Analysis Indicates The Real Cause of Recent Warming Is Not CO2

By Ed Caryl

I would like to thank a faithful warmist commenter for inspiring this research.

His claim for my mental state seen here, in The 85-year Pause, forced me to “put up, or shut up” on the subject of warming seasonality and the reasons for it. To that end, I added many more stations to my list to be examined, and plotted monthly data instead of just two seasons.

First, some discussion on CO2 warming theory, a subject on which this faithful warmist commentator claims some expertise. I searched the web for any illustrations of monthly downward long-wave radiation. I finally found them at, here. Please click on the link.

Twelve different stations are shown, with modeled and observed month-of-year data. The data for each of these stations closely matches the local insolation (sunshine hours), with a maximum in the local summer, especially for polar and middle latitudes.

One would think, that with more downward IR in the local summer, and with increasing CO2, that the resulting warming would be in the summer also. That is not the case.

Some signal theory: electrical engineers, physicists, and technicians that have ever used a digital oscilloscope will be familiar with signal averaging. When a faint signal is synchronous with a larger signal, triggering the oscilloscope on the larger signal and averaging many times will bring out the faint signal even in the presence of large random noise. In this case, the annual CO2 warming signal synchronized with the months of the year in the presence of weather.

Based on the hours of daylight for each month, here is an example of what the CO2 warming should look like for Bismarck, North Dakota, and the stations around it:

Biismarck Sunshine

Figure 1: Number of sunshine hours per month for Bismarck, North Dakota, USA. The peak is in July, the low point is in November and December. The months of the year are numbered from January with two years shown. The source link is here. Click on Climate Data For Bismarck Airport.

For all the station data downloaded from GISTemp, I split the data into two sets: from 1947 to 1980, and from 1981 to 2015. I then computed the by-month average for each set, then subtracted the first set from the second. The result is the by month average warming between 1947 to 1980, and 1981 to 2015.

Here is the result for 8 stations in North Dakota, USA. The “adjusted” data was used for all plots. As GISS uses the same adjustments for all months of the same year, the other data sets would only move these plots vertically by small amounts without changing the shape.

North Dakota Warming

Figure 2: Warming by month for 8 stations in North Dakota. Two years are shown to show a full winter in the center of the plot. The months are numbered from January. These stations are all within a few hundred kilometers of each other, so show very similar trends. The dark red trace is the average. The peak is in January.

Note that the warming in North Dakota is completely out of phase with the sunshine hours, thus out of phase with any possible CO2 warming.

Thirteen Russian stations were examined in the same fashion. This area is supposedly warming faster than anywhere else according to the temperature anomaly maps from GISS. These stations are spread across the nation from Moscow to eastern Siberia. There was more variation here, so a plot of all 13 is a classic spaghetti diagram. Here is a plot of the 13 stations with the average highlighted in black. Moscow (Moskva) is in slightly bolder red.

Russia Warming

Figure 3: Plot of the average monthly warming for 13 Russian stations. Two years are shown. The months with the most warming are January, February, and March.

Where is the winter warming coming from? It is the wrong season for the cause to be increased CO2. I was not able to find energy usage for Russia, but here is the energy usage for the U.S.:

Energy Usage

Figure 4: Energy usage in the United States by month of the year. The top curve is the average energy usage after 1980; the bottom curve is energy usage from 1970 to 1980, the only monthly data available from the U. S. Energy Administration here.

Annual residential energy usage in the United States has quadrupled since 1949, and the variation shown above is primarily due to residential energy usage. Note the increase in usage in July and August, primarily from increased air conditioning.

Here are 12 more stations in the U. S. Midwest and South. The black line is the average.

Midwest Warming

Figure 5: Plot of the by-month warming in 12 locations in the US mid-west and south. 

In the US midwest and south, the least warming (slightly cooling) is in October. The peak warming is March. All the cities have peak warming between November and March, completely out of sync with any possible CO2 warming and in sync with residential heating.

