Guest Post by Willis Eschenbach
Well, the endless curiosity of my monkey-mind led me to some interesting datasets over at KNMI. They’re a variety of meteorological measurements made in Holland since 1964. I noticed that they have global solar radiation data, so I plotted that up to see the variations. I overlaid it with the CEEMD residual, which is what is left after the various cyclical variations have been removed. The CEEMD residual shows the overall trend of the data. (For more information on CEEMD analysis see my post “Noise-Assisted Data Analysis.)
Figure 1. Monthly Dutch solar radiation, January 1964 – January 2023
This was interesting. Solar ran about level for about five years, decreased for some twenty years or so, and then increased after that.
Next, I looked at the Dutch surface temperature, in the same manner.
Figure 2. Monthly Dutch surface temperature, January 1964 – January 2023
Fig. 2 shows a general slow rise in temperature over the period in question.
Now, my choice of these two datasets was not random. Remember that the central paradigm of modern climate science is the very simplistic view that changes in temperature are a linear function of changes in downwelling radiation.
Me, I think that’s nonsense. I know of no other complex chaotic system with that kind of simple input/output relationship … but that’s what mainstream climate science claims. I discuss this idea in my post entitled “The Cold Equations“.
So … here’s the relationship between solar radiation and temperature in Holland. In both cases, I’ve used the CEEMD residuals.
Figure 3. Monthly Dutch surface temperature, January 1964 – January 2023
Hmmm … not exactly a linear relationship. It starts with solar radiation dropping and temperature rising … and then that relationship reverses.
Now, this is just solar radiation. Unfortunately, the KNMI data doesn’t include downwelling longwave radiation. However, we have that data from the CERES dataset for the period from March 2000 to February 2022. Over that time the downwelling longwave radiation in Holland decreased by -5.0 W/m2.
And this in turn gives an overall (longwave + shortwave) sensitivity for that shorter time period of … 0.1°C per W/m2 … go figure.
Can we draw any overarching global conclusions from this analysis? No way. It’s just one small country … however, it does show that over the last ~60 years, there are many more things that affect the temperature of Holland than just downwelling radiation.
My best to all.
My Usual Request: When you comment, please quote the exact words that you are discussing. This avoids endless misunderstandings.
Willis : you know I’m useless with charts, but I shall study this hard, as these are Your Charts.
Un abrazo desde Madrid
“… it does show that over the last ~60 years, there are many more things that affect the temperature of Holland than just downwelling radiation.”
Is anybody suggesting otherwise?
“Is anybody suggesting otherwise?”
Only the camp that ascribes any and all flooding, forest fires, record highs, or record lows in my region to one overarching cause.
Some idiot prime minister Rutte wants to put farmers out of business, because they use nitro fertilizers, which, regarding GW, are less than 10% of CO2 driving force
I see your trick: You are going to start enumerating the various carbonaceous demons in the sky, as proof that there are many concerns.
But they all boil down to downwelling, innit? All them little demons above radiating our defenseless little souls at night, instead of shooting their heat at the stars?
Are you suggesting otherwise!
Holland population 1960—11.4 million. GDP 12.3 billion
Holland population 2022—17.6 million. GDP 1037 billion.
Might have something to do with Holland temperature rise given UHI.
I wonder what the model of Dutch economics will project for future GDP when 20% of the farms are closed and the use of fossil fuels is forbidden?
I expect a trend reversal!
Willis: “there are many more things that affect the temperature of Holland than just downwelling radiation.”
WR: There is a super interesting study using reanalysis maps from 1836 to 2020. It is the change in the origin (!) of the air (not: wind direction) that explains the change in temperatures in the Netherlands. There is a jump in temperature from 1987-1988. See fig. 9 in the next study:
From the conclusions:
“Here we denoted the origin of the airflow instead of the direction of the final part of the trajectory of the airflow, and we showed that there often is a major difference between the origin of the airflow and the direction of the wind on the ground. Next, we have shown that the frequency of some of the weather patterns has changed significantly by comparing the periods 1961–1990 and 1991–2020, resulting in an increase of airflow coming from warmer directions. Finally, we have used the weather patterns to explain the daily temperature in the Netherlands; hereto, we used a model based on linear regression. When we extended the model with a factor representing the sea, and with factors for the TSI, AMO, and CO2, then we find an R2 value of 0.85. Interestingly enough, CO2 does not seem to play a role.” etc.
This appears to be just a manifestation of different climate regimes with changes taking place around well-known climate shifts. These changes in transport affect the amount of humidity carried by winds, and the strength of the winds too, resulting in changes in cloud cover.
