Sins of Comission, Sins of Emission

Guest Post by Willis Eschenbach

As a result of my last post looking at the question of CO2 and COVID, I came up with an interesting question, viz:

What is the lag time between changes in CO2 emissions and Mauna Loa measurements of background CO2?

I didn’t realize what a struggle it would be. I started out looking for weekly CO2 emissions data … ho, ho, ho. Now rightly or wrongly, I consider myself a reasonable searcher for data. But finding weekly emissions data turned out to be a dead end.

So I looked for weekly CO2 data. That took a while, but I finally got weekly Mauna Loa data from NOAA here. With that in hand, I went back to looking for the emissions data … still no joy.

I can’t tell you how many blind alleys I went down without finding sweet Fanny Adams. As a result, after giving vent to a cornucopia of bad words at a remarkable volume, I ended up having to digitize the data from this plot from Nature magazine …

Figure 1. Emissions from 2019 and part of 2020. SOURCE

After digitizing the emissions data, I grouped it into weekly data using the same weeks as the Mauna Loa data. Then I used a cross-correlation to figure out how many weeks there were between the emission of the CO2, and the resulting change in the atmospheric CO2.

Here are the results … which did indeed surprise me …

Figure 2. Standardized CO2 emissions and standardized Mauna Loa CO2 concentration data.

Hmmm … now there are a couple of possibilities. One is that the CO2 emissions and atmospheric CO2 each have an independent annual cycle, and each peaks at different times.

The other is that what we’re seeing above is actually the relationship between emissions and atmospheric CO2. The size of the relationship tends to support the latter. Overall, it takes 2.13 Gt of carbon emitted per 1 ppmv increase in atmospheric CO2. The relationship shown above gives a relationship of 3 Gt of carbon per ppmv, certainly within specs. However, that may just be coincidence.

Now here’s another oddity. Without much thinking about it, I’d always accepted the following explanation for the annual cycle of CO2

There’s more carbon dioxide in the winter and a bit less in the summer. That’s the collective breathing of all the plants in the Northern Hemisphere.

“Plants are accumulating carbon in the spring and summer when they’re active, and they’re releasing carbon back to the air in the fall and winter,”

However, looking at my Figure 2 above, I said … whaa? Does that seem like what I see in my figure?

Fearing I’d made a mistake, I went back to the Mauna Loa data. I averaged all of the individual months, and here’s what I got:

Figure 3. Average monthly CO2 anomaly and NH land temperature index

Curious. CO2 peaks in May, and drops after that. Sure doesn’t seem like “Plants are accumulating carbon in the spring and summer when they’re active” to me. Meteorological spring is March through May, and meteorological summer is June through August.

So according to accepted theory, CO2 levels should drop from March through August … but instead they’re dropping starting in May and don’t start rising until October.

And this doesn’t even take into account the ~ 4 month lag between surface changes and airborne changes shown in Figure 2 … if that’s the case, then a drop starting in May in atmospheric CO2 would correspond to a surface CO2 drop starting in February.

But what do I know, I was born yesterday.

Let me be clear that I take no position on all of this other than to call it most interesting.

My best to all. You’ve no doubt heard about the Lunar New Year? …

Well, this is shaping up to be the Looney New Year, so stay well in these parlous times.


4.9 25 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
Geoff Sherrington
January 6, 2021 6:14 pm

Hi Willis,
Your frustration with data is similar to mine. Here is what I wrote in May last year.

There seems to be a determination among keepers of atmospheric CO2 data to play cards close to their chests. The exception I found was Ralph Keeling. Geoff S

Reply to  Geoff Sherrington
January 6, 2021 9:01 pm
Reply to  Loydo
January 6, 2021 10:07 pm

Hey Loydo. You’re pretty good at giving links with your smarmy comments. How about a link to the paper we have been waiting for that you are prepared to stake your credibility on, and that of the author/s?

Reply to  Rockwa
January 7, 2021 12:34 am

No pleasing some people and going by the tone of your request I’ll decline.

Reply to  Loydo
January 7, 2021 1:47 am

Why the continued deflection Loydo? Just one paper.

Reply to  Loydo
January 7, 2021 3:31 am

If you provide what’s requested, it would stop any negative tone from Rockwa, or anyone else.

Reply to  Rockwa
January 7, 2021 10:33 am

Hey Rockwa. I see nothing “smarmy” about his comment. You can read it how you want and you are only complaining about the “smarmy” tone you chose to read it with.

It would be more pertinent to look at the data he linked to and point out that it was not the CO2 monitoring data Willis was complaining about not finding but weekly emissions data.

Last edited 1 month ago by Greg
Joe Wagner
January 6, 2021 6:20 pm

Well- the lag in the start makes perfect sense.. May is when the sun is almost at its highest in the Northern Hemisphere– when all plants have finally kicked into high production mode.

Also- an October end makes similar sense… with the same conditions… solar intensity (and the summer pattern) doesn’t end until late September/ early October, so your plants are still going to be using the CO2 during this time…

Reply to  Willis Eschenbach
January 6, 2021 8:35 pm

I haven’t checked wind speeds and directions at all, so take this as a suggestion only – much of the NH atmosphere that is diluted wrt CO2 has to make it around half the world or more before it reaches Moana Loa.

It should be well mixed by then though.

