I didn’t vet this before posting and have no idea as to its real strengths or weaknesses. Have at it.~ctm
J. KAUPPINEN AND P. MALMI
Abstract. In this paper we will prove that GCM-models used in IPCC report AR5 fail to calculate the influences of the low cloud cover changes on the global temperature. That is why those models give a very small natural temperature change leaving a very large change for the contribution of the green house gases in the observed temperature. This is the reason why IPCC has to use a very large sensitivity to compensate a too small natural component. Further they have to leave out the strong negative feedback due to the clouds in order to magnify the sensitivity. In addition, this paper proves that the changes in the low cloud cover fraction practically control the global temperature.
1. Introduction
The climate sensitivity has an extremely large uncertainty in the scientific literature. The smallest values estimated are very close to zero while the highest ones are even 9 degrees Celsius for a doubling of CO2. The majority of the papers are using theoretical general circulation models (GCM) for the estimation. These models give very big sensitivities with a very large uncertainty range. Typically sensitivity values are between 2–5 degrees. IPCC uses these papers to estimate the global temperature anomalies and the climate sensitivity. However, there are a lot of papers, where sensitivities lower than one degree are estimated without using GCM. The basic problem is still a missing experimental evidence of the climate sensitivity. One of the authors (JK) worked as an expert reviewer of IPCC AR5 report. One of his comments concerned the missing experimental evidence for the very large sensitivity presented in the report [1]. As a response to the comment IPCC claims that an observational evidence exists for example in Technical Summary of the report. In this paper we will study the case carefully.
2. Low cloud cover controls practically the global temperature
The basic task is to divide the observed global temperature anomaly into two parts: the natural component and the part due to the green house gases. In order to study the response we have to re-present Figure TS.12 from Technical Summary of IPCC AR5 report (1). This figure is Figure 1. Here we highlight the subfigure “Land and ocean surface” in Figure 1. Only the black curve is an observed temperature anomaly in that figure. The red and blue envelopes are computed using climate models. We do not consider computational results as experimental evidence. Especially the results obtained by climate models are questionable because the results are conflicting with each other.
In Figure 2 we see the observed global temperature anomaly (red) and global low cloud cover changes (blue). These experimental observations indicate that 1 % increase of the low cloud cover fraction decreases the temperature by 0.11°C. This number is in very good agreement with the theory given in the papers [3, 2, 4]. Using this result we are able to present the natural temperature anomaly by multiplying the changes of the low cloud cover by −0.11°C/%. This natural contribution (blue) is shown in Figure 3 superimposed on the observed temperature anomaly (red). As we can see there is no room for the contribution of greenhouse gases i.e. anthropogenic forcing within this experimental accuracy. Even though the monthly temperature anomaly is very noisy it is easy to notice a couple of decreasing periods in the increasing trend of the temperature. This behavior cannot be explained by the monotonically increasing concentration of CO2 and it seems to be far beyond the accuracy of the climate models.
![Figure 2. [2] Global temperature anomaly (red) and the global low cloud cover changes (blue) according to the observations. The anomalies are between summer 1983 and summer 2008. The time resolution of the data is one month, but the seasonal signal is removed. Zero corresponds about 15°C for the temperature and 26 % for the low cloud cover.](https://i0.wp.com/wattsupwiththat.com/wp-content/uploads/2019/07/Screenshot-2019-07-11-21.32.34.png?resize=634%2C488&quality=75&ssl=1 "Screenshot 2019-07-11 21.32.34")
Nethertheless, way less funny than the climate clownery of the IPCC GCM daft models :
https://www.youtube.com/watch?v=THg6vGGRpvA
Link from a comment by ‘Gator’ at https://realclimatescience.com/2019/07/climate-scam-collapse-continues/#comment-234479
“NO EXPERIMENTAL EVIDENCE”.
You’re telling all the false predictions from the broken computer models, even with rigged data, don’t count?
Not if they don’t want them too, it seems.
Walt D.
Computer models don´t count, because they are not evidence at all. There is no evidence in any alarmist blaablaa at all. That´s why we have always some ten years left. We are in a hurry, because ocean´s oscillations are going to give proof very soon. Cool is coming.
