI itself, Davids analysis does NOT conclude there is any change (ie a dynamic response) that causes this. Please be careful not to throw the baby out with the bathwater. The widest interpretation should be retained.
00
I guess my point is that the “modulation” is non-linear, which opens a can of worms. If the energy from the TSI passes through the albedo “modulator”, and the strength of the albedo is proportional to the notch-filtered TSI (possibly due to cosmic rays) then this effect should be shown more clearly in the drawing. Multiplying a signal by itself gives a squared component, and when you run a sine wave through a squaring function, you get added second harmonics, according to the trig identity cos^2(x) = 1/2 + 1/2 cos(2x). The waveform of the magnitude of the “cosmic ray” signal is probably not a sine wave, but more like a full-wave rectified sine wave, with its own harmonic series, which complicates things further. What I’m saying is I would like to see more detail of what is inside the “albedo modulator”, including the coefficient of the squared term. I would suspect that the “modulator” is more complicated than the “+” sign shown in figure 2 (which isn’t a modulator), something more like A * (k1 + k2 * B).
20
Yes indeed, but it is the best can of worms we have in a long time, yummy! Think of all the mathematical functions that can originate in this universe from the lesser “big bang”! Think of the most modern electronic mechanism, active, passive, resistive, inductive, capacitive. Nothing happens ’till you flip on the power supply. A temporal step function. That whole power supply, in the frequency domain is completely described by its spectrum, a 1/f power spectrum. Your 1/11-years is way down from the power in a frequency at 1/4beelion-years.
What has this circuit been doing with all that power and energy? Most of it appears to be cyclic. What if it was specifically designed to be un-understandable by earthlings?
The learned will mostly sleep. Most earthlings will get on with what “is” like most critters! Some will profit. Some will try to control other earthlings. Those that think they can teach, will claim they know, and teach what they think they know. Few will poke at “is”, to carefully observe “is”, all the while realising that one can never understand the un-understandable, but knowing that learning is its own reward. Congratulations!!
00
“The parameters for the model were found by fitting the model to the observed temperatures since 1610, when yearly TSI data became available, though focused mainly on the last 100 and 200 years. ”
Shouldn’t you “fit” the model from 1610 to say 1980, and then run it and see if it matches observed?
Or could you comment on the effects on your fitted parameters with respect to the time interval chosen?
30
Yes agreed Charles. Having worked, developed and run models in the hydrological and hydrogeological field for more than 40 years the ideal is what we term splitting the record into a ‘calibration period’ followed by a ‘validation period.’ This is the ideal method otherwise you can get caught just doing a curve fit by fudging parameters with a model that has possibly no predictive ability. But often this is not done and if the record is limited in duration and has quite variable events in the record. Under those cases one has no option but to use all of the record. Also a first pass is valuable using all data to allow, “twiddling the parameter knobs” to get a sense how the model is behaving. But clearly if that is all that is done then one can be accused of “fudging a fit” and having to wait years to see if it proves to be correct.
10
Beautiful!
11
OK, so now you have a model (and ignoring the non-physicality, or perhaps more generously assuming that you and I mean different things by a physical model), we come to:
> …are based on the solar assumption
Yes. What you’ve done only makes sense if all the forcing comes from solar.
> Now that the parameter values have been estimated, we can dispense with the solar assumption.
Its not clear what you mean by that. Perhaps the next post with the simulations will clarify. Do you mean, you’ll run your model with solar, GHG, aerosol and volcanic forcings? Its not really clear where, e.g, volcanic forcing fits in your model. It affects the albedo, so sort of belongs in the orange box inside the blue box. Perhaps.
430
All the forcing doesn’t have to come from solar. The model is emperically derived from total input and output data, so it takes into account all the Earthly systems.
140
William,
David means that we can get rid of the solar assumption that it is all based on TSI. It isn’t.
It is based on force x which alters global albedo via a different mechanism to variations in TSI.
140
Hey,
tinylittlewillyWillie,And if this model performs better at prediction than your favorite models’ projections (which have been comprehensibly falsified – Mother Nature is a b****, isn’t she?), you’ll become a
denierskeptic, won’t you? That’s what a true scientist would do.I remain skeptical of this (or any!) model, until verified by observation. Falsification. Do you even know the word?
101
Oh, what did Maynard (later Lord) Keynes (reputedly, and probably quoting someone else) say?
‘If the facts change, I change my mind. What do you do, Sir?’
?
50
If you would replace ‘falsified’ with ‘rubbished’ in my 6:37 a.m. comment, I should not object.
‘comprehensibly rubbished’?
30
Isn’t falsification what the non-classically trained scientists often get confused with making stuff up ?
10
One of my partners, happens to be a Geologist, Volcanologist, and Mining Engineer (with little interest in the climate debate).
She points out that volcanic and seismic activity is more or less constant, on a global basis.
We just notice the small percentage of dramatic variation, caused by terrestrial eruptions and significant earthquakes, that just happens to impact on human activity. But overall, such events are part of the random background noise in the Earth system, and therefore probably random noise as far as albedo is concerned.
70
Which is why http://joannenova.com.au/2014/06/surprise-west-antarctic-volcano-melts-ice/ doesn’t really make sense: there’s vulcanism, but not evidence of it changing.
328
No, you miss the point.
There has “always” been volcanism in West Antarctica, but nobody noticed it, until a) the effect was sufficient to make a sizable chunk of ice fall off the end of a glacier, into the sea, during the summer months; and b) we had acquired the technology to notice that the event had occurred.
But once noticed, the volcanism was blamed on climate change, by the vapourous luvvies! Seriously?.
It is the way you choose to interpret Jo’s posts that really doesn’t make sense.
241
I’m afraid that’s just not true
The 18th century whalers accurately mapped the warm “lakes” and persistent water vapour locations as a matter of survival – I have inspected the remaining original maps
Professionally run Antarctic tourist ships routinely stop, weather permitting, to allow the more intrepid of their passengers the opportunity to “bathe” in these small lakes
The controversy is whether this vulcanism actually affects the glacial topography or not
50
Knowledge of volcanic activity in West Antarctica has been known for a long time. In fact if you sail down the Antarctic Peninsula is is hard to miss, starting with Deception Island in the South Shetlands. The discovery of active volcanoes under the West Antarctic Ice Sheet though is fairly recent: http://www.livescience.com/41262-west-antarctica-new-volcano-discovered.html
“Earthquakes deep below West Antarctica reveal an active volcano hidden beneath the massive ice sheet, researchers said today (Nov. 17 2013) in a study published in the journal Nature Geoscience.
The discovery finally confirms long-held suspicions of volcanic activity concealed by the vast West Antarctic Ice Sheet. Several volcanoes poke up along the Antarctic coast and its offshore islands, such as Mount Erebus, but this is the first time anyone has caught magma in action far from the coast.
“This is really the golden age of discovery of the Antarctic continent,” said Richard Aster, a co-author of the study and a seismologist at Colorado State University. “I think there’s no question that there are more volcanic surprises beneath the ice.”
What has been obvious is that climatologists have bee reluctant to admit that volcanism has anything to do with the fact that while there is some melt recorded in West Antarctica, there has been a growth of ice in East Antarctica. http://joannenova.com.au/2014/06/surprise-west-antarctic-volcano-melts-ice/
Having visited Antarctica I can assure you the maps of warm lakes, and warm areas on beaches do exist, but don’t guarantee a warm swim unless you are referring to the tiniest almost ‘puddles’ which occur like spa baths from place to place on land! Otherwise the warm spots in the oceans and larger lakes are variable and move continually. So you can be sitting comfortable in a warm spot when suddenly you are engulfed in freezing water leaving you scrambling to find the new warm spot before bits of you start to drop off! 🙂
40
Jaymez, I said all that, although perhaps far more succinctly 🙂
The magic in the Schroeder et al paper “Evidence for elevated and spatially variable geothermal flux beneath the West Antarctic Ice Sheet” (May 8, 2014 PNAS) is that the exploration method of airborne radar echoes combined with logs from the West Antarctic Ice Sheet Divide ice core drillholes has produced a reasonable map of the glacial bed paths. Figure 1 in that paper is fascinating, at least for geologists, especially 1C
For those interested, the link to the paper is:
http://www.pnas.org/cgi/doi/10.1073/pnas.1405184111
00
Sorry, I thought the sarcasm would have been obvious, given the totality of the comment.
20
Will, firstly it IS a physical model, the model is based on the physical energy trasfer characteristic between the sun and the earth assuming that TSI affects temperature. It as physical a model as the model of my IR drying system I described yesterday. Yes, you can try to say, that an explanation is that the TSI doesn’t control temperature but that isn’t what David is doing, the hypothesis is that TSI does affect temperature. BTW if one says that TSI doesn’t affect temperature then one needs inconvenient explanations as to why summer is hotter than winter and day is warmer than night and why the ice ages occured. TSI obviously does affect temperature and it’s a bit of a mystery as to why sunspot induced TSI changes don’t cause 11 year cycles in temperature. There is something unique about sunspots, that David is on the hunt to find.
Finally he is saying we can dispense with the solar assumption because the model is built, now he can plug any parameter into the input and investigate the output. He is effectively removing the constraint to see what the filter tells him. At least that’s how I read it.
—-
Bobl, thanks, you’ve got it. – Jo
130
Instead of looking at climate change through greenhouse-coloured glasses, he’s effectively allocating the variance in the temperature to variance in the climate components in the order of largest to smallest influence, which means TSI is first and therefore CO2 will probably be the 2nd last priority to be added and volcanic aerosols last.
From my own attempts to model the heat fluxes of TSI, water vapour, PDO cycle, and CO2, their sizes are roughly in that descending order, since 360W/m² > 300W/m² >> ±34W/m² > 31±2W/m².
That almost sounds… helpful.
I don’t think a climate model should worry much about short term 5-year effects like ENSO and individual eruptions, but a decadal average for volcanic aerosols might be significant. If volcanic activity is proportional to solar activity, as some people have found, then there may be no need to model their aerosols separately because the variance allocated to the solar activity component will indirectly account for volcanoes. It would just mean the solar activity component will be slightly smaller, since the volcanic aerosols produce a cooling effect.
As it turns out, judging by the wording of AR5 WG1 Chap9 Box 9.2 and by doi:10.5194/acp-12-7321-2012, the world’s volcanoes are dropping surface insolation by 1±½W/m² at most. More of a radiative burp than an eruption. I wouldn’t bother modelling volcanoes at all.
00
I now understand why you spent the first 5 parts of the series on doing an introduction to the idea. A good way to get people through the “electronics” aspects of the model.
My electronics knowledge is pretty rusty, so the procedure has been rather helpful.
Can’t wait to see how it all functions.
If it predicts significant cooling, I hope it turns out wrong. 😉
A small amount of warming, or even the continuation of the current “going nowhere” climate would be the far better option for the whole world.
90
On the other hand , if I , as a farmer responsible for producing food for a certain number of earthlings, am going to experience less favourable growing conditions over the next decade or so, then the sooner I know that the better. Farms respond fairly slowly!
90
Electronics knowledge does not rust — but it can corrode.
30
It can also get globbed with solder.
40
Now that is a technical term, I have not come across before … Globbed!
I sounds like a word Rowen Atkinson would use. Glllooooobbbbed!
60
Here’s one for you. As solder ages, it can grow whiskers!
http://www.maximintegrated.com/en/app-notes/index.mvp/id/5250
10
And dry joints !
20
Just look at that funny phase change at 2 months in the whole transfer function.
How dey do dat?
00
Dr. Evans, thank you. My electrical engineering studies lasted 1 year from 1963 to 1964, then I switched to engineering mathematics. I seemed to enjoy this more, but not sure why.
My other degrees were systems and industrial engineering and civil engineering (major hydrology and water resources, finished in 1975)). I say this to let you know how old my studies and knowledge are.
But I remember working with a linear system called the Unit Hydrograph to predict runoff rates from rainfall minus infiltration. Then, the next step was the Instantaneous Unit Hydrograph which was the response of a watershed to an impulse input, and thus, the IUH was an impulse response. If the impulse response was convoluted with the rainfall-minus infiltration, the result was a predicted runoff hydrograph.
Sorry for all that detail, but here come some question from an older reader. How is a transfer function related to an impulse response? I remember taking the Laplace or Fourier transforms from input and output and then determining the impulse response from their ratio. Is that in any way related to you work?
60
Good question! The impulse response with its all uniform frequency spectrum at surface temperature is what? Any large CME is as close as you are ever to get to an impulse. What is the response? We know the impulse function here starting 8 minutes after the impulse. Time for those that can do, to express their skill!
We need no stenkin statistical mechanics here!!!
01
Stenkin statistical mechanics never have the knowledge or experience to operate an air wrench correctly! Exactly half the time, your wheels fall off!
02
And the other half of the time you can’t get your wheels off without breaking something.
00
Indeed! go find a mechanic that knows his trade, and both problems, disappear! It seems that statistical mechanics create problems, easily avoided by any earthling with skill. Is that really a branch of mathematics or only of fantasy. No slur to Sir Boltzmann, who said this applies to almost nothing but noise!
02
Know the feeling when breaking a wheel brace trying to change a staked tyre in the most isolated part of the Kimberley. Fortunately the only vehicle in 3 days came along 30 minutes later!
10
When it comes to CMEs I have noticed that Forbush events as experienced inland in southern England invariaby mean a period of sunnier days and less clouds. I don’t know if any such observed correlation can be seen in the available data however.
00
The impulse response, step response, and transfer function are the three standard ways of completely characterizing a system (if it is linear and invariant).
The impulse response is the response of the system to an impulsive input: all happening ineffectively in zero time, but whose area under the curve is unity.
You get the impulse response from a transfer function by applying the transfer function to the transform of the impulse response (which is flat, i.e. amplitude 1 and phase 0 at every frequency), then inverse-transform the result back into the time domain. Since the spectrum of the impulse is flat, the spectrum of the impulse response is itself the transfer function. So, to get the impulse response, just inverse-transform the transfer function. Or, to make it even simpler, the impulse response is just the sum of the sinusoids in the spectrum of the transfer function.
20
Thank you again Dr. Evans. So in a liner and invariant system, once you have the input, x(t), and the output, y(t), the impulse response i(t) can be defined by the Laplace transform of the convolution integral. Thus, y(t) = integral[x(tau)*i(t-tau)d(tau)] integrated from 0 to t. Take the Laplace transform of both sides and this becomes
Y(s) = X(s)*I(s) and thus, I(s) = Y(s)/X(s). Given the input, x(t) as the incoming solar irradiance, and output, y(t) as the mean surface temperature, can you determine the impulse response as the inverse of [Y(s)/X(s)] and then predict the temperature from the input radiation?
Would it make sense to remove lag you determined, apply the convolution integral, then add the lag back into the temperature? Seems simpler once you have the lag determined, but what am I missing?
20
Looks about right, without checking every detail Leonard.
No, you need to apply the delay in conjunction with the low pass filter because it interacts with the low pass filter as well. In our calculations, we either compute the whole step response and apply that to every step in the TSI (ie convolution) or multiply the whole transfer function onto the spectrum of the TSI (input) record, then inverse transform to get the temperature (output) record. (Did it both ways to check. They give same answer.)
30
Ok, David is now narrowing things down.
Force x is the cause of a change in reflectivity (albedo).
Having narrowed the effect of force x down to changes in albedo we are talking about global cloudiness.
Force x changes cloudiness but that only affects the proportion of TSI that gets to the surface, it doesn’t account for the delay.
Something else delays the appearance of the change in reflectivity turning up as atmospheric temperature changes for about 11 years or one solar cycle.
That something else would appear likely to be the oceans.
Reduced cloudiness allows a greater proportion of TSI to penetrate the oceans at all wavelengths and a great deal of solar energy gets past the evaporating layer to effectively disappear for about one solar cycle until it is released again.
So what causes the changes in cloudiness ?
Svensmark suggests cosmic rays providing more condensation nuclei.
I suggest variations in ozone concentrations in the stratosphere causing latitudinal climate zone changes which alter the length of the lines of air mass mixing to change cloudiness.
Less clouds from zonal flows which have shorter jet stream tracks and more clouds from meridional flows which have longer jet stream tracks.
Any other suggestions ?
42
Whether it is the Svensmark (cloud nucleation), UV ozone photo effect, or the Lu CRE ozone effect, is not material. It could be all three. Force X is modulating via net feedback with a delay.
As posted upthread, the impact of clouds (Svensmark hypothesis) is much more complex and potent than just albedo. Albedo is probably an unfortunately over simplistic choice of terminology.
That is what open forum feedback is useful for. This process is working in real time.
90
Thank you Rud,
Albedo is only a small part of the reflection of Solar flux from the atmosphere or surface. Albedo includes only the visible backscattered into a hemisphere convex to the sun. A definition of albedo. Rejection of Solar flux must include all power at all frequencies, reflected scattered backward or forward, still away. The energetic UV is much influnced by this, but never measured. Are you examining the trunk or the tail? You can tell by sniffing!
02
I think albedo is the correct parameter despite it being limited to the visible wavelengths.
The reason being that all other rejections of solar flux first involve acceptance of that flux into the atmosphere by absorption and then internal processing before it is radiated out.
Albedo denies a portion of the flux to the system instantly so it never even enters.
That is the correct way to ascertain the proportion of TSI that is lost.
21
What nonsense. Even in the visible another 23% is forward scattered. This is why you can still see the moon during a total lunar eclipse by then Earth.
“Albedo denies a portion of the flux to the system instantly so it never even enters.” Reflection denies a portion of the flux to the system instantly so it never even enters, including the UV and the IR that are never considered in albedo. Albedo is less than 40% of Solar flux reflected elsewhere.
03
Will,
I didn’t say that all the visible wavelengths were reflected, only that the energy within the reflected component never enters the system to cause a thermal effect.
As regards UV and IR being reflected, please could you clarify ?
Do they not get absorbed and re-radiated rather than reflected ?
UV to a solid or liquid surface heats that surface and is then re radiated. Solid or liquid surfaces don’t re radiate in the UV range so how can the UV be said to have been reflected ?
IR also gets absorbed by molecules before radiation back out as IR but that is not reflection either.
Reflection in the visible wavelengths is another matter. It constitutes a bouncing back out of energy with no thermal consequences.
“Reflection of light is either specular (mirror-like) or diffuse (retaining the energy, but losing the image) depending on the nature of the interface”
from here:
http://en.wikipedia.org/wiki/Reflection_(physics)
So I would say that what matters is only specular (mirror like) reflection of light which does NOT retain the energy.
That is what albedo describes and that is why albedo matters.
Both UV and IR wavelengths involve diffuse reflection which retains the energy.
12
No reflected energy is absorbed by definition. Mie scattering affects all wavelengths. Surface reflection is partly specular, partly diffuse. All specular reflection from the outer half of the earths disk is reflected into the hemisphere that is not part of albedo, as is any part of solar energy passing completely through the atmosphere not intercepted by the surface. Again why can you see the moon during a lunar eclipse? None of that energy is albedo.
21
“I suggest ”
I suggest possible changes in ocean penetration of UV energy..
SOOOOO many possibilities. 🙂
But that is the point here.
David is looking mostly at the OUTCOME at this stage.
It is almost unimportant what the actual mechanism is.. that is the point at this stage.
Even if the mechanism remains “not properly defined”, if the logic gives a validated answer, then the mechanism can be later investigated.
30
Its more fun to narrow down the possibilities in parallel with David’s progress through his presentation.
There aren’t many possibilities when one considers real world observations and the relative scales of the phenomena involved.
10
I agree, the process is fun 🙂
10
But is there something we are missing 🙂
00
Snow and ice are major components of albedo, but yes clouds are the big factor.
We think the delay originates in the Sun, where changes in force X lag changes in TSI by about 11 years (though, as seen from the step response of the delayed path in Figure 4 above, the effect, while centered on about 11 years, is smeared out a few years before and after).
For the reasons in Post IV, we think the delay does not originate on Earth because it is a true delay, not a time constant of a process that runs out like water out of a tank.
30
“we think the delay does not originate on Earth ”
That has been puzzling me for a while and it may be a semantic issue.
The delay under discussion arises due to a time lag between solar variations and thermal changes in Earth’s atmosphere.
Clouds being the big factor but cloudiness changes will immediately affect the proportion of solar energy entering the oceans so it would then be the oceans that cause the delay.
Alternatively one could say that the cloud changes take time to develop in response to the solar changes. In that case the delay need not be the oceans but rather the time taken for the changes in cloud cover to respond to the solar changes.
Thus a slow change in the global air circulation whilst the climate zones change their positions and the jets change their tracks.
But even then it comes back to the oceans because the climate zones and jet stream tracks are also modulated by the bottom up oceanic response to solar changes.
So, the cause of the changes originates in the sun but the delay arises on Earth because of interactions on Earth.
Or does David mean that TSI changes occur some 11 years before some other change within the sun and that that solar change then causes an immediate thermal response on Earth ?
Clarification of that ambiguity would be helpful.
00
More like your last suggestion Stephen: “TSI changes occur some 11 years before some other change within the sun and that that solar change then causes an immediate thermal response on Earth”. That’s puts it well, except that the thermal response is smeared out over a number of years around the 11 year mark.
The essence of the reasoning for that is in Post IV:
“The delay in the solar model says that today’s temperatures are more influenced by the level of solar radiation 11 years ago than by the level either 5 or 25 years ago. So something to do with climate has a memory of 11 years; the delay is not simply due to a dissipative element, like a store of heat in the ocean that declines at a certain rate.”
For that we need something with an 11 year clock, which presumably means the Sun.
In post IV I listed a few of the known properties of force X. It seems to be strongly associated with the Sun’s magnetic field, though it could be some electric influence that merely correlates well with the solar magnetic field.
Also, it’s not literally a copy of the TSI delayed by 11 years, but more a smeared out copy where the delay from a change in TSI to the corresponding change in force X is from about 3 to 15 years but mainly concentrated around 11 years, as per the step response of the delayed path in Figure 4 above.
10
Thanks for clarifying, David, but as I feared that doesn’t seem to work out because cloudiness and albedo declined when the sun was more active rather than increasing as you have proposed below.
To get over that problem I think you need to propose that force x reduces cloudiness but the increased solar energy is soaked up by the oceans, effectively disappearing from view and then progressively released over 3 to 15 years as it circulates around the ocean basins.
Would your solar model still work with that adjustment ?
00
Huh? Albedo was a minimum in 1998, after decreasing for over a decade, right when global warming stopped. Minimum albedo -> minimum reflectance -> maximum unreflected TSI pouring into the climate system.
From the Palle paper in 2008: There are conflicting studies, but some resolution seems to have been reached by 2008. The International Satellite Cloud Climatology Project (ISCCP) released its FD product in 2008 which showed small decreases in albedo from 1983 to 2000 (a period of rapid temperature rise). Satellite observations and the ISCCP FD show an albedo increase from late 1998 to mid-2000, then roughly constant albedo to 2007 (a period of near-constant temperature). [Palle, Goode, & Montanes-Rodriguez, 2008]
So for the decade to 1998 albedo was dropping, unreflected TSI was rising, and the world was warming. Then it reversed. Seems ok to me.
Further, notice that TSI peaked in 1986 along with three other solar indicators (Lockwood and Froelich, 2007) — and 11 or so years later albedo bottoms when force X peaked. Seems all ok to me. There is the delay too: 1986 + 11 ~= 1998.
(But I am fearful nonetheless that some sign isn’t flipped and the whole thing is silly! So check that I am right please!)
There are some graphs that say “% Change in Albedo: with a mnimum in 1998, and others that say “Albedo (%)” with a minimum in 1998, which have opposite implications! I think the text in Palle et al 2008 clears it up — albedo bottomed in around 1998.
70
Non-albeldo could be a answer.
00
“Changes due to force X are not all exactly 11 years after the step change in TSI, but fade in gradually after just a couple of years, reach the crescendo of the “dagger” in year 11, then build to full strength after 15 years.”
That makes sense.
A couple of years for the effect to start showing up in ENSO with the full effect to take 15 years to manifest itself in all the ocean basins.
That gets rid of a concern I had as to why the oceanic response needed to match a single solar cycle.
It doesn’t. It is spread across 15 years but is enough to remove the atmospheric thermal effect of a single solar cycle.
It also explains why some parts of the world show a thermal response to a single solar cycle better than others.
It depends on their position relative to the nearest ocean basin and that particular ocean basin’s response time.
Global cloudiness is a result of both the top down solar effect however caused (force x) and the subsequent oceanic response. The interaction between the two pushes the jets and cliumate zones to and fro latitudinally to cause what we perceive as climate change.
50
Thinking about magnetic fields, the solar field reverses every 15 years or so, but the Earth’s field does not. That means when they are aligned, they would tend to “repel.” When anti-aligned, they would merge. No?
If that’s right, what effect should we see on cosmic rays, if any. Or, what other effect might that cause?
20
Jim, the Earths magnetosphere does matter. That is why the posited Svensmark effects of x would be stronger at the poles where they matter least and weakest at the equator where they matter most. But now soppose X also works through the Lu CRE/ozone mechanism. The the reverse is true. The beauty of this is it could be some of both.
But over the observational period, the earths magnetosphere has been for all practical purposes constant, so it doesn’t play a role on these time scales.
20
The earth’s magnetic field also reverses in thousands of years. Less than that the Earth magnetic field wobbles much. Our Sun’s magnetic field reverses every?? 10 to 33 years. and effects everything from temperature, desire for intercourse, or which political party, temporarily, thinks they govern!
00
The earth’s magnetic field also reverses in thousands of years. Less than that, the Earth magnetic field wobbles much. Our Sun’s magnetic field reverses every?? 10 to 33 years. and effects everything from temperature, desire for intercourse, Color of butterflys or which political party, temporarily, thinks they govern! Try to describe the result, Sir Echenbach
01
Rud – if you read carefully what I wrote, I didn’t say the Earth’s magnetic field did not matter. I was pointing out that every 11 years the Earth’s and Sun’s magnetic fields align or oppose. And wondering what effect, if any, this alignment or opposition might have.
10
There seems to be an assumption that the notch filter has to have a delay of 11 years (is that because it starts having an effect before the change in TSI?)
But is it possible that the cooling effect is not a modulation of the warming of TSI but isntead an increase in the amount of heat being allowed to leave the earth?
Then I fail to see why it should have a delay of 11 years. In fact, it might be synchronised with TSI but start having an effect just slightly earlier … i.e. earth begins to cool more rapidly just before TSI increases and the two effects normally cancel each other out.
00
The reason for believing there is an 11 year delay (most likely, but possibly as few as 10 or as many as 20 years) associated with the notch are given in Post III.
Variations in the amount of heat leaving to space are taken into account by the low pass filter (also outlined in Post III).
10
Here’s the part I find curious about this whole expedition. David is using temperature and TSI datasets and proxies that go back to the year 1600. From these he develops a model.
Now, if I were playing this game, I’d divide the datasets in two, an early half and a late half.
Then, I’d develop the model using only the data from the early half, and see how well I could predict the late half.
After that, I’d develop the model using only the data from the late half, and see how well I could predict the early half.
And call me crazy … but I would assuredly do this kind of “out-of-sample” testing long before I published my work. That’s grade-school, basic stuff for any new theory.
Now, rather than let us in on all the secrets, David has divided his publication into parts. I find this to be very frustrating, and indeed, it strikes me as a sales ploy. It’s actually a fairly common sales tactic, you get people all excited about some hot new product, but you don’t publish the actual specs until everyone has already drunk the koolaid and has signed on … not my idea of science, and that’s a fact.
But in any case, I’ve been patiently waiting, and waiting some more, for the big reveal, so I could see how well his model does on the out-of-sample data …
Foolish me. I should have known from the hype of the sales pitch that there wouldn’t be cheese at the end of the maze.
w.
726
Yes. Of course you would do this for any adaptive model. THis is sort of system estimation 101.
This is exactly what the IPCC did NOT do and look at the ludicrous results. It’s off the rails from 2000 on.
If you don’t ‘train’ the model on some data and test on other data, you can get anything you want.
80
That’s right, they are way off from 2000 onwards- but that didn’t stop them from labelling their AR5 SPM 10 graph plot (1.2C rise from 2000-2010) as “historical” as if that ludicrous projection was the actual instrumental record. Just for good measure, they painted it black up to the present so as to make it appear as the instrumental record, superimposed on a clutch of model runs. It was just the average of those runs.
30
Nothing is “off the rails” — just, with an obviously better data model, the usual fluctuations around the trend. As Tamino showed:
http://tamino.wordpress.com/2014/02/06/cowtan-way/#more-7024
Climate models solve boundary value problems, not initial value problems. The aren’t good a short-term predictions, where small initial conditions matter, but they are good at long-term projections, which are a function of energy conservation.
221
Speculation, they will only be good with energy conservation if they have all their energy terms right, the models assume a gap in radiative energy between inbound and outbound energy must accumulate as heat. It assumes there are NO losses from that system to other forms of energy other than heat and radiant energy. That is pure fiction, there are losses, and they haven’t been quantified, therefore long term the models will run hot. What you are in essence saying is that the earth is incapable of dissipating 0.6W of gap, that the amount of heat emitted by a christmas light per square meter column of turbulent air. That’s is totally implausible to me. The more you heat, the more it will convect, the more it will radiate.
40
the models assume a gap in radiative energy between inbound and outbound energy must accumulate as heat.
No, they don’t assume that. That’s calculated from the equations that describe the physics, such as the Schwarzschild equations (also called the two-stream equations).
This part of the physics is understood very well.
37
Only by davixd!
20
This part of the physics is understood very well.
That is possible. Now how well is it quantified?
31
Your Schwarzschild equations (also called the two-stream equations), is exactly why your GCMs never work!
“This part of the physics is understood very well.”
Please show “any” two streams of eletromagnetic flux in oposite directions. No such phenomenon has ever been detected, observed, measured. You insist on calling irradiance “flux”, never is, never was!
21
That is pure fiction, there are losses, and they haven’t been quantified.
What losses?
25
All of the atmospheric effects, hurricanes, tornadoes produce only entropy, losses that must be radiated away by the atmosphere of this wonderful planet.
23
Here is a simple task for you.
Get a 1 tone load of sand delivered
Move that sand 100m by wheel barrow.
Now move it back by wheel barrow
The sand is in the same place, right?
So obviously, no energy was expended.
61
Whenever thermal energy changes form, for example in photosynthesis, nitrogen fixing etc, shortwave energy is absorbed in tbe process, if ice is melted, if thermals create wind (kinetic energy) if water is evaporated and rains out, thermal – potential to kinetic. Whenever a rock cracks from thermal stress, or a grain of sand is eroded by wind driven sand.
These losses do not always degrade to heat, there are at least 4 ways for the climate to lose energy, and 3 ways for the earth to lose energy to space. Energy can be chemically stored, energy can increase entropy (erode rocks or melt ice), energy can emit as high frequency light, and em (radio) eg lightning, and finally energy can be lost in changes in the earth’s momentum, or in stretching or contracting the gravitational forces binding the solar system.
Now as an exercise see how many ways you can conceive to either store energy, increase entropy, work against or for planetary rotation, change the axial wobble, or move the earth relative to the sun or moon. Any time a force acts against the earth, the earth moves, that pulls on the sun and moon via gravity. Just because the motion of the earth is miniscule doesn’t mean it doesn’t happen, every action has an equal and opposite reaction. And E = 1/2 m deltaV ^2
The reaction force alone of rain striking the earth if I recall right represents 10% of the 0.6W gap. Now do the same for the reaction force of wind, the reaction force of waves, the stored energy in photosynthesis, the energy lost to erosion, the energy lost to lightning, the energy lost to converting cholesterol to vitamin D in mammals, then keep going, adding up the little losses and tell me that you cant find a mere 0.6W per m2
30
The more you heat, the more it will convect, the more it will radiate.
The more it radiates, the higher its temperature. E = sigma*T^4
310
” E = sigma*T^4″
What total bullshit. E = sigma*(T1^4-T2^4) If the two T,s are the same nothing radiates.
34
I would have thought that they both still radiate, but no energy is exchanged.