These diagrams all show a large negative spike in either September (Russia) or October (in the US). Why should this be? These are the months where the air conditioners are turned off and the heating has yet to be turned on, thus they are the months with the least energy use.

There is no way that CO2 is causing this warming in the winter months. If CO2 was the culprit, we would see warming following the diagrams linked above on the SOD website. The back-radiation from CO2 is real. It just doesn’t have any effect, or the effect is offset by something else. Instead we see warming following energy use. We can keep the climate from warming. But it means freezing to death.


39 responses to “Analysis Indicates The Real Cause of Recent Warming Is Not CO2”

  1. sod

    Thanks for writing such a long post in reply to something that i said.

    I disagree with both ideas you put forward.

    I do not think, that the CO2 warming signal must be warmest in summer. Can you provide some links for anyone making that claim beyond the graphs that you linked?

    I also strongly disagree with the idea, that winter warming is caused by heating houses. This thesis should fall apart, when we compare rural and urban stations (as they should show a completely different seasonal pattern).

    I am still interested in any links, which suppose that heating houses is a major aspect of the UHI effect. Have we got any data on that?

    1. Ed Caryl

      The plots include both urban and rural stations, and they show the same shape. The problem is that there are very few truly rural stations. There is nearly always at least one building nearby where the observer works or lives.

      1. sod

        “There is nearly always at least one building nearby where the observer works or lives.”

        the house releases heat above the height of the thermometer (1.2 m).

        you need pretty strong downward wind, to blow any heat towards a thermometer that is only a few meters away(which would be the worst case scenario.

        Even the links provided by somebody else last time do not support your thesis:

        ” A gradual and constant increase of the daytime UHI intensity was detected, in contrast to the night time UHI intensity which increases only in summer, after the mid 1980s.”

        The heat effect you describe, is known from arctic settlements. This article talks about it, on the town Barrow:

        I have also seen it discussed, when the focus was on russian towns in Sibiria, with above ground heating pipes running through the city.

        The idea, that this effect is an important factor in rural places in a moderate climate is contradicted all evidence that i have seen so far. So please, provide some links!

        1. DirkH

          Why do you deny the MEASUREMENTS of IR backradiation Ed linked to? Your crazy conjecture about houses, concrete and water have no effect on the measurements.

    2. Kenneth Richard

      sod: “I do not think, that the CO2 warming signal must be warmest in summer. Can you provide some links for anyone making that claim?”

      So you do not agree with those who claim that the high Arctic sea ice melt extents in the summers (August-September) can be linked to a CO2 warming signal? Why else would there be so many warmists who claim the Arctic will be sea ice free soon – or who predicted it would have been already – if there is no warmist who agrees that Arctic sea ice loss trends are attributable to CO2?

      For that matter, can you provide an answer as to why it is that, as recently as a few thousand years ago, when CO2 levels were in the 270 ppm range, the Arctic WAS ice free during the summers? What caused the Arctic to become ice free with CO2 levels both low and stable? Why were sea levels 1 to 4 meters higher than they are now about ~5,000 years ago, when CO2 levels were so much lower than now?

      For that matter, why did glaciers and ice sheets melt more rapidly and contribute to sea level rise at much higher rates than recent decades…during the 1930s? Since that couldn’t have been anthropogenic CO2 (emissions were low and flat at 1 GtC per year from 1900 to 1945), what caused those high glacier melt and sea level rise rates? What was the mechanism?

      1. sod

        “So you do not agree with those who claim that the high Arctic sea ice melt extents in the summers (August-September) can be linked to a CO2 warming signal? ”

        I do agree with CO2 warming showing in arctic summer. But this happens, because the water stores heat.

        The UHI effect is mostly the same: concrete storing heat, which a forest or grassland does not.

        The theory presented here (CO2 having only/mostly a direct effect during days with strong sun) is much too simple.

        1. Kenneth Richard

          sod: “I do agree with CO2 warming showing in arctic summer.”