Changes in wind speed and evaporation during climate regimes.
When Arctic sea ice returns, I expect weather patterns over Western Europe to change back again to the pattern of the fifties, sixties, and seventies. Well different from the present weather pattern. Unfortunately, because the present weather pattern is exactly what the Dutch people wished for during the fifties, sixties, and seventies.
The increase in sunshine from around 1997 was to be expected, going by my own and Javier Vinos’ reasoning…
To read comments here, remember the translate bottom in the left hand bottom corner.
In the Netherlands (De Bilt) we observed extremely sunny years in the 2000s, while there was an extremely sunless period in the late seventies/early eighties: see the two columns on the right in the following link:
A lot of sun-hours coincided with a lot of high-pressure areas in spring and summer in the 2000s.
Altogether, this shows that a change in the weather pattern is able to explain most changes in temperature, if not all.
That change in weather pattern remarkably coincides with the melt in Arctic sea ice after 1987, Melting sea ice results in much more Arctic water vapor, resulting in much more low-pressure areas sucking in warm moist air over the North Atlantic. In the flow of moist air, more low-pressure areas went from the North Atlantic into the Arctic. Hence the change in weather patterns over Western Europe.
And very little wind during high pressures
Ja. Ja. It comes always back to what I said.
Excellent article, Willis and food for thought. One small correction: officially the country is called the Netherlands. It contains the provinces of North Holland and South Holland. See this nice four minute YouTube explanation for details. Note that the ç in Curaçao in this clip must be pronounced as an s, not as a k.
It is very common in the UK that particularly hot weather coincides with warmer air coming from the South from Europe and North Africa. There may well be a longterm change in wind and thus weather patterns which might throw additional light on the Dutch temperature records.
Downward (and upward) emitted radiation from atmosphere is proportional to atmospheric absorbed radiation in local profiles.
Reduced downwelling LW implies reduced absorbed LW, and therefore an increasing proportion of transmitted flux to space via the windows at this location.
Globally averaged, atmospheric absorbed energy = (3/2) OLR in radiative equivalent concepts.
In kinetic equivalence, a more apt physical description, the same relation holds.
Internal Energy = (3/2) NkT
k = Boltzmann constant, k = 1.38 × 10−23 J/K
T = temperature in Kelvin
N = number of molecules
The atmosphere behaves roughly like an ideal gas as a whole. This is useful to know. It is actually required by physics, I think.
Or, surface average radiative equivalent energy = (3/2) * Solar Absorbed.
There isn’t much you can do to change that.
This suggests the polyatomic gases contribute nothing special to the average kinetic energy (temperature) in the atmosphere. Weird.
Doubly weird because the atmosphere is also packed full of solid and liquid particles, of mostly water mass. Perhaps it is these particles which makes the relationships work in schemes devised by radiation enthusiasts.
U = (3/2) nRT
PV = NkT
I have not seen your equation for internal energy.
It is true that 3/2 nRT = 3/2 NkT for a simple ideal gas scenario in this course http://physics.bu.edu/~duffy/py105/Kinetictheory.html#:~:text=The%20internal%20energy%20of%20an,also%20the%20total%20internal%20energy.
My curiosity is to understand if/why this relationship holds for the atmosphere. If you have any insights I am interested. My interpretation so far is that the translational energy of greenhouse gas molecules is directed upwards (against gravity) due to the density gradient. And so, the relationships hold.
It is the non-ideal component that matters. The ability of water to transform to ice is the most fundamental aspect of energy control in the atmosphere as it is on land and on water.
In the atmosphere. the formation of a level of free convection due to relative buoyancy of water vapour to air is a critical process. There is a solidifying zone forming ice above the LFC and mixed zone below that creates the convective instability that powers global circulations. It also limits open ocean temperature to 30C.
You may be right, but the condensate are not gases at all.
Here, the relative buoyancy a function of the ambient atmospheric density profile. A critical parameter indeed.
Furthermore, the diurnal conversion of kinetic energy to gravitational potential energy (and back) by translational mode, as molecules are forced upward under sunlight. A nonradiative Δ storage term.
Yes – That is the “non-ideal” feature of the atmosphere that is vital to Earth’s thermal regulation..
The Dutch are an odd bunch. They find a way to combine apparent insanity with bare faced truth in ways that are unusual. They’ve played the best football and got involved in some ugly fights.
Quote:”there are many more things that affect the temperature of Holland than just downwelling radiation.
