In new research (Liu et al., 2026), scientists acknowledge clouds “markedly impact near-surface air temperatures” by altering the radiation budget and driving heating rates across the Arctic. For example, a cloudy day in winter may deliver 20-30 W/m² of cloud radiative forcing (CRF), warming the Arctic surface by 6-9°C.
However, due to massive uncertainties in estimating cloud fraction (CF) and their radiative effects, “substantial inconsistencies across datasets in trend magnitude” persist, and thus “there is still no consensus on the long-term trend of Arctic total CF over recent decades.”
Systematic errors (biases) in estimating downwelling shortwave (SW) can deviate by 90-160 W/m² from one dataset to the next. Overall, datasets disagree by 20-70 W/m² when estimating downwelling SW across the Arctic, and thus our measurements are “insufficient to meet the needs of radiation energy budget estimation.”
The increase in atmospheric CO2 is alleged to be the driver of modern Arctic warming. This attribution can be no more than speculation, however. The 20-70 W/m² uncertainty in estimating the Arctic’s surface solar radiation is 100 to 350 times larger than 10 years of accumulated clear-sky-only CO2 impacts (0.2 W/m² per Feldman et al., 2015).
It is a puzzle that Kenneth Richard goes to such trouble to give an apparently authentic presentation of the paper, with highlighting etc, when in fact he has thoroughly and selectively rearranged the text. In his pastiche, the apparent first paragraph has in fact been transported from the middle of a para on p 8. The diagram that follows is from p 7. The result is that you can’t read it as the authors meant. You are forced into KR’s world of spin.
Here is how the paper actually begins. There is no doubt that the Arctic is rapidly warming, and the greenhouse effect is the cause, with feedback effects on albedo.
Do not look at conclusions -” Over the past 2 decades, total Arctic cloudiness has
shown an overall increasing trend. However, uncertainties in CF
definitions, satellite sensor systems and data processing, cloud
detection and phase-classification algorithms, and reference
datasets used for validation have led to substantial inconsistencies
across datasets in trend magnitude, seasonal dependence, and
the contributions from different vertical layers. These disagree-
ments translate into differences of approximately 20 to 70 W m−2
in surface downwelling SW. As a result, Arctic surface SW
datasets are still insufficient to meet the needs of radiation
energy budget estimation and further climatological applica-
tions. “
Right, the paper is about uncertainties in the accuracy of surface shortwave radiation estimation due to cloud variation. It does not suggest that clouds themselves are driving the heating rates.
It plainly states in its very first paragraph (a part deceptively excised by Kenneth Richard’s, as Nick points out above) that the rapid changes in the Arctic climate system are occurring, “… in response to greenhouse warming…”.
Check it for yourself!
Just consider the lengths to which these chancers are prepared go in order to deceive their readers.
If you can’t measure the factors with sufficient accuracy to properly attribute cause and effect then assuming “ that the rapid changes in the Arctic climate system are occurring, “… in response to greenhouse warming…”.” is nothing more than attribution by dogma instead of by science.
“It does not suggest that clouds themselves are driving the heating rates.”
Of course clouds are a direct factor in heating rates. More clouds – less heating. Fewer clouds – more heating.
Do you spend *any* time at all outside or do you remain in a basement bubble 24/7?
This is irrelevant to the point being made.
This article deliberately and deceptively misrepresents what the paper says.
Yet you have nothing to say about that.
A fact that speaks volumes about you.
No, attribution *IS* the point. Your dogma tells you it *HAS* to be CO2. Science and metrology tells us that it’s all lost in the Great Unknown.
from the paper:
——————-
Their analysis revealed a CRE of 20 to 30 W/m^2 under Arctic overcast skies during September to March corresponding to surface air temperature increases of 6 to 9degC.
———————
——————
Intercomparison of reanalysis datasets shows a monthly mean Arctic SW deviations exceeding 90 W/m^2 due to CF bias.
———————
Note carefully that a CRE interval of 20 to 30 W/m^2 is wide enough that it will mask what climate science attributes to anthropogenic warming.