          So then what would have caused the Arctic to be ice *free* in summers during the early and middle Holocene, when CO2 levels were in the 255 to 270 ppm range? Did the CO2 signal just “show up” in recent decades? Have the physical mechanisms affecting sea ice melt changed?

          Back in the early 1990s, when the Arctic had been cooling/not warming for 40 years despite massive increases in fossil fuel use, scientists were wondering if we truly understand how the greenhouse effect worked at the poles.

          “Absence of evidence for greenhouse warming over the Arctic Ocean in the past 40 years” [1950-1990]

          In particular, we do not observe the large surface warming trends predicted by models; indeed, we detect significant surface cooling trends over the western Arctic Ocean during winter and autumn. This discrepancy suggests that present climate models do not adequately incorporate the physical processes that affect the polar regions.

          Do you disagree with scientists who have determined that most of the oscillations in the Arctic climate (warming in the 1920s to 1940s, cooling in the 1950s to 1980s, warming during the 1990s-present) can be attributed to natural variability?

          “[A] substantial portion of recent warming in the northeastern Canada and Greenland sector of the Arctic arises from unforced natural variability.”

          “Since the decadal variation of the AO is recognized as the natural variability of the global atmosphere, it is shown that both of decadal variabilities before and after 1989 in the Arctic [cooling, then warming] can be mostly explained by the natural variability of the AO not by the external response due to the human activity.”

        2. DirkH

          “The theory presented here (CO2 having only/mostly a direct effect during days with strong sun) is much too simple.”

          Again, you call the MEASUREMENTS of backradiation as presented on ScienceOfDooms website a lie.
          Why don’t you accept them? Are they not falsified enough for you?

    3. Kenneth Richard

      sod: “I do not think, that the CO2 warming signal must be warmest in summer.”

      By the way, sod, for the LW greenhouse effect to work, sunlight is required. At the poles, it’s dark for half the year. So why do you believe it’s possible for there to be a CO2 signal in the winter at the poles?

    4. DirkH

      “I do not think, that the CO2 warming signal must be warmest in summer. ”

      So you say that the Backradiation measurement plots at ScienceOfDoom’s website are a lie????

      You do know that ScienceOfDoom is a warmunist lawyer, right?

    5. Jamie

      Notice how the defenders of ‘settled science’, when really faced with scientific evidence, use terms like ‘I don’t think’ and ‘I strongly disagree’. Doesn’t sound like ‘settled science’ to me, rather opinion.

      If indeed global warming is caused by CO2 and if this is ‘settled science’ then why doesn’t our defender of the truth, simply deconstruct this argument, using ‘settled science.’

      One reason is that predictions of chaotic systems like weather or economic systems are never settled science when they are made, simply hypothesis. Global Warming caused by CO2 is a hypothesis … not a fact.

      1. sod

        “Notice how the defenders of ‘settled science’, when really faced with scientific evidence, use terms like ‘I don’t think’ and ‘I strongly disagree’.”

        That is garbage. I am a friendly person.

        I have been asking for evidence over two discussions now and i have seen ZERO links supporting the false idea presented here (houses heating the world in winter).

        1. DirkH

          “the false idea presented here (houses heating the world in winter).”

          Can you please point to where it was suggested that HOUSES HEAT THE WORLD IN WINTER.

          Because you are suffering from delusions.

          For your EDUCATION. A THERMOMETER is placed NEAR A HUMAN SETTLEMENT. Because humans need to be able to access it.

          And as humans are lazy especially when it’s friggin cold outside they place it CLOSE to their settlement.

          The WORLD is NOT heated by THE HOUSE. The THERMOMETER is.

          Misrepresenting arguments of others is a typically Green deceptive tactic.

        2. Kenneth Richard

          So when are you going to answer the question, sod? I’ve asked 3 times now: If CO2 is the signal cause of Arctic sea ice melt in the summer, why was the Arctic ice free earlier in the Holocene, when CO2 levels were in the 260 ppm range? Did the CO2 signal just “show up” in the last few decades?

          1. AndyG55

            Furthermore, if CO2 is globally well mixed,

            …. why has there been NO WARMING in the Antarctic for the whole of the satellite record, and no warming in the Southern ex-tropics for 20+ years?

            Either CO2 warming is global.. or its NOT !!

          2. AndyG55

            Kenneth, it seems that CO2 levels were much higher pre-1200AD, dropped until 1800AD, then started to climb again, but are not yet to the pre-1200AD levels. 😉


          3. Kenneth Richard

            Thanks, AndyG55. I own Lamb’s book, and have written out the section where this excellent graph came from:

            Cooling in the Arctic

            The cooling of the Arctic since 1950-60 has been most marked in the very same regions which experienced the strongest warming in the earlier decades of the present century, namely the central Arctic and northernmost parts of the two great continents remote from the world’s oceans but also in the Norweigan-East Greenland Sea. (In some places, e.g. the Franz Josef Land archipelago near 80 °N 50-60 °E, the long-term average temperature fell by 3 – 4°C and the ten-year average winter temperatures became 6 -10°C colder in the 1960s as compared to the preceding decades.)

            It is clear from Icelandic oceanographic surveys that changes in the ocean currents have been involved. Indeed a greatly (in the extreme case, ten times) increased flow the cold East Greenland Current, bringing polar water southwards, has in several years (especially 1968 and 1969, but also 1965, 1975, and 1979) brought more Arctic sea ice to the coasts of Iceland than for fifty years: in April – May 1968 and 1969 the island was half surrounded by the ice, as had not occurred since 1888.

            Such years have always been dreaded in Iceland’s history because of the depression of summer temperaturesand the effects on farm production. In the 1950s the mean temperature of the summer half year in Iceland had been 7.7°C and the average yields were 4.3 tonnes/hectare (with the use of 2.8 nitrogen fertilizer); in the late 1960s with the mean temperature 6.8°C the average hay yield was only 3.0 tonnes/hectare (despite the use of 4.8 kg of fertilizer). The temperature level was dangerously close to the point at which the grass virtually ceases to grow. The country’s crop of potatoes was similarly reduced. The 1960s also saw the abandonment of attempts at grain growing in Iceland which had been resumed in the warmer decades of this century after a lapse of some hundreds of years.

            At the same time the changes in the ocean have produced changes in the spawning grounds and seasonal range of migration of fish stocks – a not much publicized aspect of the international wrangles and ‘cod wars’ of recent times. With the fall by over 1°C in the mean surface temperatures off west Greenland from the peak years in the 1920s and 1950s, the cod fishery there declined by the early 1970s to a tiny fraction of what it had been in those times. The Greenland cod migrated to Iceland waters, and for a few years (1967 – 71) offset the declining stocks there; but since 1974 the spawning stocks in Iceland waters have been only a tenth of what they were in the late 1950s and the total stocks have fallen almost a half, the decline being probably due to combined effects of the change in water climate and over-fishing. Similarly, herring stocks have moved from Iceland waters to the wider reaches of the Norweigan Sea farther east, south, and north to the North Sea, while a southward shift of the southern limit of cod seems to have led to increased catches in the North Sea since about 1963.

        3. Jamie

          You don’t understand basic scientific methodology. You are defending the hypothesis that observed CO2 increases lead to out of control warming. It is your burden to present proof and demonstrate that there is no reasonable doubt as to cause and effect.

          For instance, if I introduce a hypothesis that weather can be controlled by singing ‘Evita’ outdoors — it is not for me to demand that you disprove it. It is my hypothesis, and the burden of proof is on me. Yet you warmists have the scientific method backwards — as you demand negative proofs for your hypothesis:

          “Prove to me that singing Evita does not change weather!”

          Even if you debunk this article, you are still at ground zero. You are embracing a hypothesis, mistakenly identifying it as fact, without the ability to prove that your beloved long-term forecasts are ‘settled science.’

  2. John Silver

    ” Instead we see warming following energy use”

    No, we don’t see any warming.
    What we see is increased temperature readings on UHI stricken thermometers.

    In the US, look at rural MMTS no airports. You will see statistically nada.

    1. John Silver

      It’s right here, the true number is 0.032, the lie number is 0.300.

      Ten times, an order of magnitude difference between the truth and the official fraud.

  3. Mike Gilroy

    I believe the observations, data crunching and graphs may show the warming effect of increased CO2 for the January peak anomaly. It would be caused by the angle that the sun light through the atmosphere.

    In winter, the sunlight angle is significantly lower than in summer. It therefore passes through more atmospheric CO2 than in summer.

    The CO2 absorbs infrared and then reradiates it both upwards and downwards during both seasons. Since the incoming infrared passes through more atmosphere at the oblique angle of winter, more infrared is absorbed and more infrared is reradiated downward. Additionally, there would be a carryover of downward infrared radiation well into the night.

    Increased CO2 would increase the amount of winter atmospheric infrared absorption to a greater extent than the higher angle of summer.

    Since the sun and earth are, for all intents and purposes, in the exact same positions for both data sets, it may be the increased CO2 that caused the winter anomaly.

  4. Kenneth Richard

    Ed Caryl: “The back-radiation from CO2 is real. It just doesn’t have any effect”
    Conventional theory of global warming states that heating of atmosphere occurs as a result of accumulation of CO2 and CH4 in atmosphere. The writers show that rising concentration of CO2 should result in the cooling of climate. The methane accumulation has no essential effect on the Earth’s climate. Even significant releases of the anthropogenic carbon dioxide into the atmosphere do not change average parameters of the Earth’s heat regime and the atmospheric greenhouse effect. Moreover, CO2 concentration increase in the atmosphere results in rising agricultural productivity and improves the conditions for reforestation. Thus, accumulation of small additional amounts of carbon dioxide and methane in the atmosphere as a result of anthropogenic activities has practically no effect on the Earth’s climate.
    The author associates the recently observed climate warming and carbon dioxide concentration growth in the lower atmospheric layers with variations of solar-geomagnetic activity in global cloud formation and the significant decrease in the role of forests in carbon dioxide accumulation in the process of photosynthesis. The contribution of the greenhouse effect of carbon-containing gases to global warming turns out to be insignificant.
    The current global warming is most likely a combined effect of increased solar and tectonic activities and cannot be attributed to the increased anthropogenic impact on the atmosphere. Humans may be responsible for less than 0.01 C (of approximately 0.56C (1F) total average atmospheric heating during the last century) (Khilyuk and Chilingar 2003, 2004). The global natural processes drive the Earth’s climate ‘‘Climate will change, either warmer or colder, over many scales of time, with or without human interference’’ (Gerhard 2004). Any attempts to mitigate undesirable climatic changes using restrictive regulations are condemned to failure, because the global natural forces are at least 4–5 orders of magnitude greater than available human controls.
    Correct representation of the SSTs changes is important for the Northern Hemisphere, while correct representation of stratospheric ozone changes is important for the Southern Hemisphere. The ensemble-mean trend (which captures only the forced response) is nearly always much weaker than trends in reanalyses. This suggests that a large fraction of the recently observed changes [SSTs, ozone] may, in fact, be a consequence of natural variability and not a response of the climate system to anthropogenic forcings.

  5. B Buckner

    The planet is not warmed by the sun rays hitting land, or solar radiation directly heating the atmosphere. Land does not absorb short wave solar radiation and turn it into heat, or infrared radiation, the oceans do. So sunlight penetrates water and heats it, the oceans in turn heat the atmosphere. The desert gets cold at night because there is no heat stored in the land. Your premise is wrong.

    1. DirkH

      “Land does not absorb short wave solar radiation and turn it into heat, or infrared radiation”

      Last time I was standing on a tarmac road on a sunny summer day, the road disagreed with you.

    2. AndyG55

      “Land does not absorb short wave solar radiation and turn it into heat,”

      lol…. of course not 😉

      Try walking on a beach on a hot mid summer’s day.

  6. crosspatch

    Note that the warming in North Dakota is completely out of phase with the sunshine hours, thus out of phase with any possible CO2 warming.

    Disagree somewhat here. I would expect greenhouse warming to be most evident in the moderation of low temperatures, not an increase in highs. I would expect the greatest impact of greenhouse warming to be seen in moderation of winter temperatures. A greenhouse doesn’t not produce its benefit from making daytime temperatures warmer as much as it does by moderating low temperatures by returning a portion of the LWIR back to the source and preventing its escape. This can be seen even in weather. For example: a very humid location such as Orlando Florida on any given day might have the same high temperature as Tucson, AZ but the overnight low is going to often be much cooler in Tucson because of a lack of the water vapor greenhouse provided by humidity. So greenhouse warming would raise average temperatures but not be increasing the high temperatures. It would increase the lows.

    Increased greenhouse gas might even REDUCE high temperatures. This is because IR radiation arriving from the sun would be absorbed by the GHG. Seeing increased winter temperatures is exactly what I would expect to see from increased greenhouse effect. This would have several benefits including reduction of energy costs/consumption and increased growing season length.

    1. DirkH

      You are right that it is easier to heat a cold surface – due to the 4th power of T in the Stefan-Boltzmann Law – but the backradiation measurement plots at ScienceOfDoom show that there is far more backradiation to go round in summer. If the backradiation were the important factor, as the Global warming proponents claim, this would have to manifest itself exactly the way Ed states.

    2. Graeme No.3


      Yes, rising minimum temperatures rather than higher maximums, especially in the west of the USA. Just what Reid Bryson found from records at various forts 1840-1880. All you need to do is plot those rises against the ‘rise’ in CO2 and one two thousandth of a Nobel Peace Prize will be yours.

  7. L Michael Hohmann

    I can’t fathom any more why one should even bother to consider any CO2 role in global warming since discovering the ‘Birth of Thermodynamics” as described here: and its linked corollary.

    1. yonason

      Interesting overview of Thermodynamics, but it has recently been shown that below absolute zero is achievable, with some rather odd consequences.

      Pretty “cool,” huh?!

  8. LightningCamel

    Ed, seems to me that you have a bit of a circular argument going here. Total energy use is seasonal, warming is seasonal in a similar pattern so one causes the other. I realise this is a style of argument favoured by warmunists but I wonder if a comparison of increase in energy use compared with increase in temperature may be more instructive.

    In reply to Crosspatch, you seem to be making the assumption that “greenhouse gases” operate in by a similar mechanism to terrestrial greenhouses and that is quite invalid.

    1. Ed Caryl

      I don’t know where you find any such reply to crosspatch. We all know that actual greenhouses function by blocking convection.

      1. LightningCamel

        Ed, my point exactly. I was unclear in my expression, the paragraph was my reply to him.
        I would be more interested in your thoughts on the energy amount vs change in energy use.

  9. Mindert Eiting

    In terms of the linear model of variance analysis you are talking about an interaction effect, Ed. You may have as an independent variable (absolute) latitude, but at least you have period and season, to be formalized in a saturated linear model,

    t(ij) = a + b(i) + c(j) + d(ij),

    in which t(ij) is the mean temperature as measured by all stations in period i and season j. The first term a is a grand mean. The effects b(i) are period effects, probably due to natural causes. The effects c(j) represent trivial differences between summer and winter. Finally, d(ij) is the interaction effect. The CO2 hypothesis says about d(ij) (sorry for HTML table-problems)

    Period Summer Winter
    1947-1980 – +
    1981-2015 + –

    Your alternative says

    Period Summer Winter
    1947-1980 + –
    1981-2015 – +

    These are the only solutions for an interaction effect. I would like to see the numbers. Difference scores are notoriously unreliable and you do not need them. Statistical evaluation is problematic as you don’t have random sampling.