A few tens of thousands of contrarians got the news this morning and are savoring it like a fine cup of coffee:
https://www.zerohedge.com/news/2019-07-11/scientists-finland-japan-man-made-climate-change-doesnt-exist-practice
The story also mentions a paper on the ‘umbrella effect’ which links cosmic rays to cloud cover. I think it might be this one. For what it’s worth, the paper is found at nature.com .
That’s an interesting study. The greater a proportion that is ascribed to one cause leaves less to ascribed to another.
So, would alarmists react to this study as being good news? If not, why not.
Does somebody know where they take the “observed” data from?
Joe Bastardi linked to the ZeroHedge article on his Twitter feed. I wonder if Judith Curry has seen it?
The paper seems unaware that IPCC does in fact factor clouds, in the form of aerosols, into climate models. The (negative) forcing may be overestimated though, which results in estimates of ECS that are too high, but not non-existent.
https://judithcurry.com/2018/03/11/recent-research-on-aerosol-forcing-of-the-cmip5-models/
A proper heat balance conducted in kJ/h. W = 3.6 kJ/h
As far as the Incoming Solar Energy is concerned the earth is a cross sectional disc.
1,368 W/m^2 * 3.6 kJ/h / W * PI()*6,371,000^2 m^2 = 1.275 E14 kJ/h = ISE
Energy enters only on the lit hemisphere but leaves the entire spherical surface 24/7 per Q = 1/R * A * dT. Because the lit side dT is larger than the dark side dT more energy leaves the lit side than leaves the dark side.
So:
ISE * (1-α) = In ASE = out OLE (65% * ASE lit + 35% * ASE dark)
dT and the surface temperature is controlled by 1) the albedo and 2) R = the atmosphere’s thermal resistance.
R = among several other factors including low cloud cover.
Nick Schroeder
You said, “As far as the Incoming Solar Energy is concerned the earth is a cross sectional disc.” That is a simplification because about 71% of the surface is water and reflects specularly, not diffusely. That means the total reflectivity is not uniform, but varies with the angle of incidence. Also, atmospheric absorption and scattering is controlled by the slant range, which varies with the angular distance from the center. Earth doesn’t behave like a flat disk.
https://wattsupwiththat.com/2016/09/12/why-albedo-is-the-wrong-measure-of-reflectivity-for-modeling-climate/
For the purposes of calculating kJ in – it is a disc.
For the purposes of calculating kJ out – it is a sphere.
What happens inside those system boundaries is a wash.
Nick Schroeder
You said, “For the purposes of calculating kJ in – it is a disc.” Not so! That assumes that the absorption is only a function of the kind of material at any point on the ‘disk.’ As I point out, the reflectivity (1-absorptivity) is a function of the angle of incidence for water. On the limbs of a spherical water body the reflectivity approaches unity. Therefore your disk calculation will provide an estimate of kJ in that is too high.
Clyde, you’re confusing what enters the system (kJ in) with what is absorbed at the surface. The amount of sunlight intercepted at the TOA (kJ in) is absolutely a disc.
The amount absorbed at the surface, on the other hand, is a complex function of a number of factors including solar angle, surface albedo, cloudiness, and atmospheric absorption.
Hope this clears it up,
w.
Don’t forget heat is retained by the surface, heat capacity. Energy loss back to atmosphere/space at variable rates. No infrared penetrates more than a few micrometres into water, so 71% of this down-welling radiation heats top skin of ocean & its energy overwhelmingly goes to latent heat, which causes increase in evaporative cooling of oceans. So only sunlight can warm oceans. Downwelling IR radiation does not.
The simple model you are miss-sold by establishment quickly gets nice and complex.
Didn’t Willis do a piece on tropical thunderstorms wherein hotter meant earlier thus cooler earlier? I must find my Wanchai Burberry and cycle round Singapore dockyard once more- aye de me.
Yeah. Modeling the actual effects of clouds is apparently beyond the GCM’s to calculate from basic principals. Willis Eschenbach does point out the cooling effects of thunderstorms, but nighttime cloud cover would be warming. A rather complex feedback.
Why is that I think this is the end of end, the ultimate proof man has nothing to do with the climate? Because I was waiting for something like this, sensing the perfect storm was coming, first Svenmark, then Valentina and now this. It is over.