Note carefully that a deviation exceeding 90 W/m^2 is a direct metric for the uncertainty associated with the data. Again, masking whatever differences might be caused by anthropogenic warming.
The data isn’t accurate enough to support the attribution you are making.
Do you have a special crystal ball hidden in your basement? One that can peer into the Great Unknown?
Tim, I wrote my comment before reading this by you, about the crystal ball. I’m often surprised by how much alike our thinking is.
That is your interpretation. Unlike you, I’m not going to presume to know what his thoughts or motivations were. However, a credible, alternative interpretation is that the author felt that his point about uncertainty was best made by changing the placement and therefore the emphasis. I’m surprised by how many amateur mind readers are attracted to this site.
If this conclusion is based on unreliable data, unfit for purpose, then the conclusion cannot be trusted. That is the point being made.
The problem is the attribution. If you can’t actually measure the cloud fraction with sufficient accuracy to allow proper attribution, then you DO NOT KNOW the cause-effect factors! Just ASSuming it to be a “greenhouse effect” is attribution by dogma and not by science.
The other baseless dogma is perennial citation of “albedo feedback” , which seems to make sense at the most trivial level but is never scientifically evaluated.
Less ice allows more solar to touch the water surface but mostly at glazing incidence, leading to near total surface reflection.
The counter part to that “warming” effect is less insulating ice cover; more surface evaporation ; more heat advection; more IR radiation from water surface …
They seem to think that it is sufficient speak the magic incantation “albedo feedback” and QED, you have the whole story without any scientific analysis or observation.
If there was a significant albedo driven POSITIVE feedback , the annual minimum sea ice extent would NOT have been the same in 2024 and 2025 as it was in 2007 when all this BS about “tipping points” and “run away melting” hysteria was being pushed by Al Gormless and the IPCCCP.
Despite ice extent dropping by something like 25% between 1980s and 2007, there has been zero net change in the ensuing 18 years !!
Some feedback.
The trivial conclusion to that actual data would be that there is a net NEGATIVE f/b which has stabilised ice coverage at a “new normal”.
Whenever I see the word “feedback” used by climatology the initial reaction is that a load of wafflegab word salad handwaving will follow. They don’t understand or try to use the Bode equation; nor can they use a feedback amplifier block diagram correctly.
“Its a feedback! We are in dire straights!”
Bravo.
Now per NASA.
bond albedo – how bright an object appears from a distance.
geometric albedo – how bright a object appears when viewed on the nadir, the vector between source and destination.
Hijacked and redefined to mean reflectivity. Albedo, the concise definition, includes both reflected EM energy and warm body emitted EM energy.
Too bad none of them have a clue about electromagnetics or studied radar.
Albedo also only measures light that is randomly scattered to an observer — be it bond or geometric. For energy balance analysis, one has to account for all light, not just retro-reflectance that is observed. Clouds approach Lambertian reflectance, so foreword scattering should be about the same as retro-reflection. However, that is not the case for specular reflection.
At normal incidence, the specular reflectivity of sea water is about 2%, which means that 98% of green light enters the water, with a few percent coming back out after encountering suspended sediment and plankton; thermal IR is strongly absorbed near the surface.
On the other hand, as the angle of incidence increases above about 60 deg, not only does the specular reflectivity increase to approach a limit of 100%, but the intensity of the light entering the water inversely decreases to a limit of zero. Thus, the intensity of solar light, including the thermal IR, declines to zero at the Terminator, with the spectral distribution of the reflected light matching that of the source. These subtleties of behavior are typically missed by oceanographers and climatologists who talk about Arctic “dark water.”
“At normal incidence”
For those who don’t know 3D geometry, “normal” means incident at 90deg, i.e. when the sun is directly overhead. That typically only happens directly under the sun’s path on the earth’s surface, many times simplified to the “equator”. Any point off the “equator”, i.e. latitude > 0, will never have “normal incidence” of the sun’s insolation, there will always be some amount of “tilt” involved.