Stay safe Willis … from everything

Last edited 1 month ago by philincalifornia
Reply to  philincalifornia
January 7, 2021 1:50 pm

Most deciduous trees in the NH do not come into leaf until late April early May. It would not make sense for them to breath before they have any leaves so the graph makes perfect sense.

Tim Gorman
Reply to  Newt2u
January 7, 2021 2:52 pm

The CO2 being graphed is from fossil fuel emissions and cement making, nothing from natural sources like trees, decomposition, or anything else. And it is pure guesswork based on a few sources like fossil fuel use in only 450 cities around the world

Reply to  Willis Eschenbach
January 6, 2021 9:37 pm

The “lag” on a hemisphere basis…Its likely the difference between what is turning green and how much of the previous fall’s growth is rotting…and what is growing doesn’t start to win until about May.

Reply to  Willis Eschenbach
January 7, 2021 3:51 am

I had a posh co2 meter some time back,
i put it in a sealed hole in an upturned bucket pushed well into the ground for some hrs
the readout was almost the same as someone breathing
it pulsed!
small amt of young grasses poor sandy soils
I figured it was a combo of the grass and soil bacteria doing it.

Thomas Gasloli
Reply to  Willis Eschenbach
January 7, 2021 10:56 am

Also, a lot of early spring plant growth is not dependent on photosynthesis. Think spring bulbs, growth is due to what was stored in the bulb. Think maple frees sending sugar rich sap from the roots up to the leaf buds. A May start for a photosynthetic decline in CO2 concentration makes botanical sense. The spring ephemeral plants are building the bulb for the next year, and the trees have leafed out.

Richard G.
Reply to  Willis Eschenbach
January 7, 2021 2:52 pm

The primary observing site is located at the 3,397 metres (11,145 ft) level on Mauna Loa’s north slope (19°32′10″N 155°34′34″WCoordinates: comment image 19°32′10″N 155°34′34″W) about 5 kilometres (3 mi) north of the summit Mokuaweoweo.-WIKI

Looks like the prevailing winds are usually from the Eastern quadrant. So it appears they are sampling marine environs orographic air which would indicate ocean emissions not terrestrial vegetation growth/decomposition/metabolic emissions. Have you ever seen 24 hour graphs? Isn’t science the darnedest thing.

Love your work, stay safe, happy new year.

P. S.
“The purpose of this study was to artificially enrich a cotton (Gossypium hirsutum L.) crop with carbon dioxide using a known release rate and to determine the variation in canopy carbon dioxide concentration with respect to meteorological conditions. …
Turbulence caused considerable short‐term fluctuation in concentrations, but concentrations of 450 to 500 ppm at three‐fourths plant height were maintained with a release rate of 222.6 kg/ha/hour (198.6 Ib/acre/hour).
…The daily net photosynthate production increased by an estimated 35%.”
The biosphere is hard to pin down, being highly reactive.

Richard G.
Reply to  Richard G.
January 7, 2021 3:35 pm

About diurnal and seasonal CO2 trends
“The objective of this work was to characterize the effect of tillage and irrigation on CO2 concentrations in corn and soybean canopies at 10 cm above the soil surface and midway in the plant canopy. … Effects of irrigation, residue and tillage were minor. Occasional large diurnal fluctuations with maximum CO2 concentrations at night were associated with low winds and warm temperatures. Extreme daily minimum and maximum CO2 concentrations ranged from 285 to 800 μmol mol−1 for the 2 years of this study. However, typical minimum and maximum values ranged from 320 to 450 μmol mol−1. The seasonal trends in CO2 concentrations showed that both maximum and minimum occurred around flowering, which corresponded to maximum canopy development. The results suggested adequate mixing within both the corn and the soybean canopies, and that soil respiration modified by tillage and irrigation would be inefficient for CO2 fertilization due to rapid loss of CO2 from the canopy.

Reply to  Joe Wagner
January 6, 2021 6:39 pm

It is important to note that the biologic carbon pump involves phytoplankton which are sensitive to sunlight and water temperature. Most of the earth’s land is in the Northern hemisphere, while most of the water is in the Southern. My understanding is that this is much more significant than tree leaves manufacturing O2 from CO2.
Of interest: file:///K:/temp/sustainability-10-00869.pdf

Reply to  Willis Eschenbach
January 6, 2021 8:29 pm

Willis, I’m confused by the colors in the graph. I get that it just might be the 3D perspective. It appears that the difference in CO2 concentration between north and south is what 10-15 ppm. I know that people don’t like to use CO2 estimate from the Greenland ice core records because of biological activity that could produce false reading. However, I think the distribution of CO2 production is a fascinating area of research.

Reply to  Willis Eschenbach
January 7, 2021 3:47 am

Uhm, err… uhmmm, you know there is another ninety degrees South of the equator, yah? Your graph implies a difference between tropics and Northern hemisphere, nothing about the South. Oh, wait, I see, there is so little CO2 there, it does not even warrant place on your pretty picture…

Neal in Texas
Reply to  paranoid goy
January 7, 2021 4:59 am

A little training on reading a 3D plot. To find the (X, Y) location of a point on the plot, take a straight edge and align it with the Z-axis (CO2 axis), then slide this along the Y-axis (Latitude) to intersect the plot at the point of interest (the blue corner point on the bottom in this case). Then, you need to draw a line down from the plot to the X,Y plane. This point can then be located relative to the X,Y axes. For this case, no additional work needs done since the straight edge intersects the X,Y axes and shows that the minimum (date, latitude) point is at, roughly, 1990 and 90 degrees south.