Might just be my interpretation.
20
Which is another reason why using the average TSI for a global energy balance is not very sensible.
30
the Griss June 22, 2014 at 8:40 am
“I would have thought that they both still radiate, but no energy is exchanged. Might just be my interpretation.”
Sigma T^4 is “irradiance” W/m2, a potential for radiation. The difference between two opposing irradiances is called radiative flux or”radiation” a power exchange. The radiative flux is always at most unidirectional.
Appell and the modelers use Schwarzschild equations (also called the two-stream equations), and is exactly why the GCMs never work!
“turnedoutnice” blaims the two stream mess in Carl Sagan 1960.
Appell et all, call this well understood. It is now taught to physics students, but EEs just giggle.
13
And the more it convects.. the more heat will escape,.
I dare you to try this simple test of heat radiation vs convection.
Light a candle.
Bring two fingers as close as you can to either side of the tip of the wick.
Now try to get one finger the same distance ABOVE the tip of the wick.
I dare you. !!!
40
Are you an idiot or what?
03
That’s fine Willis. But there is nothing for sale here. if you want to use the analogy , then this is market research. or in other words , it’s enquiry.
There is a market for intelligence (intel) ;what sort of intel does that market want.
I want it all , and I want it now !!
61
Well, they certainly don’t want it all, or now.
They only want it, if it will be significant, and they want it last month.
(Their bad emphasis, not mine)
20
farmerbraun
June 21, 2014 at 6:30 am · Reply
Certainly there is something for sale, one brand new top-down climate model, only driven on weekends. If you don’t think that science is inter alia a crowded marketplace with scientists promoting their own ideas … well, look around. Activists are trying to sell their ideas to the public.
I don’t mind any of that, it’s just part of the game. What I’m objecting to is that I don’t like the kind of sales pitch that David is using. The good scientific sales pitch is something on the order of “I have the best data, and the best method, and the clearest results, and here they all are with the code to show well they work”.
The second-rate sales pitch is like the first, with a slight difference … they provide everything the first method has, except the code and data.
This method of David’s, of revealing and explaining one bit, then another, with lots of hints of things going on behind the scenes, there’s hindcasting that is claimed to be good but we’re not allowed to see it, there is a secret transfer formula to convert the sunspot Fourier transform into the temperature Fourier transform (which is basically a straight line) … sorry, but to me that yields nothing but the constant frustration of a bad strip-tease.
w.
417
Fair enough, you don’t like it. But with all due respect, you don’t have to like it Willis. For whatever reason(s), David and Jo have decided on this “sales pitch”. Either you wait for all the chapters to be revealed, or you spend your time elsewhere.
,
That’s not quite accurate. You are allowed to see it all in good time. The following was stated in the very first post of this series, viz
It’s a 170 page document plus an excell spreadsheet that people can play around with. It was circulated to various people. Have you considered asking David for a copy?
regards
140
Like emergent phenomena as the one true thermostat, for example. I’m beginning to think you know what you’re talking about when it comes to climate buskering not that it has anything to do with David’s expository. It might improve your combative image if you give David the same opportunity to discuss his discoveries that you were afforded when you revealed your thermostat.
80
I reckon that I’m gonna get to see it all ; and at a pace that I can handle, given that I’m still putting in a 13 hour day on the farm. And I can see any criticism in real time, thanks to you and others .
I mean , I really get to give the tyres a damn good kicking.
And at the end of the whole appraisal process , if i like it , I can buy it . . . for free.
It’s not a bad deal.
http://www.youtube.com/watch?v=Tzhp3jWnpbs
50
Willis,
Nothing for sale. Just free altenate ideas for your consideration.
Please try to consider.
54
You don’t have to like it, Willis. And given your nasty attitude, who cares if you like it or not. I know I don’t.
52
I’m objecting to is that I don’t like the kind of sales pitch that David is using. The good scientific sales pitch is something on the order of “I have the best data, and the best method, and the clearest results, and here they all are with the code to show well they work”.
So what you object to is – “I have the best data, and the best method, and the clearest results, and I will give it to you in easy digestible bites with the code. Because this is such a new way of looking at things with climate science I want to take it a step at a time. We will hold nothing back including the whole 170 page paper. But, since there is time you won’t get everything all at once.”
So your objection Willis is that you are not getting everything all at once.
For some one who can’t see a 12 to 14 dB notch in a Bode plot (unless you have recently noticed) I think you could use the time to your advantage. No rush. Study the materials. Understand them better than David does.
“You cannot be sure you are right unless you understand the arguments against your views better than your opponents do.” Milton Friedman
Just another misgiuded EE – Simon
41
How do you know he hasn’t, or will not do so? Heck, we haven’t even read all the parts yet, nor seen any fit to anything except to the transfer function.
This approach is useful for building understanding one step at a time in this forum. Works for me, and I am usually very impatient since a quick study. Has given me time to brush up on the underlying physical interpretations of this model. And how it could be extended. Is proving richly rewarding.
Your last sentence, Wilis, is not like your usual self elsewhere. Almost as disappointing as when you trashed my Climate Etc posts on the implications of the 2008 IEA oil field decline rate study, and on the egregious error committed by Maugeri in his 2011? Harvard paper, Oil, the next Revolution.
71
Rud Istvan
June 21, 2014 at 6:35 am · Reply
Rud, he may well have done so … and in that case, the fact that he has NOT published his results becomes quite damning. So I made the assumption that looked better for him. However, your assumption is certainly possible.
So … you’re saying that if it was all written as one post, it would have been less understandable? OK … I guess … I don’t see how that works, but if you think the slow drip method works better in your case, I’m not gonna argue.
So I’m a genius when I agree with you, and when I disagree with you I’m “very disappointing”? Funny how that works … take a number, Rud, you’re far from the first to laud me when they agree with my words, and say they’re disappointed in me (or worse) when they disagree.
And your accusation with no link? I find nothing at Climate Etc. even remotely resembling what you describe. This is why I have asked you and others, over and over, to QUOTE MY WORDS when you disagree with them. Making such a claim, with no link and no quote, is an underhanded technique—making such an accusation without backing it up is just flinging mud at the wall and hoping it sticks.
In any case, hang on, let me see what my “last sentence” was, I’ve written other comments since then … OK, here it is:
Look, Rud, I do my best to call things exactly as I see them. In this case, the presentation has many points in common with a sales pitch. I said nothing before, because I thought that when he (finally) got around to publishing his model, he would also publish the results that you say he may already have in hand. But now that he’s published the model and STILL hasn’t published results, I’m sorry, but the resemblance to a sales pitch is uncanny … I mean, look at all of the adherents this theory of his already has, and he hasn’t published a single result! Is this what passes for science on your planet? Because it sure doesn’t on mine …
I expected that he’d publish results with the model because as they say, “The proof of the pudding is in the eating” … but he’s given us nothing to eat, nothing to chew on, nothing of substance. Or, in my previous metaphor, there’s “no cheese at the end of the maze”.
Now, you’re correct when you say we don’t know if at some future date he will publish the results. Indeed he may do so, and I certainly hope he does so soon for his sake. And we don’t know if at some future date he’ll get the Nobel Prize for solving the climate puzzle, and if he does I’ll be the first to cheer him … but so what? How does that help us today?
I’m just pointing out that to date, there’s no cheese at the end of the maze, and it sure looks like a sales pitch …
w.
620
Willis, as much as I admire lots of what you have posted elsewhere, the mere length of this convoluted reply to my rather simple comment suggests a ‘treffer’. Please read the silly management advise book “Who moved my cheese?” then let’s get back together for another confab. Meanwhile, I somehow am quite certain that this little kerfuffle will improve the future exposition of the ideas from Jo and Dave. Quite, quite certain. Thank you. More proof that this open forum stuff works at even quite technical and quite ovloque angles levels.
Highest regards to a fellow traveler, even if not on the same exact path.
62
Rud Istvan
June 21, 2014 at 2:22 pm · Reply
Rud, you made unpleasant accusations without any attempt to back them up. When I protested in detail, rather than back up your claims, and rather than answer even one of the other issues I raised in response to your comment, you claim that the length of my reply suggests a “treffer”, which I find out is German for a hit or a punch.
So … next time, if I just say “Bugger off, fool!”, will you claim that the length of my reply suggests that I’m right? I mean, if my long response means I’m wrong, what does a short response mean?
Is this some complex joke that I’m not getting? Your claim that someone answering your baseless accusations at length means that they’re wrong is beneath you, Rud. You’re a better man and a better scientist than that.
w.
418
Perhaps some day you will, “get it”
34
Treffpunkt means meeting place,
Treffer can mean point scoring..
Pass.
21
“You’re a better man and a better scientist than that.”
Willis, I’ve read many of your posts over the years and you’ve always been most congenial. Yet here you’ve turn decidedly nasty. I sure hope your self import isn’t turning you into another W.Connolly clone. Please, don’t.
41
Why are you being so antagonistic?
This is not what I thought you were like..
Have you got sore toes or something ?
FFS, sit back and see where this leads… its not worth you having a heart attack !!!
71
“Then, I’d develop the model using only the data from the early half, and see how well I could predict the late half.”
I think he is going to do something like that: “In the next post we will run some climate simulations, to see how well the model does at hindcasting.” So he is going to take the inputs from some period in the past and see how well the model response compares to observation — or that is what I believe it is saying.
70
Do I sense that someone is having a jealousy attack?
110
Rod I think that’s confirmed
60
Willis,
David has done the “Grade School” Stuff. If he wasn’t happy with those results, we wouldn’t be publishing anything at all.
David has done this unpaid. He could’ve, say, possibly been paid well to work on other modeling projects (like he was, on the one he used to do for the Australian Greenhouse Office).
So it’s cost our family thousands, we’ve had no family holiday in over year, our dining chairs are coming apart, the kitchen cupboard doors are falling off, but we’re giving the whole project to the world because we think we’ve got something that will advance our knowledge. Forgive us, we hope it might make the world a slightly better place for honest people.
So if we want to release this large project bit by bit, perhaps you could find the goodwill to allow us to do it on our schedule? If it’s too frustrating, sorry, you can always stop reading, and we’ll happily email you a full copy when it’s over. Would you prefer that?
We’re really enjoying the feedback, the conversation, and this is it’s own reward. It seems like a small ask to me. We want the project road-tested bit by bit. And if we get an improvement or hit a problem, we’d rather find out before we release it. Once it is out there, we’ll lose some control. Fixing bugs is easier at this stage.
In any case, we’ve tried dropping the full 170 page report on people. As a communication format, it mostly doesn’t work. eg Lubos read just enough to get it wrong, you can see how helpful his feedback this week was. So we decided we wanted a conversation, and this is it.
This is open science at work. We’d like you to join us, but you don’t have to be here.
261
Joanne Nova
June 21, 2014 at 4:21 pm
Great. Can we see “those results”? And why on earth do I have to ask you pretty please if you’ll release your results as if you were Phil Jones or Michael Mann? If he’s done it and he is happy with it, why is he keeping it a secret?
Jo, first, I’m sorry to hear about the personal cost to you. And assuredly, as you point out, you’ve paid your dues and you’re perfectly free to play the game any way that you wish.
I’m just reporting on what it looks like from out here.
From here, it looks like you’re doing all this partial releases and giving us dribs and drabs when you haven’t even published the completed “grade school stuff” that you’re happy with, and you haven’t divulged your code, and you haven’t published the function at the heart of the model. And despite that, you’re making claims like that it’s done well on hindcasting …
It doesn’t look pretty. You’re making claims, and you need to back them up. Otherwise, it’s not science.
As I said, play it how you want, that’s your choice. I’m just letting you know that as a format, your chosen path is no different from the “science by press release” practiced by the alarmists. Yes, your intentions in the doing are different and innocent of ill intent, but the outcome is the same—look at all of the adherents you have already gathered, who are busily defending a theory that has not yet been detailed or tested. How is you doing science by press release different from the alarmists doing science by press release?
So now, NOT publishing your code and data is “open science at work”? Would you like to reconsider that claim?
You’ll have to explain that, because on my planet, “open science” means full transparency of data and code, and you are functioning at something close to total opacity of data and code.
Jo, please be clear that I hold that your website is one of the best on the web.
But your choice of format for your foray into “open science at work” is deeply troubling.
Best regards,
w.
PS—Regarding your claim that “Lubos read just enough to get it wrong”, Lubos said:
So Lubos absolutely disagrees with your characterization of whether he read the document … and for a man that you claim didn’t read it, his analysis is quite penetrating. I can only echo his conclusion (emphasis mine):
However, I invite readers not to take my word for it, or to take Jo’s word for it, but to actually read Lubos’s long, detailed and interesting analysis of David’s “notch filter” hypothesis and make up their own minds. To me, the problem you had with Lubos wasn’t that he didn’t understand what David was saying … it was that he understood it all too well, and came to the exact same conclusion I came to—that the hypothesis was without merit.
618
Willis,
You lose, bye bye!
65
G’damn, you are an impatient son of a *****
Are you really finding it that difficult to follow ???
93
ps.. I use to think you had something to offer..
that is rapidly changing to contempt. !
81
Patience Willis, Patience.
I am happy with starting at the beginning and moving to the end and so should you. Jo and David have promised data and code, and I have no reason to doubt that we will get it.
51
Big ENSO swings in near antiphase to solar cycles mess up the correlation and damp the *apparent* response (in the surface temp data), but there is something there.
https://tallbloke.files.wordpress.com/2014/06/ssn-vs-t.png
61
Willis,
Sorry my blogging strategies and timetable are not to your liking. The regular contributors here are emailing me to say how much they are enjoying it.
Because Phil Jones and Michael Mann get your taxes. We don’t. That’s why.
There is no secret. There is just a different schedule to what one occasional commenter would prefer. After five years of unpaid work, I apparently have earned enough goodwill and trust amongst thousands of readers so they enjoy the journey.
You evidently have little faith that I will keep my word and release everything as I have promised. 🙁
As for our friend Lubos, please read my reply again (more carefully this time). Lubos had a bad day. He may have spent hours reading — but clearly did not read enough. It is black and white (you cannot fault my quotes) that he did not understand the central point of the thesis, even when the brief exact answer was personally emailed to him on April 11. That email was five minutes of reading that sadly didn’t happen. Otherwise, his paraphrasing bears no resemblance of my words at all. I’ve given him permission to quote my emails of June 18-20 exactly. An apology would be graciously accepted.
131
His misogyny is showing and he isn’t interested in listening to you. That should be patently obvious. For similar behavior, google “open letter to” site:wattsupwiththat.com and read the comments where his insults continue. Specifically, Carol Costello and the “strikingly good looking” Marcia McNutt. His whack-a-mole conversations with Dr. Curry on her site re-enforce this view. His appeal to authority via his association with Lubos is also an established tactic likely reflects his inner insecurity.
I think you’re dealing with a mancave pig (sic) who is also a parasitic blogger, gaining popularity traction on successful people’s sites rather than hosting his own, but that and all my contributions are just my opinion.
—-
I think you mean this article? – Jo
20
I dunno. His attitude seems independent of sex. IMO.
10
I think you are being extremely unfair Willis. The drip feed release is extremely helpful to those of us who would feel swamped by too much in one hit. David is doing a great job of explaining the building of the model.
You talk about ‘marketing ploy’ – if ‘marketing’ came in to the equation, the last thing you would do is dump a 170 page paper at a blog like this and expect everyone to read it. The drip feed release makes it more likely people will read this, and allows us the opportunity to raise questions and have them answered as we go.
Certainly this approach is novel and ground breaking, but I don’t see it as negative.
We all know one of the problems with the massive UN IPCC reports is that people only read the politicised ‘Summary for policy makers’ rather than the detail within individual chapters which often give quite a different story.
I like this format!
150
I’ll second that, Jaymez.
Willis talks about “marketing” as though it were a nasty thing.
On the contrary, when I studied marketing at Massey, Prof R. Buchanon used to tell us “EVERYTHING is marketing, and marketing is EVERYTHING”.
It’s true, when you are designing a new system, you are trying to develop something no-one else has. That’s marketing. When you interview for a job, you are marketing. When you are attempting to hire employees, you are marketing. When you are trying to convince someone of the merits of your work, it s marketing. What the HELL would you call that which the IPCC does?
Willis appears to have closed his mind to anything with which he is not familiar. This presentation is excellent, and one reason for that is that it IS marketing. It has everyone on the edge of their seat. I’ll wager that there are folks that get up an hour earlier than usual to see the next episode. This sort of suspense engages the mind.
Perhaps Willis’ biggest problem is that someone has recognised something he didn’t? If that is the case, we ought to feel sorry for him. David and Jo have bent over backward to portray this idea in the best possible fashion, and the fact that many of us are enjoying it is obvious. If old sour puss wants to spit the dummy and take his bat and ball and go home, will he be missed?
110
Sales and marketing?
In a sense, everything beyond the spreadsheet on my computer is marketing, for communicating to other people. The main document, the blog posts, the shorter documents to come, they are all “marketing” an idea (though not a product). Without them, no one would know about this development except Joanne and me.
As explained earlier, we are releasing this as a series of blog posts so as to have a more productive discussion. There are a lot of aspects to the theory, so if it was released all at once then the discussions would be all over the place.
We found that presenting people with the whole work at once usually does not work. Most are put off by the length, and even those who do read it often get it wrong. This was clear from the feedback.
(Again, a big thank you to those hardy souls who did provide feedback, even if they did not get what was going on — that was valuable to know too, and the presentation got better as a result.)
Case in point: Lubos Motl. Lubos is bright guy and he spent several hours reading the main document, yet he completely did not get some of the main points that we cover in the first few blog posts of this theory. They are spelled out in the main document, both in the executive summary and the relevant section, both in simple conceptual terms and with equations, but he missed them.
Another case: Miskolczi’s theory. I printed out his main document but it is long and complex. It sat on my desk for 6 months a few years ago, but I never did more than skim it, and I’ll confess I still don’t know what his theory is, though I may have picked up one or two themes by osmosis. If he had written a series of simple blog posts, concepts and diagrams but no equations, I reckon we’d all know what he had figured out.
After these blog posts, almost no one will come to the main documents of the notch-delay solar theory without at least some idea of what it is about and what the main ideas are. When the school system teaches a complex subject with depth, like say maths, they first give some simple ideas, then repeat next year but with some more complexity, and so on. If you gave a calculus book to someone with no previous exposure to mathematics including arithmetic, they are very unlikely to make much headway. I’m not suggesting the notch-delay theory is on the same scale, but the principle applies. It is big enough that, as we found, the all-at-once approach does not work.
231
I think you should ignore Willis altogether and carry on as you wish.
70
Sometimes an employee is so bad you fire them without explanation.
40
Willis, we haven’t got to the hindcasts yet and you want the splits?
Actually I did do splits. Not for the current model, but a previous version (the same without the immediate path, which was added later when I realized it obviously needs to be there). I showed the splits to a few people, but no one showed any interest. When I switched to the new model I chucked out all the old model. So no, I don’t have splits to show you, but we did them.
The splits show broad agreement; they worked. But there is a major problem with uneven data quality — the data before the satellites era (1979 and on) is problematic, especially before 1882 (when the Lean monthly TSI data starts, and the land thermometer network is fairly extensive — HadCrut starts in 1880). We have volcano data from GISS, but only back to 1880, so the model can only try to accommodate the huge eruptions in 1815 and 1836 with TSI data. TSI data is missing during the Maunder minimum from about 1650 to 1700 – we just know that there were no sunspots. And fitting to proxy temperature data before 1850 is of dubious value except in the broadest smoothed sense.
What became obvious is that the issue of the extent of the temperature rise in the land thermometer era is important. The model trains on that temperature data, but if the temperature rises are exaggerated then the fitting will arrive at model parameters that make the model too sensitive. A topic for another day.
152
Beautiful.
44
“if the temperature rises are exaggerated ”
chuckle… well said , David.. 😉
51
Random assignment of data points in time to induction and deduction would solve the data quality problem. I assume the analysis can cope with missing data. I’m sure I once knew but these days …
00
Willis,
You received a lot of red thumbs for that and I can understand your frustration as someone familiar with the field.
Myself, I’m not familiar and likely won’t be much the wiser after all is revealed, unfortunately.
However, I have spent most of my life in sales and I am most certainly familiar with that field. I can assure you David is not doing a sales pitch.
A sales pitch concentrates on selling the ‘sizzle’, not the sausage.
David is plainly laying out the sausages for all to see and it’s up to yourself and others to evaluate the ‘sizzle’.
He’s already said these sausages will go into the bin if they don’t ‘sizzle’.
I don’t think David and Joanne can be fairer than that.
Patience you must have my young padawan.
50
EXTREMELY well described, Raven.
Marketing is one thing, but SELLING is only a part of it.
20
Thanks Wills well put.
without those splits this is a joke
08
There are no splits on the Sunday (US day) menu. That does not mean they are not on another menu. Both you and Willis are behaving like imperious asses because you are treating the current unavailability of something that is out of sequence as if it is never going to be available in spite of assurances what you ask is on the schedule.
50
willis ,i would almost bet my house that david has already done what you suggest .let us wait and see what the next part brings .for those of us not blessed with the smarts,the slow leak process has served a purpose.
100
bit chilly
June 21, 2014 at 6:27 am · Reply
Thanks, chilly. If David has done as I suggest, then he’s mad not to have published the results as soon as he had them in hand. Publishing the model first is just more of the come-on.
However, I’m a patient man, I’m willing to wait some more.
w.
810
Willis,
It’s probably an engineer thing, I do the same thing as David, that is I start from the beginning and proceed to the end. It can seem frustrating to those that want the executive summary first. Engineers like to lay out the reasoning in a way that the evidence leads to their conclusions so they can be cross checked because for us, if we are wrong, 1/2 billion dollar law suits are possible. If you publish the result first, reviewers will brush over the meat, mistakes are made and heads roll.
100
bobl
June 21, 2014 at 1:11 pm · Reply
Thanks, bob, that’s good to know, but … as opposed to what? Starting at the end and working backwards?
You say:
I like to do the same, but he hasn’t published the conclusions, so how can we do that?
I never said he should publish the result first. I said that at the same time that he publishes the model he should publish the results. Otherwise, it’s just an advertisement with no scientific value.
w.
316
So what? It is his choice not yours!
63
All I’m saying here Willis is that David is being thorough in an engineering sense, he is asking us to critique his ideas in a way that allows him to take on board criticism a piece at a time, on HIS timetable and in his order. If he works like me, then David will pull in any valid criticisms that he encounters ( there have been one or two) and update his paper as he goes. At the point he is through this proofing exercise he can then release the paper with the changes included already. To some extent Engineers do this to get things right the first time, because it’s too expensive for engineers to iterate. You saw what happened to the Watts paper for example when it was dumped online, been a much longer path for that paper because of it.
We need to have some respect for David and allow him to work through it in the way he wants even if it isn’t the way we want. I know you want to go through the actual model, so do I but David wants to cover every base to make sure it’s right BEFORE it gets out, that’s a very Engineer thing to do
30
Ps.
Lets me give an example. The commentary here shows that A. Many people miss the point that David is explicitly assuming TSI affects temperature and is proceding to model that, and the consequential very clear need to have an opposing influence in the case of sunspots. This needs to be very clear in the paper.
B. Many people are jumping to the conclusion that Force X must be separate and distinct from TSI or that Force X must be a dynamic feedback, it might be but it is not NECESSARY. A lot of that readership understanding comes from his probably inappropriate use of the word albedo, which is probably technically wrong.
If I were David I’d be reviewing these parts of the paper to remove the ambiguities.
Let’s afford him the opportunity
30
Willis,
On occasion, I am asked to give an opinion on something to some senior people in the Political arena. I have learnt the hard way, that I have to present all of the considerations up-front, and discuss the method and boundary conditions of the approach, before giving them the “Executive Summary”, immediately followed by the “Conclusion”.
If I give them the Executive Summary first, they tend to jump to their own Conclusions, which are invariably different to the “considered opinions”, in my Conclusion. This especially true in contentious situations, where there are multiple agendas present, and several existing belief systems.
Given that Climate Science has become highly politicised, I think that David is being very astute in his approach. It is certainly the one I would follow, if I had his expertise.
41
. . . and make courageous decisions? 😉
21
You been there too …?
40
Willis you do bite hard, seems to me if you see a nail, you have to drive it home, just maybe that nail is there for a reason, like hanging your hat on. Patience is a virtue, you once bit my head off for suggesting the sun gives us our climate. I reminded you that the Earth has but one heater, the sun. Looking at the chaos on Earth and pretending what gives us our varying climate has been an exercise in futility. David is working hard here to give us a way forward from the futility, please be patient and stop biting people, Aussies give everyone a fair go, you are in Aussie territory here.
90
I think you are asking us to believe that:
* The dominant influence of solar variations is not the radiation but something else
* That something else is determined NOW by the state of the sun 10-20 years ago
* and that something else notches out 11-year cycles
My gut feeling is that its too contrived and far-fetched, but I can see how the “slow” variations in TSI can be
amplified to match the temperature variations, but without the appearance of unobserved 11-year cycles.
Are there any simpler versions, maybe that don’t fit the data quite so well but which are more believable?
Maybe with the 11-year cycles being suppressed by a modified low-pass filter (no need for a notch) and lost in the noise (sometimes),
and maybe without the separate delay,
which would leave just TSI being low-pass filtered (implying a delay) and scaled.
06
It would seem that for you at least, there will be no believable version. Mikky, I think you should wait to see how well this model can predict the past and then the future before you judge it.
Remember, Dr. Evans found all of this so far by analyzing actual temperature data — all three filters.
100
I am trying to be constructive, and think it would improve the presentation if each part of the overall filter is justified in terms of temperature data.
So, here is the best fit of the simple version (low-pass filter only), works poorly?,
so here is an improved filter that fits better, etc.
There is a big danger of being too ambitious with the curve fitting, adding more complexity than is believable.
Another point of presentation:
Is it really necessary to dress the thing up in elaborate signal processing theory?
The non-causal filter shown will make some doubt even that the theory is sensible.
People will judge this on “how good is the fit to temperature data” versus “how many adjustable/ad-hoc parameters are there”.
I don’t think many will judge it on its theoretical foundations, which are in danger of detracting from the eagerly anticipated money shots.
00
I quoted two sentences from your comment. I will agree on your second sentence.
It should be judged, as should any theory of how the Earth’s climate works, on whether it predicts accurately. That’s why I’ve adopted a wait and see attitude about it.
Frankly, if pulling random numbers out of a hat could predict the climate today from the past temperature records and do it consistently every time, I would be pushing that theory. It’s results that count, not how you get them.
I don’t know what your first sentence means. When something works and is proven to work then how it works should be of interest but It can hardly be a matter of controversy if that’s what you’re hinting at.
50
My gut feeling is that its too contrived and far-fetched
The wave-particle duality seems contrived and farfetched.
00
Congratulations on the new theory. There is now a sparring partner to go up against the GCMs, if you don’t mind the pugilistic language. I hope you score a knockout, but I do have a question. Why do you hypothesize that the notch filter is connected to the Sun? There is some elegance in that, because we already know that the Sun has cycles of the approximately 11 year span that you would be looking for. However, there is an alternative. As you point out, climate on Earth does not vary appreciably with variations in TSI accompanying the solar cycles. So anything which cancels increases in solar insolation would do the job that you require, including processes on Earth. It so happens that there are also extant theories showing that various processes involving clouds cancel out a lot of the effects of solar radiation. Theories of Spencer, Eschenbach, and Lindzen come to mind, but I have come across others which esccape me at the moment. One advantage of these theories is that since the effects they invoke are virtually instantaneous they don’t require new forces or a delay.
Far be it from me to cast doubt on any theories at the moment, but my question is whether there was empirical evidence that led you to exclude cloud feedbacks as the factor nullifying variations in insolation? Apologies if you have already addressed this point somewhere.
10
David did not “exclude cloud feedbacks as the factor”, although “feedbacks” can never be defined. David explicitly said something between TSI and ground temperature is futzing things up really weird! It would good to know what else is happening between Solar flux at the Sun, and poorly measured TSI, especally in the UV.
13
The notch requires a delay, and the delay apparently can only come from the Sun. Post III.
There are clouds feedbacks in the RATS multiplier. Since the system is linear for small perturbations, those albedo changes presumably superpose (or act alongside without essentially interfering) the force X albedo changes.
30
Thank you. That just pulled together and answered a lot of ill formed questions in my mind. The key aha (proving that sometimes the obvious in hindsight is not in foresight) is that the to most most important purely atmospheric feedbacks are water vapor and clouds. And these are obviously related (clouds produce precipitation that lowers water vapor, water vapor produces clouds). And these are both related to convection cells, mostly over the oceans. And therefor are obviously ‘fast’ feedbacks operating on scales of hours to days , and certainly faster than seasonsmwhichnyour model cannot resolve. Lindzen’s adaptive iris, and Eschenbachs thermoregulation hypothesis, both say the net effect is strongly damping. So net net consequence is in RATS.
This leaves only the second order possibility that RATS may not be a constant, but itself a function of temperature. That question is perhaps for a second refining generation of the model, if the first generation here is found to work reasonably well.
10
Another though from someone who has built a lot of different models of different sorts. To the extent your model is proved to work. There is a way to sort attribution into three piles using the period roughly from 1965 to the present, in which we have a temperature rise followed by a pause nearly now half as long. Postulate three components: solar, CO2, and ‘natural Earth’ ( stadium wave, ocean, unknown, whatever). Now postulate that the natural component changes sign but not strength of impact on deltaT. We know that the GCMs are CO2 only forced. Then by subtracting your model result (solar, from data also incorporating natural) from their result (CO2 tuned to past everything including natural) one gets a rough approximation( missing cross products, of course) estimate for natural residual. This also allows forward predictions for a few years of the magnitude of each component on T.
Reason this osminterestingmif viable is that the most reasonable physical reality IMO is that all three are on the mix, and knowing the proportions enables more refined estimates of CO2 sensitivity, which appears to be the biggest policy driver.
I would note in passing that the regulatory policy meme in the US is already moving moving to carbon pollution including health effects of PM2.5, since globe isn’t warming, and climate change is just weather, and the extremes are not increasing. Pointing to the next battle over fossil fuels When ( not if ) skeptics win this one.
00
Thought exercise:
Inner tube with a hole in it. Air pump feeding air in at a constant rate. Tube begins to inflate until equilibrium is reached. Now, measure flow of air out of the hole, pressure inside the tube, volume inside the tube. Calculate what the pressure and volume will be if you increase the rate of air flowing into the tube in a step change. You will probably be wrong and your calculation will likely be too high (too much volume, too much pressure in the tube). Because as the tube expands, so does the hole and the rate of air escaping increases, too. In fact, if you step the air input up instantaneously, you might see the pressure of the tube increase rapidly, then reduce a bit, and then gradually start filling again if you play that at very slow speed because of the time lag it takes for the rubber to stretch. The air will go in instantly increasing the pressure but the hole has not expanded yet.
40
Yes! This is the modulation I’m talking about. With enough modulation and a resonant system caused by time delays, you get an oscillator. This is why farts have a pitch!
22
When you come up with an example of a THERMAL system acting like that, let us know. I’ve asked David if he had an example of a thermal notch filter, saying that I knew of none … he claimed that my question was immaterial to the case, viz:
From that I assumed, as anyone would, that his aggressive circumlocution meant no, he’d never seen a thermal notch filter either … so do you have such an example?
w.
212
I believe he is saying he believes the system behaves as if there is one. He isn’t (at least so far) claiming to know what it is at this stage. From how I read it so far, he is saying that given certain inputs the output is behaving as if there is such a notch filter. WHY it is behaving that way has so far been beyond the scope of what is being presented.
If the model of the behavior is good enough, one doesn’t even need to know why in order to predict a good result. For example, if I have a box with a crank on one side and a wheel on the other and if I crank the crank three turns and the wheel makes 5 turns, I know that overall whatever is inside the box produces a 3:5 turns ratio. There can be any number of ways of accomplishing that result from very simple to very elaborate but one doesn’t really need to know what is going on inside the box to know that if they crank the handle 6 times the wheel will make 10 turns.