(Because of the actual shape of the earth and tilt of the earth, the sun’s path is not at the “equator”. )
In optics, it is the practice to measure the angle of incidence from the vertical at the point where the incoming ray intersects the surface. That basically matches latitude. At the Equinoxes, over the Equator, at solar noon the angle of incidence is zero degrees; at the geographic North Pole it is 90 degrees. At solar noon, at the Solstices, the sun will be directly over the Tropic of Cancer in the north and the Tropic of Capricorn 6 months later in the South.
Yep. The term “normal” can be confusing to many. It doesn’t mean “usual” or “typical” when used as a term of 3D geometry. It means 90deg or completely vertical.
So CO2 is melting ice. Got it.
“No doubt” except that the minimum annual ice coverage occurred in 2012, and has generally been increasing for the last 13 years.
Visualize the trend of the red line since 2012:
“There is no doubt that the Arctic is rapidly warming, and the greenhouse effect is the cause, with feedback effects on albedo.”
Had you left out that conclusion, you would have made a valuable contribution.
That one sentence cancelled your contribution. Net Zero.
I am very disappointed. A number of years ago, you wrote an enlightened article, published on WUWT, concerning feedback.
Now you are embracing the Tran-Reality Activist lexicon definition. Also “anomalies.” Also “greenhouse effect.” Also “albedo.”
And you insert a bit of hyperbole of “rapidly.”
“Had you left out that conclusion”
I am simply summarising what Liu et al said in that intro. Including “rapidly”:
“The rapid changes in the Arctic climate system…”
““The rapid changes in the Arctic climate system…””
This is *NOT* an attribution statement for what is causing the rapid changes. Your claim that it is due to anthropogenic warming is attribution by dogma.
Can you please put a Facebook Share icon on your material, Mr Watts. It’s excellent, and I’d love to share it, regularly!
I get sick of all this talk about “downwelling SW radiation”.
Radiation does not “well” up or down: it RADIATES.
If I lay on the beach all day I don’t suffer from all that radiation “welling over my skin”, I get burnt by UV RADIATING through the atmosphere.
Put that in the basket with all the other stupid terms they invented like the perfectly normal temperature “anomalies” and the physically meaningless concept of “average surface temperature” which happily adds a non-extensive property of matter as though I was a valid proxy for an extensive property ( heat content ). Adding land air temps to SST is junk science even at the most basic level.
Let us expand a tad on your wonderful remarks.
The sun’s incident EM radiation is a spectrum that extends well beyond short wave. UV through uW and beyond.
The bulk is in the short wave regions, but the Plank distribution extends well beyond near IR.
Very interesting. As I usually when I don’t fully understand a scientific article I ran it past Grok.
Grok initially wanted to play the role of gatekeeper as to which parts of the liu paper it would permit Kenneth to use. Quite similar to the stylized distractions made by Stokes.
I eventually managed to get grok to focus on the science. It seemed a bit discombombulated and spat out lots of words before drawing a conclusion which didn’t seem to follow from the premises of the argument it made.
So I went through paragraph by paragraph. Grok tells me it has no objections to the first three paragraphs of Kenneth’s article. It balked at the fourth. It seems to me that it didn’t like the synthesis of a new conclusion from the presented facts.
I was about to go through the fourth paragraph sentence by sentence when my time expired (like it always does when grok is cornered).
I will pick up the conversation with grok when it allocates me further time.
For what it’s worth I think Kenneth is right.
I’m surprised to read that Grok rations time. I was recently trying to regain some storage on my 128GB SSD and Copilot made some mistakes that dragged the effort, which I naively thought in the beginning might take an hour or two, out to two days with only breaks for food, bathroom necessities, and sleep. If it had suddenly said, at an inconvenient time, “Oops! Time is up for today,” I would have been in a real pickle. I might have lost the ability to re-connect and even lost the ability to boot the system. This is an issue I hadn’t considered, namely that the AI agent might leave me hanging before solving a problem.
Does the removal of SO2 by the clean air acts significantly increase solar forcing in the arctic?