Reply to  Willis Eschenbach
January 7, 2021 8:54 pm

I have often wondered and asked how CO2 measured at one spot in Hawaii can be representative of global CO2. Replies have always been something like this. ‘It is representative because CO2 diffuses quickly around the globe.’ The above graph shows CO2 varies with latitude. Does it also vary with longitude? Does it vary across the oceans and lands?

I am confused and wonder if CO2 should be measured at as many points as temperature. Or at least at many points around the globe.

Reply to  Willis Eschenbach
January 8, 2021 4:18 pm

Thanks Willis. That adds to my understanding. I will have to ponder it a while to fix the background CO2 relative to other measurements of CO2 in my understanding. Great information in your “Under the volcano..” post. Thanks

Richard G.
Reply to  Willis Eschenbach
January 9, 2021 5:49 pm

I realize this thread is going stale but a comment on Under the Volcano is in order.

You state that the readings are made at night to obtain catabatic flow samples, which are intended to measure upper atmosphere levels as the air descends from above the island, to avoid convectively lifted surface air. I am with you in theory. Never the less, look at the readings from Moana Loa 72 hr graphs concerning wind speed and direction over the diurnal cycle.

Wind speed drops consistently at night to below 5 M/S (but rarely -0-) and consistently swings from east to north north west. The observatory is located 5 kilometers (3 Mi.) North of the summit. If the night time airs were catabatic they would be from the summit (south) down slope, not from the north (up slope). They are rarely from the south at night. (Disclaimer- I have never been to the summit to personally evaluate the terrain features.)

My point is that it appears the air is being consistently orographically lifted throughout the diurnal cycle, not transitioning from orographic to catabatic. I am perfectly willing to take the data set for what it is, I just fear that it is not the gold standard of upper atmosphere values that they are striving for. But what do I know. I’m just a guy(biology teacher) with a keyboard some where trying to push science forward.

Warm regards.

Richard G.
Reply to  Willis Eschenbach
January 9, 2021 6:00 pm

Another thought, it would be nice if NOAA would add a graph of the diurnal CO2 readings over 72 hrs. Then we could see how lively the readings are.

John Moore
Reply to  Enginer01
January 6, 2021 7:36 pm

Oceans both lead and buffer atmosphere components.

Reply to  Enginer01
January 6, 2021 8:43 pm

Is it possible then that once a significant amount of ice has melted in the Arctic that, combined with surface plant life, plankton growth takes off, soaking up enough CO2 to outweigh the emissions from other sources? Our is it possible that the atmospheric level of CO2 is always an equilibrium level between air and ocean? Or did I just repeat myself using different words!

Reply to  Enginer01
January 6, 2021 9:19 pm

You gave us a link to one of your computer’s drives.

Reply to  Enginer01
January 7, 2021 12:16 am

Then there’s a lag in the temperature, reaching a minimum around February. Also at what latitudes are the most northerly forests? I would hypothesise that the forests of Sweden, Finland, Russia and Canada don’t start having an impact until the sun reaches a certain elevation and daylight time is lengthening.

Jim Ross
Reply to  JohnC
January 7, 2021 8:35 am

Sorry, missed this comment previously. As I mentioned below, this hypothesis of yours is consistent with the view of Scripps.

January 6, 2021 6:35 pm

Global man made emissions are only about 4% of natural emissions, i.e. not significant on a short time scale.

Reply to  Scissor
January 7, 2021 1:21 am

That’s what I thought as well. And humans emit about 10.000 Gt per annum (the three decimal places follow figures given in the literature). So how does Nature magazine come up with a graph where the pre/during Covid graph of CO2 emissions shows a disparity greater than 10?
And why are we analysing the time difference between CO2 emissions and CO2 levels on just one year?

Tim Gorman
Reply to  Capell
January 7, 2021 4:44 am

Here’s how the data was concocted by Nature:

Carbon Monitor is a frequently-updated daily CO2 emission dataset, to monitor the variations of CO2 emissions from fossil fuel combustion and cement production since January 1st 2019 at national level with near-global coverage. Daily CO2 emissions are estimated from a diverse range of activity data, including: hourly to daily electrical power generation data of 29 countries, monthly production data and production indices of industry processes of 62 countries/regions, daily mobility data and mobility indices of road transportation of 416 cities worldwide. Individual flight location data and monthly data were utilised for aviation and maritime transportation sectors estimates. In addition, monthly fuel consumption data that corrected for daily air temperature of 206 countries were used for estimating the emissions from commercial and residential buildings.

In other words, there is no relationship between the data presented and overall CO2 emissions by all sources.

Tim Gorman
Reply to  Willis Eschenbach
January 7, 2021 2:57 pm

I didn’t say they did anything wrong. They are only looking at a limited set of metrics to estimate CO2 emissions. They never really state in the article that they are only looking at limited data leaving it to most readers to assume they ae talking about *all* CO2 sources. For instance they only look at the transportation effects for about 400 cities worldwide. That is hardly a complete picture of the entire reality.

John Garrett
January 6, 2021 6:39 pm

Weekly (or any time scale you choose) CO2 emissions? You’ve got to be kidding. Talk about a squishy number! I don’t even want to think about how they (whoever “they” are) come up with it. If I had to guess, I suspect it’s mostly a made-up number.