That’s a bit mean, they’re called people. They have feelings, they’re not ‘things’
Going about their businesses of being farmers, city builders and drainage engineers
Or as I’m trying to figure an extent of in the landscape I now find myself living – Greenhouse Constructors.
But it was the Dutch drainage lads who made The Fen liveable.
My puzzle, which I’ve flagged up before, concerns this graph and those abrupt spikes in the temperature plot through the night between 3AM and 6AM
And many other Wundergrounds had very similar plots, same dates, same times and all the ones randomly scattered inside a square-shaped patch of about 25 square miles.
But those spikes completely vanish on other nights when temperatures are well (more than 5 Celsius) above Zero Celsius.
Being there a lot of farmers around here, of Dutch origin, there are ‘some number’ of greenhouse/glasshouses so straight off I says,
“There go their Frost Protection Systems – ramping up some heat to protect delicate (literally) little flowers”
(I kid ye not a place name of ‘Floral Farm’ is not far away – owned & operated by a guy name of ‘Prins’)
OK, I then tries to correlate where Wunderground says the stations are and where there are any sizeable glasshouses on the ground using Google Map.
abject complete fail
In fact the nearest large glasshouse is 60,000 square metres (OMG – think of the radiational grief that must catch) and the approximate centre of it is a little few hundred metres beyond my front door.
It is The Largest Area of glass for miiiiiles around. You’d have never guessed just driving past.
yet my datalogger,bang next door to this behemoth barely flickered compared to the Wunderground ones that are miles away.
Apart from the UHI Effect, whether caused by actual Urban or Agricultural Urban has a much greater and wider influence than anyone imagines
Just before submit I realises: It might be ‘Cold Stores’ doing the same – especially needed as they grow huuuuuge amounts of potatoes around here. Yes they need to be kept cold but if they freeze that’s a disaster for them
Point being: But all those things will be going on to even greater extent in Holland and have been for quite a long time.
‘things’ eh, what are they like?
Meanwhile: The search continues = any excuse to go exploring, burn some diesel while I still can and ‘have a little dance’
edit to PS Fun Fact
A little patch of dirt, not especially imaginatively named ‘New Holland‘ is about 75 miles North of here, on the southern banks of The Humber, not far downstream of where the mighty Trent joins in
I cannot figure why they called it Humber, the Trent a pretty major, if not the major, drain for most of The Midlands of England.
It’s where Longitude was invented.
New Holland is a brand of farm equipment.
Peta, off topic but there was Notice in the Grauniad (17th Feb) of an application to build the Cottam Solar Project in an area east and south of Gainsborough, consisting of Cottam 1,2,3a,3b. Total capacity over 50MW.
Covers quite a large area from village of Corringham, just north of Gainsborough to village of Coates well south and east of Gainsborough.
Anywhere near you?
I like the way your mind works.
Willis: “we have that data from the CERES dataset for the period from March 2000 to February 2022. Over that time the downwelling longwave radiation in Holland decreased by -5.0 W/m2″
WR: The decrease in downwelling longwave radiation in the Netherlands in the 2000s would comply with more frequent high-pressure areas in spring and summer, with some very sunny and very dry periods, and less water vapor in the air.
I must be missing a trick, but how do you arrive at the 0.1 W/m2 increase in the period from 2000 to 2022? Looks like the irradiation increased by approximately 10 W/m2 (of the 14.6 total) in that period, which you then correct by the -5 W/m2 change in downwelling radiation.
Per, it’s “0.1°C per W/m2″, not “0.1 W/m2”.
The Elfstedentocht years in your chart are 1985, 1986, 1997. 2012 was a nearly year but the race did not go ahead when icing was insufficient after a 10 day freeze. Is that enough to debunk the idea that it needs a year with low solar radiation?
Elfstedentocht is Eleven towns skating race over the canals.
Another significant contribution from Willis. The increased in solar radiation he reports here is occurring elsewhere in Europe and at my Texas site. This has been discussed in various papers, and the general conclusion is that the increase in sunlight is due to the reduction in sunlight-absorbing emissions from coal-fired power plants. This is very obvious in my 33-year measurements of aerosol optical depth, the first 30 years of which are described in my latest paper in Bulletin of the American Meteorological Society. You can find this paper by searching on: BAMS 30-year climatology Mims. Forrest M. Mims III
As always, Forrest, your contributions are both interesting and well-documented. Thanks.
That is not true. This “paradigm” was given up a while ago.
Very unlikely. Given “downwelling radiation” is mainly a function of temperature, which has increased, it will rather be the opposite.