Dr. Evans is saying that the response to a change of input is behaving like a notch filter. Exactly why that is might (or might not) take a lot more research to figure out.
80
Just imagine the wonderful music if all in the universe were farting in acordance with the “director”
11
Ahhh, all that yummy CO2 and CH4 (and olfactory ester)
11
It’s called loss of precision, Jo. And you’re entitled to call it the first cousin of GIGO, garbage in, garbage out. 😉
60
Where is the feedback. Even if CO2 is ignored, H2O must surely be included as a feedback. It is physically unrealistic with no feedback.
04
Not relevant, TSI in, temperature out (after all feedbacks are applied). David places the boundaries of the system at points so the feedbacks get incorporated into the filters.
40
So are you saying that the notch and delay filters are “active filters” rather than passive. That would make sense but it certainly isn’t obvious from the text.
David Evans perhaps you might like to elaborate on this in the document. Cheers.
01
bobl is right. And we’ll get to feedbacks, especially water vapor feedbacks, soon. Btw, see this.
30
Hey,
tinylittlewillyWillie.Why don’t you respond?
Scared of a debate with a real scientist?
Maybe any other model is superior (let’s confine things to prediction, shall we?)?
All others? Including wiggle matchers?
[Willis already smacked you down in his response Hamish, but anymore of these sorts of useless comments and you’ll go into the sin bin. mod oggi]
02
You open your comment by claiming I have a tiny Johnson? Gosh, Hamish, I was unaware that you cared so much about the other guy’s junk … I used to get lots of spam encouraging me to enlarge my Johnson, maybe I shouldn’t have trashed them all.
However, grade school taunts? That’s your idea of how to open a discussion with someone? And you accuse me of not being scientific?
Why don’t I respond to what? A quote, citation, or link would make this question meaningful. As it stands, it is babble, with no subject.
Which real scientist are you referring to? Can’t be you, I’ve never had a real scientist that fascinated by the size of my Johnson.
And it can’t be David, since I’ve been discussing this with David for the last three posts, and he certainly strikes me as a real scientist, as does MKelly … you should try to follow the bouncing ball. By the way, David didn’t seem all that concerned about the size of my Johnson …
Say what? What does that mean? Maybe any other model is superior to what? Superior to your model? My model? David’s model? The IPCC model? Details, Hamish, details are critical, without them you’re just blowing smoke.
Again, I fear that is incoherent. Who are the “wiggle matchers” when they’re at home?
Let me offer a bit of friendly advice, Hamish. If you would slow the flow of your unpleasantness enough to actually let us know what it is that has your knickers in such a twist, I could respond to it.
But as it stands, there’s not the slightest indication of what it is that you think I should respond to.
w.
[Marks for very amusing response! – Mod]
711
Ummmm …. I assume he’s trying to taunt Connolley, not you. I’ve seen this elsewhere a while ago.
72
Queue Helen Reddy. Wrong Willie, W.
20
dp
June 21, 2014 at 12:58 pm · Reply
Thanks, dp … and you know this how?
In any case, the moderator has already warned him, so my job is done.
w.
PS—I do believe you meant “Cue Helen Reddy”, although it’s totally unclear just what she might have to do with this …
41
Senior moment – s/Helen Reddy/Carly Simon/. It was a great old song.
Given the target of the insult I doubt a lot of sympathy is going to be extended, but given an adjacent post on civility, good for Jo for the corrective action.
21
Thank you for beautifully illustrating the scientific method.
The unrealistic goal of consensus science is the final answer.
A much smarter person correctly concluded many centuries ago,
“To know that you do not know is best.
To pretend to know what you do not know
Is a disease.”
60
Can we ever get the clowns to admit ignorance, and get the money back?
54
For those tempted to “toss out” this theory before seeing it clear through to the end, here’s a little true story.
I once bought a paperback book at the local news stand. I had no idea what it was but I wasn’t finding anything else that looked interesting and the blurb on the back was at least “passable”. So I took it home and started reading. After just a few pages I said to myself, “How can I take this nonsense seriously?” I was about to toss it but then I remembered I had bought it with cash out of my pocket and thought, OK, I’ll go on a little farther and see what develops. What can I lose? The money was already spent.
After the first chapter I couldn’t put it down.
It would have been a great loss to have tossed away that first volume of Lord of the Rings just because I judged the whole adventure by the first few rather nonsensical pages and I would have missed the best reading I’ve ever had.
Over the years I’ve read that trilogy many times and I still can’t put it down once I get started.
So maybe a word to the wise is sufficient? And if at the end it turns out to be nothing or questionable I’ll be glad to read your complaints.
40
PS:
I’m rather enjoying seeing how David’s mind works as he develops his thesis based on a whole different way of looking at the climate and what influences it.
Among other things, it’s clear that a lot of careful thought and a lot of work went into this and it seems a bit unfair to judge it before seeing the whole thing. 🙂
60
I am kinda scared of the whole thing! Riots everywhere!
20
Who will have any need to riot? I think we need not worry about that.
The ones with the financial interest in vilifying CO2 will have a big problem. But they’ll hardly be able to solve it by rioting.
30
Trust an old pilot not to panic 😉
10
Amen, Yonni! 🙂
10
Jo I suggest very few people are looking at this and the other posts about D Evans work because it is too scientific and probably extremely relevant. You will have to wait for it to sift through the system.In the end it will be proven to be correct. In about 5 years.
00
I disagree.
There are at least 250 to 300 scientists and engineers that haunt WUWT and I am sure many have hopped over here for a read of Dr. Evan’s work.
You just aren’t going to see much discussion until his presentation is completed.
I have certainly seen lengthy discussions over at WUWT that were ‘Above my pay grade’ as far as the math was concerned although I could usually follow the discussion.
110
I disagree. This presentation method is masterful as it allows the student (me) to internalize these ideas in bite sized chunks. It is also written at a level that I comprehend and I am able to review areas of the work after reading some of the more pithier comments. I sense that one commenter is worried that this might blow his life’s work and fortune out of the sea. Don’t worry, many of your articles at WUWT are stretched out over several posts.
40
Agree, a day or three to consider each part is great.
A single huge article, or 6 articles in a day would swamp me. I couldn’t read it.
But this is riveting.
10
Remember the tale of The Arabian Nights.
Young lady kept the guy on the hook for weeks.
10
Technical question on the method used to fit parameters.
It looks to me like you adjusted the parameters of the model to try to fit to the global temperature record using the imputed TSI record as input.
An alternative and in many ways simpler approach would be to adjust the parameters of the model to try to fit the model transfer function to the calculated transfer function.
Could you comment on these two methods. Whay did you choose the former approach over the latter?
00
Interesting question.
Answer Whatever works well!
00
I did both simultaneously, driving the model with composite TSI and fitting the temperature record then produced to the composite temperature record, and fitting the transfer function to the empirical transfer function, giving priority to the latter.
Did them both because the solution has to fit both of those constraints.
The transfer function was only fitted to the *amplitude* of the empirical transfer function, because we do not know its phase. There are many transfer functions whose amplitude fits the empirical but whose performance on temperature is woeful — because the phases of the transfer function of the model are presumably very wrong. So just fitting the transfer function amplitude is not enough, need to fit temperatures.
If we just fit temperatures we can get some fits that do quite well but have moved a fair way off the empirical transfer function — and we know those cannot be right either.
30
With careful choice of parameters you could use the GCMs to model a doctor’s prescription.
10
But not the handwriting.
30
If you are going to do this all the way back to the 17th century, I have a hunch (and only a hunch) that the altitude of the CO2 emission layer has risen since then. Don’t know how that will impact things, but I would expect the point at which the atmosphere goes opaque at CO2 absorption frequencies would rise as more CO2 is added to the atmosphere.
10
Is this concept valid? Why would a trace gas exist in a “layer”, if that layer were not immediately on the surface in the absence of any air movement. (Consider the way in which H2S pools on the floor of a building).
It seems to me that at 400 ppm it would be equally distributed throughout a homogeneous atmosphere. How would a layer of a trace gas exist in even a light breeze?
00
Dr Evans is not talking about a ‘layer’ of CO2.
Visualize the CO2 in the atmosphere as layers of thin translucent paint applied over a window. The ‘layer’ in question is the point where the additional layer of paint causes the window to be totally opaque. The corresponding point in the atmosphere is the height where CO2 wavebands are saturated.
If I have that wrong someone please correct.
00
I would visualize it like this using the optical wavelengths as an example:
Imagine a planet where you could see the surface clearly from orbit. Now you start pumping smoke into the atmosphere. At first you can still see the surface but it is getting hazy. At some point you can no longer make out the surface but maybe you can see airplanes flying 10,000 feet above the surface. If you keep adding smoke, eventually you can’t see those, either, but maybe you can still see planes flying at 30,000 feet.
Same would happen if your vision were in the wavelengths that CO2 absorbs as you add more CO2 to the atmosphere. I would assume water vapor hasn’t changed enough to make any difference in that layer over time.
00
And we also need to understand another thing in that we aren’t dealing with “layers” here so much as for the first 8km or so we have a mixture of water vapor AND CO2. Above 8km, water vapor tends to be the dominant operator. But below 8km it is a mixture of both and water dominates at all altitudes because it is a more efficient absorbent of LWIR over a wider range of frequencies.
This is one reason why a conventional greenhouse is a bad analogy for Earth’s atmosphere. In a greenhouse you have convective barrier at the top AND that convective barrier is also a barrier to LWIR. It absorbs heat from both radiation and convection and re-radiates half of that back from the top. Earth’s barrier to LWIR starts at the ground where it is densest and thins at altitude. Instead of solid glass barrier, imagine a “greenhouse” filled with aerogel that is most dense at the bottom and thins on the way up and is made out of two different materials that absorb different wavelengths of LWIR but one of the materials is in much less proportion to the other and a convective barrier at the top that is transparent to LWIR. It’s a more difficult problem than simply a convective barrier that is ALSO a barrier to LWIR.
30
I consider the atmospheric GHG effect (H2O+CO2+CH4+…) to be like a greenhouse with half its roof panels removed. These missing roof panels correspond to the transparent window in the atmosphere. If you include the effects of clouds, less then 1/4 of the panels are removed.
The transparent window is often understated because when considered globally, you must include the effects of clouds. Line by line simulations of the standard atmosphere demonstrates a window size near 50%, that is, for the clear sky, half of the energy in photons emitted by the surface leaves the planet without being absorbed by a GHG molecule or cloud. Considering that about 66% of the surface is covered by clouds and cloud pass an average of 20% of the surface power (from ISCCP satellite data which lumps mostly cloudy with cloudy), the net transparent window can be calculated as the window relative to clear sky,
T = (1-.66)*.50
plus the surface emission window relative to cloudy skies,
+ .66*.2*.50
whose sum is .236, or about 24%.
From this, we can easily calculate what the surface gain should be, that is, the ratio between emitted surface power (Ps) and incident post albedo power (Pi) which is given by,
gain = Ps/Pi = 2/(1 + T) = 1.62
For Pi = 239 W/m^2 @ 255K and Ps = 385 W/m^2 @ 287K, the measured gain is 385/239 = 1.61
For those who are interested in where 2/(1 + T) came from, consider that what is not passed must be absorbed by the atmosphere and clouds and given the omega term of the RT differential equation , energy entering the atmosphere from the surface leaves by both returning to the surface and being emitted to space, thus half of what enters from the surface is emitted to space and the other half is returned to the surface.
If we define A = 1 – T to represent the fraction of surface emissions absorbed by the atmosphere and half of that is returned to the surface, we can calculate the gain, Ps/Pi as 1/(1 – A/2) which after the substituting A with a function of T is equal to 2/(1 + T).
Finally, to close the loop, the power entering the surface is equal to Pi + Ps*A/2 = 386 W/m^2, which given the relatively loose precision of the specified window, cloud transparency and cloud coverage is pretty close to the 385 W/m^2 of average surface emissions leaving the surface, which of course must be the same for a surface in LTE.
George
00
At ground level 300ppmv CO2 absorbs 98% of 14.6 micron energy within 6 meters. At an altitude of 10 km that much is absorbed in 800 meters. Looking up the 14.6 apears as at ground temperature. Looking down from above 14.6 apears as the tempereature of the thermopause. Adding more decreases the distances some, but not the aparent temperature.
11
Yes, that is correct crosspatch, you have it.
CO2 changes are only a mechanism within the (TS in, temperature out) system, which we are assuming is invariant, so our analysis method is ok. Of course it’s not quite invariant, and changing the CO2 concentration is one way the climate system might be different to what it was 200 years ago. Whether it is different enough to affect the conclusions we essentially do not know except by seeing of the model works — which we will find out in a few years.
40
I am greatly surprised at the naysayers. Do you really think David and Jo would lead you on a wild goose-chase if they didn’t already know your questions and had answers at hand? Didn’t they say they spent 18 months of research(with external review) to get to this point where they were ready to publish?
This is like parents trying to educate their kids on some issue they already think they know….8>)
Patience is a virtue… It is also much easier with a bottle of wine/scotch/beer.
Have a good weekend all!!
70
I have no problems with the exercise David has embarked on. But is is interesting how almost all of the ‘gurus’ here e.g. Stephen Wilde are obsessed with finding a top down source for Force X (even if it is suggested it has ‘something to do’ with cloudiness
IMHO there is a stunning degree of parochiality (?) and/or hubris (?) and/or plain naivety (?) involved here when purporting to deconvolute the mechanics of a system which originated and began evolving from approximately 3.5 billion years ago, dealing rather effectively, with every scale and frequency of perturbation – cyclic (Milankovitch), semi-cyclic (volcanic, plate tectonic) and tightly cyclical (solar) ever since…..
I think Lovelock did us all a big disfavor by injecting a certain amount of hippy jargon or 60s spirituality into the field? It is clear the Radio Shack generation just can’t hack even the faint whiff of it.
Wherefore art thou, Homeostasis? To have Evolved or not to have Evolved – that is the question!
11
Top down modulated by bottom up.
Not’ obsessed’, merely noting that it fits observations rather well.
Hoping that it remains consistent with David and Jo’s work to the end but meanwhile looking at each new element with some trepidation in case I’ve been wasting my time 🙂
30
Edison didn’t have to know WHY a current flowed inside a light bulb when one placed a plate with a positive charge relative to the filament inside of a bulb, he simply documented that it did.
31
David, have you published the actual function for your model anywhere? Not the functional diagram that you post above (which is helpful), but the actual function where you feed the sunspot numbers in at one end and temperature comes out the other. I’ve looked around, but I haven’t been able to find it.
Regards,
w.
410
And are the data, spreadsheets, programs etc available for public review and auditing?
[They will be David. On another matter to readers generally – why give thumbs down to people who simply ask questions? – Mod]
513
How about doing a bit of reading before opening your mouth?
21
How about answering the question after you open yours?
313
FW !
21
David Appel, YELLOW CARD!
They have said it would be. Said it at the very beginning. Not yet posted is just part of their reveal style. Please go sit in the patience penalty box. And if you wish to come back on the playing field early, please consider that I will take it as a personal honor to play directly against you here and at Climate Etc. Oh, and I can play for keeps, since trained by the same folks who trained OBummer.
H/t to the World Cup.
151
David Appell
June 21, 2014 at 11:24 am
Rud Istvan
June 21, 2014 at 2:54 pm
Huh? David Appel’s question is perfectly valid, and quite relevant. Yes, David Evans and Jo said at the start that they would publish them.
As a result, asking, as David did, if they have made good on their word is not out of line in the slightest.
However, your sensitivity on the subject, a sensitivity which is strong enough to make you pretend to have the power of a referee who can interdict serious and significant questions that you don’t happen to like, is indeed revealing …
w.
116
Willis , in case you miss my other reply ;-
http://www.youtube.com/watch?v=Tzhp3jWnpbs
10
Get the mechanism right and the numbers will follow.
David remains open minded about the precise nature of force x. Until that is clarified precision is not possible.
However, predictions of direction of trend should be possible and that will be an advance on GCMs.
80
Stephen, are you claiming that to date David doesn’t have the numbers? Because Jo’s already claimed success in hindcasting with the model, viz:
So how are they doing that without the formula and the numbers?
Note, however, that while I’d like the parameter values, I didn’t ask for them.
Instead, I asked for the full function to get from sunspot fourier to temperature fourier. He can put in “P1”, “P2” for the various parameters if he wishes.
w.
w.
014
Wills I am sure if you go back to the first post you will read that David did ask a number of people to do initial private reviews I assume of the whole 170?page full document.
If you get on Jo’s mailing list you might get the hole thing now to save you waiting and postulating on half information.
50
ColA
June 21, 2014 at 3:48 pm
Thanks, CofA. I’m not interested in what happens within the Seekret Cirkel of David’s Climate Besties, whom I am assured are all poring over the Seekrit Document.
I’m interested in transparency. If it’s not transparent, it’s not science … and to date, it’s not science.
Yes, it is common practice to circulate such documents BEFORE publication. What is very uncommon is to publish all of the claims, all of the teasers about how well it’s performed for hindcasting, all of the hype and the claims of future reveals, and not publish any the details … while revealing the details only to the inner circle of your friends.
That, to me is a tactic designed to get lots of supporters. They are all signing on to buy a pig in a poke, an idea for which they have none of the details, only the broad outlines. It’s science by press release before the study is published, the exact same anti-scientific practice used by the alarmists. The alarmists make all kinds of claims in the PR statement, and the idea gains force and goes around the world as truth, despite the underlying study not being released … does this sound familiar?
And unfortunately, it’s a very effective technique. Once you have people doing that, once they’ve bought in to a vague idea, you can put almost anything in the poke and they won’t complain … because since they have already bought in, and bought in very publicly, they’re very, very unwilling to admit that they were wrong. It’s not unusual or unique, in fact we have a folk story of this exact problem … it’s called “The Emperor’s New Clothes”. People hate to be wrong, they hate to admit they were fooled, they’ll do anything other than that … and as a result, David has already gathered a bunch of adherents who are extremely unlikely to publicly change their minds.
Now, was that their conscious intention in all of this? I doubt that greatly, neither David nor Jo strike me as being that way … but had they paused to think about it, it is certainly the predictable outcome.
For me, it’s all just too tawdy to see skeptics taking up the alarmists’ practice of science by press release. Publish and be done with it! If the review of the Seekret Document is not complete … then why on earth have they started the PR campaign already?
And if the review is complete … then why on earth haven’t they published it?
w.
[There is no hidden agenda other than to properly explain a model which will no doubt receive a lot of scrutiny from antagonists. David has promised to make all data and coding and the maths publicly available as part of this process. Something almost unique in warmist circles! Be patient Willis. – Mod]
025
I’m sorry Willis,
I’ve read over the years many of your thought bubble flights of fancy at WUWT, and have never seen you held to such a snarky account as you are attempting with David. I suspect you are getting somewhat territorial in your old age, and seem unhappy with someone taking a different approach to that you consider he should according to your own personal credo. All that you suggest might be reasonable to assert if it weren’t for the fact that our host and hostess have said initially in their posts (and repeatedly thereafter) how they were going to approach this, and have addressed your concerns by stating that they intend to release all the data and coding that is required to evaluate their theory and its practical application, yet you still persist even when the moderators and DE and JN have repeatedly replied to you to the contrary.
Now, you have asked people to indulge you in your posts at WUWT in the interests of giving you a fair hearing, so my question to you is: What business is it of yours what process David goes through in releasing his hypothesis? Especially since clearly it will be scrutinised to the hilt by not only warmists, who will try to tear it to shreds (even if it is correct- such is their desire for the quest for truth!), but also by sceptics wishing to be protecting their turf in the battle against the political and academic establishment, as well as their position in the pecking order which their egos seem incapable of reconciling with their natural intellectual curiosity.
The truth will out, and if David’s hypothesis is not able to be sustained then that will become evident in the fullness of time when the full 170 page paper is finally released. We are sceptical by nature, Willis, and none of us are approaching this as anything more than an interesting, novel approach to describing the earth/solar connection and its manifestation as our global climate. Whether it holds up to scrutiny, only time will tell. If it doesn’t so be it, but if it does seem to hind cast well, then (and only then) will the test as to its predictive capability be the final arbiter of whether we accept the framework David is constructing. In the meantime, you’ve made your point so let’s just wait and see how the cards are dealt, Willis. Take it from a fellow sceptic, your responses do you no credit if only for their argumentative tone, and the insulting assumptions you make about the denizens here, who are not half as gullible as you seem to be suggesting.
350
See this comment
“Bernie Hutchins
June 22, 2014 at 3:05 am · Reply
David –
You keep claiming that your Fig. 2 of Part 3 is correct. But no one can check it because you haven’t said how you got it. A step response should be the inverse Laplace transform of (1/s)T(s). But you haven’t even given us T(s). Such Laplace inversion integrals rank with pure torture, and are easy to mess up. This is why I did the simple EXPERIMENT instead, one I have done for 40 years, and got the expected and proper causal result. (No need to quote Feynman about the relative value of theory vs. experiments here!) Your Fig. 2 is at least BACKWARD in time, and offset vertically. Yours appears to be critically damped (Q=1/2?). Mine was slightly underdamped intentionally (Butterworth) so that it rang slightly, as a better illustration.”
How do you expect people to make critical evaluations when the data release is piecemeal
NOTHING prevent them from releasing the data and code to support every installment of the series.
10
I think this statement is at the root of Willis’ problem: “They are all signing on to buy a pig in a poke, an idea for which they have none of the details, only the broad outlines. It’s science by press release before the study is published, the exact same anti-scientific practice used by the alarmists.”
He is so accustomed to the traditional (in the last couple of decades) of publishing that he can’t fathom what is going on here. Perhaps if he had been around during many of the discussions here about the faults in that process he would understand.
I don’t believe any of the regulars here have yet “bought in”. We are all sufficiently skeptical to understand that it’s not over ’til the fat lady sings.
I think what the majority of the regulars are sold on is this novel approach of introducing something new. It was Richard Feynman that said ” Science is the belief in the ignorance of the experts”. The phony “peer review” or worse the “pal review” process is in fact a very inefficient way of arriving at the the Truth, the Whole Truth, and nothing but the Truth. There are many aspects to this novel approach that I feel appealing, and I sense that I am not alone. This is Liberty and Democracy in its finest form. Those that have an interest, and those of us that have life experiences that may perhaps shed a bit of light on the process, not to mention those whose expertise is in this chosen field, will arrive at the Truth faster, I believe, than a handful of narcissistic “experts”. Furthermore, it is exciting, and it is fun.
And it is that novel approach, I believe, that Willis does not understand and it is that that makes him so angry.
150
If you don’t like what is being sold to you Willis, the door is over there ->.
My respect for you is disappearing rapidly, all within one post. Why can’t you let it unfold as it will and value the idea on it’s merits.
It’s unlikely to be correct in it’s first incarnation, which is why feedback and questions are so valuable. Your hostile rather than curious.
50
Oh they also said they will post all the data and the equation on Excel – that will be fun!
30
Phil, Phil Jones… where are you… I need your help on the Excel stuff, dude !!!
51
No, the notch-delay solar model has not been published yet.
As explained in post I, this series of posts is the first time anything about the project has been made public, and:
“All the data, model, and computations are in a single Microsoft Excel spreadsheet. It runs on any pc with Excel 2007 or later; it runs at least partly (and maybe fully) on any Mac with Office 2011 or later. This is completely open science—every bit of data and every computation is open for inspection. We will be releasing this towards the end of the series of blog posts.”
See this answer.
120
Edison had the benefits of Gauss’ Law, Ampere’s circuital law, Maxwells equations etc., etc. Whether he fully understood them is not our business.
Everyone here has the opportunity to e.g. to take out a subscription to Nature Geoscience etc.
00
And he was quite unaware of any of those and the electron wouldn’t be discovered as a separate particle until nearly two decades after he discovered he. He was just trying to understand why filaments of his bulbs blacked more at one terminal than the other.
The point here is that Dr. Evans has not attributed a specific cause, instead calling it “X”, but is describing how the system behaves. It is basically a black box. I expect him to come under attack and expect derision from many angles because there are a lot of vested interests at stake in the CO2 charade. Every dollar spent “fighting global warming” lands in someone’s pocket and they are going to fight to keep it. It also enables the policies of those would use fear of it to enact a global economic agenda of “global redistribution of wealth” from developed to developing countries using strict environmental regulations in developed countries to force economic activity to flee to areas without them.
Even more important is going to be the fallout of some of the establishment political parties who have staked so much on the issue. Political parties can play very dirty pool and this risks them looking like the charlatans they really are. It is going to risk knocking very powerful people out of power and making a mockery of some who are considered celebrities in those circles. Most of all it will anger the people when they realize the extent of the deceit.
So the political class will throw the scientists under the bus. They will use the excuse that “97% of them said so” while diverting attention from the fact that those “97%” said so because they were making money hand over fist for saying so and risked losing their jobs if they said anything different.
I believe this might be the beginning of the end of the greatest robbery ever perpetrated in the history of the human race.
120
Crosspatch said, “So the political class will throw the scientists under the bus. They will use the excuse that “97% of them said so” while diverting attention from the fact that those “97%” said so because they were making money hand over fist for saying so and risked losing their jobs if they said anything different.
Whilst I can understand the position that people find themselves in when having to conform to the prevailing tyrannical (co2) religion, as time goes on my view has hardened and I think that if large numbers of scientists end up getting ‘thrown under the bus’ then I shall feel little sympathy for them.
40
Steve, some us have. Or the Equivalents. Years ago.
Are you implying we ‘deniers’ are unread ? If so, please riddle me this. Lindzen’s adaptive iris hypothesis was published in what year, in what journal? Zweir’s ( as co author) statistical refutation of GCM validity given the pause, to 95% confidence, was published in which 2013 Nature subjournal?
I was taught not to bring a knife to a gunfight. A lesson you need to learn.
30
David –
Step Responses of Notch Filters are Causal.
Please explain why you are using an anti-causal step response? Contrary to what you said back in Part III, notches are causal. At least on my bench. Really – it couldn’t possibly be otherwise! It just rings. Here is my actual experiment.
http://electronotes.netfirms.com/NotchStep.jpg
I used a standard state-variable notch as seen in the jpg. The notch position was about 800 Hz (not important) and the Q is about Butterworth. (Yes I can do a notch with just one op-amp.) And the asymptotic level long after the step should be to the step height (The LP step).
What am I missing?
10
Bernie I’ve no idea and am not going to debug and figure out what you have done exactly.
The step response to the notch filter in Figure 1 here is non-causal, as shown in Figure 2.
The step response of the complete model above, in Figure 4, is causal — but without the delay it is not.
00
David –
You keep claiming that your Fig. 2 of Part 3 is correct. But no one can check it because you haven’t said how you got it. A step response should be the inverse Laplace transform of (1/s)T(s). But you haven’t even given us T(s). Such Laplace inversion integrals rank with pure torture, and are easy to mess up. This is why I did the simple EXPERIMENT instead, one I have done for 40 years, and got the expected and proper causal result. (No need to quote Feynman about the relative value of theory vs. experiments here!) Your Fig. 2 is at least BACKWARD in time, and offset vertically. Yours appears to be critically damped (Q=1/2?). Mine was slightly underdamped intentionally (Butterworth) so that it rang slightly, as a better illustration.
You said in Part III:
“in any electronic notch filter without an accompanying delay, the step response is blatantly non-causal. Notch filters by themselves are intrinsically non-causal.”
and in a reply comment to ControlPhD (who also correctly pointed out, before me), that your Fig. 2 didn’t make senses you said:
“It is well known that notch filters are non-causal.”
WHERE (specifically) did you get that idea? Sorry, but I think you are just WRONG. You really don’t need to worry about “debugging” my offering. Fix your own Fig. 2.
I hope doing it right won’t mess up the rest of your theory.
12
Perhaps you are using a different sign convention for phases. Have you tried changing the sign of the phase?
A Butterworth filter is causal but it is a low pass filter, not a notch filter.
Even if the causality argument was wrong, the delay works and is independently corroborated.
00
I’m with Bernie here, totally confused as to why you have CHOSEN to have acausal filters.
People only do that when processing recorded data.
You could have CHOSEN to implement a notch filter causally, like the ones that suppress mains hum.
Yes there is a delay, but you get that with all filters.
Your delay looks like something needed to get you out of the hole you created by CHOOSING an acausal filter.
00
In simple terms (my zone)
>”The step response of the immediate path”
Fine, the “fast” response path. No problem there. No departure from anything known already.
>”The step response of the delayed path”
Fine again, the “slow” response path. Out to 15 years, entirely consistent with anyone else who has calculated a planetary delay (land+ocean) by other techniques e.g. stats, thermo principles. Abdussamatov calculates 14 +/- 6 years by thermodynamic principles, Scafetta 12 years by stats, Trenberth 6, and 10+ years out to 100 years oceanic delay simply by observation I think. To name some.
But then, apparently, slow response “is due to force X”.
Slow response (without the notch) is simply due to the thermal characteristics of surface material i.e. thermal inertia of: ocean (mainly), land, lakes, rivers, rocks, sand, soil, grasslands, vegetation, foliage etc and human modification of all that.
So the model is thermally conventional up to this point despite the introduction of the term “Force X” to describe the entire slow response which to me is fallacious.
The only departure from convention is the notch (requiring Force X), identified by an absence of a cyclic “pulse” of temperature in the globally averaged surface datasets at around 11 years by temperature spectra analysis. Plenty of literature covering how the pulse does translate to climate using other techniques and datasets. At the surface, in the troposphere, in the stratosphere, at the sea surface, in the subsurface, and underground down nearly 30m. Not necessarily as a temperature pulse at 11 years though but there is evidence of it.
Nevertheless, the notch is established on the narrow scope of globally averaged surface datasets that, intuitively surely, wouldn’t be sensitive enough even with filtering to reveal that cycle because only a fraction of the stations in the global set would have the local conditions allowing such sensitivity. I’m basing my opinion on the work of Tatiana Barlyaeva who says she identifies a cyclic pulse of “about” 11 years in some specific stations in some timeframes. Not conclusively in my opinion but I think that is where to look rather than in global averages.
So my main question is: over the long-term (say steps of 100, 200, 400, 1000 yrs), how significant is the notch?
And in the short-term (multidecadal)? This seems to be where Force X is significant. How does the model perform with vs without Force X (or notch transfer)?
Could the model do without the notch if the basis for it is just too flimsy and still be validated by the fast path and slow path delay system?
I’m inclined to think so but I’m certainly willing to be proved wrong if the model does actually track temperature progression in the short-term (to end of 2017 should be enough).
00
Force x is needed to change albedo BEFORE the proportion of TSI entering the oceans changes and THEN the oceans cause the delay.
Assuming force x from the quiet sun has increased global cloudiness via a more meridional global air circulation pattern then less energy has been entereing the oceans since 2000 than before 2000.
We can see that since 2000 El Ninos have become less powerful relative to La Ninas.
So far the decline in solar activity has been enough to stop warming but as yet not enough to start a decline.
David’s model (and mine) both predict direction of trend but not precise scale or timing of that trend.
If there is no further warming by 2017 despite increasing CO2 then that would support the solar models.
I think Salvatore del Pret has suggested how weak for how long the sun must become for an actual downward trend to begin and we are not there yet.
The next half of the current solar cycle should tip the balance towards a measurable downward trend.
The recent solar maximum has slowed down the timing of the shift to cooling.
David and Jo have suggested 5 years for validation or not and I concur.
71
Stephen Wilde #33.1
Thanks for reply Stephan.
>”Force x is needed to change albedo BEFORE the proportion of TSI entering the oceans changes and THEN the oceans cause the delay.”
Firstly, see #33.2 for evidence (just some of it) of an 11 year pulse in temperature. The basis for Force X is the absence of this pulse. I say it exists and point to the literature for evidence (via UCAR in this case).