Do they have green eye-shade wearing clerks buried somewhere in the bowels of The Swamp whose sole purpose in life is to collect and cumulate data reported by fossil-fueled electricity generating plants, refineries, manufacturers, and ___???___ ??

Mr. Eschenbach, please don’t spend any appreciable time (or any time, at all, for that matter) responding to this comment. You do wonderful work; you’ve got better things to do.

Submitted with my highest regard. HNY.

Reply to  Willis Eschenbach
January 6, 2021 9:22 pm

Yes, but why on earth do they only compare to the year before? Why not show at least 5 other years and their average so we can filter out economic differences.

Tim Gorman
Reply to  John Garrett
January 7, 2021 4:45 am


Here’s how the data was concocted by Nature:

Carbon Monitor is a frequently-updated daily CO2 emission dataset, to monitor the variations of CO2 emissions from fossil fuel combustion and cement production since January 1st 2019 at national level with near-global coverage. Daily CO2 emissions are estimated from a diverse range of activity data, including: hourly to daily electrical power generation data of 29 countries, monthly production data and production indices of industry processes of 62 countries/regions, daily mobility data and mobility indices of road transportation of 416 cities worldwide. Individual flight location data and monthly data were utilised for aviation and maritime transportation sectors estimates. In addition, monthly fuel consumption data that corrected for daily air temperature of 206 countries were used for estimating the emissions from commercial and residential buildings.

In other words, there is no relationship between the data presented and overall CO2 emissions by all sources.

John Garrett
Reply to  Tim Gorman
January 7, 2021 4:56 am

It’s a joke, a SWAG (Sophisticated Wild Ass Guess). It’s like many government statistics— a guess based on an estimate on top of a conjecture.

January 6, 2021 6:46 pm

What does it say when public data is inaccessible?

January 6, 2021 7:01 pm

I expect you will find that the CO2 aligns reasonably well with water vapour. Both reach their peak when the global energy accumulation peaks around April/May. The factors involved here are the exposure of the wet hemisphere to sunlight; the insolation is past its peak for the year and the thermal inertia of the oceans. The accumulated energy is near its peak for the annual cycle.

The sea surface temperature data is noisy, being unreflective of the the thermal energy in the oceans because it peaks in July/August due to the river outflow in the northern hemisphere. That fresh water is already warmer than the ocean it enters and it does not mix well due to its lower density.

Reply to  Willis Eschenbach
January 6, 2021 9:03 pm

I should have stated Global Ocean – thought it was obvious since I was implying outgassing of CO2 being consistent with water vapour over oceans.

Reply to  RickWill
January 6, 2021 9:22 pm

TPW over oceans peaked in July in 2017, 2018 and 2019. So later than the CO2 peak.

Tim Gorman
Reply to  RickWill
January 7, 2021 5:00 am

The data Nature used only includes estimates of man-made CO2. Nothing about oceans or natural CO2. It’s totally understandable that man-made CO2 peaks in May. That’s when construction starts in earnest (i.e. cement making). It may be when transportation also begins to peak, I haven’t researched that. Peak electricity generation is in Dec. and Jan. so I’m not sure why we don’t see any kind of a peak from fossil fuel electricity generation.

January 6, 2021 7:09 pm

Ole Humlum at his site has been looking at CO2 in the atmosphere for some time and updates his original paper every month. He opens with this:


1: Observed average global air temperature change last 30 years is about +0.15oC per decade. If this change rate remains stable, additional average global air temperature increase by year 2100 will be about +1.2oC.
2: Tide gauges along coasts indicate a typical global sea level increase of about 1-2 mm/yr. Coastal sea level change rate last 100 year has essentially been stable, without recent acceleration. If change rate remains stable, global sea level at coasts will typically increase 8-16 cm by year 2100, although many locations in regions affected by glaciation 20,000 ago, will experience a relative sea level drop.
3: Since 2004 the global oceans above 1900 m depth on average have warmed about 0.07oC. The maximum warming (about 0.2oC, 0-100 m depth) mainly affects oceans near Equator, where the incoming solar radiation is at maximum.
4: Changes in atmospheric CO2 follow changes in global air temperature. Changes in global air temperature follow changes in ocean surface temperature.
5: There is no perceptible effect on atmospheric CO2 due to the COVID-related drop in GHG emissions. Natural sinks and sources for atmospheric CO2 far outweigh human contributions. 


The whole point of science is to question accepted dogma.   Freeman Dyson

Reply to  DMA
January 6, 2021 7:30 pm

“Since 2004 the global oceans above 1900 m depth on average have warmed about 0.07oC.”

Really? Did he know what the water temperature of the ENTIRE ocean water above 1900m was in 2004 and in 2020 to the hundredth of a degree C? It’s a guess. When are we going to stop putting out numbers that are just fabricated? Put up a grid with all the possibilities between +2C and -2C in hundredths of a degree and let my monkey Deadeye throw a dart. It will be just as possible.

Reply to  DMA
January 6, 2021 9:12 pm

“Since 2004 the global oceans above 1900 m depth on average have warmed about 0.07oC. The maximum warming (about 0.2oC, 0-100 m depth) mainly affects oceans near Equator, where the incoming solar radiation is at maximum.”