The primary negative feedback is the AMO, which is warmer when the solar wind is weaker, directly via negative NAO/AO, and lagged through negative NAO associated El Nino conditions. That drives the decrease in low cloud cover and the increase in lower troposphere water vapour. In the UK the annual increase in sunshine hours has been around 8% since the mid 1990’s.
Ulric
That’s interesting. it would be even more interesting to know the sunshine hours back to the start of CET. I have the wind data from Hubert Lamb but I am not sure how you would ever calculate old sunshine records
tonyb
It looks quite thin indeed.
My apologies for not being more discriminating. I was fooled by the Cornell University flag.
As for the CBC, there is plenty of instances where their “information” was incomplete and biased. Until next time…
CONs being CONNed. Don’t you love it.
“I haven’t vetted this but,….” It makes for a good headline so print it and spread it to all the other CONservative sites. TRUMP’S CHUMPS will love it!
“The basic problem is still a missing experimental evidence of the climate sensitivity.”
How will there ever be experimental evidence? You might as well ask for experimental evidence that the sun is powered by nuclear reactions. Large-scale climate effects are just not something that can be replicated experimentally. All we are ever going to have are estimations and models. It’s important to recognize the impact this has on strength of evidence, but complaining that experimentation doesn’t exist is a weak ass cop-out.
One of the implications of this is that Decision-making Under Uncertainty needs to become a core competency of anyone making recommendations on what course of action we should take. Tragically, many of those calling for action don’t even recognize this an area of expertise that exists. Thus they at the functional level of a child playing with blocks to spell three letter words. They’ve never heard of decision trees, much less created one. To them, the do-nothing scenario isn’t a base case upon which to measure actions, but an evil thing that good, right-thinking people ignore.
Has this paper by Holmes been discussed here?
http://article.sciencepublishinggroup.com/pdf/10.11648.j.earth.20170606.18.pdf
It’s peer reviewed and comes to a similar conclusion to Nikolov and Zeller. Basically atmospheric composition has little or no effect on temperature (leaving aside condensing gases such as water vapour ).
This paper is simple to understand and makes some very interesting and important points and is well worth reading in my opinion.
Huh? Near as I can tell, all that they’ve done is to prove that the Ideal Gas Law (PV = NRT) actually works. That is to say, if you know P, V, N and R at the surface of any planet, you can calculate T.
Be still, my beating heart …
I’m sorry, but that’s a whole lot less than impressive.
w.
“Has this paper by Holmes been discussed here?”
Yes, it has, and by Willis, no less. Correctly tagged by WUWT as Bad Science.
1978-1987 is the coldest ten-year period in Denmark in a hundred years. It was also the period with the least sunhine (1374 hours sunshine, average temperature 7,3).The following twenty years the sunshine went up 25% and the temperature 1.7 oC making 2000-2009 the warmest period ever (1723 hours of sunshine and 9,0 degree Celcius). High values were also found in 1930-1939 (1325 h and 8,2 oC).
I’ve been waiting for papers like this for decades. I remember in the early days when the models were running hot, they would increase the aerosol levels to try to bring the temperatures down. Unfortunately, the aerosol levels that were needed were way beyond any normal level. It was clear then that the idea of CO2 being the climate control knob was just crazy.
All Brits know that it can be very hot in the UK when there are blue skies and the sun is beating down. But when a passing cloud gets in the way it can turn quite cold in seconds. I always find it astonishing when so called climate scientists dismiss clouds as being of limited importance. Their inability to model them for reasons of model resolution and lack of understanding should not mean that nature’s temperature control knob is effectively ignored. Why is climate science such an amateurish mess that seems to defy common sense, logic and observation?
Because it has nothing to do with what is factual.
From the article: “The climate sensitivity has an extremely large uncertainty in the scientific literature. The smallest values estimated are very close to zero while the highest ones are even 9 degrees Celsius for a doubling of CO2.”
This is what is described as “settled science” in some quarters.
Of course, the science is not close to being settled, as should be obvious, and we shouldn’t be making Trillion-dollar decisions based on this level of uncertainty.
Does anyone know if the authors actually exist? The observation that sensitivity is hard to estimate seems to be accurate though. If there is anything to it this merits a proper paper.
I think they are physics professors at the University of Turku in Finland.