E. Schaffer said:
That is absolutely true. It is the basis of the idea of “climate sensitivity”, which is the claim that
∆T = λ ∆F
where “T” is temperature, “F” is forcing, and “λ” is climate sensitivity. If they’ve “given up” on the paradigm why are they still arguing about climate sensitivity.
Look, all I can tell you is what the CERES dataset says, which is the decrease listed. Don’t like it? Go argue with the CERES folks.
And the idea that “”downwelling radiation” is mainly a function of temperature” is a joke. It’s also a function of cloudiness to a very large extent. Less cloudiness = less downwelling radiation.
Note that CERES says that “net” downwelling longwave radiated power at the surface is negative – about -53 W/m^2 globally on average. That’s because the surface is warmer than both the atmosphere and outer space, normally, so it will always lose energy at those wavelengths to those two masses of cooler matter. As cloudiness increases, that upwelling value can decrease towards 0. (“Net” is what they use to refer to “measurable” power, which in regular physics is just called plain “power”, since there is no such thing as non-measurable power)
Willis, you wrote: “Solar Radiation (W/m^2)”. However, as you know, solar (or any) radiation is not measured in Watts. It is measured in Joules, or degrees, being a form of energy, as you have told us before. So in order to get Watts, via the S-B equation, you need two temperatures (and emissivities/absorptivities): the source and the target. In the case of solar radiation, of course the source is the Sun, and we know roughly how hot that is at its surface, and it’s essentially a blackbody too. But the target is what, exactly? The ground? That temperature can vary by quite a bit, anywhere from 243 K in winter to 303 K or more in the summer. A 60 K variation in the target temperature might not sound like much when the source temperature is closer to 6000 K, but it is still on the order of 1%. Since your graph is looking at variations that are also on the order of 1%, the difference is significant. Can you supply the temperatures used to make these Watt measurements, so that we can see whether KNMI knows what they are doing or not?
Then you said “…downwelling longwave radiation. However, we have that data from the CERES dataset”. Yes, we do, even though it is not a direct measurement. But nevertheless they came up with a “net” global surface DWLWIR power of -53 W/m^2 (i.e., this is an upwelling power). Was that what you were referring to? It roughly matches what you would get if you averaged the unadjusted SURFRAD pyrgeometer measurements, so it sounds like it’s in the right ballpark. That means we can finally stop referring to NOAA’s fake 340+ W/m^2 DWLWIR numbers, right?
Finally, you also wrote “the central paradigm of modern climate science is the very simplistic view that changes in temperature are a linear function of changes in downwelling radiation.” Since downwelling radiation is itself a temperature, this amounts to “changes in temperature are a linear function of changes in temperature”. If we leave out the “linear”, which does indeed sound way too simplistic, then we have more or less a tautology, which I can subscribe to. But I don’t think that’s what the climate scientists are actually selling us. I think what they are actually selling is “temperature is a (linear or otherwise) function of changes in atmospheric CO2 concentration, so stop emitting CO2, you peons“, which is much less obviously correct (to put it charitably).
stevekj February 17, 2023 1:42 pm
Willis, you wrote: “Solar Radiation (W/m^2)”. However, as you know, solar (or any) radiation is not measured in Watts.
From the source document:
Don’t like it? Go argue with KNMI.
I’m not the one relying on stuff they made up. You are. Can you defend your words, as you claimed you could, or not? Remember, all it takes to convert radiation energy into a power measurement is two temperatures. Do you have those temperatures? Did KNMI publish them? Without them, 1 Watt is just as likely to be correct as 1000 Watts, and both are useless for making any conclusions about climate factors. (At least you did get that last part correct.)
The monthly ground level insolation ranges from 20W/m^2 to 250W/m^2. The monthly surface temperature ranges from -5C to 20C. You will find that the temperature response to the sunlight over that range is highly correlated, better than 95% fit, when you make allowance for the one month thermal lag.
Once you produce anomalies, you have lost most of the useful information. It is termed anomalous for a reason.
An image attached to make my point. Three years of sunlight and temperature compared. Temperature obviously follows insolation.
Your smoothed monthly chart is comparing the noise not the signal. Anomalies are noise.
The surface conditions in the Netherlands changes dramatically over the seasons. Accordingly the surface response to solar input is seasonal. You cannot detrend this data and make meaningful conclusions.
You are making the same mistake that has winter warming of the Greenland plateau being interpreted as “global” warming.
Sorry, Rick, but anyone who seriously believes that anomalies are noise is “far above my poor power to add or detract”.
You cannot compare across seasons where the ground conditions change so much. It is pointless. You have highly reflective snow to high absorption ploughed fields. They respond quite differently to sunlight.
Insolation and temperature are HIGHLY correlated across the seasons but removing seasonal trends just creates noise because the ground conditions are not the same across the seasons.
You don’t even understand the difference between anomalies and the removal of seasonal variations. Discussing climate with you is an exercise in futility.
Do your comparison across selected seasons. – say July and January to look at the extremes. That is the best way to removal seasonal trends. Then you start to get some understanding.
Willis – you have no idea what is going on. You are looking at noise.
I have attached two correlation from the data you are examining. One shows all the data; temperature versus surface insolation. The regression coefficient of 87% is a bit lower than typical but is still highly correlated. To suggest there is no correlation between surface insolation and temperature is naive.
The second is just July temperature and June sunlight. The correlation not so strong but is solidly positive.
How do you allow for the thermal lag when you deseasonalise the data? You are comparing apples and oranges? Average land temperature lags surface insolation by about a month.
The Atmosphere cannot heat itself, what an absurd concept. Down-Welling Radiation indeed!
CO2 only affects the way the Atmosphere radiates to space. More CO2 raises the altitude at which the Atmosphere is not opaque to 15-micron radiation, thus lowering the temperature at which the Atmosphere radiates to space, thus retaining more energy in the Atmosphere. But no one can measure the magnitude of this effect, nor calculate it.
Ask your friend Mosher, I had to school him on this, now he understands.
Do you know the difference between a dipole moment and an induced dipole moment? Look it up, I have been watching a lot of NCIS New Orleans, “Learn things!”
MichaelMoon February 17, 2023 9:08 pm
It’s “Mr. Eschenbach”, or “Willis”, thanks …
Please quote where I’ve said that the atmosphere can “heat itself”.
Indeed. DLR (downwelling longwave radiation) is measured daily and even hourly all over the planet. In fact, we have an entire networks of SURFRAD stations which inter alia measures DLR every minute.
We have these curious things called “satellites” which can indeed measure the amount of upwelling longwave energy (ULR) absorbed by the atmosphere.
Nah, I’ll pass on that.
Water has a dipole moment, in other words a charge. Water is a bent molecule, angle of 108 degrees, produces snowflakes with six edges. CO2 is not a bent molecule, straight, only has a charge when excited by a 15-micron photon. It quickly gives up its charge low in the atmosphere, due to bumping into other molecules.
Higher in the atmosphere it will re-radiate this photon, which will either go up or down or sideways, and be re-absorbed and re-radiated many times before reaching the surface, where it will not warm anything already warmer than itself.
I am trying to help you, the other side points out your many technical errors.
Did you have any idea that charge and electricity is involved in this? You never mentioned it….
You are “trying to help me” by showering me with meaningless incorrect statements?
Pass. Please, I implore you, go “help” someone else.
Cold matter, such as the Top of the Atmosphere, cannot heat Warmer things, such as lower in the Atmosphere, If there is only one thing we know for sure, it is the 2nd Law of Thermodynamics, many billions are spent on heating building, moving vehicles, generating electricity, and we know what we are doing.
Please read my post “Can A Cold Object Warm A Hot Object“. You are clueless and spreading falsehoods.
Thanks from Sweden with a similar development in sun and temperature.
Up 11% since 1983.
With 0,1 deg C per W/m2 this will render a rise in mean T of 1.1 degree.
It is easy to correlate temperature to sun during summer.
Willis, your Fig. 3 is simplistic, omitting the ~50% 0.8C ocean warming effect on land temps.
…and Happy Belated Birthday oldtimer
Glad to see you are posting Willis. I have no pull with Elon, but have repeatedly requested your reinstatement. Do you have your own site or somewhere else where one can follow you? Thanks, the Health Skeptic
Thanks, Kevin. My blog is called “Skating Under The Ice“.
Actually, you’ve demonstrated quite convincingly, that if you smooth data enough, you get smooth curves. You’ve also demonstrated quite convincingly that correlation does not prove causation.
Forgive me for asking, but with so many well read people in the forum, it surprises me that with all the talk about irradiation there is no mention of cosmic rays and their impact on cloud cover. I find that whole theory quite convincing….and would not be surprised if variations in cosmic rays over time could disturb a solar periodic cycle here and there. What is the problem with that (GCR) theory.
Per, I’ve written dozens of posts looking for some sunspot-cycle-related signals in surface weather data. Cosmic rays vary with the sunspot cycle, so if such an effect were significant, it should have turned up in my analyses … no joy.
My studies on this subject are here. Plus there’s a new one published just a half hour ago, it’s here. Enjoy.