Interestingly, The UHI tropical oceans SST has dropped to zero anomaly

Reply to  fred250
January 7, 2021 5:42 pm

0.07oC ??? That’s ridiculous ! The entire ocean temperature has increased 0.07oC , I would love to see the methodology behind that figure . Whats the error +- .5oC ? This is a mathematically derived figure and only exists in the mathematics not in the real world .

Chris Hanley
January 6, 2021 7:34 pm

That the word ‘estimate’ (and derivatives) appears fifty-seven times in the Nature article and the claimed variations are such a small fraction of the overall individual countries’ emissions as accurately reported by them (or not), I’m skeptical.

January 6, 2021 7:34 pm

Willis, your articles spurred me to go check out the data from the Mauna Loa site.

This of course led me to this site

ESRL Global Monitoring Division – Mauna Loa Observatory (

I really just wanted to check out the temperature data for Mauna Loa from a primary source.

At the bottom of the web page there is a link to additional data. Clicking it brings you to the this page

Global Monitoring Laboratory – Data Visualization (

So I choose Temperature to graph. The only choice is hourly average data. Temperature is measure continuously. I choose the entire date range (1977-2020). I fiddle with the graph options and submit. What happens? It tells me its working but never produces the graph. Strange. Of course I choose a shorter period and tried again thinking there were too many data points. Once again it told me it was working. So I set my computer down and went to have dinner. I came back to the graph below. I’m wondering if anyone else can produce a graph for the entire period. Also, does any else find it strange that the only choice is average hourly. There is a lot of data to be had from all the various sites. I find it interesting that the CO2 data from Mauna Loa that people point to to explain warming apparently has little effect on temps at the Mauna Loa.

Reply to  Nelson
January 6, 2021 7:59 pm

I recall plotting the temperature against CO2 at Mauna Loa a few years ago and if I remember correctly the correlation coefficient was about zero or even slightly negative.

Reply to  Nelson
January 6, 2021 9:15 pm

Do my eyes deceive me, or do I notice a slight rise to about 2012, then a slight falling trend ?

John F Hultquist
January 6, 2021 7:51 pm

Just asking: When in the NH does decomposition begin, peak, and than taper off?
If there is a decent flow of air through compostable materiable the aerobic microbes will thrive and the process of decomposition will be fairly quick. Materials will break down cleanly into CO2, H2O, minerals and humus.

My assumption is that temperature warming in spring allows decomposition to accelerate, cold in fall does the opposite.

Clyde Spencer
January 6, 2021 8:13 pm

After a few decades of watching, it is my impression that NH deciduous trees generally start to leaf out about early-May, with variations with latitude and elevation, and are pretty much fully leafed out by June. However, the heavily forested areas I’m familiar with, May seems to be a good time to put the stake in the ground as to the start of photosynthetic activity. (Outdoor planting of corn for central Ohio is roughly the first two weeks of May) Thus, I’d expect CO2 draw-down to start in May and continue until the leaves change from green to their Fall colors in October. All the while, bacteria and fungi are decomposing the leaf litter on the ground until it is pretty much used up, which I’d guess to be somewhere around July or August. Those subjective perceptions pretty much explain the NH CO2 concentration behavior.

Pat from kerbob
January 6, 2021 8:21 pm

Isn’t it likely the delayed reaction is based on moana loa remote location from the great bulk of northern hemisphere landmass and vegetation
There must be some time lag before a drop in atmospheric CO2 shows up there?

Glancing at google maps much of the middle lower latitudes are desert, not much foliage, a vast majority of landmass is farther north than me here in calgary and here there is little greening until early may

Steve Case
January 6, 2021 8:39 pm

Curious. CO2 peaks in May, and drops after that. Sure doesn’t seem like “Plants are accumulating carbon in the spring and summer when they’re active” to me. Meteorological spring is March through May, and meteorological summer is June through August.

For those of us living close to 90°W 45°N the growing season starts in May which agrees with your figure 3. Regarding the temperature, this chart that I like to splash up here at WUWT every chance I get, is May through October because that six month period shows decline in TMAX. The cooler part of the year does not. Our friends on the left monotonously paste up heat wave images of dry cracked mud and dead cattle ignoring the IPCC that says that the warming will be at night, in winter and the higher latitudes.

January 6, 2021 8:53 pm

Timing of atmospheric CO2 rise and fall also appear to coincide with arctic sea ice extent max and min, suggesting sea ice extent as inverse influence to CO2 absorption in north polar region.

Reply to  JCM
January 7, 2021 2:34 am


“suggesting sea ice extent as inverse influence to CO2 absorption in north polar region.”

The late Vincent Gray posted an analysis based on this idea but I have lost the link to his comments.
However there is this work by Dorte Haubjerg Søgaard which shows that the relationship between CO2 and sea ice is complex.

Søgaard, D.H., 2014. Biological Activity and Calcium Carbonate Dynamics in Greenland Sea Ice: Implication for the Inorganic Carbon Cycle: PhD Thesis. Greenland Institute of Natural Resources.

January 6, 2021 9:47 pm

While the NH is warming and greening in the summer, the SH is cooling in its winter, and its rain and its oceans are absorbing CO2.

Geoff Sherrington
January 6, 2021 10:22 pm

In May last year, Ken Stewart of kenskingdom blog fame did quite a good analysis of the atmospheric CO2 changes between several stations over time with most work on Barrow, ML and Sth Pole.