Secondly, oceanic thermal lag is longer than planetary thermal lag i.e. planetary lag (10, 12, 14 years or whatever +/-) takes into account all response delayers (heat sinks) over the entire surface to come up with the most significant lag in terms of current climate. Oceanic lag is out around 20 years (Abdussamatov) and 10 – 100 years (Trenberth). That period spans multiple albedo (cloudiness) changes that basically change sign each nominal decade (e.g. 1990 – 2000, 2000 – 2010) according to latest studies (Wild et al 2012 I think, for one).
Yes I concur that there has to be some explanation for albedo/cloudiness changes on a decadal scale but I don’t think it is Force X because the underlying basis for it is flimsy at best from what I’ve seen.
Force X does not explain why there was basically a decrease of cloudiness 1990 – 2000 and an increase of cloudiness 2000 – 2010 (unless it’s the sun’s magnetic polarity, possible, but I doubt that so convince me).
>”…less energy has been entereing the oceans since 2000 than before 2000″
An assumption. We have ARGO OHC metrics for each ocean basin (doctored by Josh “too cold” Willis) that don’t support that. Certainly not the Indian where transports have elevated OHC considerably since 2000. The upper Pacific does appear to be losing rather than gaining heat so I concur on that at least.
10
“Force X does not explain why there was basically a decrease of cloudiness 1990 – 2000 and an increase of cloudiness 2000 – 2010 ”
Before 2000 the sun was active, ozone in the stratosphere declined and the stratosphere cooled, the jets became more zonal, cloudiness declined and the climate zones shifted poleward by about half a mile per annum by one account.
Since 2000 all those processes have stopped with some regional reversals.
If the sun stays inactive for much longer the process sahould begin to reverse more generally.
“Certainly not the Indian where transports have elevated OHC considerably since 2000. The upper Pacific does appear to be losing rather than gaining heat so I concur on that at least.”
The timing for each ocean basin is different. It starts with the Pacific.
40
Stephan #33.1.1.1
>”Before 2000 the sun was active………..Since 2000 all those processes have stopped with some regional reversals.”
Are you including the sun in “all those processes?
Up to end 2005 the sun was as active as the previous 3 SCs (see below). I think 2009 is generally considered the end of the Modern Grand Maximum because there was a perceptible drop in activity and lengthening of the cycle after about 2005 to end of cycle 2009:
http://www.giss.nasa.gov/research/news/20120130b/619623main1_solar_irradiance_graph-670.jpg
But if there is a thermal lag in the system (there is, I favour Abdussamatov’s thermo principles based calc of 14 +/-6) of 11 years, the full effect of the lower activity will not be exhibited in temperature until 2016 at the earliest.
I’m inclined to think at least 2019 but from now on SH SST the best early indicator:
http://www.woodfortrees.org/graph/hadsst3sh/from:2000
There will be the weaker “fast” responses meantime but they are competing with ocean oscillations (PDO/IPO, AMO) which are acting to maintain temperature (the “pause”).
>”If the sun stays inactive for much longer the process sahould begin to reverse more generally.”
Yes I agree but there’s an enormous amount of accumulated heat in the ocean from the last 60 years or so to influence atmospheric temperature for some time yet as the heat is released rather than accumulated.
00
Re #33
>”Not necessarily as a temperature pulse at 11 years though but there is evidence of it”
Like this:
Solar signal detection: Wavelet analysis of 85- and 11-year cycles in solar activity and Mann et al. (1998) Northern Hemisphere surface temperature (Details: see Oh et al. 2003)
http://www.assessment.ucar.edu/paleo/img/forcing_solar_2.jpg
From:
‘Natural External Forcing in the Climate System’
http://www.assessment.ucar.edu/paleo/past_forcings.html
00
Re #33
>”Not necessarily as a temperature pulse at 11 years though but there is evidence of it”
Like this (and see #33.2):
‘Eleven-year solar cycle signal throughout the lower atmosphere’
K. Coughlin and K. K. Tung (2004)
Abstract
[1] A statistically significant atmospheric signal, which represents the influence of solar radiation changes on our climate, is found in global data (1958–2003). Using a nonlinear, nonstationary time series analysis, called empirical mode decomposition, it is shown that atmospheric temperatures and geopotential heights are composed of five global oscillations and a trend. The fourth mode is synchronized with the 11-year solar flux almost everywhere in the lower atmosphere. Statistical tests show that this signal is different from noise, indicating that there is enhanced warming in the troposphere during times of increased solar radiation.
http://onlinelibrary.wiley.com/doi/10.1029/2004JD004873/full
Read the entire (full) paper at the link.
And this:
‘Observed Tropospheric Temperature Response to 11-yr Solar Cycle and What It Reveals about Mechanisms’
JIANSONG ZHOU AND KA-KIT TUNG (2012)
ABSTRACT
Using 54 yr of NCEP reanalysis global data from 1000 to 10 hPa, this study establishes the existence and the statistical significance of the zonal-mean temperature response to the 11-yr solar cycle throughout the troposphere and parts of the lower stratosphere. Two types of statistical analysis are used: the composite-mean difference projection method, which tests the existence of the solar cycle signal level by level, and the adaptive AR(p)-t test, which tells if a particular local feature is statistically significant at the 95% confidence level. A larger area of statistical significance than that in previous published work is obtained, due to the longer record and a better trend removal process. It reveals a spatial pattern consistent with a ‘‘bottom up’’ mechanism, involving evaporative feedback near the tropical ocean surface and tropical vertical convection, latent heating of the tropical upper troposphere, and poleward large-scale heat transport to the polar regions. It provides an
alternative to the currently favored ‘‘top down’’ mechanism involving stratospheric ozone heating.
http://depts.washington.edu/amath/old_website/research/articles/Tung/journals/Zhou_and_Tung_2013_solar.pdf
00
“It provides an alternative to the currently favored ‘‘top down’’ mechanism involving stratospheric ozone heating.”
We need both processes constantly interacting on different time scales to explain observations.
Has anyone suggested that ?
30
Coughlin and Tung’s criticism:
[4] We have decided against using subjective methods such as Fourier analysis for detecting the solar cycle. Fourier analysis is subjective because it assumes a priori that the signal should have a constant period and a constant amplitude throughout the length of the time series. Neither of these assumptions is applicable to the solar cycle index or to the atmospheric signal correlated with it. This has the potential effect of reducing the amplitude of the extracted signal. Given that our length of record is already short, the statistical significance of the signal cannot be demonstrated at all pressure levels. Figure 1 shows the Fourier spectra of the 600-hPa geopotential height (January 1948 to February 2004) data averaged from 20° to 90°N. The statistics are naively calculated, with the mean (magenta line) fitted to the spectra of the data without the annual cycle and the 95% confidence interval calculated with a student t test. In Figure 1, there is a Fourier spectral peak near the 11-year period. Its amplitude is reduced from that of the “true” signal because of its entanglement with other atmospheric signals. This Fourier signal is also difficult to see everywhere over the globe. At 700 hPa, there is less noticeable power at 11 years, and when only the years after 1958, where the data are considered more reliable, are used, the power at this period also diminishes. The statistics calculated here do not take into consideration the choice of height or the choice of years included in the spectrum. This would also diminish the significance of the signal. A more objective method should be one which can disentangle the atmospheric signals from one another.
http://onlinelibrary.wiley.com/doi/10.1029/2004JD004873/full
00
Although the influence of the 11-year TSI peaks cannot be found in the global surface air temperature record (HadCrut, UAH, etc), it will presumably be present in many other climate records.
When the TSI peaks, we are proposing that force X troughs — a shortish period of less warming due to force X. Force X acts on albedo, so that means more reflection when TSI peaks. Presumably, force X causes more clouds (or snow and ice) when TSI peaks.
So the rainfall records might be expected to show peaks in rainfall every 11 years, which might effect say wheat prices. The hydrological records in Africa show a strong 22 year cycle, peaking every 22 years, I seem to recall from Will Alexander.
In general, every 11 years there is a peak in TSI (and albedo) and a dip in force X. While those influences apparently largely cancel out in global surface air temperature, presumably many other climate variables will be effected predominantly by only one of those influences and therefore will show 11 year peaking.
That might be a good direction for research: catalog all the variables demonstrating 11 year peaks or dips that are synchronous with the TSI peaks, and from that we might be able to deduce how force X works or even what it is.
30
“Force X acts on albedo, so that means more reflection when TSI peaks. Presumably, force X causes more clouds”
Hmm,
Except that Svensmark (and and others) propose that there are more clouds when solar activity is low.
During the late 20th century warming spell with high solar activity global cloudiness reduced. It is now recovering according to the Earthshine project.
Yet David considers the Svensmark hypothesis as a possible candidate ?
It will be interesting to see how David squares that off.
The hydrological records along a line of latitude are of little help because the effect at any particular location depends on which side of the nearest rain bearing band it is situated and whether that band is shifting poleward or equatorward.
The ITCZ moves north when the sun is active and south when the sun is less active.
00
Force x producing LESS clouds at a time of high TSI and the delay then occurring within the oceans would produce the observed result.
00
At a TSI peak force X is cooling (or less warming than usual). Force X works on albedo, so it cools by causing more reflectance, which requires more clouds.
Check: More clouds -> cooler (the next effect of clouds is cooling), a greater % of incoming TSI is reflected back out to space. Required to counteract the surge in TSI.
Ok, seems right. Force X produces more clouds at TSI peaks.
The strength of force X during this TSI peak is proportional to the amount of TSI during the previous TSI peak (the delay).
At TSI peaks the Sun’s magnetic shields are weak, so there are more cosmic rays and thus more clouds. Thus force X is weak when the Sun’s magnetic fields are weak.
30
But we observe less clouds when the sun is active and more clouds when it is less active. See the Earthshine data which someone else linked to. Albedo has increased since 2000 with the less active sun
Nor is there any ‘surge’ in TSI. The variation is miniscule. It disappears compared to internal system variability in any event.
On the other hand the thermal effect of an albedo change is far greater.
I think you are onto something with the notch and that it introduces thermal effects at a timscale of 11 years and multiples of 11 years but it isn’t needed to counteract any surges of TSI.
That force x exists and you have found it for which you are to be congratulated but I’m not convinced that your interpretation of it is correct as merely a process whereby the single cycle TSI variations are smoothed out.
It is correlated to the TSI peak but it is operating at a completely different level and I think it accounts for climate changes on a scale way beyond what TSI variations can achieve.
As Leif Svalgaard constantly points out the TSI changes even across multiple cycles are far too small to account for our observations of climate changes from MWP to LIA and LIA to date.
On the other hand, changes in force x can do it.
Your discovery is probably bigger than you yet think.
However, you have to get the process to match observations and to do that you must accept less clouds when the sun is more active.
Having taken that step the only way to delay the effect of less clouds is to involve the oceans in the manner I have suggested.
Have you anything coming up in future posts that deals with the apparent discrepancy with observations of cloudiness changes?
11
Stephen, we are getting confused across three threads.
Please check out this answer, especially the note at the bottom about the misleading axis label on one of the albedo graphs.
When you say “sun is active” do you mean when there is more TSI? Do you mean the year or two of the TSI peak, or the larger scale such as the last 50 years of “Grand Maxima”? By “surge” in TSI I meant a TSI peak, one or two years.
The low pass filter takes care of the thermal lags in the climate system. The albedo is just the gate that lets in more or less unreflected TSI into the climate system.
Yes, force X is way more influential than immediate TSI changes — those just indicate what force X is going to do in a few years. We’ve got to get the interpretation and implications right though, so keep on thinking it over please.
I’m pretty sure: TSI peak -> less force X (notching) -> more clouds.
40
“At TSI peaks the Sun’s magnetic shields are weak, so there are more cosmic rays and thus more clouds”
See here:
http://en.wikipedia.org/wiki/Total_solar_irradiance#Cosmic_ray_flux
“The outward expansion of solar ejecta into interplanetary space provides overdensities of plasma that are efficient at scattering high-energy cosmic rays entering the solar system from elsewhere in the galaxy. Since the frequency of solar eruptive events is strongly modulated by the solar cycle, the degree of cosmic ray scattering in the outer solar system varies in step. As a consequence, the cosmic ray flux in the inner solar system is anticorrelated with the overall level of solar activity. This anticorrelation is clearly detected in cosmic ray flux measurements at the Earth’s surface.”
A more active sun results in LESS cosmic rays at the surface of Earth.
That is in accordance with the Svensmak hypothesis which proposes less clouds when the sun is active and in accordance with my hypothesis involving UV effects on ozone.
Your model still looks sound to me if you just reverse your cloudiness assumption, forget about single cycle TSI variations because they are lost in the system anyway and use force x effects on albedo to anticipate trends in energy getting into the oceans to drive the climate system over the subsequent 3 to 15 years.
Unless there is still something to come which demolishes the above 🙂
10
Stephen, yes you are quite right. Cosmic rays hitting the Earth and global cloud cover both dip at TSI peaks. Wow. This is great. That implies the cosmic ray effect runs counter to force X, so cannot be part of force X. Thanks, this is good to know.
We’ve got a post planned on interpreting force X soon.
00
” catalog all the variables demonstrating 11 year peaks or dips that are synchronous with the TSI peaks, and from that we might be able to deduce how force X works or even what it is.”
That’s just what I was thinking while I was feeding out the silage to the cows last night. The Manawatu River which surrounds my farm on three sides has a diverse catchment extending both North and South of here , and both sides of the main divide , so it could be a useful river to study. I was thinking about it because we haven’t had an extreme flood event since 2004, and we usually get a ripper about every ten years or so.
10
David #33.3.3
>”Although the influence of the 11-year TSI peaks cannot be found in the global surface air temperature record (HadCrut, UAH, etc)”
That is because there is no such thing as the “11 year cycle”. That is simply a term of convention due to the long-term average length of 11.2 years.
I’ve replied to the last comment in Part II (Willis) with the the reasoning (recent actual SC lengths etc) here:
http://joannenova.com.au/2014/06/big-news-part-ii-for-the-first-time-a-mysterious-notch-filter-found-in-the-climate/#comment-1492355
That comment then refers back to this Part VI thread where the relevant papers are current (#33.3).
Those papers show how, if you discard the spurious notion of an “11 yr cycle”, warming at the SC peaks when they actually occur is clearly identifiable.
00
Given our assumption that the climate system is a linear invariant system, sinusoids (and thus Fourier analysis) is of a lot of interest — sinusoids are rather special to such systems. See from “The way to analyze a linear and invariant system is with sinusoids ” in Post II.
00
_
Ka-Kit Tung has natural quantitative intuition on a level rarely seen anywhere in the climate discussion. That doesn’t necessarily ensure perfect interpretation of analyses (for example there may be some cross-disciplinary knowledge deficits here & there), but I strongly agree that some of his publications deserve or even demand careful attention. Links (no paywall = free access) to his papers:
https://depts.washington.edu/amath/faculty/tung/publications.shtml
Caution: To ensure sensible interpretation, remain vigilantly aware of where reanalyses are being analyzed.
10
Without realizing Tung & colleague Yang had published this analysis in 1995, I’ve illustrated the same patterns for the climate discussion community before:
On the phase propagation of extra-tropical QBO in observational data
http://depts.washington.edu/amath/old_website/research/articles/Tung/journals/yang-tung95.pdf
When you see clarity of vision like that, it really stands out. The important thing to note is the tropical – extratropical phase contrast.
10
So they detect some effect from the 11 year cycle in one of their papers. Interesting stuff. The effect just doesn’t seem to show itself in the global near-surface temperature record.
01
As explained in Post III, a true delay is required because a notch filter without an accompanying delay is impossible. The true delay is not a time constant of something that dribbles out, like heat out an ocean reservoir, but an actual delay like in a clock.
As argued in Post IV, this seems to point to the Sun as the origin of the delay and thus of force X. Note the “delay” in the model — it means literally that, it is not a time constant for a process where something slowly empties.
10
David #33.4
>”The true delay is not a time constant of something that dribbles out, like heat out an ocean reservoir, but an actual delay like in a clock”
Yes I noted that subtle distinction when you made it at the outset.
I’m disputing that your “clock” delay is “true”.
00
David,
I would suggest it may be premature to dismiss the oceans. Due to the nature of larger circulation patterns such as thermohaline circulation, warmer or colder water can return to the surface in bulk rather than a “trickle”. The oceans do meet the criteria for a system with “memory”.
In looking at energy accumulation in the oceans it is important not to do as the AGW crowd has done and treat the oceans as a “near blackbody”. The oceans are a “selective surface”*, and both the strength and depth of radiation penetration has a great effect on the rate of energy accumulation. Further as a ‘selective surface” the oceans are particularly complex, with differing regimes of non-radiative transport, above the thermocline (diurnal overturning) and across and below the thermocline (thermohaline circulation)
*The distinction between “near blackbody” and “selective surface” is not just a minor theoretical point. It is the reason that the oceans are 33C warmer than theoretical blackbody calcs show, not DWLWIR. Were it not for evaporative cooling, SSTs could top 80C.
13
Thanks Konrad, I am quite mindful of that. Yes, the answer may turn out to be in the oceans. We spent months trying to figure out how that might be, without coming up with anything very concrete.
On the other hand, force X is synchronized to TSI. The easiest explanation is that they both come from the Sun, but it could be that something on Earth responds to the rise in TSI at the beginning of a TSI peak.
The oceans heat storage and delay is captured, at least for global surface temperatures, by the low pass filter. I agree there are deeper storages and delays, but they don’t show up in the transfer function (and won’t without centuries of good data, perhaps).
20
David,
with regard to an ocean mechanism I would suggest variance in UV absorbed below the thermocline / diurnal overturning layer is very plausible. The issue here is not TSI variance of only 0.1% but spectral variance in UV frequencies, some up to 100%.
Energy absorbed below the overturning layer can accumulate, and it is the shorter frequencies that penetrate to these depths. It is also these frequencies that vary most between solar cycles. It is notable that surface UV has increased ~ 10 – 20% in the last 30 years, but stabilised since the mid 1990s –
http://www.nasa.gov/topics/solarsystem/features/uv-exposure.html
It should also be noted that UV-A still has the power of ~10 w/m2 at 50m depth.
Changes in ocean heat content due to this mechanism would be slow and cumulative below the thermocline. This would allow short diurnal and seasonal signatures to occur in SSTs, with 11 year signatures masked or smeared due to deeper circulation patterns. Depending on how much of this energy is lost to the diurnal overturning layer, this mechanism could account for 0.8C in 150 years.
22
Over the long term the notch is unimportant because it only attenuates sinusoids with periods over about 40 years (see Figure 1 of Post III) — so you can ignore it on timescales over about 100 years say.
However, as noted in Figure 4 above, the step response of the delayed path is an order of magnitude greater than the step response of the immediate path — so force X is powerful.
We observed the notch, from which we deduced the delay, from which we figured that there was a force X coming from the Sun. We noticed that TSI peaks coincide with troughs in the solar magnetic field, which gives us a possible notching mechanism.
20
David #33.5
>”Over the long term the notch is unimportant”
Thanks for your reply on this, I was very curious because I’m interested in solar activity in say 1000 yr blocks over the last 11,000 years (i.e. the Holocene).
The recent Modern Grand Maximum ending 2009 was the highest level of solar activity since about 11,000 years ago. I’ve stated previously that (intuitively) we should expect a commensurate temperature response.
>”However, as noted in Figure 4 above, the step response of the delayed path is an order of magnitude greater than the step response of the immediate path…”
Yes, “slow” responses have far greater effect than “fast” responses. That is well known already.
>”We observed the notch”
You, Joanne, and others accepting your methodology observed the notch. I don’t accept your methodology (premise, analysis tools, datasets) but even so I also observed the notch by your method.
But by other premise (no “11 yr cycle”), other analysis tools (EMD, CMD, adaptive AR(p)-t test), and other datasets (e.g. NCEP reanalysis of the entire troposphere), the notch is not observed. On the contrary, warming is observed at the SC peals when they actually occur.
00
Re #33.5.1
>”On the contrary, warming is observed at the SC peaks when they actually occur”
I should point out that Coughlin & Tung and Zhou & Tung identify the very faint and very minor “fast” temperature response (immediate TSI path in David’s terms) to SC peaks.
But the major response is “slow” and delayed by thermal inertia of the surface heat sinks. The delay is months, years, decades, and even centuries. To relate the delay to current climate the delay can be described as a thermal lag around a dozen years (I prefer 14 by Abdussamatov) plus or minus several years.
So the major temperature response to solar change is “smeared” over a very long period and impossible to detect as a spike in temperature records except the difference between Grand Min and Grand Max (H/t Griss):
http://therionorteline.files.wordpress.com/2012/08/2000-years-of-global-temperaturessp.jpg
For example, the Modern solar Grand Maximum (a series of SCs) peaked in 1986. The temperature effect of that is “smeared” over the period following so that the years coinciding with the lag (e.g. 2000 and either side @ 14 yr lag +/- 6 yrs) experience the effect of the maximum much more than 1986 does. This is what the guys at Skeptical Science don’t understand on their solar case rebuttal page.
Similarly, SC 23 peaked December 2001 (by F10.7 monthly average). The temperature effect of that is yet to be experienced to it’s fullest extent (2001/02 + 14 = end of 2015 +/- several years).
For this reason it is spurious to go looking for a recurring 11.2 yr average period temperature spike coinciding with TSI at SC peaks as evidence of the solar effect on temperature – or not..
That temperature spike exists as a faint “fast” response but is an all but negligible signal. To focus on it (the “fast” component) as the temperature response evidence (or not) to solar change, is barking up the wrong tree.
10
Re #35.5.1.1
>”…the Modern solar Grand Maximum (a series of SCs) peaked in 1986. The temperature effect of that is “smeared” over the period following”
The lagged accumulating heat content effect of that series of SCs is graphed here:
Change in heat content: ocean + land + ice + atmosphere
http://www.skepticalscience.com/graphics/Nuccitelli_OHC_Data_med.jpg
SkS attribute all that to GHG forcing of course – a classic case of miss-attribution.
00
See this answer, Richard.
00
David #33.5.1.2
>”See this answer [hotlink], Richard.”
Hotlink to answer #33.3.3.1 (my emphasis):
Yes, these are your assumptions and your methodology. I get that.
You were replying to my comment #33.5.1 (emphasized this time):
The “warming” referred to is simply the very faint and very minor “fast” response.
My point here is that just pointing me back to your assumptions and method does not address my criticism.
00
Re #33.5.1.2.1
>”My point here is that just pointing me back to your assumptions and method does not address my criticism.”
More on my criticism at #33.5.1.1 here:
http://joannenova.com.au/2014/06/big-news-part-vi-building-a-new-solar-climate-model-with-the-notch-filter/#comment-1492481
00
Richard, I suspected you might have mixing up time and frequency domain concepts, so was suggesting you reconsider what happens in the frequency domain, especially the superposition and non-interference of sinusoids at different frequencies.
Your criticism talks of slow and fast responses. In the frequency domain, that just means the level of the transfer function amplitude at low and high frequencies.
In frequency terms, the thermal lag of the system is fully expressed in the low pass filter — which passes low frequencies (slow responses) but attenuates high frequencies (fast responses). This is all the thermal lag required or observed for the model to work, and has a time constant of about 5 years. Of course there are much longer time constants for very long term warming and cooling of the oceans, but in terms of surface temperature we didn’t see them in the datasets.
On the other hand, the delay is a true delay, not a thermal lag.
Yes, the main response to TSI change is slow, and takes a few years — as governed by the low pass filter. The delay is something new we have detected and deduced — the full force associated with the TSI makes itself felt about 11 years after the TSI changes. This is force X, not the TSI, and it is after a true delay, not a thermal lag. That’s what is new here; it’s a different theory, it’s not the same as Abdussamatov’s.
TSI peaked in 1986. This signaled that force X would peak in about 1997, but TSI has hardly any effect beyond some very minor immediate effect. We are saying that force X peaked 11 years later, in about 1997. Force X acts immediately. Force X stopped increasing in 1997, so the planet stopped warming in about 1997.
Of course there is also a thermal lag that smooths out both the effect of the TSI and the effect of force X. See Figure 2 above.
00
David #33.5.1.2.2
>”In frequency terms, the thermal lag of the system is fully expressed in the low pass filter — which passes low frequencies (slow responses)”
I’m disputing that the thermal lag of the sun => ocean => atmosphere system is a “frequency” (i.e. not necessarily a sinusoid) in Part II here:
http://joannenova.com.au/2014/06/big-news-part-ii-for-the-first-time-a-mysterious-notch-filter-found-in-the-climate/#comment-1494630
>”has a time constant of about 5 years”
Definitely disputing this in Part II above.
>”Of course there are much longer time constants for very long term warming and cooling of the oceans, but in terms of surface temperature we didn’t see them in the datasets”
I show in Part II above that, over about the last 45 years or so, we DO see the atmospheric response to oceanic heat accumulation (due to elevated solar activity) in the temperature datasets.
>”Yes, the main response to TSI change is slow, and takes a few years”
It has taken 45 years plus some. See Part II above.
>”TSI peaked in 1986″
Disputing this in this Part IV thread (#86.1.1.1.1) here:
http://joannenova.com.au/2014/06/big-news-part-vi-building-a-new-solar-climate-model-with-the-notch-filter/#comment-1494668
00
I’m disputing that the thermal lag of the sun => ocean => atmosphere system is a “frequency” (i.e. not necessarily a sinusoid)
Go back and re-read what David E. said:
Richard, I suspected you might have mixing up time and frequency domain concepts, so was suggesting you reconsider what happens in the frequency domain, especially the superposition and non-interference of sinusoids at different frequencies.
Most EEs (having spent a lifetime doing it) can switch between time and frequency domains almost naturally. It is not so easy for those without that experience. And of course it makes it hard for engineers too “What is it you don’t understand? This is all obvious stuff.”
It is likely that what is confounding you is that you are having trouble accepting that – time series = frequency response – is an identity.
00
MSimon #35.5.1.2.2
>”Go back and re-read what David E. said”
I’ve read it and I know what he’s on about but you, apparently, have not read my contention at the Part II link above.
Are you seriously saying an 11,000 year spike in solar activity (far above others in the intervening time) that via the ocean exhibits as simply a rising trend in either OHC or atmospheric temperature so far, shows up at some frequency (what?) in a temperature frequency analysis?
Maybe over one million years a solar phenomenon coinciding with the Modern Grand Maximum occurs every 11,000 years but I doubt that, let alone the temperature response frequency.
Maybe the spike is one of several others of lessor magnitude over the Holocene but what is the frequency? The word “quasi” is used in generalizations but as with “11 yr cycle” the period is varying even if a recurring spike can be identified. OK lets have look at one:
‘The influence of the deVries (∼200-year) solar cycle on climate variations: Results from the Central Asian Mountains and their global link’
O.M. Raspopov, V.A. Dergachev, J. Esper, O.V. Kozyreva, D. Frank, M. Ogurtsov, T. Kolström, X. Shao (2006)
5. Conclusions
Review of published palaeoclimatic data shows that the ∼200-year climatic variations are evident in the oceans and different regions of the Earth (Europe, North and South America, Asia, etc.). Analysis of the results obtained in simulation of the effect of long-term variations in solar irradiance has shown that, because of an essentially nonlinear character of the processes in the atmosphere–ocean system, the climate response to external long-term solar forcing, including the 200-year variation, differs in different regions of the Earth and can manifest itself in various climatic parameters.
http://gacc.nifc.gov/sacc/predictive/SOLAR_WEATHER-CLIMATE_STUDIES/2008%20De%20Vries%20200yr%20Solar%20Cycle%20Influence%20on%20ClimateRaspopov_etal_2008.pdf
According to David’s rationale and methodology (which you appear to consider rock solid because…EE), if the deVries cycle doesn’t show up in globally averaged temperature, there must be a notch delay in the system.
Never mind Raspopov et al (2006), “climate response to external long-term solar forcing, including the 200-year variation, differs in different regions of the Earth“.
00
Dang, forgot to tick “Notify me of followup comments via e-mail”
00
And again I didn’t tick – Aaaagggh!
This time.
00
I have been following this closely since the start and I am thoroughly enjoying it. I look eagerly for each new part of the story and I appreciate the value of building the understanding, step by step even though I struggle a little with the electronic concepts.
I think the criticism of the method of disclosure is unwarranted and petty and suggests ruffled feathers to me. It is up to Jo and David how they report their work. Those who have a problem with that should stay away until all the details and spreadsheets are made available.
Let us get the comments back to developing an understanding of the physical processes that are taking place between the input and the output. That would be much more constructive.
50
Im still struggling with how feedback from eg H2O is accounted for. bobl above indicated that this is accounted for in the filters. I dont see any reference to this in the text, so if anyone can advise me that will be great. As I see it, that woud work if the filters were active filters, and would make good sense but the text doesnt imply this, leading to the assumption that they are passive filters. Anyone ?
00
Terry, the model so far does not care about the internal mechanisms of the black box, only its behavior.
However, soon the posts will be discussing feedbacks and water vapor.
20
Let me see if I can explain, if David will allow me some latitude. Imagine a black box,TSI goes in, and temperature out, both based on measured values, what happens in the box isn’t relevant to the ultimate effect that TSI has on temperature, all the internal responses between TSI and temperature output that cause the ultimate temperature are part of the innards of the box that transforms TSI into temperature.
That a partticular TSI gave a particular temperature characteristic is just a fact of life, the feedbacks are part of what causes that fact of life, the innards of the box.
Using the engineering math, David has shown how the innards of the box, – all the innards including feedbacks behave after all is said and done … it’s a macro model. Whether the innards are active or passive doesnt matter, they are what they are, but from the analysis we can examine how they behave to an outside observer.
Having characterised how the TSI to temperature box behaves with all it’s intricacy, David has built a model that behaves the same way, and it involves a parallel path (Force X) and a delay (phase shift) there is a need for a physical interpretation for that later, but we don’t need that yet, we just know we can’t create that transfer function without it. Having built a simulation that gives us the same relationship between TSI and temperature as found in the wild David can begin to make predictions. If the predictions are useful then it has some value.
For example David can plug in TSI and various other parameters for Force X to try to deduce what Force X is.
40
Nicely said.
It as often struck me that Engineers tend to work from the generic to the specific (that is what I do). Scientists however, seem more inclined to work from specific observations to the general principle, ultimately arriving at a Law.
Both are obviously valid, but sometimes communication can be difficult.
30
Exactly bobl.
00
Hey.. all you guys wanting this and that….
David’s telling the story..
Let him tell it at his pace..
Don’t be so darn impatient !!!!
Gees its like a child asking, “are we there yet” every two minutes on a 5 hour trip !!!!!
121
“Are we there yet, are we there yet?” absobloodylutely!
I wouldn’t be surprised if David said “One more impatient question and I swear to god I’ll turn this model around, and don’t think I won’t do it!”
70
This is the graph for albedo produced by The Earthshine Project that was posted at WUWT.
Note the inflection point at 1997/1998 (Super El Nino)
Cycle 23 started in 1996 and lasted 12.5 years.
Cycle 22 was the last of the “super sunspot” cycles. GRAPH
Cycle 21 started in June 1976 and Cycle 22 started in September 1986 and lasted 9 yrs/ 8 mo.
If the change in the sun (factor X) occurred during the solar maximum of cycle 22 then you are looking at 5 years before the Albedo change.
If the change in the sun occurred during the solar minimum of cycle 22 then the lag is one to two years.
If the change occurred during the minimum before cycle 22 then you are looking at ~10 years. This is the time period when factor X occurred according to Dr. Evans.
00
I’m wondering whether David is saying that the albedo change occurs around ten years after the solar change or whether he is saying that the atmospheric temperature change occurs about ten years after the albedo change.
As far as I can tell so far David says that the maximum TSI coincides with whatever causes the notch and it is the notch that causes the thermal effect of more TSI to be delayed by around ten years.
If that notch is then the cause of a rapid albedo change then the delay occurs on Earth and not in the sun although the indirect cause of the delay was in the sun.
00
Sorry, ‘the thermal effect’ and not ‘the thermal effect of more TSI’ because we have already removed TSI as the cause and replaced it with force x.
00
See this answer.
00
David,
Agreed on the interpretation issue. It is critical.