Proving something (positive manner) is extremely rare if not impossible.
Disproving a hypothesis is common and routine and how science moves forward.
Finding evidence to support a hypothesis is routine, but other explanations not thought of at the time are the scythe of truth in a dry wheat field of ideas, most of which are wrong and need to be cut down before they go to seed.
Maths have proofs.
So proofs in science are almost (or entirely) nonexistent.
I’m with Mosh on this one. No code, no data, no science. He doesn’t mention the source of his “low cloud” data anywhere. Nor does he define what he’s calling “low clouds”—how low, what is the threshold for inclusion/exclusion.
Next, he seems to assume that there is a cause and effect going on. However, climate is rarely that simple. For example, as he says, the temperature is often a function of the clouds. But particularly in the tropics, the clouds are a function of temperature. I call this a “chain of effects”, in which we cannot distinguish cause from effect.
This can be explored using the math of what is called “Granger causation”.
X is said to “Granger-cause” Y if and only if having information about X allows us to improve our prediction of Y. Here’s the interesting part. There are three possible outcomes, viz:
1) X Granger-causes Y
2) Y Granger-causes X
3) X Granger-causes Y AND Y Granger-causes X
In the case of clouds and temperature, it turns out that it’s 3) — clouds Granger-cause temperature, AND temperature Granger-causes clouds.
Finally, even the relationship between cloud area and temperature is far from simple. Here is the correlation of the two variables:
As you can see, in some areas when temperatures go up, clouds go down, and in other areas when temperatures increase, clouds increase as well.
All of which, I fear, renders his claims less than viable …
Best to all,
w.
Clouds can act as a parasol, reflecting heat and as a blanket, retaining heat. Typically, the former would take place during the day and the latter in the evening. I think that this challenges your final point. The cloud effect is not simple.
Altitude also matters.
Net effect of clouds since 1952 appears to be cooling:
http://scrippsscholars.ucsd.edu/jnorris/files/caltechweb.pdf
Might I suggest you have a look at the relationship between temperature and cloud in 2016 during the hottest year of the El Nino? Better still get in touch.
Your global net negative result validates the claim.
Ulrich, a correlation of 0.02, whether positive or negative, doesn’t “validate” anything. That’s lost in the noise.
w.
So what percentage of cloud change is that?
Do me a favour and write my name as it is written.
Ulric Lyons July 14, 2019 at 7:33 am
I have no clue what that means. What I measured is a correlation, not a measure of change.
My apologies. You are the only person I’ve ever known with your name, so it’s not familiar to me.
w.
I’m not satisfied that is ‘lost in the noise’. Plus the regional inverse responses are actually ideal for greater mutual temperature control in the same direction. With the negative correlations strongest in the Horse Latitudes where there is less water vapour, and a positive correlation in the Arctic.
Ulric, if you truly think that a correlation of 0.02 is somehow significant, you need a new line of work …
w.
0.02 is very weak, but it’s not zero. The regional responses are much more important than the global average. Low cloud needs to decrease in the mid latitudes and increase in the Arctic to achieve a greater temperature rise for the hemisphere. In both cases the change in cloud cover drives the temperature change. The tropics are the exception. There are also opposing changes between low and high cloud cover regionally to consider. I like this line of work.
Ulric Lyons July 14, 2019 at 5:48 pm.
So are correlations of 0.002, 0.0002, 0.00002, and 0.000002 … all are very weak but not zero.
So what?
w.
You are right the 0.02 average is not very significant. But the mid latitude and Arctic correlations are significant, and to average them is meaningless because of the opposite effect of low clouds in the two regions.
3) X Granger-causes Y AND Y Granger-causes X
This sounds like part of a recipe for chaotic emergent pattern formation, of some kind.
Something like a Turing reaction.
Or the Belousov-Zhabotinsky (repeatedly self-reversing) reaction.
Something like real climate and real life.
Nothing at all like any GCM climate model, somewhere over the rainbow in linear-land.
You have not defined low clouds there either Willis.
Ulric, that was simply a correlation between cloud area and temperature. Since I didn’t use “low clouds”, I didn’t have to define “low clouds”.
w.
High cloud and low cloud have different effects on surface temperatures.
True. What’s your point?
w.