The relationships are complex. There is a lot more that could be drawn out of global CO2 data because there are now many stations monitoring it.
Your 3D graph at 7.16pm is rather sweet. Maybe it is asking for more stations to have their data added in your leisure time. Overall, this post of yours is just superb. Geoff S

Martin Cropp
January 6, 2021 10:46 pm

Dear Willis
It is impossible to understand the the controlling and influencing factors of any atmospheric issue with just 2 to 3 data references, no matter how accurate they are.

Atmospheric Entanglement

Kind regards

Bill Parsons
January 6, 2021 10:46 pm

Local Denver based weather man Mike Nelson in his ongoing “We-done-it” presentations to school kids, likes to repeat the claim that

“We know that the increase [in CO2] that we’re seeing is not from any other source than from the burning of fossil fuels,” Nelson said, because CO2 has a chemical signature like a fingerprint, and scientists can determine where it came from.

How sound is that claim? If it were true wouldn’t the attribution to humans an open / shut case a long time ago?

The World’s Littlest Book on Climate: 10 Fact… (Paperback)
by Nelson, MikeBanks, Michael

Tim Gorman
Reply to  Bill Parsons
January 7, 2021 5:06 am

ROFL! Yep, each CO2 molecule gets a micro-etch (like they do on grains of rice) from the generating source!

M Courtney
Reply to  Bill Parsons
January 7, 2021 6:06 am

It’s not a silly idea. The thinking being that the ratio of carbon 12 and Carbon 13 differs between carbon sources that are buried underground for thousands of years and those that are circulating now. Carbon 13 has a half life of about 5000 years.

So it’s not stupid. But it’s not a slam-dunk either. That’s because modelling decay and replenishment is hard.

Here’s a paper related to it. None of the models seem to agree.

Jim Ross
Reply to  M Courtney
January 7, 2021 6:35 am

M Courtney,

I think you are mixing up 13C which is a stable isotope, like 12C, with 14C which decays. However, you are certainly correct that the isotope data can give us some clues to sources. The most recent paper I have seen on trying to match the observational stable isotope data, as characterized by the 13C/12C ratio, is by Keeling et al (2017):

January 7, 2021 2:35 am

‘There’s more carbon dioxide in the winter and a bit less in the summer. ‘

Ja, but it is not due to vegetation or emission. It is because the sinc area in the SH has remained virtually constant over the last 40 years. Therefore the same (big) amount CO2 is dissolved during the SH winter.
CO2 + 2H2O + cold => H3O+ + HCO3-
The arctic is getting warmer and warmer and as long as the temperature goes up there the overall trend upward of CO2 will remain….because the sinc area in the arctic is getting smaller!!

Paul C
Reply to  HenryP
January 7, 2021 6:46 am

I have seen it postulated that sea ice (when approaching 100% coverage) blocks the dissolution of CO2 into the alkaline seawater below, and that meltwater also has a similar effect. The system may exhibit complex non-linear behaviour.

Reply to  Paul C
January 7, 2021 7:18 am

compared month- to -month of the previous year I think the ice in the antarctic stays more or less the same. My worry is the arctic where you can see it gets warmer like it does every 1000-1100 years or so.

Anyway, my other worry is the waste water of 7 billion people + factories + animals which is mainly acidic:
H3O+ + HCO3- = > CO2 + 2H2O 
which also causes more CO2 in the atmosphere.
CO2 from emissions is a non-issue….really

January 7, 2021 4:21 am

I’m more interested in the sins of omission where the science of climatheology is discussed.

Loren C. Wilson
January 7, 2021 4:34 am

Willis, the reference states that it is for fossil fuel and cement production, which is only a fraction of anthropogenic CO2 and that is a very small percentage of the total global production of CO2. Mauna Loa measures CO2 from all sources so their data will not correlate with these two small sources, which are perhaps 2% of the total. I don’t think anyone has global production and consumption data of all CO2 from all sources. That is one of the fundamental flaws in the “science”. It’s like having a broken water main and a leaky faucet. They can estimate the leak from the faucet but ignore the broken water main, and claim they can model the flow of water through the entire network.

Tim Gorman
Reply to  Willis Eschenbach
January 7, 2021 2:44 pm


Actually, the changes in the MLO CO2 can be calculated quite closely from the anthropogenic emissions alone. Here’s my calcs, based solely on anthropogenic CO2 with an exponential decay.”

Huh? If natural CO2 sources emit 20 times the amount of CO2 that humans do then how are the “changes” at MLO all due to humans?

As more and more land is tilled for food, left over stubble will add to atmospheric CO2 from decomposition. Is this also anthropogenic CO2?

If, as the CAGW advocates claim, the oceans are warming then they are a huge natural source of CO2. Or do you consider that to be anthropogenic as well?

Tim Gorman
Reply to  Willis Eschenbach
January 8, 2021 5:13 am


As temp goes up the ocean emits more CO2, right? How do you separate that increase from the man-made increase?

Tim Gorman
Reply to  Willis Eschenbach
January 8, 2021 1:02 pm

But the CO2 change lagged the temperature change. Where is the lag for anthropogenic CO2. And what caused the temp rise in the first place?

Gordon A. Dressler
Reply to  Tim Gorman
January 8, 2021 1:15 pm

Tim, you stated: “as temp goes up the ocean emits more CO2, right?”