You referred to this:
“There are some graphs that say “% Change in Albedo: with a mnimum in 1998, and others that say “Albedo (%)” with a minimum in 1998, which have opposite implications! I think the text in Palle et al 2008 clears it up — albedo bottomed in around 1998.”
Albedo bottomed around 1998 after a long run of very active solar cycles. That means that the extended period of active sun gradually reduced cloudiness from one active cycle to the next.
Your interpretation would expect an increase in albedo with the more active sun.
You need to forget single cycle TSI because the variations are trivial and have led you up a wrong path (I think).
The solar effect is slow and cumulative and relies on force x whatever it may be.
Solar (force x) from the top down modulated from the bottom up by the delayed ocean response (smeared from 3 to 15 years) and single cycle TSI variations lost in the noise.
That way you can get the right sign for cloudiness variations AND retain your force x insight.
00
David, you can still use TSI variations, albeit tiny, as a proxy for the subsequent much larger force x effects to achieve better predictive skill as regards global climate trends than can the GCMs.
My New Climate Model proposed using the net latitudinal position of the climate zones and/or the zonality/meridionality of the jet stream tracks or total global cloudiness as suitable proxies but since force x follows TSI but greatly amplified then TSI will serve as well.
Many have bben looking for a suitable amplification effect above and beyond the power of TSI variations. Your work appears to have provided evidence that there really is a force x separate from raw TSI behind it all.
“Total Solar Irradiance (TSI) varied so little over that period that it cannot explain the warming that was observed. I don’t yet accept that TSI tells the whole story because it is ill defined and speculative as regards it’s representation of all the different ways the sun could affect the Earth via the entire available range of physical processes”
from here:
http://www.newclimatemodel.com/the-death-blow-to-anthropogenic-global-warming/
Published by Stephen Wilde June 4, 2008
01
Without going into every detail, that seems to jibe pretty well with the picture emerging from the nd solar model. Pretty good for 2008 🙂
00
TSI reached a maximum in about 1986 (after an 11 year smoothing) then gently declined. Force X is a warming force that basically lags TSI by about 11 years and affects albedo.
So a minimum in albedo in 1986 + 11 =~ 1998 followed by an increase in albedo fits the pattern.
00
THanks, gal, I had a problem finding it with all the thread layering.
00
Re #38 (gai)
>”Cycle 23 started in 1996 and lasted 12.5 years”
Hence papers with predictions based on SC length, for example:
‘The long sunspot cycle 23 predicts a significant temperature decrease in cycle 24’
Jan-Erik Solheim, Kjell Stordahl, Ole Humlum (2012)
http://hockeyschtick.blogspot.co.nz/2014/06/new-paper-predicts-temperature-decrease.html
00
@Willis
dp was right. I wasn’t aiming at you (or Will J, for that matter. Should’ve made my target clearer.
(BTW [and this is not sucking up, but genuinely felt], admire your wrtings at WUWT).
@mods
Sorry. Won’t happen again.
60
This is relevant to the albedo issue:
http://wattsupwiththat.com/2007/10/17/earths-albedo-tells-a-interesting-story/
00
Perhaps it’s a great big fluke of nature, that Earth is here and supporting intelligent life at all (well yes, you might wonder sometimes) and that cancellation whatever it is, is one more thing that is necessary to make it possible. Another inherent stabilising mechanism without which Earth might have gone the way of all other planets, unable to sustain life at the very least.
11
Is it fair to say that force x is the TSI transmission control knob? Turn up force x and more solar heating gets through by suppressing cloud formation. Turn it down and we get cooling. This is capable of producing periods like the MWP and LIA if a trend continues over several cycles.
Linked to this is the peak in TSI that occurs about 11 years earlier, but this is when force x is low, so that extra heat is met by more cloudiness.
All of that makes sense to me, however, I have trouble seeing how the notch fits in and whether TSI peaks have much of a role. I could imagine having the whole thing based on the solar magnetic field.
00
Schrodinger’s cat,
Interesting point.
Altering force x alters global cloudiness so it works on the proportion of TSI that reaches the surface. It therefore mimics a change in TSI and can warm or cool the system.
A change in actual TSI also changes the temperature of the system.
Furthermore, the change in TSI in one cycle is a proxy for the scale of the amplification factor that will be wrought by factor x some years later but smeared by the ocean response over the next 3 to 15 years.
However, in both cases the additional warmth is countered by a negative system response from convection aided as necessary by the water cycle.
So, either way, the effect on the global temperature is minimised (possibly negated) by emergent phenomena worldwide (not just in the tropics as proposed by Willis Eschenbach).
However, there is an effect in the form of climate zone shifting and changes in jet stream tracks but the portion attributable to TSI is lost in system noise whereas the portion attributable to factor x is a magnitude or more larger and can accumulate over time hence MWP and LIA et al.
The notch would be a statistical representation of force x and entirely separate from the TSI change. Having lost the thermal effect of the TSI change in the noise of the system the notch then predicts the scale of the atmospheric thermal response that will occur smeared across the next 3 to 15 years as noted by David.
Hope I’ve got that right. If not, no doubt someone will say.
10
That’s a possible interpretation of the model. Not one we thought of, but it may be part of the answer IMHO.
00
Yes, force X influences the tap or albedo. While I’ll bet force X turns out to have much less energy than TSI, it is like the tap on the firehose of TSI.
By the way, the model logic only finds force X affects albedo, not necessarily clouds. Snow, ice, dust and biological agents might be involved.
00
the Griss
June 21, 2014 at 7:17 pm
Thanks for the kind words and well-meant advice, Griss. And regarding your concern, I’m not upset or in heart attack mode in the slightest.
But “sit back and see where this leads?” As near as I can tell, this bus is driving right off the edge of a precipice, and is about to take the plunge and crash into the canyon below.
Now perhaps in that situation your habit is to wait to say anything until the train hits the bottom of the canyon … mine, not so much. When I see someone I admire, and be clear that I have great admiration for Joanne, who is about to drive their bus off of the edge of a cliff, I say something, and the earlier the better.
Not only that, but Jo and David have said that they wanted to do it this way specifically because they wanted early feedback. As Jo said:
And I agree with that part. They have asked people to tell them early on if they are driving the bus off of the cliff, because mid-course corrections are easier the earlier that they occur. In science, however, this is generally done behind the scenes and prior to publication, but they’ve chosen a more public path.
But when I say “Ummm … yer bus? It’s headed right for the cliff …”, I get people like you telling me to shut up, that my feedback isn’t wanted, that I should “sit back and see where this leads”, that I should wait until the final echo of the bus crash has died before I open my mouth.
The message I’m getting, from you and lots of others, is that y’all DO want feedback … but only laudatory, positive, ain’t-it-great feedback saying how great the theory is and how impressed you all are.
Sorry, not my style. I call them as I see them, friend and foe alike. And I do my best to warn my friends when I see them about to take a wrong turn. Jo and David have asked for feedback. I’m sorry you don’t like it, but that’s my feedback—I strongly advise them to stop faffing around and immediately publish the dang results and the exact formula for the model and the code to run the model and the data and the out-of-sample tests and whatever else they have.
My other feedback, of course, is that sad to say, and it does sadden me, I find their method not only unbelieveably complex, but totally without supporting evidence. As far as I can see, the only “evidence” presented to date is Figure 5, the diagram showing the “notch”.
But you will get an identical notch if you use their exact method to compare sunspots to white noise. Do you not see what that means? (Let me note that a few folks other than myself, mostly signal engineers, have pointed this out just as I have and we have been roundly ignored for our troubles.)
As a result, the “notch” is not evidence of anything except the fact that the 11-year sunspot cycles do NOT show up in the temperature data … a lack of solar effect that I have been advocating with little success and much opposition over at WUWT. In particular, they are NOT evidence or even indirect support for the “notch filter” theory. All that the notch means is that the solar changes are not reflected in the earth’s temperature.
And as Lubos Motl noted, that leads to a curious paradox. David and Jo are using evidence that the sun does not affect the temperature (no 11-year cycles) to advance a theory that the sun controls the temperature ,,,
w.
46
“Now perhaps in that situation your habit is to wait to say anything until the train hits the bottom of the canyon … mine, not so much”
If you are worried about where the train is going, you can always get off. 🙂
I’m not telling you to shut up, just saying you seem unnecessarily antagonistic and impatient, and was wondering why.
Trip planned for next week or something ? 🙂
70
But Willis, if you look at the Maunder minimum, with its very low sun spot count, then the current slightly warmer period, with very high sunspot counts (until now), …
There does seem to be some sort of link between the sunspots and global temperature over a longer period.
If the current snooziness of the sun does lead to a cooler period, there would seem to point to being quite a solid link.
Certainly not an immediate link, (as David has shown) but one well worth investigation.
30
Another link that seems to show that sunspot numbers and temperature are quite well related over time..
http://wattsupwiththat.com/2013/05/03/sunspot-cycle-and-the-global-temperature-change-anomaly/
20
And another..
http://hockeyschtick.blogspot.com.au/2013/11/the-sun-explains-95-of-climate-change.html
10
the Griss
June 22, 2014 at 7:17 am
Thanks, Griss, but to make that claim you need some data. The Maunder sunspot minimum is generally taken as being the 70-year period from 1645-1715. I was unaware that we had temperature records covering that time … so:
• What are you using in place of such a thermometer-based dataset in order to make your claims about the Maunder minimum?
• Please let me know the average temperatures for say 1610-1645, 1645-1680, 1680-1715, and 1715-1750, so we can see how much cooler (if at all) it was during the Maunder Minimum compared to the periods on either side, and whether the low point occurred near the end of the minimum (as you’d expect if the cause were solar) or whether the low point occurred near the start of the minimum (which would indicate a non-solar cause).
Best regards, I look forwards to your data.
w.
04
The folks at NASA seem to think so.
But then…………
10
http://therionorteline.files.wordpress.com/2012/08/2000-years-of-global-temperaturessp.jpg
11
” you will get an identical notch if you use their exact method to compare sunspots to white noise. ”
Right , several posters have expressed similar concerns. You want that addressed now.
00
Thing is, (as I said above), there is a very low solar period during the Maunder minimum, and the current slightly warmer period had a heap of strong solar peaks,
So there would seem to be some kind of link., sort of like a volume control on the white noise.
21
I can’t see any point to this argument at all though.
You could compare sunspots to pirates, and what would that tell you?
Compare sunspots to a straight line, white noise, pink noise, the loudness of an AC/DC concert. It’s a “so what” question, surely?
A comparison must be a meaningful comparison.
01
Greg, it is a bit repetitive.
In Post II I said “The initial aim of this project is to answer this question: If the recent global warming was associated almost entirely with solar radiation, and had no dependence on CO2, what solar model would account for it?”.
Associated with “solar radiation”, not pirates.
10
But “sit back and see where this leads?” As near as I can tell, this bus is driving right off the edge of a precipice, and is about to take the plunge and crash into the canyon below.
There is no bus, there is no precipice, there is no canyon.
There is only an opinion (supposedly supported by evidence we will be shown in time) being expressed.
Your Apocalypse Now mode of expression is a sign of you antagonism or attachment to something else. Possibly both.
People make mistakes all the time. Why the Alarmist verbiage? David is actually less invested in his discovery than I am and I’m open to evidence proving him wrong – once all the cards are on the table. Probably in a week or two.
Ordinary citizen: Why is there time?
Astrophysicist: So everything doesn’t happen all at once.
70
It seems to me David is actually less invested
30
Willis,
You have repeated your opinion maybe 10 times already? Read and noted….
Go back to the initial posts where they say “THEY DID SEND IT OUT FOR REVIEW AND COMMENT”. They apparently did not send it to you, and based on your current attitude, I don’t blame them one bit. Do you really think they would be doing this if they had been notified of a flaw that they could not scientifically work through?
They are placing their professional careers on the line with what they are doing, and their livelihood. I applaud them for it, as should everyone here!
Do you, or anyone else here, have that much skin in the game?
110
Brad
Well said.
I have had some slight misgivings about how David and Jo have gone about it. But after some thought and reading the comments now believe they have done exactly the right thing. They obviously tried the “traditional” approach when they sent the full, very large report out to scientists they believed in but the lack of response obviously told them that if interested scientists could not respond they had absolutely no show with the public or more importantly the politicians. So like their model a new approach was needed. Nothing will be lost because in the end everything will be out both in “block form” and the “drip feed form”.
I find it instructive that Anthony Watts has said nothing on WUWT –he has probably made the decision to wait until it is all out before making comment and also to not take any attention away from Jo’s blog ( a sign of respect in my view).
Also I think maybe there are many who realise that this is a very large, difficult piece of work for most to understand but it is potentially very significant so commenting and critiquing early would ruin the impact of it all.
110
Brad
June 22, 2014 at 6:51 am
Indeed, as have David, and Jo, and MKelly, and a host of others, each of us focusing on different aspects of the question with each succeeding comment … so?
Let me suggest that you don’t like my attitude because you don’t like my results and claims, and if my results and claims were in their favor, you wouldn’t care in the slightest about my attitude. In any case, attitude is not the issue here, science is the issue.
As to whether they should have sent it to me, I have stated many times that the proper people to send your work to are your worst enemies … because if they can’t poke holes in your work, no one can. And although I’m in no sense Jo’s enemy or David’s enemy, quite the opposite, I am known for incisive analyses … so yes, it would have been smart to send it to me.
In any case, they sent it to Lubos Motl, who trashed it as badly and for exactly the same reasons that I have … and in return, they blew his opinion off totally. Given that, sending it to me might have made no difference at all.
How would I know what Jo and David would do in hypothetical situation X? Half the time, I couldn’t even tell you what I would do in hypothetical situation X. There are far too many unknowns in that question for it to be answered.
I’m sorry, but that’s a bit melodramatic. If they’re wrong, they’re wrong, but lots and lots of scientists have been wrong. In fact, science is built upon one idea being knocked down and replaced by another. I don’t think their livelihoods are on the line.
And you don’t seem to understand that I am doing the best I can to prevent them from suffering damage to their reputations and their livelihoods. I think they have made a mistake hitching their star to this model, and I’m doing my best to point out why and do so early in the game so that they can climb down gracefully.
How on earth would I (or you) know how much “anyone else here” has at stake? You are anonymous, as are many people here. I have no clue how much skin you have in the game. Might be a little, or your whole career might depend on it, how could I possibly tell the difference?
Regards,
w.
09
Willis,
David and Jo acknowledged your comments the first time around, so why repeat them? Other people’s repeated posting is your justification? “He did it first?” Really?
Only “smart” people send their papers to you for your “royal stamp of approval”?
You really believe that what you think is more important than what any one else thinks… hmm, what group also has that attitude? Should we all sit back and let uncle Willis “school” us? Why haven’t you solved this problem 20 years ago?
I know an Engineer such as David would not spend years struggling on a concept to post false premises. He knows there are hundreds of people all over the world waiting to pounce on every word in an attempt to tear it to shreds and discredit him.
I am not anonymous, I have posted all my data on WUWT repeatedly, even on your posts.
Brad Weaver, PE.
Owner, Northwest Energy Consulting.
I am self-employed and have spent the last 8 years trying to educate people on economic-based resource conservation in buildings. The “Green” crowd is my biggest challenge, talking people into spending millions on bad projects.
Regardless of where this goes or how it turns out, this is science at a very high level, and is being presented in a class-oriented manner. I would even like a quiz after each chapter, just to make sure I understand what is being said.
Brad
100
“I know an Engineer such as David would not spend years struggling on a concept to post false premises. ”
that settles it
011
Mosh,
Not sure I remember you commenting on earlier posts? Did you read how this evolved? That their work has been funded by donations only?
So, you are implying David (and Jo by extension) are unethical?
60
Mosh, How does it feel to know that you were hired by BEST as a mouthpiece ?
Never as a scientist.
40
The “Green” crowd is my biggest challenge, talking people into spending millions on bad projects.
The “Green” Building standards are unconcerned with the productivity in the buildings they certify. All they look at is energy consumption.
As people get experience with these buildings “Green” will get a worse name.
30
MSimon,
As a preface I am not sure who you are, or what you do to “justify your existence” :). (See here – http://en.wikipedia.org/wiki/Black_Widowers.)
You can look me up online and shoot me an email if you want further discussion.
Actually, tenant satisfaction (productivity) is supposed to be the prime goal, supported by energy efficiency and resource sustainability.
Tenants logically should only lease space in buildings that maximize their productivity (profit).
Many “green” standards have been hijacked by marketing and sales, just like the Climate debate. All based on reducing a persons’ carbon footprint.
IMO, the biggest corollary of my industry to the AGW industry is poor modeling – most commonly used computer models CANNOT accurately portray energy use in a building. They build the building envelope thermally, and add in internal loads like people lights, and tenant power.
They also “select” code-compliant outside air levels. They then select the “appropriate” NOAA weather station data.
When the computer-generated annual energy use does not match reality they “tune” the model to make it do so. (I have seen 50 or so user-input parameters considered as fair-game for tuning.)The most commonly used parameter is horizontal (2-D) infiltration through the envelope.
The process implicitly accepts that the building is operated perfectly with no variations by the operator, or abnormal equipment conditions.
How many failure points do you see in this description?
Now the fun part is trying to get the program input and output data in its raw form, instead of the 200-page report with 27 8-1/2″ x 11″ glossy pictures with circles and arrows, that is either TCCR (Too Complex, Can’t Read) or TLDR (Too Long, Didn’t Read)for most people. (How many people would even admit this applies to them? America’s Affordable Care Act, Obamacare, comes to mind.)
Back on the post subject:In contrast to the above, David says he is going to release the program and data, whether the ball gets through the goal posts or not. I have not seen any action by him that indicates he won’t follow through.
20
I justify my existence with this: LPC812 Devl 20 March 2014.
But I have also done commercial – industrial – and aerospace. I’m a large fan of control systems and audio gear. My favorite computer language is Forth. Real Time Control.
00
Willis,
As i understand it the notch is supposed to be showing that at the same time as there is a peak in TSI there is something else going on that has an opposite effect on Earth’s temperature but the effect of that opposing effect does not show up in the temperature of the Earth’s atmosphere until somewhat later. It is suggested that the effect is spread across the subsequent period of 3 to 15 years.
David also says that the opposing thermal effect is a magnitude greater than the TSI thermal effect and manifests itself via global albedo changes that sffect the proportion of solar energy reaching the surface.
Therefore your summary would not seem to be accurate in that David suggests that the notch shows that a solar effect other than TSI is wiping out the effect of TSI and, indeed going further in a negative direction which to my mind suggests net cooling every time TSI rises which cannot be right.
Instead I think that David should reverse the thermal sign of force x so that it operates in the same thermal direction as TSI but that Earth system dynamics then largely mitigate the effect via emergent phenomena worldwide and not just in the tropics as per your theory.
After all, if force x changes the proportion of solar energy entering the oceans as a result of albedo changes then that mimics a change in TOA insolation and the system temperature must change in the same way as if one adjusted the Earth’s distance from the sun.
Thus force x could warm the system when the sun is active and cool the system when the sun is less active which would account for the changes MWP to LIA and LIA to date.
But to get that result David needs to reverse the sign for the thermal effect of force x.I do not know whether his methods could accommodate that.
Note that this proposition is independent of David’s graphics.
Even if David’s graphic is meaningless for the reason you suggest then the scenario that I suggest could still be in operation and would fit in with your theory of emergent phenomena but extended globally.
That still begs the question as to WHY the theory of emergent phenomena works and in that respect I favour processes based on atmospheric pressure but I realise that you do not accept that and it is off topic here.
20
Or something is filling the gap when the TSI is in a trough.
11
the Griss
Yes, I think that covers one point that I was concerned about. Troughs would offset peaks and peaks offset troughs with cooling every time TSI rises and warming every time TSI falls. That implies some median point where the net effect switches over.
Force x operates against the effect of TSI whether TSI rises or falls but only on the 3 to 15 year timescale. On other timescales it seems not to suppress the TSI effect.
Force x is supposed to be a magnitude greater than TSI so how would one get warming instead of cooling when TSI is high if the effect is opposite to that of TSI ?
High TSI does warm the world over multiple strong solar cycles.
David needs force x to amplify TSI effects in the same thermal direction as TSI to produce MWP and LIA but he currently has it amplifying in the opposite thermal direction.
A puzzle.
10
or maybe force X acts to regulate the effect of TSI ?
breky time 🙂
20
Regulation suggests damping down but it is amplification in both directions that we need in order to to fit observations.
Better await comments from David when he wakes up and has had time to think.
00
If you have something that LEADS the TSI peaks, it might explain the 1940’s temperature peak (now erased).
That peak was always a sticking point for the climate scientists 😉
21
If you look at http://jonova.s3.amazonaws.com/cfa/solar-radiation-peaks-magnetic-field-b.gif.
you can see that the magnetic peak around 2005 seems to be leading the TSI drop for the current cycle.
Do we have magnetic data going back further? Something from say 1920 onwards might be interesting.
21
Posting this because its cool !!
http://landshape.org/enm/files/2011/08/dosolar.gif
And another interesting link.
http://www.ice-age-ahead-iaa.ca/scrp/pipf000.htm
21
That implies some median point where the net effect switches over.
I have seen a graph recently (linked in the comments in one of the posts – good luck with that) that said the turning point was a SSN of 40.
00
Stephen, force X acts in the same thermal direction as TSI, for instance to “produce MWP and LIA” as you point out.
My interpretation that around TSI peaks the cloud cover peaks was wrong. That’s just interpretation, and now being corrected.
10
Thanks, David.
It was important to me to get that sorted out because I would very much like my conceptual New Climate Model to fit in with your numerical New Solar Climate Model.
If they can fit together then the combination could present a useful alternative to the GHG based model.
My model has often been criticised for the lack of a suitably ‘sciency’ technical underpinning which is a fair point but I think it fits observations nonetheless.
10
David
I have the same concerns as Stephen Wilde regarding the albedo being at a minimum during solar maximum. However, I still think that the magnetic reversal may have a part to play, coming as it does right in the middle of a solar max- and predictably so.
Before I continue, this graph from Voyager 1 data will interest you. It sums up everything that follows and would certainly appear to support your suggestion that magnetic reversal is the key. It was culled during the peak of this solar maximum (summer 2012) and in the throes of the magnetic reversal:
http://blogs.scientificamerican.com/basic-space/files/2013/06/Screen-Shot-2013-06-27-at-18.41.05.png
Up to 30meV means Voyager 1 was detecting a distinct influx of GCR’s when the solar magnetic field dropped. Moreover, it cycled in the same manner over several days and did so hour by hour. This was recorded at the most salient time in the cycle. The relationship couldn’t be more robust.
This issue of correlation of GCR to solar min or max and the semantics of what solar max means is confusing. This is because the solar magnetic field reversal means the field is at or near zero for a short period during what we would call peak solar activity. Shouldn’t that mean GCR’s come flooding in from interstellar space just at the right time to increase cloud cover and as TSI peaks? Apparently not according to the earth-based neutron count (Oulu Neutron Monitor shows anti-correlation of neutron count with the solar cycle). And yet the Voyager data suggests that GCR’s flood into the solar system when the magnetic field is weak.
The following Tallbloke’s Talkshop post grappled with a related paradox back in August 2013:
http://tallbloke.wordpress.com/2013/08/25/rog-tallbloke-carbon14-and-solar-activity-wikipedia-has-it-wrong/
The post concerned the long-term C14 count and whether it correlated or anti-correlated with temperature in the long term. However, the same issues as you have here with correlation in the short term were being discussed. Stephen also commented on this post a few times. There is a lot of information and links in the comments regarding GCR’s, the nature of solar versus galactic cosmic rays, their distinctively different energy levels and the altitudes at which they spallate. There seems to be a lot of uncertainty and perhaps that is why NASA has been studying GCR’s with a vengeance during this solar maximum by sending planes over the East Pacific on their ATTREX research sorties.
I made a comment in support of Rog Tallbloke’s assertion that C14 should correlate with the long term temperature record as opposed to the conventionally accepted anti-correlation. I did so by paying attention to the 11 year cycle as a short-term version and then extrapolating to the longer term. Using the 2012 Voyager 1 data, I posited that the flip in the solar magnetic field at solar maximum meant that the field was always weakened at solar max and so, paradoxically, a strong sun meant more GCR’s. Rog Tallbloke correctly pointed out the Oulu graph which certainly appears to shoot that theory down. (I was also led astray by the theory that the strong solar wind at maximum would weaken the Earth’s field enough to allow GCR’s in).
However, I wonder where in the solar system the incoming ‘Voyager GCR’s’ ended up. Is Oulu missing what may be nuanced variations in neutron generation, being at ground level and at one particular latitude (65 degrees North)? Is this why NASA also sends balloons high over the entire South Pole solely in order to hunt for GCR’s? Could it be that a short-lived spike of GCR’s set against a background low means that they all get gobbled up high in the polar stratosphere? The secondary neutrons would also be gobbled up surely. That way, they would all be missed at Oulu but detected by balloons. NASA definitely has a preoccupation with GCR’s even though they play it down in their press releases.
Anyway, I hope that gives some food for thought. I don’t have a full picture but I do agree that the magnetic flip could be the key.
Scute
More links:
My comment at the Talkshop deals with the Voyager data more fully than here and has links to articles and 3 NASA scientific papers on the subject:
http://tallbloke.wordpress.com/2013/08/25/rog-tallbloke-carbon14-and-solar-activity-wikipedia-has-it-wrong/comment-page-1/#comment-58190
ATTREX paper on GCR research:
https://espo.nasa.gov/missions/attrex-epo/content/SOLAR_INFLUENCES_ON_CLIMATE
NASA balooning for cosmic rays (old NASA article- the South Pole mission was a year or so ago):
http://science1.nasa.gov/science-news/science-at-nasa/2001/ast15jan_1/
10
Scute,
It does seem possible that cosmic rays could flood in during the short period of solar magnetic reversal but would that really more than offset the reduced cosmic rays through the entire period of high solar activity either side of the reversal ?
Anyway, the link I referred to:
“overdensities of plasma that are efficient at scattering high-energy cosmic rays entering the solar system from elsewhere in the galaxy”
so it is the plasma density and not the magnetic field changes which causes the cosmic ray variations.
Two reasons why I’m not convinced.
I very much prefer the idea of wavelength and particle effects on ozone distribution in the stratosphere to the cosmic ray concept.
00
NASA definitely has a preoccupation with GCR’s even though they play it down in their press releases.
They have people not protected by the Earth’s atmosphere to safeguard as best as they can.
20
willis ,in terms of perception,intelligence and grammar ,you are light years ahead of me ,so i write this post with reticence .
my impression of what david is saying is there is indeed no eleven year temperature signal ,but the eleven year solar cycle is driving the mechanism(s) (force x) that dictate what the temperature will be in eleven years ,a subtle but distinct difference to me.
if this is incorrect or does not make sense please accept my apologies.
20
I get the impression that force x influences the cloudiness of our atmosphere and the clouds determine how much heating or cooling takes place. This is the primary influence on our climate.
TSI is relatively small in comparison and when it peaks there is a small direct heating and also a delayed effect that is modified by force x.(i.e. cloud level)
However, at this stage we are only talking about the input side of the black box.
I think David said that force x is proportional to the previous TSI peak. I guess all of this is saying that the solar cycle strongly influences our climate, but we have to specify which cyclic property we mean. It seems that the key one is the magnetic field strength. Sunspots are obviously a consequence of the contortions of the magnetic field but the point of reversal is a more precise indicator.
This model gives a boost to Svensmark’s cosmic ray theory, but I note that while the magnetic field is a likely candidate for force x, the model is open to other possibilities.
I seem to remember that a few weeks ago No Tricks Zone showed similar trends for global temperature and solar wind speed. I could be wrong, but it seems likely that solar wind speed and the solar magnetic field strength are linked.
30
I’m half suspecting the TSI to trail the forcing.
01
To All. We hear fairly consistently that the suns activity cannot be seen in the temperature record.. But for those who have Vahrenholt and Luning’s book ‘The Neglected Sun’ turn page 69 and look at Figure 3.5. There we see plotted on the same time (yearly) axis Sunspots; Total Solar Irradiance (TSI); UV (Top of Atmosphere);Solar Magnetic Field; Neutron Monitor (Cosmic Rays); Total Solar Irradiance (Earth Surface, South Pole) and finally Temperature. All the graphs lined up vertical beneath one another and all show a close relation of positive and negative pulses. This includes the temperature record. That is, the pulses can be clearly discerned in that record. Looks pretty convincing to me or is this an illusion?
Any comments?
10
Frankly,
On time scales longer than a single cycle, there are correlations.
However, detecting the thermal effect of a single cycle remains problematic except in certain regions which are highly sensitive to variations in the water cycle. Some of them show single solar cycle correlations to the movement of nearby rain bearing weather systems.
It was reports such as that which long ago alerted me to the significance of shifting climate zones and jet stream tracks. They are the true manifestations of climate change and represent the negative system response to forcing elements.
Linking them more tightly to solar variability would a real step forward.
00
Shrodinger’s Cat said:
“I get the impression that force x influences the cloudiness of our atmosphere and the clouds determine how much heating or cooling takes place. This is the primary influence on our climate”
We do seem to be centering on that issue but currently David thinks there are more clouds when the sun is active which is not what we see.
I’d like to know whether his methodology could accommodate a sign reversal without destroying his thesis.
I hope so 🙂
00
“We do seem to be centering on that issue but currently David thinks there are more clouds when the sun is active which is not what we see.”
I got the opposite impression but maybe I misread something.
20
Shrodinger’s Cat is right, I believe (although the reasons won’t be given until much later in the series). No, the interpretation about a dip in cloud cover at TSI peaks is wrong — it is the other way around.
Btw the modeling is independent of the interpretation of force X, so the methods and conclusions from this series of posts are independent of the interpretation of force X.
10
FS- It sounds plausible to me. I’ve not got the book, but I have seen many convincing looking examples of the sort that you describe.
I think that the CO2 supporters are often keen to emphasize the trends that sometimes don’t agree, such as sunspot counts and the TSI cycle, which, according to David, does not show up clearly in the temperature record.
What I like about this model is that it explains why the cycles need not show up. It also explains how relatively long periods of heating or cooling can come about and it also explains the time lag, another factor that has maybe been used to disprove a solar effect.
10
Stephen – I’m a layman on this stuff, but as I understand it, when force x (e.g. magnetic field) is weak then the solar wind is weak and cosmic rays reach the earth in greater numbers and cause more cloud formation > cooling.
I agree that we associate cloudiness and cooling with low solar activity, i.e. few sunspots and less solar wind.
David was talking about force x being weak when TSI was high, meaning they are out of phase. Is that the cause of confusion? I don’t remember him saying that there are more clouds when the sun is active – which implies many sunspots, strong solar wind, less cosmic rays and fewer clouds > heating.
I agree this is important and needs to be resolved.
10
“I don’t remember him saying that there are more clouds when the sun is active”
Sorry, SC, I inadvertently directed my response to crosspatch so I’ll repost it here for you:
David said:
“David Evans
June 22, 2014 at 1:13 am
At a TSI peak force X is cooling (or less warming than usual). Force X works on albedo, so it cools by causing more reflectance, which requires more clouds.
Check: More clouds -> cooler (the next effect of clouds is cooling), a greater % of incoming TSI is reflected back out to space. Required to counteract the surge in TSI.
Ok, seems right. Force X produces more clouds at TSI peaks.”
So, that is a problem since observations show less clouds when the sun is active, especially across multiple strong solar cycles. The evidence is declining albedo/reducing clouds until 1998 as shown by the Earthshine data.
Furthermore, force x must amplify both the warming effect of high solar cycles and the cooling effect of low solar cycles in order to achieve the MWP and LIA.
To do that the idea that force x reduces clouds when the sun is active works because during the MWP the climate zones were more poleward and the jet stream tracks more zonal for a warmer world.
The opposite in the LIA.
David is saying that force x offsets changes in TSI which won’t give that result and requires the wrong sign for the cloud response.
The question is whether David’s methods can survive that sign reversal and I await hearing on that point when he has had time to think. I am not good enough at statistical methods to know either way.
Is the notch a true indication that a force x other than TSI exists (as I and others believe it does) or is the notch just a statistical artifact which has been misinterpreted rather like the MANN hockey stick?
I very much hope that David can adapt his work to deal with the problem because statistical evidence for the existence of force x would be very helpful even though it is not strictly necessary for the force x conceptual narrative to work.
00
Dear All,
>
> We would like to invite you to the upcoming workshop entitled
>
> Workshop on Maximum Entropy Production: Earth, Life and Physical
> Approaches?
>
> 10 – 12 September, 2008
> Yukawa Institute of Theoretical Physics Kyoto University, Japan
>
> This workshop is aimed to bring researchers from different disciplines
> together to discuss new developments and ideas relating to a principle
> of maximum entropy production (MEP) suggested for time evolution of
> statistical mean states of nonlinear systems which are kept far from
> their thermodynamic equilibrium. While such MEP states have been
> detected in a number of nonlinear non-equilibrium systems, from small-
> scale turbulence to large-scale planetary circulation as well as
> biological systems, the basic physical mechanism remains to be
> explored.
> The objective of this workshop is therefore to review the state-of-
> the-art applications and theory of the MEP principle to natural
> phenomena, and thereby investigate future prospects of the MEP
> principle and associated formation mechanisms of spatio-temporal
> patterns in nature.
>
> Invited speakers:
> Axel Kleidon (MPI-Jena, Germany)
> Ralph Lorenz (Johns Hopkins Univ., USA) Garth Paltridge (Univ.
> Tasmania, Australia) Jiangnan Li (CCCma, Canada) Yasuji Sawada (Tohoku
> Institute of Tech., Japan)
>
> We encourage in particular the following topics for contributions:
> 1) Physical and theoretical studies on the MEP principle
> 2) Experimental studies – validity and limits
> 3) Earth and climatic aspects
> 4) Planetary and cosmological aspects
> 5) Fluid dynamics – turbulence and upper bound theory
> 6) Biological aspects – Gaia and its evolution
> 7) Pattern formation in non-equilibrium systems
> 8) Self-organized criticality and its relation to MEP
>
> Further information is available at:
> http://www.yukawa.kyoto-u.ac.jp/english/contents/seminar/detail.php?
02
Suomi’s Paradox: The characteristic jet speed in planetary atmospheres increases despite a decrease of thermal forcing going from Earth to Neptune.
Refer: Coupled Evolution of Earths Atmosphere and Biosphere
by David C. Catling
in
Non-equilibrium Thermodynamics and the Production of Entropy: Life, Earth and Beyond
Axel Kleidon and Ralph D. Lorenz (Eds)
Springer-Verlag 2005
00
Here is another study, posted at Tallbloke, that might tie into yours. It indicates that droughts can be traced back to the times of genesis in the bible and that they occur at 19 year intervals. The part that might tie in is that the drought was not necessarily caused by the lack of rain but the intense heat of the sun I.e. The lack of cloud cover. Wouldn’t you know the author is from OZ.
H. C. Russell: The Periodicity of Seasons – Lunar Timed Drought Cycles
Posted: June 21, 2014 by Rog Tallbloke in Celestial Mechanics, climate, general circulation, History, weather, wind
H/T to Phill for this paper by H.C. Russell reproduced in The Queenslander (Brisbane, Qld. : 1866 – 1939) Saturday 27 June 1896
. But I think the evidence that history furnishes 188 droughts, every one of which fits into the cycle, justifies us in assuming that the 19 years’ cycle has been running for at least 1000 years; and may be trusted^to continue,^ and justify forecasts based upon it for some time to come.f
1. Gen. xxi, 1, in the time of Isaac, 1804 8.C., 3652 years before D, equals 19 multiplied by 181 plus 3.
2. Gen. xlvii., 13, Pharaoh’s (seven years), 3534 years before A, Equals 19 multiplied by 186 exact
3. 11. Sam., xxi., 1, David’s time, 2849 years before A, equals 19 multi plied by 150 plus 1.
4. I. Kings, xvii., 1, Elijah’s drought, 2736 years before A, equals 19 multi plied by 144.
5. 11. Kings, viii., 1, Elisha’s drought, 2717 years before A, equals 19, multi plied by 143.
00
Willis has made this about Willis so I’ll join him. There is zero chance I will accept his opinion given [SNIP..technically ad hom]. His comparison of this process to a motor car speed controller is classic nonsense. And he appears here now under double standard conditions, self promoting even as he badgers an in incomplete work. He has offered his forecast of where this series of posts is going so let him be out and wait for the end game with the rest of us.
70
Joanne Nova
June 22, 2014 at 4:56 am · Reply
Jo, thank your for your thoughtful reply. I fear my point about Jones and Mann wasn’t understood, no doubt my lack of clarity, so let me try again.
My wonderful high school science teacher who taught us so much, Mrs. Henniger, insisted that we do two things on each and every problem— carry through the units, and show our work. And if we failed to do the latter, she’d scrawl it at the head of the page in capital letters with her dreaded red pencil, SHOW YOUR WORK … and science is no different. It means nothing if you aren’t showing your work.
You are asking us to believe you while refusing to showing your work. That is absolutely no different at all from what Michael Mann and Phil Jones did, they asked us to take it on faith, to believe them despite their refusing to show their work. Why do you think that’s OK for you but not for them? Do you think it’s OK for you because you’re one of the good guys, which assuredly you are? In my book you’re definitely on the side of the angels … but science requires transparency whether or not you are on the side of the angels.
And yes, as you say, I can see that your adherents like that fact-free style … sorry, but that wouldn’t impress my science teacher, she’d still whip you with her red pencil for not showing your work, and them as well for applauding it.
Faith? I thought this was about science, not religion. Mann and Jones also asked us to take them on faith. Science is about transparency today, not faith that there will be transparency tomorrow.
Do I believe that you will release the documents? While I certainly hope you do so and will not be surprised if you do, I neither believe you nor do I disbelieve you. I have no reason to do either, so I have no opinion either way. As far as I know, you have no track record in the question of promises made while putting forward a new scientific theory, so I have no evidence that you will or won’t. I would like very much to believe you indeed will … but unfortunately I’m a guy who likes evidence, and in your case, I have none.
However, consider what would happen if I wrote a post on WUWT, told the people I had all these great results, and asked them to take the results on faith … can you imagine the hiding I’d get? And rightly so, if I did that I’d deserve it. Faith has no place in science. Unless and until I am willing to show my work, I have no business advertising my results and asking you to take them on faith … and without data and code, that’s all you have—an advertisement for your results.
I’m sorry, Jo, but I can’t make sense of this. You say “please read my reply again (more carefully this time)”, but the only thing I can find that you said about Lubos in your reply was this single sentence, which I’ve just re-read very carefully:
I’m afraid a careful reading of that hasn’t helped, and I’m way out of the loop regarding your personal emails with him … a citation to further information and a copy of your emails and his replies would be much appreciated, I’m lost here.
Jo, as you point out, you are perfectly free to do what you have done, to publish while refusing to provide data, code, results, tests, out-of-sample runs, the equation of your model, the number of tunable parameters, the exact type of filter used, and all of the rest of the things you’re holding back at present.
I’m just saying that the optics of that suck, regardless what your regular contributors might assure you … and as Mrs. Henniger would tell you, it’s not science.
And that’s about all I have to say, Jo. I do wish you and David all the best, and curiously, I find myself hoping that I’m wrong about your theory … and me hoping I’m wrong doesn’t happen very often.
So I’m going to leave the field to you. There have been a number of people making the same points that I’ve been making about the “notch filter” model, including Dr. Lubos Motl. You’ve paid little attention to either them or to me. I’ve done my best to dissuade you from what I see as an upcoming train wreck, without visible effect.
So I’ll put my pen back into the electronic inkwell, and I’ll check back in whenever you finally decide to release the “out-of-sample” testing that you say has already been done, along with the rest of the things that you’re holding back. The proof of the pudding is in the eating. I’ll hold my peace and wish you well and wait until there’s something on the table.
My regards to you, and my best wishes to everyone here, commenters and lurkers alike.
w.
310
That’s a sensible point you made, and one that Evans needs to address at some stage
For the rest, please drop the corny, salt-of-the-earth, wise old uncle persona. The condescension in it is insulting. We are not convinced of anything as yet. We are simply not that stupid, although we are patient
130
Its funny Willis,
since 2007 we fought together to bring climate science to account.
unpaid of course.
And what were the things we asked for.
1. The data as used
2, the code as run.
3. Standard practices like reporting out of sample testing.
4. Showing things you tried that didn’t work
And when we badgered Jones and mann, well that was fine and dandy.
Go get em Willis!!! Good one Moshpit!!, Way to go Mac!
remember those days?
remember when we demanded code from Hansen and he finally relented?
And the whole choir shouted amen.
Funny how things have changed.
417
You have changed, that’s for sure.
Funny what funding from the climate trough will do to a person, hey Mosh . !
131
nope. what has changed is the target. not me.
principles dude.
you are all for the early release of drafts for the IPCC ( me too)
you are all for open review ( me too)
EXCEPT when it comes to your own.
02
roflmao … seriously !!
20
And you got them all immediately that you asked for them ; absolutely no problem.
I don’t know who you think you are fooling.
60
So you cannot read either Steven Mosher. David and Jo said right at the start they would provide everything you ask for but it would not be published in the traditional way. If you was provided with the 170 page document plus all the associated data , would you read everything ??
I accept David and Jo’s word so I’m sure you will get the data, unlike many other researchers
90
” If you was provided with the 170 page document plus all the associated data , would you read everything ??”
Most certainly. When hansen released his code I spent a good deal of time reading it and trying to get it to run. Eventually CCC got it worked as did ron broberg
When gavin pointed me at ModelE code I read that. took a year
When 2000+ climategate mails dropped in my lap I read for three days straight
When climategate III mails were released to me.. well thats over 100K
still reading.
I read huge amounts of papers and poor over terabytes of data.
170 pages is nothing.
Might as well release it now.
15
Mosher: Things changed when one Steven Mosher helped produce yet another hockey stick, heavy promoted by opportunistically playing off of debunking of Mann and something was sorely left out, namely your own points 3 and 4, and as it stands your promise of a reply to Willis’ inquiries to you about this remain unanswered:
http://wattsupwiththat.com/2014/06/10/why-automatic-temperature-adjustments-dont-work/#comment-1662422
Your own background was this week revealed as being itself in more of the marketing realm rather than the scientific one, I note for the record:
http://www.populartechnology.net/2014/06/who-is-steven-mosher.html
What *changed*, funnily, is that you are now promoting climate alarm using the Berkekey brand in dishonest fashion with an organization unassociated with the college except through the personal leadership of a brazen liar who promoted a knowingly false narrative that he was a skeptic who was converted to climate alarm through the results you helped him create. Furthermore you have become a notorious troll on climate blogs since the old days when you and Muller were in your own marketing phase of playing off of Mann in order to obtain your own portion of climate attention and in Muller’s case, huge media attention. For what? For a ridiculously over zealous slicing and dicing black box with so many parameters that nobody can fathom the internals since you won’t answer any questions about those internals.
190
As I said before.. A used car salesman.. Dodgy Bros personified.
He’s taken the money and made his boat.. now he can sink with it. !:-)
11
Nik,
thanks for that link. “technical paper about flight simulators he co-authored”
Makes it easy when a circle in a Venn diagram only has very few people in it 😉
“Prefer the Sikorsky S-58 myself, having built a flyable simulation w/simple joystick/collective on a hybrid (analog/digital) computer that a military helicopter pilot trained me to fly.” – old comment by “Trick”
10
” refusing to showing your work. ”
Got a reference for that Willis?
Just a quote will do. 🙂
70
Willis,
We knew that once any of this work was released the discoveries would slow as same old counterproductive Internet battle would rage as you know all too well. It’s proving true, since my days are now consumed putting out pointless fires, and responding to arguments which have nothing at all to do with the science, like marketing, and styles of releasing it. Hm?
Sometimes real science needs undisturbed silence to progress. I know you prefer to think-out-loud with group help, and that’s a successful technique too.
Jo, but I can’t make sense of this. You say “please read my reply again (more carefully this time)”
I asked you to read the whole post reply to Lubos because you have evidently missed (twice now) that most of what he says does not apply because he’s attacking something we are not suggesting. I am surprised you find it hard to acknowledge that. It is black and white.
Neither have you managed to quote something substantial or useful from him.
As far as the emails go, if he was not a good guy, I would have viewed his insults as a gift and had quite a lot of fun tearing them apart in public. As it is, I don’t enjoy scoring points off skeptics. I’m hoping we can find another way.
As for the teacher and exams, we are showing you the work as we go. The exam is not over.
We have not told anyone to take our work as a fait acomplit, nor to change global laws because of it, nor to blindly accept it. We’ve invited them to a series of unfolding discussions and the web traffic stats and feedback tell me people really enjoy that. My readers, being skeptics, know this. The only “faith” they need to have is that reading my posts and waiting will not prove to be a complete waste of time.
You claim to be patient and wish us the best, but the content of most of your words reflects neither.
250
Joanne Nova
June 22, 2014 at 4:48 pm · Reply
Huh? You never asked me to read your reply to Lubos, you said:
So I read your reply to me (and quoted your reply to me) in my response, viz:
I’m more than happy to read your reply to Lubos, Joanne, but I can’t read your mind. I have no clue where you replied to him, and I gave up guessing what’s in someone’s mind long ago, that’s a mug’s game.
Which is why I clearly asked you for a link to your comments, a link which, despite you once again telling me how wrong I am, you have neglected to provide. Instead, you’ve continued to accuse me of negligence for NOT READING A PIECE OF YOUR WRITING THAT YOU HAVE NEITHER CITED, LINKED TO, NOR IDENTIFIED IN ANY WAY.
And you think I’m wrong for asking for details of your work?
This is a perfect example of why I ask … because without details, citations and references, vague claims are useless. So let me ask again what I asked before, a request which you have ignored completely:
w.
[Willis did you read this? I think this might be the ‘reply’ Jo was referring to. – Mod]
011
Willis you are not reading carefully at all.
My comment #13.6.1.6 did link to the reply. (You quoted that comment, but you missed the link.)
Lubos’s post which you found so “interesting” also includes a link to my reply.
After a short fruitless discussion with Lubos over the weekend, I’ve now made my full emails available. I had hoped I could save him the embarrassment.
He apparently will not publish my sole comment on his site linking to those emails.Credit to him. He has now.As someone who (like me) asks people to quote directly, perhaps you’ll join me in asking Lubos to quote David’s work, and my emails exactly instead of his mistaken, and bizarre interpretation?
I remain baffled at his behaviour — and yours.
80
Joanne, thanks for your patience with me. My bad, I’m not used to your site’s color scheme and I missed the link. My apologies for saying you hadn’t linked to it, I was wrong and you were 100% right. Mea maxima culpa.
Now, having read your link, it definitely seems like you and Lubos are talking past each other. However, as you know I have retired from further comment on your hypothesis and model until you release the hidden data, code, results, out-of-sample tests, and whatever else has not yet been made public. So I fear cannot take sides in the issue.
Regarding the question of quotations, I will gladly join you in asking anyone to quote the exact wording of whatever it is that they disagree with. My goal is to apply the same standards to everyone. I do my best to quote what I don’t think is correct, as it avoids all kinds of misunderstandings.
As to being baffled at my behavior, if by that you mean my request that you disclose the hidden results and methods. I have asked the same of everyone—reveal the data, reveal the codes, practice normal scientific transparency. That’s what I have repeatedly asked Science magazine to enforce. That’s what I asked of Michael Mann. That’s what I asked of the NSF. That’s what I asked of Phil Jones. That’s what I asked of Scafetta.
And that is what I’ve asked of you … and here is your reply:
I’m sorry, but I’ve put in five years of unpaid work as well, and more, as have Anthony Watts, Steve McIntyre, and a number of others. Guess what? That doesn’t buy any of us an exemption from the normal requirements of scientific transparency. And it has nothing to do with whether the work is tax supported. My work isn’t tax supported either, but for my studies to be considered as science I have to post up my data and code just like Mann and Jones.
Your adherents may indeed love it that you’re not revealing your work … but to date, Joanne, what you have published about your hypothesis is not science in any manner. To date, it’s just an advertisement. Heck, we don’t even know how many tunable parameters your model has. As I said before, when you decide to join the ranks of transparent scientists and release everything, I’ll take another look at your hypothesis. Until then, if I want advertisements, I’ll just watch Mad Men.
Or perhaps that’s not what you mean by you being baffled at my behavior, in which case your meaning isn’t clear.
But regardless how all of this plays out, my best regards to you, and again my thanks for your part in the struggle against alarmism, and for the good done by your blog.
w.
03
I would not expect there to be a direct correlation between TSI and Earth’s temperature because the variation is not, in and of itself, enough to cause the changes that seem to be correlated with solar changes. So it isn’t the TSI itself but the TSI is yet another manifestation of what causes variations in our climate. The TSI change is another effect, not the cause. I personally believe we see variations in clouds caused by cosmic rays but even these aren’t 100% predictable. I would expect space around the Solar System to fly through areas with varying numbers of these particles with varying energies. I would look more at the neutron counts from cosmic rays than using sunspots or TSI and see how those correlate with temperature changes. I believe there are proxies for cosmic ray activity we can use (14C and 10Be) and those can be correlated to temperature proxies from ice cores and speliothems though I don’t know if we have sufficient resolution to see changes on decadal scales.
That there seems to be a 10 year lag in some aspects of our climate has been pointed out by many people over time. That seems to be the amount of time it takes changes in tropical oceans to be reflected in the polar regions. Earth is also a thermal flywheel or a thermal where it tends to resist changes in temperature. If you increase the amount of heat into the system, the oceans will tend to absorb it. If you decrease the energy arriving, the ocean will dump heat to atmosphere but I believe the short term response is different from the longer term response and the response to cooling different than warming (changes in convection would cause changes, particularly in the North Atlantic that would impact the entire circulation system of the oceans).
One other data point to throw out there. I was reading a rather massive paper a while back ( http://deepblue.lib.umich.edu/bitstream/handle/2027.42/62273/waddelin_1.pdf ) on ocean circulation. There is reason to believe that during the last glaciation, areas of the South Pacific became very stagnant at the bottom and pools of pure liquid CO2 (kept liquid by great pressure) might have formed and sat relatively undisturbed for over 10,000 years at a time. So not only do we have the expected property of ocean temperature changing the amount of CO2 released into the atmosphere and being held dissolved, during glacial phases we might have an enormous sequestration of CO2 going on naturally at the bottom of the ocean in deep stagnant areas. Once ocean circulation picked up again, these CO2 lakes in the abyssal deep were stirred up, mixed back in to the water and transferred back into the atmosphere (but it would take somewhat less than 800 years to completely dump that CO2 to atmosphere as we would have to wait for a ventilation cycle to exchange gasses). Anyhow, there do seem to be some potential points for resonance (or damping depending on phase of things) at 10-ish year and 800 year periods.
20
crosspatch,
David said:
“David Evans
June 22, 2014 at 1:13 am
At a TSI peak force X is cooling (or less warming than usual). Force X works on albedo, so it cools by causing more reflectance, which requires more clouds.
Check: More clouds -> cooler (the next effect of clouds is cooling), a greater % of incoming TSI is reflected back out to space. Required to counteract the surge in TSI.
Ok, seems right. Force X produces more clouds at TSI peaks.”
So, that is a problem since observations show less clouds when the sun is active, especially across multiple strong solar cycles. The evidence is declining albedo/reducing clouds until 1998 as shown by the Earthshine data.
Furthermore, force x must amplify both the warming effect of high solar cycles and the cooling effect of low solar cycles in order to achieve the MWP and LIA.
To do that the idea that force x reduces clouds when the sun is active works because during the MWP the climate zones were more poleward and the jet stream tracks more zonal for a warmer world.
The opposite in the LIA.
David is saying that force x offsets changes in TSI which won’t give that result and requires the wrong sign for the cloud response.
The question is whether David’s methods can survive that sign reversal and I await hearing on that point when he has had time to think. I am not good enough at statistical methods to know either way.
Is the notch a true indication that a force x other than TSI exists (as I and others believe it does) or is the notch just a statistical artifact which has been misinterpreted rather like the MANN hockey stick?
I very much hope that David can adapt his work to deal with the problem because statistical evidence for the existence of force x would be very helpful even though it is not strictly necessary for the force x conceptual narrative to work.
10
I think you have the logic backwards.
“At a TSI peak force X is cooling (or less warming than usual). Force X works on albedo, so it cools by causing more reflectance, which requires more clouds.”
This is saying that at TSI peak there are MORE clouds. Higher albedo, more cooling going on at the surface of the Earth. TSI peak is sunspot peak when sun is most active. Counter-intuitively the TSI peaks when there are great black spots on the sun because the areas around the spots are quite bright at various bright patches on the surface where a spot hasn’t quite erupted yet.
Here is TSI graphed with sunspots: http://www.climatedata.info/Forcing/Forcing/sunspots_files/BIGw02-sunspots-and-solar-irradiance.gif.gif
And here is TSI graphed with cosmic rays. Note the cosmic ray graph is inverted (higher on graph is lower cosmic ray count): http://www.climatedata.info/Forcing/Forcing/sunspots_files/BIGw04-sunspots-and-cosmic-radiation.gif.gif
00
Yes, David says that at TSI peak there are MORE clouds.
Observations show that during the high TSI cycles of 21 and 22 up to 1998/9 cloudiness reduced (decreasing albedo).
When weaker cycle 23 began cloudiness started to increase again.
Do you not see the problem?
During weak cycle 24 cloudiness has been pretty stable but the downside of 24 and an even weaker 25 would increase clouds once more, contrary to David’s assumpttion.
Focus on cloudiness and TSI. Ignore magnetic field and cosmic rays.
David’s force x offsets TSI variability and then some which should produce an opposite sign net response to TSI variations
Real world force x augments both high TSI AND low TSI
The difference helps us to pin down the nature of force x.
If force x augments TSI in both directions as a result of cloudiness changes then the only available buffering effect must be from the oceans which places the delay on Earth and not in the sun as proposed by David.
00
I’m not sure where you are getting that from.
All force x needs to do is ‘hold’ the effect of the peak ‘on’, thus filling in the gap of the low TSI periods.
I don’t see why it has to augment the high TSI period.
I’ve played a lot with audio x-o’s, so when I saw that graph of solar peaks and radiation, my mind did an “automatic-addition” of the curves,…
producing something which would be pretty darn “un-bumpy”, but unfortunately. dropping down significantly in recent times.
(I would prefer warm to cold, and raised CO2 to subsistence levels)
10
How would ‘holding the effect of the peak on’ lead to cool periods such as the LIA when TSI is low ?
To do that it must also be capable of ‘holding the effect of the trough on’.
Force x needs to amplify very small solar variations in both directions similarly to fit observations.
00
It never ceases to amaze me how ruthless the world of science is to new schools of thought.
(Reuters) – An Israeli scientist who suffered years of ridicule and even lost a research post for claiming to have found an entirely new class of solid material was awarded the Nobel Prize for chemistry on Wednesday for his discovery of quasicrystals.
50
Stephen, I still don’t fully understand your concern.
You say “Ok, seems right. Force X produces more clouds at TSI peaks.”
Force x is low when TSI is high and vice-versa because their cycles are out of phase with each other. This happens every cycle, regardless of how active the sun is at the time.
“So, that is a problem since observations show less clouds when the sun is active, especially across multiple strong solar cycles. The evidence is declining albedo/reducing clouds until 1998 as shown by the Earthshine data.”
When the sun is active means a high number of sun spots, a strong force x (magnetic field) fewer clouds, more warming.
I don’t see any contradiction.
Are you assuming that an active sun means higher TSI? For me, an active sun means more sunspots equals strong mag field. Low solar activity = low sunspots = weak magnetic field.
The sunspots are a consequence of magnetic field contortions and there are more sunspots when the field has high amplitude. Now TSI peak height may or may not be related to solar activity, but we know that changes in TSI amplitude are relatively small, hence the concept of TSI being a solar constant. Force x can vary a lot, with the power to cause warming or cooling regardless of TSI.
That is my understanding but I could be wrong!
10
Stephen -look at Part IV figure 1.
10
Usually I am not complaining when there is a typo, but now it’s been there for so long and I am waiting for the next post in the chain and it’s easier to see when the right post arrives if the number in the headline is right. ：-） This post has number VI = six not as it should IV = four (In the headline). Amazing it’s not spotted, but guess people is more interested in the content !
Bye the way, this is interesting and so far very easy to follow. Well written, and a great way to publish something. Who needs pal review ? Lubos is a good example that it doesn’t works always. Even you asked for a real test of your idea. Guess you will get a better work and response from him when he done his homework.
10
SC
A more active sun means higher TSI.
The magnetic field is in antiphase so low magnetic field coincides with active sun. As per Part IV Fig 1.
David says more clouds for an active sun (low magnetic field) but in fact we see less clouds at such times.
Note that the magnetic field acts on charged particles rather than cosmic rays.
Cosmic ray amounts are affected by solar plasma effects which are high when the sun is active and allow less cosmic rays to reach the Earth from outside the solar system when the sun is active.
Less cosmic rays for an active sun (low magnetic field) and less clouds (if one assumes cosmic rays are a factor). David says more clouds.
That is a problem.
In any event it has not been established that the magnetic field is force x.
I think force x is variations in the mixture of wavelengths and particles from the sun affecting cloudiness via ozone interactions above the tropopause.
00
Stephen Wilde,
Am getting a bit confused with how you use the term low activity of the sun.
Take 1998. This was a low TSI and low albedo period and a HIGH earth T – empirically. I imagine the cosmic rays entering earth were lower and cloud formation was lower (as albedo was LOW).
This is in keeping with Svensmark’s ideas except that David is delaying the effects with a notch filter. This would mean that TSI is of the opposite sign to sunspot activity (low SS means hi TSI).
I thought Schrodinger’s Cat had it right.
10
tonyM
The low albedo and high Earth T in 1998 was the cumulative result of many years of high solar activity (high TSI). It was not a result of the fact that TSI was then low (bottom of trough between cycles 22 and 23). Indeed, albedo began to increase (more clouds) during cycle 23 which was less active than cycle 22. That is the opposite of David’s contention
As regards solar behaviour I understood that high SS and high TSI go together. It is magnetic field strength that is in antiphase.
00
Stephen Wilde,
Thanks for clarifying your point which forced me to go and look it up. 🙂 Sure enough in Svensmark’s own paper it is confirmed:
“ ….Reid suggested that the solar irradiance may have varied by approximately 0.6% from 1910 to 1960 in phase with the 80-90 year cycle (the Gleissberg period) of solar activity represented by the envelope of the 11 year solar activity cycle.”
So it is in phase and indeed the data shows a minimum sunspot activity at around 1997/8. Given this then it seems David has the signs wrong.
I don’t think it is calamitous as I imagine a consistent sign change to his model and formulae would fix it. Hope so.
I guess this vindicates Jo and David’s idea to release it in sections to get feedback so that modifications can be made. I would like to add my congratulations and appreciation for their hard work and inviting us on their journey in exploring this hypothesis wherever it leads. David made it very clear from the outset that it is a hypothesis subject to testing along the lines of the scientific method; he is not wedded to it.
00
A double hump (amplitude) at the peak might work out. In other words some kind of quasi (few harmonics) square wave.
The climate seems to have preferred temperatures that changes about every 30 years.
10
A lot of the posts upthread were not worthy of the lengthly discussions they triggered. No need to name names. Style has little to do with substance here. The substance is the nature of the black box model, notions about force X (Wegeners continental drift, exhaustively proven using four difference lines of evidence by 1916, was not accepted until plate techtonics was identified as force X), and possible model limitations (e.g. Upthread RATS may not be a constant). Since all ill eventually be published (data and code), style matters even less.
I would suggest Jo and David spent as little time as possible on irrelevant style complaints, and as much as possible on the interesting suggestions concerning possible substance.
30
Stephen
Thanks, I read the wiki article you linked. I do hear you on the fact that the record shows low GCR’s at solar max. I was just wondering that seeing as the Voyager 1 graph shows such an impressive anti correlation between GCRs and the solar field, there might be some way they were being guided in via a magnetic reconnection to the Earth’s poles or some similar mechanism and so weren’t being detected.
Also, you said in a later comment, “Note that the magnetic field acts on charged particles rather than cosmic rays.”
But cosmic rays really are charged particles and the word ‘ray’ is a misleading term stemming from the early days when they thought they were electromagnetic ‘rays’ (like x ‘rays’). This from Wikipedia:
http://en.m.wikipedia.org/wiki/Cosmic_ray
This article says: “The term ray is a historical accident, as cosmic rays were at first, and wrongly, thought to be mostly electromagnetic radiation. In common scientific usage[4] high-energy particles with intrinsic mass are known as “cosmic” rays, and photons, which are quanta of electromagnetic radiation (and so have no intrinsic mass) are known by their common names, such as “gamma rays” or “X-rays”, depending on their frequencies.”
By that reckoning, the solar magnetic field should repel (or guide) GCR’s because, as you say, a magnetic field acts on charged particles.
Having said that, the paradox is that the GCR count is high when the magnetic field is high (low TSI) but if the magnetic field is supposed to repel GCR’s they should be low at this time. This graph was in the Wiki article you linked and shows the Moscow-based cosmic ray count anticorrelated with TSI and therefore correlated with the solar magnetic field (not so much for you as to show other readers):
http://en.m.wikipedia.org/wiki/File:SpaceEnvironmentOverview_From_19830101.jpg
To confound the issue further, the Earthshine data does appear to show that there was a rise in albedo from mid 1999 rising sharply to 2001 and remaining steady all the way to 2007. That means more clouds during the solar max which is what David was saying all along. (My apologies to you, David- I was assuming Svensmark’s posited cloud cover would time-track the Oulu (or Moscow) GCR data). See this Earthshine page, graph in figure 3. It shows the increase in albedo, rising with the solar cycle and the caption says that the rise is ‘a climatologically significant change”.
http://www.bbso.njit.edu/Research/EarthShine/
From the above reasoning it seems to me that David has it right on the low magnetic field and the high cloud cover during solar max but will perhaps find it difficult to link the two via a GCR mechanism. But the mechanism is secondary to the notch and, as far as I remember, David made it clear that the Svensmark theory was only one possibile theory to churn through the black box, so to speak.
20
Scute said:
“Earthshine data does appear to show that there was a rise in albedo from mid 1999 rising sharply to 2001 and remaining steady all the way to 2007. That means more clouds during the solar max”
That’s what comes of cherry picking 🙂
There was a long period of declining cloudiness across several strong solar cycles to 1998/9. The opposite of David’s contention.
The sun became less active at the end of cycle 22 and cloudiness started to rise. The opposite of David’s contention.
You are suggesting on single cycle 23 and saying that the rise in cloudiness from 1999 to 2001 was due to the rising first half of cycle 23. For your approach to hold true there should hasve been no plateau from 2001 to 2007. Cloudiness should have dropped again on the downswing of cycle 23 but it didn’t.
So, on balance I don’t think you or David have it right.
In general an active sun gives less clouds not more clouds so David needs to deal with that and I hope he can.
01
“That’s what comes of cherry picking” I think that comes not from “cherry picking” but from a peer reviewed paper that was published covering that period. Here is a longer term graph:
http://wattsupwiththat.files.wordpress.com/2007/10/earth_albedo_bbso.png?w=640
Mr Watts had an article on the subject at his site, too:
http://wattsupwiththat.com/2007/10/17/earths-albedo-tells-a-interesting-story/
The time frame mentioned seems to be the same as this paper on the subject.
http://bbso.njit.edu/Research/EarthShine/literature/Palle_etal_2008_JGR.pdf
10
The American Thinker has an article that appear to be using David and Jo”s Model, with some interesting graphs. http://americanthinker.com/2014/06/a_cold_dawn_coming.html.
10
Yes, and the article contains a pointer to “BIG NEWS part VI…”
“Well, we now have a way of cross-checking the tree-ring based predictions. A “just released climate model” using a notch-delay filter has the promise of providing much higher resolution in climate forecasting. Using historic TSI data, the model can see out to 2025:”
00
What’s kinda interesting is their fig 2, which shows predictions out to 2100, presumably from this model. Clearly American Drinker gets early sight of results that folks here have to wait for.
17
Mr. Connolley,
I am still trying to determine who the players are on these websites so I often write impressions of the more negative and unhelpful writers, IMO. I took my impressions of your comments and googled the results and ended up at wiki at this definition:
Malignant narcissism is a psychological syndrome comprising an extreme mix of narcissism, antisocial personality disorder, aggression…[1] Often grandiose, and always ready to raise hostility levels, the malignant narcissist undermines organisations he/she is involved in, as well dehumanising the people he/she associates with.[2]
32
Re #67 Evans model prediction in American Thinker.
I don’t find that scenario plausible but we’ll know by end 2017 whether it’s valid or not. I think David’s “clock” delay (based on spurious observation and fixation on the minor “fast” response IMO) produces a temperature response far too soon and far too abruptly. Yes historically, there is evidence of abrupt temperature falls when solar activity becomes relatively weak (the IPCC’s Mike Lockwood acknowledges this) but this is a radical fall.
If the delay was derived by thermodynamic principles, the major “slow” delay is over a range of years and decades but concentrated on 14 years +/- 6 years (by Abdussamatov’s calc) i.e. the effect of solar conditions in December 2001 (SC 23 peak) has yet to experienced to its fullest extent (2001/12 + 14 = end of 2015 + 6 = 2021). That peak was at near Grand Maximum levels of 1986. The “slow” effect of the relatively weaker SC peak February 2014 wont be experienced until 2028 +/- 6 yrs = 2022 to 2034. Sure there will be a faster response in the meantime but faint and competing with ocean oscillations.
We’ll see who’s got it right over the next 3 years or so.
20
David,
I think your transfer function is flattish, apart from the “notch”, because the trends in temperature (T) and TSI are similar.
Imagine a simple linear trend for both T and TSI over the analysis period. The resulting transfer function would be perfectly flat,
because matching Fourier amplitudes would be identical, apart from a scaling factor.
Now imagine adding 11-year oscillations to TSI.
A peak appears in the TSI spectrum, and a notch appears in the transfer function, even though T is still just a linear trend.
10
Note that my point about the correct sign for cloudiness changes does not necessarily detract from the significance of David’s notch filter solar climate model.
It only affects the way it should be interpreted in order to identify the nature of force x.
I am not in a position to judge the validity or otherwise of the notch filter itself. Whether it is valid or not our observations show that force x (as an alternative to TSI as a solar driver of Earth’s climate) does exist anyway.
Whatever the outcome of the statistical exercise David and Jo have given a powerful boost to the concept of a solar based climate driver other than TSI.
30
Question for Jo Nova and Dave:
Might your empirical notch filter indicate the same energy source for TSI and for cosmic rays (e.g., Peter Toth’s suggestion of a pulsar solar core)?
Nature 270, 159-160 (10 Nov 1977): http://www.nature.com/nature/journal/v270/n5633/abs/270159a0.html
00
Quite frankly, there are too many “cheerleaders” here (who wouldn’t know a notch-filter if they got one as a prize in a Cracker-Jack box) who are not appreciative of the extreme caution engineers usually (at least should) take when they are presented with a new device, tool, or theory. So many of the best skeptics are engineers precisely because of their normal practice of wanting to know, exactly, how things work. Any actual crackpot theory is dismissed out of hand, and properly so, so I contrast the asking a lot of inconvenient questions as really the sign of healthy progress. All of us are perhaps guilty at time of not being patient either as proponents of a theory they have been working on for a year, or as witnesses waiting for particular details, and (understandably) not understanding. If it were easy, non-engineers could do it.
I don’t know yet about David’s full theory or even the individual installments. One cannot be comfortable if even one of the many component parts is obviously wrong. In comment 32 above, I pointed out that David’s idea of a notch filter being “inherently” non-causal is wrong. It is wrong, or David is talking about something entirely different which he has not specified, although anyway, non-causal is something it is hard to justify embracing as an explanation in the physical world. David himself says (just below Fig. 2 of Part III):
“Notice that the step response starts several years before the step-up, which violates causality — it is impossible. In our universe, a response can only come after the corresponding stimulus. The non-causality of the step response of the notch filter in Figure 2 is not a fluke: in any electronic notch filter without an accompanying delay, the step response is blatantly non-causal. Notch filters by themselves are intrinsically non-causal.”
Fig. 2 is clearly a calculated response, and it is apparently rotated 180 degrees and offset from reality, giving the appearance of being non-causal while actually it should be asymptotically ringing about the step height for positive time. We are not given the corresponding (Laplace) transfer function which should be something like T(s) = (s^2 + 1)/(s^2 + Ds + 1) where D=1/Q is the “damping”. The step response is the inverse Laplace transform of (1/s)T(s). This is sophomore/junior-level EE. (Not to suggest that anyone can actually do inverse LT’s correctly! )
We have neither David’s T(s), nor anything about the calculation of Fig. 2. (Some program package?) It does seem from Fig. 2 that there is a single “lobe” shown moving left. It does seem to be asymptotically approaching a zero, strongly suggesting that David has chosen a “critical damping” (D=2 or Q=1/2), a special case. Critical or not (with ringing, or never reaching 0) the anti-causal response goes on indefinitely to negative infinity. With the critical damping, it might LOOK as though a non-causal response could be made causal by using a delay. In a practical case, why not? Three problems, the first of which is that there is nothing to fix – it is not non-causal. Secondly, it does not correct the erroneous flipping of time direction. Thirdly, we would have to specify critical damping (D=2), and have yet another “just-so story” parameter (along with Force-X, exact notch cancellation, a special Fourier transform).
My own contribution here was to simply throw together a working notch (classic circuit – from memory) on my bench and fire it up. The result was, or course, causal. I knew that but had to breadboard the circuit to take a scope photo. It is repeated here from my Comment 32:
http://electronotes.netfirms.com/NotchStep.jpg
David replied in a perfunctory manner:
“Bernie I’ve no idea and am not going to debug and figure out what you have done exactly.”
I thought I had made it as simple as possible, using the “state-variable” signal flow-graph form, three summers and an integrator, rather than one of the half-dozen or so hard-to-analyze single op-amp possibilities. I was expecting something like a comment that that was not what he was doing.
Another even more classic form of a real notch filter is the series RLC. There is a nice web calculator for this which I just found:
http://sim.okawa-denshi.jp/en/RLCbekeisan.htm
For convenience, enter C=1, L=1, and then R as the value of D you want (for example, R=D=2). Out comes the step response. Does it look like David’s? Nope – it looks almost exactly like mine – causal and all. Of course, it had no choice.
It is possible that reversing time as David did could be compensated by a subsequent sign change later. (There is still the offset to worry about). But you can’t just truncate and shift (delay). Time is still backward. That’s different, and not unlikely, very different.
David – please take this seriously. This comes from a friend (see for example, your tip-jar receipts over the last few years!). We don’t need to hand the alarmists a softball which they will find later.
Bernie
41
Wow, a real engineer, I’ve only ever done these things in software for digital systems.
I too am baffled as to why David chose to use acausal filters.
You can easily do a causal one, thereby not frightening the horses, plus a separate time delay.
10
Indeed. My comment was already too long, so I did not want to get into a discussion that filters can be non-causal in the discrete-time (sampled) case, usually called “digital filters”. You can and do use them for things like 2-D pictures where space rather than time is the independent variable, and for things like oversampling music players where the music file might have been recorded 5 years ago and no one cares about an additional 1 second delay. (Someone above mentioned the “already recorded” aspects.)
We also DESIGN non-causal digital filers of the FIR [Finite (duration) Impulse Response] type, using for example, an inverse DTFT (a Fourier Series in frequency) and then shift the whole thing to the right, putting in a linear phase all at once. It’s just easier mathematically. More honestly, we probably just do it wrong, hoping in this one case phase won’t eat our lunch, say ooooops, and put in the delay.
No one uses non-causal continuous-time (analog) filters because – well because there aren’t any such real-world filters.
10
No one uses non-causal continuous-time (analog) filters because – well because there aren’t any such real-world filters.
This is a DIGITAL analysis not a control system. And the playback delay can be days with out mattering a bit.
The big error possibility is that it will cause a misdiagnosis.
10
A DIGITAL analysis? Do you mean “discrete-time”? Where is the sampling?
00
Bernie, See the answer here.
My time is extremely limited at the moment, I am already a day behind in answering questions, and I cannot spend hours debugging other people’s set ups. Sorry.
I think you need to check what you’ve done, in particular the sign of your phase (different sign convention? try reversing it, could account for “time backwards”). Butterworth filters are low pass. Certainly check my calculations when the spreadsheet is released. In the meantime, I am pretty certain that the step response that I am responsible for is correct, and that is is non-causal unless teamed up with a delay of several years.
20
David –
I do appreciate that you respond, but you need to consider that you have this wrong. I think it makes significant differences. Further, doing it right probably not only corrects an error, but also simplifies the results.
(1) I did no calculations, so there is no assumption about phase. Mine was an EXPERIMENT – components on a bench. Independently, the website I found DID do the calculation, and agreed with my experiment. In as much as the notch is just the sum of a LP and HP (with same poles), the notch step response is the sum of their step responses, widely published, and easy to sum.
(2) I assumed you were familiar with filters, but it appears you are not. Butterworth CAN be low-pass but it can also be band-pass, high-pass, or notch. Butterworth refers not to the “type” but rather to the damping “characteristic”, the reciprocal of “Q”. Butterworth Q (for second order) is 1/sqrt(2). Your Fig. 2 response looks to my eye to be Q=1/2.
(3) You said “Even if the causality argument was wrong, the delay works and is independently corroborated.” Well – if the causality is wrong, you don’t NEED the delay, and putting in an artificial delay misleads you further. Your Fig. 2, because it appears you chose critical damping, or nearly so, leads you to suppose that you can remove the anti-causal part with a time shift to the right (a delay). [In theory, since the anti-causal tail goes on to negative infinity, even with critical damping, you could not ever do this.] But even in a “for all practical purposes” scenario, this would NOT fix the direction of time. Because the step response is really causal, you not only don’t need a delay, but the delay you add IS destructive to a correct result. And perhaps “destructive” is exactly the right term – in summing immediate and delayed paths. (Is that perhaps where your “dagger” comes from in the overall response?).
11
Part 4 says
1. When force X is strong the earth warms
There seems to be some confusion in the comments
00
Which of course implies that force X is not a force that cancels the TSI peak, but a force that augments between the peaks.
00
the Griss,
Absolutely, you are on the right track but note that to produce the MWP with an active sun and LIA with an inactive sun force x needs to augment TSI in the SAME direction. It must augnent BOTH peaks and troughs of TSI.
Otherwise the more more active sun results in a greater negative force x response and the LIA should have occurred when we actually got the MWP.
One can only get the right sign of response if an active sun reduces cloudiness to allow more solar energy into the oceans to produce the MWP or the 20th century Warming Period.
David proposed more clouds when the sun is active (high TSI) which cannot be right in light of observations.
00
I have come around to your point of view. It is an augmentation that is out of sync with the SSN – maximum augmentation happens at SSN minimum. The amount of augmentation is related to the previous SSN peak.
00
There seems to be confusion about active sun – TSI – more clouds – less clouds and what David said about it.
I’ve searched a lot of data and to be honest, I haven’t got the time to resolve this. TSI alone seems to be very difficult to measure and depends on which satellite data you pick.
I realise this is anecdotal, but one way of looking at it is the Maunder minimum = cold = low solar activity = low sun spots = probably lots of clouds = low transmission of heat to earth surface regardless of the TSI.
Now, I tend (maybe wrongly) to associate low sunspots with lower magnetic field amplitude and reduced solar wind (hydrogen, helium and other nuclei). This reduction of the solar wind means less shielding of the earth to Cosmic rays.
These mis-named particles tend to be protons and nuclei of the same charge as the solar wind but they are said to be very high energy and according to Svensmark, good at seeding cloud formation by some indirect mechanism.
So a reduced solar wind allows more cosmic particles and creates more cloud cover.
I can live with all of that. I’ve got a suspicion that Stephen Wilde’s definition of high solar activity = high TSI may in fact be correct, but misplaced. By that, I mean that the TSI effect is small compared with the effect on cloudiness.
Also, I seem to remember that David said that force x is proportional to the previous TSI.
So, my overall feeling is that force x = transmission of heating from the sun to the earth. The TSI peak corresponds with low transmission of heating. Later on, the transmission is poor but then rises to a peak around 11 years later. The heat from that TSI is a combination of the strength of solar irradiance modulated by its transmission to earth (force x aka inverse of level of cloudiness).
In this balance of powers, force x as a factor is bigger than TSI by an order of magnitude but force x is proportional to the previous TSI peak.
It seems that TSI is actually a small player, diminished by force x to the point that it is more than cancelled or enhanced, depending on the value of force x.
10
“Also, I seem to remember that David said that force x is proportional to the previous TSI.”
I think this might be a heat transport issue. Again, he is simply describing how the system behaves, not why it does so. Basically he’s taken the climate, put it in a black box, observed the inputs, observed the outputs, and is stating what he believes the function is that the black box is performing.
I would say for the sake of these articles so far, ignore any ideas about why or how and try to focus on the behavior only without attributing causes for it.
It has been noted in many places in the past that there seems to be a 10 year lag in response to tropical changes and I believe that has to do with the transport of enough heat from the tropics to the rest of the planet, mainly oceans. The same factors that can increase albedo and lead to cooling in the tropics can also lead to lead to LESS cooling in the polar regions. Clouds might be a net cooling factor in the tropics and a net warming factor at the poles. It depends on where the clouds form and when. And some things are also a bit counter-intuitive such as ENSO cycles. During La Nina, there is generally less cloudiness over the equatorial Pacific. More solar energy is being absorbed by the water and that water is being pushed into the Western Pacific Warm Pool by strong trade winds. La Nina is basically the charging of the system with energy. In El Nino cycles, the trades relax, the warm water sloshes back across the equator and dumps its energy to the air so we get warmer air temperatures even when there are more tropical Pacific clouds. El Nino is a discharging of the energy from the ocean into the air.
The function inside the box is very elaborate with a lot of different things operating on overall climate and how something works in one place might not be the same as how that very same factor operates in another place.
40
Crosspatch,
Thanks for your view from a 1,000′ flyover, it helps keep the discussion at a layman’s level.
Can’t wait for the next chapter!
Question: if there is an 11-year delayed response in an 11-year cycle, wouldn’t the instantaneous signal appear reversed?
00
>”observed the outputs” (crosspatch #72.1)
That’s the problem I have with David’s rationale.
1) A false premise that there is an actual “11 year solar cycle” based on average length, not actual length of each recurrence of the phenomenon false.
2) A very incomplete understanding of the major “slow” thermal delay and response vs the minor “fast” response he looks for as a result of 1) and this incomplete understanding in 2).
3) From 1) and 2), a false assumption that 11 year periodicity “should” be observed in temperature at a significant level (also a false premise).
4) Then, even in the narrow and spurious scope of the minor “fast” response, he uses inappropriate analysis tools and datasets thereby missing the weak “pulse” in temperature at SC peaks due to the fast response when other analysts using different tools, datasets, an not constrained by a fixed period of 11 years DO find the solar signal of the minor fast response in temperature.
01
Should be:
“…not actual length of each recurrence of the phenomenon”
And,
“…DO find the solar signal [in] the minor fast response [of] temperature.
00
“1) A false premise that there is an actual “11 year solar cycle” based on average length, not actual length of each recurrence of the phenomenon false.”
He has making no such “premise” that I can see. There is certainly an 11 year cycle to TSI and sunspots and cosmic rays. We can see that documented in any number of places. There also seem to be significant variations in the strengths of individual cycles and these variations in relative strength. It isn’t so much that the numbers vary on the 11 year cycle so much as the variation in the strength of them. So when “force X” gets “strong” we might see a reduction in clouds and in albedo and increased energy to the surface well before the sunspot cycle even begins to ramp up significantly.
“2) A very incomplete understanding of the major “slow” thermal delay and response vs the minor “fast” response he looks for as a result of 1) and this incomplete understanding in 2).”
Again, it appears to me that you are reading more into what he is saying than what he is actually saying. He has so far not gone into any speculation on causes and is simply showing so far the data of the change in inputs and the change in outputs. So far he has only stated that when you feed that change in input and observe the output, the result looks a lot like the filter he describes.
“3) From 1) and 2), a false assumption that 11 year periodicity “should” be observed in temperature at a significant level (also a false premise).”
Did he say it “should” be observed or did he say simply that it “isn’t” observed? Big difference.
“4) Then, even in the narrow and spurious scope of the minor “fast” response, he uses inappropriate analysis tools and datasets thereby missing the weak “pulse” in temperature at SC peaks due to the fast response when other analysts using different tools, datasets, an not constrained by a fixed period of 11 years DO find the solar signal of the minor fast response in temperature.”
Opinion. What data do you have to show the analysis tools and data sets are “inappropriate”? Why are they so? What we do seem to see, though, is a response to a more general change when the cycles themselves get stronger or weaker and the influence of a strong or weak cycle doesn’t seem to be felt until after some 10-ish years of lag. That could well be just a coincidence. Due to current configuration of the continents and the ocean circulation currents, we might see a 10 year lag in a global data set (though we might see local changes much faster, I don’t know. Maybe the tropics would reflect the change faster than a global climate average). If the continents were in a different configuration or if there was no “Antarctica” or Arctic Ocean and the poles were in the middle of wide open ocean, maybe the response would be quite different.
But I don’t believe he is trying to ascribe causes to things at this stage. He’s only describing behaviors.
10
crosspatch #73.1.2.2
Re 1) >”He has making no such “premise” that I can see.”
His premise from Part II (my emphasis):
“vary between 8 and 14 years”, is that really an “11 year cycle”?
And from my reply to Willis #74 in Part II:
http://joannenova.com.au/2014/06/big-news-part-ii-for-the-first-time-a-mysterious-notch-filter-found-in-the-climate/#comment-1492355
>”There is no sign of the 11-year solar sunspot cycle”
Why bother looking for one in the first place?
SC 22 9 yrs 8 months
SC 23 12 yrs 6 months
SC 23 peak to SC 24 peak 12 yrs 2 months.
I think the long-term average is 11.2 yrs.
The “cycle” is a recurring phenomenon conventionally described as an “11 yr cycle” because of the average length but that is not what it actually is (as with the other longer “cycles”).
In other words, there is no such thing as an 11 year solar cycle to go looking for.
Coughlan and Tung (2004) decided against Fourier analysis for this reason. See their criticism of Fourier analysis wrt solar vs temperature in the Part VI thread here:
http://joannenova.com.au/2014/06/big-news-part-vi-building-a-new-solar-climate-model-with-the-notch-filter/#comment-1491912
Paper linked above that at #33.3
Using EMD they clearly identify SC peak warming from the surface right up through the troposphere in IMF 4.
Similarly but with different methodology with Zhou and Tung (paper also linked at #33.3).
Re 2) >”Again, it appears to me that you are reading more into what he is saying than what he is actually saying. He has so far not gone into any speculation on causes and is simply showing so far the data of the change in inputs and the change in outputs. So far he has only stated that when you feed that change in input and observe the output, the result looks a lot like the filter he describes.”
Yes he has speculated on causes, see Part IV ‘Physical Interpretation of the Notch and Delay’ but only in terms of his notch and delay. From what I can gather he has (by his rationale) eliminated the ocean by low pass filter. David in Part IV in the same section:
He seems to assume that a low pass filter will detect the oceanic response (by far the greater of “fast” and “slow” responses). I think this is a fallacious assumption. Stephen Wilde says at #77.1 (I don’t know his basis for this, my emphasis):
I replied to that at #77.1.1 pointing out Abdussamtov’s 20±8 year oceanic lag and that:
David appears to eliminate the ocean then suggests (according to Stephen) reintroducing it.
Re 3) >”Did he say it “should” be observed or did he say simply that it “isn’t” observed? Big difference.”
He said, specifically (see 1) above), “we’d expect to see stronger sinusoids around 11 years”. I infer from his expectation that he thinks he “should” observe the signal, hence his surprise when he doesn’t.
I wouldn’t expect to see the (very faint and very minor) “fast” signal in globally averaged datasets and I’m not surprised he didn’t observe it there. Look elsewhere with more appropriate datasets and analysis tools and the “fast” signal IS observed and documented in the literature.
Neither do I expect to observe the “slow” oceanic response in temperature, certainly not as a sinusoid and certainly not at 9±6 year period (“3 to 15 year oceanic buffering” attribution unclear).
Re 4) >”Opinion. What data do you have to show the analysis tools and data sets are “inappropriate”?”
Good grief, see 1) above.
00
1) Ok, right. I didn’t read it’s meaning that way. Basically I thought he was saying that in a “flat” system, if Earth was a pure black body, the changes in TSI would result in changes in temperature. But we don’t seem them. That is the entire point of the article, to describe what the REAL response is. At least that is how I’ve taken it. He was describing the system and I didn’t take that to mean a literal expectation of temperature variation. I took that to mean something like “if there were no filters, we would expect to see Earth’s temperature vary with the cycles”. I didn’t take that to mean he literally expected it to vary on 11 year cycles because it is widely known that it does not. It was said for illustration purposes in introducing the concept.
2. “He seems to assume that a low pass filter will detect the oceanic response (by far the greater of “fast” and “slow” responses). I think this is a fallacious assumption. ” Ok, that’s opinion. I happen to be of a different opinion. I believe the oceans DO act as a low pass filter, sort of like a thermal flywheel. They moderate short variations in energy by absorbing or releasing it but at some point, if the energy input is sustained, the temperature will begin to rise.
3) ok, again as in number 1, I don’t believe he meant that in a literal sense and he already knew that the fact that we don’t see 11 year changes due to solar cycles and was using that as illustrative. If Earth were a bowling ball with no water and a clear atmosphere, it would see temperature variations with those changes. I took the point to noting that was to lay the groundwork to explain why we don’t see them.
10
crosspatch #73.1.2.2.1
Re 2) (my emphasis)>”They [oceans] moderate short variations in energy by absorbing or releasing it but at some point, if the energy input is sustained, the temperature will begin to rise.”
Yes, except the short variation of a single SC peak is not sustained. It is not until there is a sustained variation over a series of SCs that a change in the heat content of a heat sink (indicated by temperature) is apparent.
The heat sink is not necessarily ocean. I’ve already cited PERRIER et al (2005) elsewhere but here it is again:
LONG-TERM CLIMATE CHANGE AND SURFACE VERSUS UNDERGROUND TEMPERATURE MEASUREMENTS IN PARIS
http://onlinelibrary.wiley.com/doi/10.1002/joc.1211/pdf
See page 1639:
Figure 12. Underground temperature records ( diamonds) below the Paris Observatory ( depth 28 m) compared with sunspot number:
(a) data from Cassini. (b) data from Arago using the Lavoisier thermometer
The temperature response of the heat sink to solar change is clearly identified in (b) 28 m underground over 3 SCs.
Similarly, the accumulation of heat in all planetary heat sinks from a sustained level of consecutive SCs (4+ – the Modern Grand Maximum) is only now reaching peak:
http://www.skepticalscience.com/graphics/Nuccitelli_OHC_Data_med.jpg
In other words (in planetary terms), lagged accumulated heat is the response of the heat sinks to a sustained increase of solar energy input over decades.
The reverse for a sustained decrease of energy input after an increase.
Oceanic release of the recent heat accumulation will also take decades.
00
crosspatch #73.1.2.2
Re 4) >”(though we might see local changes much faster, I don’t know. Maybe the tropics would reflect the change faster than a global climate average)”
Exactly, varying responses in localized regions so look local – not global averages.
I suggest you read comment #79 which quotes excerpts from the paper:
‘Sensitivity of the surface temperature to changes in total solar irradiance calculated with the WRF model’
http://joannenova.com.au/2014/06/big-news-part-vi-building-a-new-solar-climate-model-with-the-notch-filter/#comment-1493205
Not because I view model simulations as the be-all-and-end-all answer to your knowledge gap (certainly not) but because the paper does go some way to explain how local signals are lost by averaging, as you surmise.
Keep in mind there are reanalysis datasets that don’t average to the single time series of datapoints that David has analyzed. Model simulations as above attempt to resolve locally all over the globe. If successful (V&V, verified and validated) they mimic the local regional observations that form the basis of reanalysis datasets.
00
SC, crosspatch,
David said high TSI = more clouds.
In fact, observations suggest high TSI = less clouds.
The TSI effect is indeed small compared to the effect from force x. Both I and David agree on that.
I am sorry that confusion has arisen as a result of the involvement of magnetic field strengths and cosmic rays but the basic discrepancy is there and needs attention.
That is not a derogation from David’s basic finding of a notch. If substantiated, the notch is helpful but I am neutral on the issue of substantiation.
My interest is in the interpretation of the notch (if substantiated) and the identification of force x.
To correctly interpret the notch and/or identify force x we must resolve the apparent discrepancy concerning the relationship between solar variations and global cloudiness trends.
I think David has got the point and is working on it and I await his comment when he is ready.
01
I’m astonished that people think they can deduce how the climate works from 2 highly noisy, uncertain and “short” time series.
There are so many speculative and uncertain links between uncertain variables, each of which has uncertain effects on the climate.
That way madness lies, sanity is treating this as semi-empirical:
If I filter this time series it closely resembles another time series, which may (in the fullness of time) be of significance for climatology.
00
Can you please quote the segment where he said that? I haven’t been able to find it.
“David said high TSI = more clouds.”
10
The part you quoted in an earlier posting:
“At a TSI peak force X is cooling (or less warming than usual). Force X works on albedo, so it cools by causing more reflectance, which requires more clouds.”
Says that TSI peak has more clouds, not less. At TSI peak there is cooling. Force X works on albedo … which requires more clouds” He is saying there are more clouds at TSI peak, I think.
Let me re-word it:
At a TSI peak, force X is causing cooling. Force X causes changes in albedo. Force X cools by causing more clouds”.
00
Ok, I think I have figured out the confusion. We are talking about two different things.
When TSI is high, clouds generally are less than when TSI is low. In other words, when the sun is inactive, more clouds. When the sun is active, less clouds. But that has nothing to do with “Force X”.
Now when force X is more active, there is still fewer clouds at TSI maximum but MORE than there would have otherwise been. So if you look at a very active cycle of solar activity, you would see low cosmic ray counts and low earthshine albedo. If you look at weak cycles, you will see more cosmic rays and more earthshine. So yes, we do probably have fewer clouds now and lower albedo than we had during the last solar minimum but we probably have more clouds and higher albedo than we had the last solar maximum or the solar maximum before.
30
Stephen, thank you for participating in this ongoing discussion. I enjoyed the challenge of trying to compose a logical argument regardless of whether I am right or wrong.
Such debates stimulate the brain and the thinking process and that is very valuable for the sake of the science and for the thinking processes of the individual.
Maybe David will consider our attempts to understand his model and offer clarification in a future post. In the meantime, I have really enjoyed the debate and feel that I have learned more about your view as well as my own.
Good fun.
00
Thanks SC.
Civilised debate is rare in this field.
30
David and Jo I have to say I am really enjoying these threads. Even the rants by your ‘contributor’ “Basil Fawlty” Like the husband and wife psychiatrists where one turns to the other, and referring to Basil, says ” There’s enough material there for entire conference”(lol)
10
This new theory is beginning to explain how we ‘see’ oscillations in temperature datasets.
http://www.marinesim.eu/Tisdale_modified.jpg shows a Bob Tisdale graphic with a step response overlaid from 1988,(blue line) showing the typical output from a PID controller when controlling a system with built in delays.
Delays greater than one year can be explained with this new theory.
10
Delays greater than one year can be explained by oceanic buffering of solar input by more than one year.
David suggests buffering smeared over a 3 to 15 year period and suggests that the delay occurs in the sun.
I think the delay occurs on Earth and the smearing represents the time required for the initial thermal result of cloudiness changes to spread through all the ocean basins.
That fits with Bob Tisdale’s oceanic analyses.
I would expect to see upward temperature stepping at 30 year intervals (one positive Pacific Multidecadal Oscillation -PMO- to the next)during a period of increasingly active sun such as LIA to date and downward temperature stepping at 30 year intervals (one negative PMO to the next) during a period of increasingly inactive sun such as MWP to LIA.
If David reverses the sign of the cloudiness response to force x and places the delay response in the oceans then his theory will fit observations.
20
Stephan #77.1
>”Delays greater than one year can be explained by oceanic buffering of solar input by more than one year. David suggests buffering smeared over a 3 to 15 year period……”
Abdussamatov puts the oceanic component of planetary thermal lag at 20±8 years (page 2 at link below):
“Difference between the incoming Ein and outgoing Eout radiation is described by the following equation:
Е = (S+ΔS)/4 – (А+ΔA)(S+ΔS)/4 – εσ(Тр+ΔТр)4. (1)
Here S is the TSI, ΔS the increment of the TSI, A is the albedo of the Earth as a planet (Bond albedo), ΔA is the increment of the Bond albedo, ε is the emissivity of the Earth-atmosphere system, σ is the Stefan-Boltzmann constant, Tp is the planetary thermodynamic temperature, E is the specific power of the enthalpy change of the active oceanic and atmospheric layer [W m–2], which can be considered as the energy balance of the annual average budget in the debit and credit of the thermal power of the planet.
Quasi-bicentennial variation of the solar radiation absorbed by the Earth remains uncompensated by the energy emission to space over the interval of time that is determined by the thermal inertia of the World ocean (20±8 year).”
http://scienceandpublicpolicy.org/images/stories/papers/originals/grand_minimum.pdf
Also (page 3),
“The formula for the increment of the Earth’s effective temperature due to the increments of the TSI and Bond albedo can be obtained from:
ΔТef = [ΔS(1 – А – ΔА) – АS]/(16σТ3ef). (2)”
# # #
If you look at Figure 2. ‘The prognosis of natural climate changes for the next hundred years’ on page 6, a 0.6 C fall in temperature occurs by 2035 from Abdussamatov’s approach.
David’s approach produces a 0.6 C fall by 2016 according to Archibald:
http://admin.americanthinker.com/images/bucket/2014-06/192604_5_.png
A 19 year difference in time to reach the same level of temperature, David’s 19 years in advance of Abdussamatov’s.
Assuming David’s model does not yet have this oceanic buffer (does it?), the introduction of 3 to 15 year oceanic buffering (9±6) wouldn’t make up the 19 year difference to that of 20±8 year lag but it would go some way towards it.
00
Stephen Wilde
June 23, 2014 at 7:41 am
I would expect to see upward temperature stepping at 30 year intervals
That would make 30 years an integral part of the strange attractor. Ocean thermal resonance would do that. The strange attractor seems to be temporal and a function of energy balance with trip points. The climate (at least short term) has temperature preferences.
Now if all the oscillators go negative at about the same time? They will tend to lock for a time with the higher frequency oscillators falling out of sync first. Probably lower amplitude oscillators will also stay in sync longer.
What makes all this so exciting is that there is a little time to prepare. If you stock up on food the added demand will tend to increase productive capacity. We will need all the capacity we can get.
10
Yes, I agree with the 2 x Steve’s (and Bob Tisdale) that the delay occurs largely on (and above) the oceans. I also think the following near-perfect annual anticorrelation is what initiates the (causal) delay chain. It also possibly explains the levels of contrast between the NH and SH over the warm and cold periods of recent millenia. I suggest looking at the following animation then jumping or stepping backwards and forwards between July 2002 and April 2014.
http://earthobservatory.nasa.gov/GlobalMaps/view.php?d1=MY1DMM_CHLORA&d2=MYD28M
The following is also an example of what may be happening (e.g. during the current ‘pause’ in ‘AGW’):
Procedia Environmental Sciences
Volume 13, 2012, Pages 626–631
Changing trends and relationship between global ocean chlorophyll and sea surface temperature
J.F. Feng and L. Zhu
With the MODIS Aqua satellite-derived ocean chlorophyll concentration and sea surface temperature(SST) time series data, recent changing trends were evaluated by linear regression analysis. Furthermore, the time frequency space and regions in time frequency space where the chlorophyll and SST time series show high common power and phase angle were explored and found with Continuous wavelet transform (CWT) and Cross wavelet transform(XWT) methods. Results showed that during 2003-2009, the global ocean chlorophyll concentration is increasing while the SST is decreasing. There is a stable one year frequency in the global ocean chlorophyll for the whole time period but a half year frequency only in 2007-2009. To the SST, there are a stable one year and half year frequency during 2003-2009. There is a stable one and an unstable 0.5 year resonance frequency between global ocean chlorophyll and SST. The phase angle between global ocean chlorophyll and SST increased from 130° to 180 °for one year resonance frequency and phase relationship changed from in-phase to anti-phase for 0.5 year frequency after 2006.
00
Analysis of the “fast” response (model NOT coupled with an ocean model):
‘Sensitivity of the surface temperature to changes in total solar irradiance calculated with the WRF model’
Carolina Cipagaut, Blanca Mendoza and Jorge Zavala-Hidalgo (2014)
Results
The monthly mean surface temperature
differences (D_HR, D_LR, D_FMR and D_IC), from September 1983 to August 1984, for each simulation are shown in figures 2, 3 and 4. Larger changes occur in latitudes higher than 30º N. The changes are observed on the
Eastern US and Canada and Northwest Africa.
In low latitudes (5º – 10º N) the differences are smaller compared to those observed at higher latitudes. Larger positive differences occur in November and June on the Eastern US
and Canada and negative differences, also in November, occur on Northwest Africa.
In these plots, variations of surface
temperature are noticeable for each case during the year of study. However, the variations observed for D_HR, D_LR and D_FMR are of the same magnitude as those corresponding to variations in the initial conditions D_IC. Then to determine if the model is more sensitive either to TSI changes or to changes in the initial conditions, we use the technique of dimensionless parameters.
In Figures 2-4 we also see variations in
some continental zones in low latitudes.
Those changes are usually smaller than
in higher latitudes. When we calculate the monthly surface temperature averaged over the full area of study (see Figure 5a), i.e. the simulation grid, the five simulations do not
show meaningful differences; i.e. the base, HR, LR and FMR have the same monthly mean temperature over the full grid. This can be due to the fact that the zones where there are large differences are small compared to the zones where the differences are smaller.
If we compute the monthly average for each one of the five zones (A, B, C, D and E) we found larger monthly differences. For instance in zones A, C, D and E there is a difference of at least 1ºC in November and June, while in January and March the differences are smaller (see figs. 5b, 5d, 5e and 5f).
[…]
These results show that changes in the TSI modify the temperature as well as other variables. Figure 8 shows that the largest changes in these variables also occur at higher latitudes. Changes of latent heat, sensible heat, and surface pressure occur mainly in the ocean rather than in the continents.
Conclusions
[…] we conclude that changes in
TSI produce changes similar to those of a disturbance in the initial conditions, but do not lead to significant changes in the conditions of the average temperature. Probably significant temperature variations will be produced if larger TSI changes occurred. To have more conclusive results concerning the sensitivity of the model to TSI variations leading to effects on weather, it will be desirable to couple the WRF model with an ocean model in order to allow the ocean heat content change under different TSI conditions. Our results are relevant for climate change experiments using the technique of downscaling [Giorgi et al, 2009] because it highlights the importance of using coupled ocean-atmosphere models to get a better temperature setting.
http://www.geofisica.unam.mx/unid_apoyo/editorial/publicaciones/investigacion/geofisica_internacional/anteriores/2014/02/4_cipaguata.pdf
Yes I know – simulations. Point being globally averaged temperature is not the place to look for the solar signal in temperature. And neither can you leave out the “slow” response of the ocean.
00
Another bit to throw into the bucket.
Earth’s magnetic field protects us from some of the charged particles and cosmic radiation. Our magnetic field is always changing and it doesn’t change evenly. For example, between January and June 2014 our magnetic field weakened over North America, strengthened over the southern Indian Ocean, stayed steady over most of Australia but weakened a bit northeast of Australia. This might (or might not) cause changes in cloud formation even without any change in solar magnetic field strength. We might get more clouds over the Americas due to a weakening of Earth’s magnetic field and less cloudiness over the Indian ocean and Asia due to strengthening there. So while solar changes could result in global climate changes on Earth, there might also be local changes within Earth that cause local changes in climate.
References:
http://www.esa.int/Our_Activities/Observing_the_Earth/Swarm/Swarm_reveals_Earth_s_changing_magnetism
http://www.nmdb.eu/?q=node/172
http://www.esa.int/Our_Activities/Observing_the_Earth/Swarm/Swarm_reveals_Earth_s_changing_magnetism
00
crosspatch
June 23, 2014 at 5:23 pm
Another bit to throw into the bucket.
Nice.
00
There are some interesting comments on the Lubos Motl blog post, in particular this written by Lubos:
“The required response function we need here *cannot* result from *any* physical mechanism because it’s not causal”.
So, has the MISTAKE made about causality come from Lubos, in prior discussions?
He talks about Fourier transforms of notches, and says they involve non-causality.
Engineers like Bernie Hutchins and myself (who have implemented such filters without needing time travel) know that to be bollaux.
David mentioned the simplest notch filter in an early post: 2 poles and 2 zeroes.
Such a filter involves only the current and 2 PREVIOUS inputs,
and the 2 PREVIOUS outputs.
I don’t think the notch has been “proven” in any way, but it is a sensible way of removing the 11-year oscillations in TSI,
and it can be implemented without violating causality.
Such a notch filter is probably too unphysical for a thermal system like the Earth, as it is related to resonant “ringing”,
(just look at its impulse response in the time domain)
but not at all unphysical for the sun’s magnetic field, which appears to “ring” with 11-year cycles.
Bottom line: talk more to engineers, less to theoretical physicists.
30
Mikky – thanks
Remind me again that I am getting old ;). I vividly recall the days when continuous-time was automatically assumed. Now is it discrete-time that comes first? Why not.
I take David’s notch to be continuous, else his frequency response (he calls it a transfer function) would be periodic, and more indicative, his step response would be discrete (with what sampling interval, I have no idea). But you and I have probably eschewed much continuous-time analysis in favor of numerical integration with tiny time increments. A hybrid case I guess.
So I agree with your notion that the second-order digital notch depends only on previous inputs/outputs. Causal and IIR. The main difference I see with the analog continuous-time case is that there, NATURE enforces the causality.
10
I think David is using digital data with a sampling interval of 1 year, as that is how the TSI data is defined
(averaged over a complete orbit of Earth).
00
Mikky –
Perhaps so. Of course the the dt in any differential equation would not be compelled to be 1 year. With fs=1/1-year, the 1/11-year notch would have its zeros at what, +/- 33 degrees. Poles = ? Bandlimited (antialiased) to 1/2-years by averaging prior to sampling? Perhaps.
Of course still causal.
Good to keep in mind.
00
Mikky –
I have looked for the full context for your Lubos quote:
“The required response function we need here *cannot* result from *any* physical mechanism because it’s not causal”.
and can’t find it at the Reference Frame. It is often hard to look at things there – at least for me – everything jumps around too much. But onward……
The implied agency that is “requiring” the response as in “required response function” is not clear. If Lubos is talking about what David is showing as a step response (required) then Lubos’ concerns may be as direct (simple-minded!) as mine.
Or is it that the mathematics (Nature Herself) “requires” non-causality, if we assume a meaning, in the sense of what most of us call a “transfer function” (suggesting actual causative path existing) for what David calls a “transfer function” but which is actually in David’s usage just the ratio of some Fourier transforms. Such a ratio could well (would?) have a negative time side, but no necessary meaning.
Actually, it seems unlikely that Lubos is the source of any mistake, although he may be unclear.
00
Is this the X factor Jo has alluded to?
http://www.esa.int/Our_Activities/Observing_the_Earth/Swarm/Swarm_reveals_Earth_s_changing_magnetism
10
In case anyone missed it in the thread layering we can now move on from my query about the sign of the cloud response to force x
David Evans said at
June 23, 2014 at 10:28 pm
“Stephen, force X acts in the same thermal direction as TSI, for instance to “produce MWP and LIA” as you point out.
My interpretation that around TSI peaks the cloud cover peaks was wrong. That’s just interpretation, and now being corrected.”
To which I responded:
“Thanks, David.
It was important to me to get that sorted out because I would very much like my conceptual New Climate Model to fit in with your numerical New Solar Climate Model.
If they can fit together then the combination could present a useful alternative to the GHG based model.
My model has often been criticised for the lack of a suitably ‘sciency’ technical underpinning which is a fair point but I think it fits observations nonetheless.”
Looking forward to the next instalment 🙂
10
Firstly, well done Jo and David for taking on this considerable research effort, outside of the well-funded climate consensus machine.
As a Chemical Engineer (particularly one with an interest in automation and control) the approach of trying to determine a transfer function that explains the relationship between a time dependent input and a time dependent output makes a lot of sense. I always thought that Electrical Engineers had it easy when they wanted to do this – they would put their “black box” on the bench an excite it with a signal generator whilst monitoring both input and output on an oscilloscope. By sweeping the signal generator over the entire range of interest and recording the results, it is a fairly mechanical task to plot the transfer function as an amplitude ratio of input to output and a phase shift from input to output, each as a function of frequency. Using appropriate log and linear scales you get the Bode plots that are bread and butter to control engineers.
Life is not that easy for Chemical Engineers. The “black box” might be a 10m high distillation column containing large quantities of valuable and dangerously flammable liquids and with a production schedule to adhere to. There is also no direct equivalent to a signal generator in this instance. You get the idea. But Chemical Engineers and Control Engineers have not just given up, because it has been shown to be a very useful to be able to have a good idea of the transfer function of a complex process that you may not fully understand. Some people questioning the utility of David’s fundamental approach are probably not aware of the extensive body of work in this area. Whether it eventually yields valuable insights for the TSI to Surface Temperature case is a different issue and we will never know unless the idea is properly followed through – more credit to Jo and David.
Does the experience of Chemical and Control Engineers help throw any extra light on what Jo and David are presenting? Hopefully yes.
As a slight aside, Control Engineers like to stimulate systems with steps or impulses and observe the responses. This often provides very useful information on the transfer function. An impulse (sadly often less of a practical option than a step change) can reasonably approximate the mathematical ideal of a “Dirac Delta” function. The mathematical consequence of this is that the Laplace transform of the response of the system (to this impulse) is the transfer function. This means that the Bode plots of the transfer function can be very easily generated by taking the Fourier Transform of the response of the system to an impulse. Step and impulse tests are still impractical in many real situations where Chemical and Control Engineers want to determine transfer functions. As a result there have been efforts to try and determine transfer functions from normal operating records of process plant equipment. This is the work that has most relevance to the approach used by Jo and David. I have not worked directly in this field for a while but I think it would be very worthwhile to get some input from specialists in this area.
A few interesting papers (from some substantial time back) are:
Gallier, P.W., Sliepcevich, C.M. and Puckett, T.H. (1961). Some practical limitations of correlation techniques in determining process frequency response. Chem. Eng. Prog. Symp. Ser. No. 36, 57, 59.
Murril, PW, RW Pike and CL Smith (1969) Frequency response data from statistical correlations, Chem. Eng. May 19 1959, pp 195 – 200.
MM11 Murril, PW, RW Pike and CL Smith (1969) Random inputs yield system transfer functions, Chem. Eng. April 7 1959, pp 151 – 154.
There are two particular issues that I think are worth raising, based on work in this area.
If you are relying on inputs to the system being simply those that happen to be available, the input signal needs to have enough variability in it to be able to test the system over the range of frequencies that are of interest. One can express this very loosely as wanting the signal to have enough “colour”. To extend this analogy, if you want to get a detailed idea of the colour of an object (ie a full colour spectrum) you cannot do this by looking at it when it is lit by a single colour (or a light with a limited colour spectrum). Hence the need to check colour under a proper white light. The equivalent in terms of time dependent signals is that you really want the input to the system to be white noise – a signal that is a combination of every possible frequency, all of equal amplitude. White noise of a sufficient amplitude (or a practical approach to it) is usually not present in practice – so there is an approach of adding noise to the input signal. A practical form of noise is pseudo random binary noise (PRBN) and this has been shown to be useful in real world situations.
Clearly we can’t go about adding PRBN to the TSI over the past few hundred years. So to the big question; “Is there sufficient colour in the TSI signal to tease out the transfer function from TSI to surface temperature?” There clearly is a strong signal with a period of around 11 years (a single colour illumination) but is there enough strength in the rest of the input spectrum? At the risk of misrepresenting someone else’s point of view, I think that it is this issue, expressed in different ways, that is ultimately troubling people like Lubos, Willis and Jonathan Jones.
Saying this again, using the colour analogy: if you illuminate an object with the narrow frequency red light from a laser and see no reflection, it is not robust to assume that the object has maximum absorbance in this region and minimal absorbance elsewhere (ie it behaves as a notch filter). Particularly if the measured light of other frequencies is just broad spectrum, low amplitude noise on your measurements of both incident and reflected light, because when this noise is analysed as signals, it will tell you that the object is equally reflective at all other frequencies other than the red laser light. In reality the object may be completely black or some specific colour other than red, or you could be right about the deep absorption band at that red frequency – you just wouldn’t know. I am not in a position to make a judgement on Jo and David’s work in this regard, but I think this is an aspect that needs careful analysis and consideration.
Even if one cannot rationally deduce a “notch filter” from the available data, it does not mean that the idea does not have merit – particularly if it turns out to have predictive power that turns this into a falsifiable theory! I really think we need to stick with Jo and David on this one and see where it leads.
A second aspect of the “Chemical Engineering” approach to determining the transfer function from normal operating records is based on an interesting mathematical relationship. If you determine the auto correlation of the input signal (A(t)) and the cross correlation between the input signal and the output signal (C(t)) these two new signals are related by exactly the same transfer function that relates the input signal to the output signal. Roughly speaking it enables one to take long records of input and output signals and “average” them down to manageable lengths that cover the frequencies of interest before applying Fourier Transforms to calculate the transfer function. Unfortunately the records for TSI and surface temperature are not that long, but this approach may still have some value (perhaps in terms of handling noise) when applied to this particular problem. I think it is worth a try.
As a last comment, and a personal aside, I have never come across anything approaching a “notch filter” in my experiences as a process/chemical engineer. My familiarity with the concept comes from when I was about 13 years old and my brother and I pooled money accumulated from Christmas gifts over a number of years and bought a second hand communications receiver. It was a Halicrafters SX 111 designed for radio amateurs (Hams) with the narrow ham bands spread beautifully across the wide dial. It had a BFO for resolving SSB signals. A really nifty feature (explained nicely in the manual) was the tuneable notch filter that could be used to neutralise a signal that was almost on top of the one you wanted to monitor. Who knew that I would be discussing notch filters in the current context so many years later?!
00
+10
I’m a retired Chemical Engineer with electronics, audio, as hobby, I too wish I could have hooked up a signal generator and a spectrum analyzer to many process plants. I fully agree with sugar daddys comments, I’ve lost count of how many times I had to grab a newly minted engineer by the nether regions to stop them disappearing down a rabbit hole before comprehending the “big picture” of an operating units input/output relationships. One comment, need to consider not just white noise but also pink noise.
I’m enjoying the “reveal” by David and will wait until the total picture is revealed before forming an opinion.
00
David,
I think there is too much speculation going on, based on false conclusions from your data analysis:
1. The data analysis does not establish a notch, just an absence of 11-year signal.
You would have to see a dip in the SPECTRUM of the temperature data to establish a notch, and there is no such dip?
Absence of 11-year signal could probably have mundane explanations, for example averaging due to thermal inertia.
2. A notch filter does not require a time delay, unless you choose to implement it non-causally, which is not necessary.
To me, the notch filter is just the method you have used to remove the 11-year signal from the TSI data, leaving the long-term variations.
You could have used instead a flat 11-year average, which I suspect would give similar results.
Assuming that there is a good correlation (with WOW factor?) between temperature and notched-and-scaled TSI data,
then you can determine (roughly) the time delay between those 2 time series,
giving you interesting scientific results without any theoretical speculations, and something with predictive power.
The reason(s) for the absence of 11-year signals could be left as speculations, rather than proposing a specific notch filter that some are finding hard to accept.
01
Mikky. A reduction of amplitude can be caused by a notch with lower attenuation than the signal. EG: signal = 0.8 Notch = 0.6 You may be used to looking at log vertical scales. The attenuation here is not so large. I suspect the low pass filter has a longer time constant if there are not three seperate filters (Air, Sea and land) with more combined attenuation. That would be part of the reason but you should look at what happens when you add the effects in Fig 1a and d here. It looks very like what David got.
http://www.sensorsmag.com/sensors/electric-magnetic/an-introduction-analog-filters-1023
As for noise and the LPF being the sole reasons, what about distortion as well? What is the source of the noise? If the noise is real and not data errors then the notch is real. If the reason is fudgusted data ENHansen then this may prove that has been commited.
00
Micky said:
“To me, the notch filter is just the method you have used to remove the 11-year signal from the TSI data, leaving the long-term variations.
You could have used instead a flat 11-year average, which I suspect would give similar results.”
That seems to be a good point.
Since there is no significant thermal response from the 11 year cycle something causes the 11 year variations in TSI to be smoothed out.
It is such a small level of variability anyway that I think it gets lost in the noise of the system and in the thermal inertia of the oceans such that a thermal response to solar variations only shows up across multiple cycles.
However, even across multiple cycles the changes in TSI are too small to produce the temperature changes that we observe so something else, force x, must be amplifying the thermal effects of the changes in solar activity.
Several commentators have previously suggested possibilities in other blogs so the question was being asked even before David produced his charts. It was even clear to some that albedo variations must be at the heart of it
Now, what I need to know, being not good with statistics, is whether David’s notch is representative of a real world phenomenon or just a consequence of his starting assumptions.
If one knows that TSI varies and should therefore have a thermal effect then it is easy to draw a chart with an opposing factor that cancels it out which is what David seems to have done.
Does that help to identify the nature of force x in any way ?
Does that ‘prove’ the existence of force x bearing in mind that simple logic applied to observations already suggests that there is a force x independent of TSI that has an influence on climate.
I suppose I am asking what is new or helpful in the New Solar Climate Model presented here if one still has to ascertain the nature of force x.
One helpful feature has come to light in that the timing of the various parametrs of solar behaviour taken together with the sign of the albedo response suggest that cosmic rays are not part of force x as proposed by Svensmark but I’ll leave David to expand on that if he deems it appropriate.
00
Answering my own question I note that the notch is only one part of a greater whole.
That greater whole presents a simple and elegent way of restarting from first principles so I think that it is a useful base from which to ask all the important questions about the nature of force x.
Even without knowing what force x is it should be possible to predict temperature trends in broad terms from solar variability.
The proof of force x will come not from the drawing of the notch but from superior predictive skill as compared to the GCMs.
Ascertaininmg the nature of force x should be possible from examining all the features of climate that do show changes of trend in tandem with solar changes and there are a lot as I have pointed out elsewhere.
The period around 2000 was a time when declining solar activity put a stop to the earlier warming trend.
A host of climate phenomena all changed trend at the same time but I’ll leave a discussion of them to a later stage in the process.
20
“The period around 2000 was a time when declining solar activity put a stop to the earlier warming trend.”
Maybe. Debatable whether it was the solar activity. Can also say that the period around 2000 (and up to at least 2004) was a time when increasing cloud cover/albedo put a stop to the earlier warming period.
R. T. Pinker, B. Zhang, E. G. Dutton [5/6/2005]
Long-term variations in solar radiation at Earth’s surface (S) can affect our climate, the hydrological cycle, plant photosynthesis, and solar power. Sustained decreases in S have been widely reported from about the year 1960 to 1990. Here we present an estimate of global temporal variations in S by using the longest available satellite record. We observed an overall increase in S from 1983 to 2001 at a rate of 0.16 watts per square meter (0.10%) per year; this change is a combination of a decrease until about 1990, followed by a sustained increase. The global-scale findings are consistent with recent independent satellite observations but differ in sign and magnitude from previously reported ground observations. Unlike ground stations, satellites can uniformly sample the entire globe. [Abstract]
So, according to Pinker et al., 2005, surface solar irradiance increased by an average 0.16 W/m^2/year over the 18 year period 1983 – 2001 or 2.9 W/m^2 over the entire period.
This change in surface solar irradiance over 1983 – 2001 is almost exactly 1.2% of the mean total surface solar irradiance of the 2000 – 2004 CERES period of 239.6 W/m^2 for which the mean Bond albedo has been claimed to be 0.298 and mean surface albedo to be 0.067 (Trenberth, Fasullo and Kiehl, 2009).
The ISCCP/GISS/NASA record for satellite-based cloud cover determinations suggests a mean global cloud cover over the 2000 – 2004 CERES period of about 65.6% and over the entire 1983 – 2008 27-year period a mean of about 66.4±1.5% (±1 sigma).
ISCCP/FD and Earthshine albedo data for the 2000 – 2004 period enables a crude estimation of the relationship between albedo and total cloud cover and it is best described by the simple relationship:
Bond albedo (A) ~ 0.353C + 0.067 where C = cloud cover. The 0.067 term represents the surface SW reflection (albedo). For example, for all of 2000 – 2004; A = 0.298 = 0.353 x 0.654 + 0.067
According to ISCCP/GISS/NASA mean global cloud cover declined from about 0.677 (67.7%) in 1983 to about 0.649 (64.9%) in 2001 or a decline of 0.028 (2.8%).
This means that in 1983; A ~ 0.353 x 0.677 + 0.067 ~ 0.305
and in 2001; A = 0.353 x 0.649 + 0.067 ~ 0.296
Thus in 1983; 1 – A ~ 1 – 0.305 ~ 0.695
and in 2001; 1 – A ~ 1 – 0.296 ~ 0.704
Therefore, between 1983 and 2001, the known reduction in the Earth’s albedo A as measured by ISCCP/GISS/NASA should have increased total surface solar irradiance by ~ 200 x [(0.704 – 0.695)/(0.704 + 0.695)]% = 200 x (0.009/1.399)% = 1.3%
This estimate of ~1.3% increase in solar irradiance from cloud cover reduction over the 18 year period 1983 – 2001 is very close to the 1.2% increase in solar irradiance measured by Pinker et al (2005) for the same period.
The period 1983 – 2001 was a period of claimed significant global (surface) warming. However, within the likely precision of the available data for the above exercise – probably of the order of say ±0.5% at ± 2 sigma, it may also be concluded that it is easily possible that the finding of Pinker et al (2005) regarding the increase in surface solar irradiance over that period was due to an almost exactly functionally equivalent decrease in Earth’s Bond albedo resulting from mean global cloud cover reduction.
10
TSI peaked in 1986. This signalled that force X would peak in about 1997, but otherwise the TSI changes had hardly any effect on temperature. Force X duly peaked 11 years later, in about 1997. Force X acts immediately. Force X stopped increasing in 1997, so the planet stopped warming in about 1997. Force X acts on albedo, so the nd solar theory finds confirmation in the observation that albedo bottomed in about 1997.
00
David #
>”TSI peaked in 1986″
As determined by minima – not maxima. See List of Solar Cycles:
http://en.wikipedia.org/wiki/List_of_solar_cycles
Using the sunspot proxy, SC 19 exhibited the most activity of the Grand Maximum (201.3, Mar 1958 vs 164.5 and 158.5 either side of 1986 minimum).
>”This signalled that force X would peak in about 1997″
There was no solar signal in 1986 except that it was the peak of the bicentennial trend as determined by minima. This graph demonstrates the relative activity in the SCs 19, 20, 21, 22, and 23:
http://www.sidc.be/images/wolfmms.png
How are you able to infer a TSI “peak” in 1986 from this?
00
Should be:
David #[86.1.1.1]
00
Lockwood and Froehlich 2007 established the TSI peak in about 1986. See their paper for method.
http://www.atmos.washington.edu/2009Q1/111/Readings/Lockwood2007_Recent_oppositely_directed_trends.pdf
00
>”Lockwood and Froehlich 2007 established the TSI peak in about 1986″
1366.03 W/m2 approx by smoothing, not actual peak and not before PMOD. This is little different to the bicentennial trend being described by a line tracing minima.
L&F07 page 5 (my emphasis):
Page 3 wrt to Figure 3:
Figure 1, page 5 shows actual TSI peak concentrated around 1366.5 in SCs 21 and 22 either side of the 1986 minima and up to about 1367.2. But that’s only PMOD from 1976. SSNs indicate considerably higher activity in SC 19 1958.
Sure enough, see Figure 4, page 7: Centennial variations revealed by running means over the solar cycle length L since 1890 of (a) the sunspot number, R;(c) the open solar flux FS from geomagnetic activity data (available from WDC-C1, Chilton, UK).
FS is higher around 1958 than 1987.
Now see Page 11:
In other words, there were in fact two activity peaks, the former (1958ish) at a higher level than the latter (1987).
00
Stephen #86.1
>”The period around 2000 was a time when declining solar activity put a stop to the earlier warming trend”
Also when PDO/IPO was going negative again:
http://upload.wikimedia.org/wikipedia/commons/thumb/0/09/PDO.svg/640px-PDO.svg.png
PMOD
http://www.woodfortrees.org/plot/pmod
Don’t want to dispute solar “decline” from 2000 too much (I see what you see) but I see the decline more from 2005 than from 2000.
10
Also re Stephen #86.1
Using minima (the bicentennial trend), TSI decline started from SC 22, not 23.
http://nextgrandminimum.files.wordpress.com/2012/11/figure-2-tsi-variations.png?w=693
10
Timings are not precise due to so many interacting parameters.
It was around 1998/2000 that the decline in solar activity started to show up in more meridional jet stream tracks followed by a change in trend of global cloudiness.
The negative PDO/IPO was also a factor but I say that responds to solar variations too.
00
>”It was around 1998/2000 that the decline in solar activity started to show up in……”
I was seeing something else (see below).
I take from my emphasis that you consider the observed start of the decline of actual TSI (not some perceived effects) to have started prior to SC 23.
SC 22?
List of solar cycles
http://en.wikipedia.org/wiki/List_of_solar_cycles
I contend that the TSI decline from SC 21/22 1986 was negligible until 2005 (see PMOD #86.1.2).
Even to end of SC 23 Jan 2008 there’s negligible difference between SC 23 and SC 22 average (1365.82 vs 1365.99 W/m2, Abdussamatov).
I do however see a change after the SC 23 2001 peak in the way the power was delivered i.e. the concentration of pulses seems less intense even though the average power was all but the same.
00
Richard,
I agree with all that.
At present I am dealing with the generalities in order to anticipate trend changes. I think David’s solar model is proceeding similarly.
The detail will follow confirmation of the correct diagnosis.
00
>”I contend that the TSI decline from SC 21/22 1986 was negligible until 2005″
Change “seen” at surface obviously not negligible as per Steve at #86.1.1.
00
“Having discovered the pattern in the way TSI is tranformed into temperature (sic), David builds the model with the filters to produce the same “transfer function” as he found in empirical data.”
But isn’t the problem here that the frequencies of surface temperature responses, in particular the absence of a temperature response (anticorrelation) to TSI around a frequency of 11 years were built up by examination of only a small number of ‘modern’ literature ‘records’ of TSI?
What about the response/non-response frequency information from the ensemble (sigh) of all other TSI records (inclusive of modelled elements or not)e.g.
Modern instrumented period:
Stott et al. (2003) Do models underestimate the solar contribution to recent climate change? J. Clim. 16, 4079-4093
Benestad and Schmidt. (2009) Solar trends and global warming. J. Geophys. Res. 114
Historical – high inferred surface temperature dependence on TSI:
Shapiro et al. (2011) A new approach to to the long-term reconstruction of the solar irradiance leads to large historical solar forcing. Astron. Astrophysics 529, A67
Historical – inferred moderate surface temperature dependence on TSI:
Steinhilber et al. Total solar irradiance during the Holocene. (2009) Geophys. Res. Lett. 36, L19704
Historical – inferred minimal surface temperature dependence on TSI:
Wang et al. (2005) Modeling the Sun’s magnetic field and irradiance since 1713. Astophys. J. 625 522 – 538
This list goes on (and on and on). Notice these are all papers published in the last 10 years – and I still haven’t yet listed the Svalgaard et al. ‘its all damn flat for TSI’ stuff (most of which appears at conferences and colloquia….with all that that means in terms of the strength and affiliations of review or not…).
So, my conclusion is perhaps that the frequency curve maybe should really have been made up of the ensemble (sigh) of all inferred responses? That’s a question please note.
Relevant to also note that back in 2010, using the Wang et al (2005) TSI – temperature record for the last 400 years), and another means of frequency analysis Glassman could not find any trend at all associated any notch and delay around 11 – 22 years. I know that Glassman had also looked at the earlier Lean et al papers around 2000.
This also leads into even more radical recent stuff like:
Van Hateren (2012) A fractal climate response can simulate global average temperature trends of the modern era and and the past millenium. Clim. Dynam. 40, 2651 – 2670
and
Schurer et al (2014) Small influence of solar variability on climate over the past millenium. Nature Geoscience. 7, February 2014 (published online 22 December 2013)
Are we just dancing around on the head of a pin here?
Notice I am carefully avoiding the dreaded and IMHO naive ch**** p****** slur here.
00
And yet one more bit for the bucket. Leif Svalgaard at Stanford University has discovered an inconsistency in the sunspot record that is generally agreed at this point needs correction. Many people are operating on sunspot counts from the old record. It is my opinion that one should try to consider the corrected record in any discussion of sunspots over the observational records. Prior to the observational period, we must use proxies like 14C and 10Be
Here is a link to the paper.
http://journals.cambridge.org/download.php?file=%2FIAU%2FIAU7_S286%2FS1743921312004590a.pdf&code=bcf767f207b546e33f42f347c677729f
20
Thanks crosspatch
Yes, love him or hate him, Leif Svalgaard is surely the guru of the Sun. At least he has the decency to post frequently at WUWT and inject the periodic piece of cold hard solar science. The immature words ;’denier’ or ‘denialist’ never sully his posts either. Like Nick Stokes we ignore their decent objectivity at our peril. One thing you have to hand to the Stanford crowd (and I have several close buddies who are Stanford alumni) they seem to manage to avoid the ‘mutual masturbation little boys clubs’ (clustered of course around one or more titular ‘gods in their own minds’) which characterize many of the State universities (and a goodly number of UK red brickers as well). You know what I mean – e.g. the ‘hide the decline’ crowd.
What a more precise set of clues around the notch and delay filter we might we have seen if David had run Svalgaard’s corrected TSI record of the last 300-odd years. At the very least; the Dalton Minimum is sitting therein as a nice little precursor of our current position (methinks) replete with global thermometer records courtesy the Royal Navy etc. (think ‘Master and Commander’).
00
You would think that working out the heat of a small highly reflective and stable orbit moon would be easier. Nope. Saturn’s moon Enceladus defies most theories.
https://www.youtube.com/watch?v=gTS0Vv3yS6U
10
Part two.
http://youtu.be/ia3_VsEAvk8
Part one with adifferent link.
http://youtu.be/gTS0Vv3yS6U
10
Time for the next episode !:-)
20
Ref: Figure 3: The transfer function of the notch-delay solar model, I have a query:
The red line, ‘amplitude’ displays classic ‘forced’ notch filter characteristics ie smooth response till near the notch frequency, then a great reduction in the amplitude etc.
As this model is representing a physical reality, where we are considering the sun – earth mass, inertia, distance plus force X, would the notch be so pronounced?
Could it be a gentler notch with the amplitude and phase responding more like the feedback within an Wein bridge circuit?
Where the amplitude varies smoothly and the phase alters explicitly around the notch frequency?
Just a thought!
SR
00