Well the case of a reduction of low clouds and an increase in high clouds both driving surface warming.
Willis,
I would think that the correlation between clouds and temperature would vary almost minute by minute, even reversing at times, and highly dependent on the previous state of temperature, humidity, cloud cover, etc. I remember one afternoon at my cabin, at 7,000ft altitude and about 15 miles east of the Colorado continental divide, the sky was bright blue and cloudless. I was sitting in the backyard looking westward when I noticed a single little puff of a cloud developing over the divide. Soon, it started to slowly move eastward toward me. Once it got over my house around 20 minutes later it was probably a quarter mile in diameter and moving eastward at a fairly fast clip. By the time it had moved another 15 miles eastward it had developed into a large cumulonimbus and was dropping 6 inched of golf ball size hail on a friends house out in Louisville. Within another hour or so it had moved out of site and there was maybe 20-25% cloud cover.
It has to be a hoax as delta cloud cover is expressed in % and delta temperature in K. No scaling is applied so it would be all too much of a coincidence if the graphs match. Sort of see if we can get fake science noticed, and they did.
Jyrki Kauppinen’s research while affiliated with University of Turku and other places
I’m coming to the conclusion that our average global temperature is a function of the atmospheric mass via auto compression. In addition to that, solar heating of the oceans provides a heat reservoir that takes years, decades or even centuries to work its way through the system, depending on currents, winds, earth topography and other factors.
GHG concentration plays a small role, with some radiative warming, but photon emissions are less frequent than molecular collisions, so kinetic energy transfer enhances convection much more than heating by downwards radiation. In effect, more greenhouse gases aid cooling.
Solar modulation of GCR radiation affects cloud cover as predicted by Svensmark and this is another important influence on our climate. Cloud coverage has the potential to be the dominant effect.
There may be other, as yet not known or poorly understood factors, but these are sufficient to provide the natural variability that is the hallmark of our climate, whilst keeping the range of temperatures within narrow limits ensuring stability over centuries.
If one had a magic tap to turn down the sun’s radiation. At what % of today’s insolation would clouds no longer form? What difference to surface temperarure would that lead to? Ok, assuming CO2 level stays as it is today and again assuming CO2’s level was at preindustrial level.
97% of scientists should be able to answer this as the science is settled.
There were probably clouds when solar output was only 70% as powerful now, ie 3.3 billion years ago.
Most likely even 4.1 Ga, when oceans already existed but the sun’s radiation was around 63% of today’s strength.
It helps to look for independent analysis (completely different method to determine planetary cloudiness) and different authors to see if it supports the concept.
There are other studies that support the assertions that there has been a reduction in low level cloud cover and that the warming due to the reduction in planetary cloud cover is as much or more than the IPCC estimated/assumed CO2 forcing.
The moon reflected light study supports the satellite data conclusion that there was a reduction in low level cloud cover.
The moon refection analysis paper calculates a 7.5 W/m^2 warming would occur due to the reduction in the earth’s albedo compared to a 2.7 W/m^2 warming from a doubling of CO2. This interesting as it supports to conclusions 1) That the earth resists rather than amplifies temperature forcing changes and 2) It is possible the majority of the recent warming was caused by ‘natural’ mechanisms as opposed AGW.
This is a direct measurement of changes to the earth’s albedo by measuring changes in the light reflected off of the moon that supports the assertion that has been a change in planetary albedo and the warming due to that change would be 7.5 W/m^2.
http://bbso.njit.edu/Research/EarthShine/literature/Palle_etal_2004_ASR.pdf
The Earthshine Project: update on photometric and spectroscopic measurements
“Our simulations suggest a surface average forcing at the top of the atmosphere, coming only from changes in the albedo from 1994/1995 to 1999/2001, of 2.7 +/-1.4 W/m2 (Palle et al., 2003), while observations give 7.5 +/-2.4 W/m2. The Intergovernmental Panel on Climate Change (IPCC, 1995) argues for a comparably sized 2.4 W/m2 increase in forcing, which is attributed to greenhouse gas forcing since 1850.
Still,whether the Earth’s reflectance varies with the solar cycle is a matter of controversy, but regardless of its origin, if it were real, such a change in the net sunlight reaching the Earth would be very significant for the climate system.”