This is definitely NOT true if one assumes that gaseous CO2 that was absorbed previously by the cooler ocean is in an equilibrium solution state such that warmer ocean water would result in this absorbed CO2 come out of solution based on solubility being an inverse function of temperature (as is the case with soda beverages).

A very common misconception is to view CO2 as dissolved in ocean water at a pH of 8.1 to 8.2 as analogous to soda water. It is far from it. There is essentially no dissolved CO2 that exists for very long as a dissolve gas just waiting to be outgassed from saltwater as its temperature increases or as the gas pressure over the liquid water decreases. 

If you really dig into the chemistry of how CO2 enters and chemically reacts with sea water (understanding such things as the Revelle factor and, most importantly, the Bjerrum plot), you will find that CO2 gas “solubility” into the ocean is NOT a reversible process, depending on water temperature, as long as the water pH remains above 8 or so. At the current ocean average pH of 8.1-8.2, far less than 1% of the CO2 entering the ocean remains available as dissolved CO2, a portion of which might possibly come out of solution gradually with increasing ocean temperatures.

Basically, an insignificant amount of CO2 remains as a gas in aqueous solution in the world’s oceans, and might therefore be available for temperature-induced “outgassing” (see Figure 1, the Bjerrum plot, at for the detailed, elegant explanation.

Earth’s ocean MAY “release” some significant extra CO2 from biological processes increasing with warmer water temperatures, but not from purely physical chemistry processes.

Tom Abbott
Reply to  Willis Eschenbach
January 8, 2021 11:52 am

I like your “Equilibrium” example. 🙂

Jim Ross
January 7, 2021 6:19 am

For what it’s worth, I recall a comment from someone at Scripps that the northern hemisphere atmospheric CO2 character was dominated by the Boreal forests, which might explain (i) the fact that CO2 is highest at Point Barrow, Alaska (PTB) and (ii) why it starts to fall in May, rather than earlier in the spring.

Another related point can be seen in this plot:
comment image
Note that in almost every case, the changes of atmospheric CO2 level at PTB ‘leads’ or coincides with (on a monthly basis) what is seen at Mauna Loa (MLO). Is it possible that what we see at MLO is mainly a consequence of what is happening in the atmosphere further north? Weekly data at PTB would help to compare directly with weekly data at MLO, but that seems to be lacking.

Reply to  Jim Ross
January 7, 2021 2:25 pm

If the industrial and domestic heating demand in the northern hemisphere was integrated it would produce something close to the blue line; peaking around April then falling till October. Attached shows cyclic gas demand in UK. Separating demand and production of energy is not that easy. I expect wood gets stored for the boreal winter during the previous boreal summer.

The red line is just a damped (filtered) response from the blue line so the blue line is closer to the source.

Screen Shot 2021-01-08 at 9.18.38 am.png
Reply to  RickWill
January 7, 2021 4:24 pm

Yes I agree the seasonal heating demand must be looked at.

Jim Ross
Reply to  RickWill
January 8, 2021 2:30 am

There seems to be some confusion here. I am not questioning your plot or Willis’ graph of emissions, but the idea that variations in emissions could explain the seasonal cycle at Mauna Loa is untenable. Estimated emissions are always positive, whereas atmospheric CO2 at Mauna Loa actually drops during the months of spring/summer; it does not simply increase at a lower rate. Emissions cannot explain this with or without a lag. In addition, emissions would lead to a decrease in atmospheric O2 (as seen in the longer term data trend). During the spring/summer, however, atmospheric O2 increases (this is photosynthesis in action) as does δ13C.
comment image
Now that’s what I call a correlation. (Note: δ(O2/N2) scale is inverted and changes in O2 in ppm terms can be approximated by dividing by 4.8, data source and more information available at the Scripps O2 program:

January 7, 2021 8:27 am

Well, part of the problem is that the atmosphere is more than 1 meter thick and the emissions data is taken IN that one meter but Mauna Loa’s measurements are taken at the 3397th meter. So that could be messing with your numbers. Also the wind blows up over the Big Island before it gets to Mauna Loa, after a trip of 4,800,000 meters across the ocean that takes it about nine and a quarter days. And the CO2 sensor at Mauna Loa wasn’t put there for the purpose of global weather prediction. You put CO2 sensors on volcanoes like you put thermometers beside runways – because you expect the reading AT the sensor to be different and the difference to be important.

Jim Ross
Reply to  Willis Eschenbach
January 8, 2021 5:40 am

That would be Charles David Keeling, I believe.

Reply to  Willis Eschenbach
January 8, 2021 7:42 am

It was a fallacious placement prior to understanding the way the Oceans actually worked and it is still an incompetent choice for a “background” reference location. Largely the Bahamas would be far far far more reliable for such a placement. Hawaii was and still is downwind of several major slash-and-burn agricultural economies. Placing a “background” gas sensing station there for measurement of combustion products was abjectly unscientific and stupid. Even Tenerife or Terceira would have instantly been a better choice accounting for costs as well. Any scientist who chooses a location for a sensor because it’s where the scientist wants to be is a waste to research and produces only anecdotal output. CO2 may disperse in the atmosphere but that literally occurs either by heating or by turbulence, both of which are ALSO geologically anectedized results. We need a shitload more sensors located in places far better selected to get an actual background reading. The placement of it on Hawaii literally removes it from the background.

Not understanding these facts and arguing against them is incompetent.

January 7, 2021 10:41 am

And this doesn’t even take into account the ~ 4 month lag between surface changes and airborne changes shown in Figure 2 …

I really don’t understand what you are reading into this lag relationship. Emissions is not “surface changes” it is emissions. The surface changes need to include plants which you do not do in the data , only in you assumptions about when plants emit/absorb.

Trivial statements about “meteorlogical summer” are not that informative. Plants take time to grow. They may poke out first sprouts in March-April but the bulk of leaf material arrives much later.

Similarly for decaying matter. When do leaves fall , now long does it take the microbial decomposition to reach full swing. I do not see anything in the graph which is obviously in contradiction to the use account.

What is clear is that human emission cycle has little to do with atm CO2.

It has been shown a number of times here that short term change in atm CO2 is primarily a function of SST.

Reply to  Greg
January 7, 2021 11:22 am

It has been shown a number of times here that short term change in atm CO2 is primarily a function of SST, with the annual variability removed.
comment image?w=800

Reply to  Greg
January 7, 2021 11:34 am

Another interesting comparison is Arctic ice cover ( rather lack there of in the summer ) and trough in atm CO2. Cold water absorbs CO2. The annual swing in CO2 is largest in the arctic .
comment image?w=800

January 7, 2021 12:02 pm

By an amazing coincidence there is just today an article in ‘No Tricks Zone’ about a paper that claims that CO2 concentration reduces as temperature increases, and vice versa.
Could this be the source of the summer reduction in CO2 concentation?

Tim Crome
January 7, 2021 1:03 pm

I’m not convinced CO2 is that well mixed, certainly not if the numbers here are correct.,22.91,185/loc=-39.991,19.533

It can range from around 410ppm to 440 over relatively short distances.

Lots of data, also back over several years, could be a source of almost endless analysis and speculation!

Ulric Lyons
January 7, 2021 3:07 pm

The AMO is usually relatively cooler around May and relatively warmer around September, which is the right kind of timing, but on the face of it, the wrong sign temperature wise.

Fred Souder
January 7, 2021 3:25 pm

I believe that the overwhelming majority of “active” biomass on the planet is bacteria. Plants contain the most biomass, but it is mostly inert. Thus, I’d look at cycles of bacterial growth rather than plant growth for possible seasonal carbon sinks.

January 7, 2021 4:08 pm

Nature can’t get much of anything right if they can inject human guilt.

Figure 1. Emissions from 2019 and part of 2020. SOURCE

Initial reports based on a limited sample of power plants and indirect satellite observations of atmospheric pollutants<sup>13,14</sup> suggested an significant drop in global emissions.”

<a href=>Note 13 includes:</a>

“In terms of attributing any changes to the on-going pandemic, a long list of confounding factors cloud the picture. This means it is hard to attribute a changing indicator solely to coronavirus, given multiple reasons why fossil-fuel demand in March 2020 might have fallen, relative to the same month in previous years.

The mild winter across Europe and North America has cut demand for heating in the first quarter of the year, for example, making it cheaper to burn gas for power and industry. Temperatures also affect electricity demand. Adjustments to account for this are possible, but add complexity.”

Note 14 includes:

Update: 30 March 2020

Coal consumption at power plants and oil-refinery utilisation bottomed out in early March and returned to a normal range by the fourth week of March, about seven weeks after the country was originally going to return to work, on 3 February.”

What is actually used are cherry picked postulated estimates of estimates.
These numbers are prepared by Carbonbrief! Not by a government agency vested with supplying the military with serious estimates.

Even their cherry picked sources were back at normal rates early in the year!

Under the Methods in the nature article:

The CO2 emissions and sectoral structure in 2018 for countries and regions are extracted from EDGAR V5.0, and the emissions are scaled to the year 2019 based on the growth rates from Liu et al. and studies by the Global Carbon Project.

For countries with no current estimates of emission growth rates in 2019 such as Russia, Japan, and Brazil, we assume their growth rates of emissions were 0.5% based on the emission growth rates of the rest of world.”

Given the large uncertainty of CO2 emission in China, we calculated China’s CO2 emissions separately. For China, the energy consumption of coal, oil and gas in 2000–2017 are based on energy balance tables from China Energy Statistical Yearbook. However, due to the 2 years lag of the publications of China Energy Statistical Yearbook, we project the energy consumption of coal, oil and gas in 2018 and 2019 by multiplying the annual growth rates of coal, oil, and gas reported on the Statistical Communiqué”

On and on, fudge factor after fudge factor…

If mankind’s emissions dropped as they claimed, then Mother Nature sure filled the vacuum.

Cristiano Griggio
January 8, 2021 12:57 am

Dear Mr. Eschenback,
I found this paper on another site
Maybe their work on how CO2 varies in time can help to understand your interesting findings?
Best regards and my greatest esteem for your excellent comments.

January 10, 2021 8:25 am

Thanks Willis.

Me thinks nobody really knows. Except perhaps in Plato’s other world populated by High Priests like David Sleazuki.

Note that in the Northern Hemisphere heating of buildings is a major emitter of CO2 in winter, and to lesser extent lighting which often is fueled by coal. (Automobile driving has several factors – more personal trips in summer, more warmup and idling in winter. Much construction equipment in summer, transport of supplies to construction sites will vary with amount of preparation. (Pipelines get pipe in place early so they can roll when able, though then have to pay to guard it.)

%d bloggers like this: